This internal working paper is prepared for STAFF USE ONLY. The views expressed are not necessarily those of the World Bank. INTERNATIONAL GRAIN TRADE, 1950-80 AGREP Division Working Paper No. 64 Prepared by: Eric A. Monke (Consultant) Economics and Policy Division Agriculture and Rural Development Department January 1983 Table of Contents Page I. INTRODUCTION 1 II. GROWTH OF THE GRAIN TRADE, 1950-80 3 III. CAUSES OF INCREASED TRADE 5 IV. GOVERNMENT POLICY AND INTERNATIONAL TRADE 13 V. PRICE BEHAVIOR ON THE INTERNATIONAL MARKET 20 VI. POLICIES AND INSTABILITY IN TRADE 30 VII. CONCLUSION 36 REFERENCES 39 APPENDIX: Table 1. Estimates of Population and Total Cereal Production 43 Table 2. Declines in Cereal Production per Capita, 1948/52-1979/81 44 Table 3. Grains: Direct and Feed Consumption, 1978 46 Table 4. Estimates of Net Protection Coefficients for Cereals, 1975-80 48 Table 5. Grain Trade Policies in Selected Countries, 1970s 55 Table 6. Grain Prices, 1950-80 60 Table 7. Ratio of Cereal Import Costs to Total Commodity Export Earnings for Selected Developing Countries, 1970-80 61 List of Tables Page 1. Annual Average World Grain Trade 3a 2. Crop Yields 6a 3. Cereal Production and Trade 7a 4. Sources of Growth in Import Demand for Cereals 8a 5. Consumption and Net Protection Coefficients for Cereals 13a 6. Grain Trade Policies 18a 7. Concessional Exports of Cereals 22a 8'. Cross-Price Correlation Coefficients for Grains 28a 9. Ratio of Total Cereal Import Costs to Total Commodity Export Earnings in Selected Countries 31a 10. Coefficients of Variation of Prices Under Hypothetical Domestic Self-Sufficiency Programs 35a List of Figures 1. Income and Grain Consumption 12a 2. Nominal Grain Prices 21a 3. Trade Policy and Domestic Objectives 25a 4. Grain Prices, Wheat 27a Maize 27b Rice 27c I. Introduction Few aspects of the post-war development of the grain economy have received more attention than international trade. While few would argue with the need -to trade grain within national borders, international trade in so basic a commodity appears to many as a sign of political fallibility or a faltering agricultural economy. The patterns of trade have also led to controversy. The prominence of developed countries on the export side of the market and less-developed countries on the import side has targeted international grain trade as a prime example of the dependency in North-South relations. Patterns of trade have also thrust grain into the East-West conflict, if not as a political weapon, then as an economic one in which export cartels would fix prices to maximize benefits to Western grain producers and governments. How has international trade come to occupy such a controversial position in the world grain economy? While the patterns of trade are readily observable and well understood, the same cannot be said of the economic forces which are responsible for those patterns. As Johnson (1973), Josling (1974) and others have observed, international trade and policy can be no more than an extension of production and consumption policy at the domestic level, a connection which is frea'uently forgotten or misrepresented in the shrillness of political debate. This survey of the international grain market focuses on these relationships. Part of the reason for the interest in trae has been its extraordinary rate of growth, described in Section 2. Annual trade in grains increased from about 40 to 210 million mt over the 1950-80 period, a rate of about 5.7 2 percent per year, while total production increased at an annual rate of only 2.6 percent. Much of the growth in import demand emanated from the less- developed countries (LDCs), and this pattern has fueled the popular belief that population and floundering production performance were principal forces behind trade expansion. As the third section demonstrates, however, these factors can only explain a small portion of the increase in import volume (principally in Africa and the Far East). Estimates of the relationships between income, direct and indirect demand for grain implicate income growth as a far more significant factor, and suggest that the international market has served largely to augment consumption rather than aupplant domestic production. A salient feature of the international market is the overwhelming dominance of governments as economic agents. The fourth section of the paper describes the pattern of government controls over international trade, and implicates trade policy as a cornerstone of domestic grain policy in almost all countries. Government monopolies or quantitative controls on imports/exports mean that international prices are of only limited relevance to most consumers and producers. Thus price behavior on international markets, described in section five, says as much about government policy actions as about consumer-producer response. Many of the concerns voiced about international trade, such as the substantial degree of price fluctuation and foreign exchange constraints, emanate from government policy-makers rather than domestic consumers and producers, as the vagaries of the international market complicate the management of domestic grain policies. The sixth section investigates the empirical validity of some of these complaints about trade. Conclusions are provided in the final section. 3 II. Growth of the Grain Trade, 1950-80 Data for total cereal trade comfirm two popular conceptions about trade-- its rapid growth and the dominance of the developed"market economies in both imports and exports. Table 1 provides estimates of exports and imports of total cereals, wheat, maize and rice for the period 1948/52 through 1979/80.1/ The latter three grains comprise about 85 percent of total cereal trade and 80 percent of total. cereal production, shares which have renained relatively stable throughout the survey period. 3arley, rye, oats, sorghun and millet account for the remainder. Total trade volume increased from 42 to 210 million metric tons (mt) over the three decades, with most oF the increases occuring since 1960. World grain trade grew more rapidly than trade in most other categories of agricultural products during the post-war era. FAO trade indices, for example, indicate that 'otal agricultural trade doubled during the 1961-80 period, while grain trade volumes more than tripled. Developed market economy exporters, primarily the US, Canada, France and Australia, accounted for 150 million of the 170 million mt increase. By 1980, these four countries were providing nearly 85 percent of total cereal exports./ The developed countries also dominate the import side of the market, although their trade share has declined from 68 to 57 percent over the . -FAO data are usually based on official statistical reports from individual countries. Estimates for the early periods and the centrally-planned economies are the least reliable among the data represented in Table 1. Estimates of year-end in-transiz shipments represent an additional difficulty for the preparation of calendar year data. Totals of exDorts and imports are reasonably consistent, however, and differ rarely by More than two or three percent. Exports in 1980 by the US, Canada, France and Australia were 113, 22, 20 and 19 million mt, respectively. TABLE 1. ANNIJAL AVERAGE WORLD GRAIN TRADE, 1948/52 - 1979/80. a EXPORTS IMPORTS Developea Developing Total Developed Developing Total Year Market CPEs Market CPEs Mfarket CPEs Market CPEs (million netric tons) TOTAL CEREALS 1948/52 30.7 0.5 10.2 0.2 41.5 28.2 na 12-.9' 0.1 41.2 59/61 49.4 8.1 11.8 - 69.3 37.8 7.9 21.5 - 67.2 69/71 79.8 10.3 18.2 1.8 110.1 57.9 11.0 33.0 6.8 108.7 79/80 179.1 6.0 23.3 2.0 210.4 74.6 46.3 71.1 19.6 211.6 (HEAT 1948/52 22.4 na 2.9 na 25.3 16.9 na 7.8 na 24.7 59/61 31.5 5.9 2.6 - 40.0 16.7 6.2 14.6 - 37.5 69/71 44.2 8.0 2.5 - 54.7 18.6 6.8 22.4 5.3 53.6 79/80 79.8 4.8 6.1 - 90.7 20.0 18.9 40.6 12.9 92.4 MAIZE 1948/52 2.8 0.2 1.5 0.1 4.6 4.4 na 0.5 na 4.9 59/61 7.6 0.6 3.4 - 11.6 10.6 0.6 0.6 - 11.8 69/71 19.1 1.2 9.0 - 29.3 24.2 1.6 2.4 0.6 28.8 79/80 69.7 0.7 7.4 0.2 78.0 37.8 19.6 15.1 5.1 77.6 RICE 1948/52 0.8 na 3.8 na 4.6 1.0 na 3.2 0.1 4.3 59/61 1.2 0.1 4.3 - 5.6, 0.9 0.8 4.6 - 6.3 69/71 2.9 - 3.8 1.7 8.4 0.9 0.6 6.1 0.7 8.3 79/8) 4.6 - 5.9 1.8 12.3 1.6 1.0 9.2 0.6 12.0 Sources: Food and Agriciture Organization, Trade Yearbook, 1959, 1963, 1972, 1975, 1978, 1980. - less than 50,000 jit na not avaiilable TABLE 1. NOTES a The country classification follows tliat cirrently uised by FAO, with earlier data ad.justed to nalntaln consistency. The develåped market economies include the countries of estern Europe, Canada, the United States, Australia, New Zealand, Japan, Israel and South Africa. Developed CPEs are the countries qf Eastern Europe and the USSR. Developing CIEs include the PRC, Vie#.uam, the Khner Republie, Laos, and Nortt Korea. ieveloping market economies conprise the remaining coi:atries. tv 4 period. This decline would have been much more substantial if not for the increase in imports by the USSR and Eastern Europe. In absolute terms, the greatest increases in imports were registered by the developing market economies (58 million mt), followed by the developed market economies (46 million nit), the developed CPEs (approximately 45 million mt) and the developing CPEs (19.5 million mt). Table 1 also demonstrates the significant changes which occurred in the commodity composition of cereal trade. Wheat was the dominant traded grain at the onset of the period, accounting for 60 percent of total trade volume. By 1980, however, maize trade had grown almost as large as that of wheat, increasing from five to 78 million mt. More than 50 million mt of the increase in maize imports was due to demands of the developed countries, reflecting the importance of increased meat consumption. The slower growth in wheat trade was due largely to the stagnant demand among the developed market economies, Low or negative income elasticities of direct demand, the limited use of wheat as an animal feed and policies to promote domestic production were significant factors in forestalling substantial growth in imports within this category. In contrast, wheat imports in the developing economies increased sharply, and by 1980 this group of countries accounted for nearly 60 percent of total wheat imports. Rice remained a minor grain throughout the period, and the share of rice in total cereal trade declined from 10 to 6 percent. Although aggregate production is roughly equal among the three grains, only about three PercE&t of rice production enters world trade, while about 20 percent of wheat and maize production are traded. Changes in rice trade volumes followed the pattern of wheat, as developing countries registered the largest increases in imports, while developed countries accounted for most of the increase in exports. 5 III. Causes of Increased Trade Changes in the patterns and volume of trade can only be understood in the context of changes in consumption and production. Population growth is perhaps the most obvious factor behind increased consumption. World population increased by about 75 percent during the 1950-80 period, as annual growth rates averaged about two percent in the developing countries and one percent in thc developed countries. While population growth rates were unprecedented, the growth of cereal production was even more extraordinary, doubling over the period. In part, unprecedented rates of population growth are themselves responsible for production growth, due to the dominance of agriculture as an economic activity. Estimated agricultural population increased from 1.38 to 2.04 billion between 1950 and 1980,3/ and thus population growth alone can explain as much as half of the increase in world cereal production. The realization of production potentials depended on a host of factors, such as availability of uncultivated land, substitution for other crops, expansion of multiple cropping and yield increases. Land expansion aonears a less important factor, as harvested area of cereals increased by only 25 percent between 1950 and 1980, from 595 to 743 million hectares (ha). Most of this increase appears due to the substitution of cereals for other crops and increases in multiple cropping, rather than increased utilization of uncultivated land. FAO data for the 1970-80 period, for example, indicate FAO data. Early estimates of Chinese agricultural population are particularly uncertain. FAQ estimates for 1950 of 377 million (FAO, Production Yearbook, 1963, p. 18) are close to the estimates of Tang and Stone (1980, p. 43) of 373 million. 6 that total agricultural cropped area increased by only 28 million ha, while harvested area of cereals increased by nearly 70 million ha. Yield increases were roughly three times as important as land area expansion in the determination of increased production. Average cereal yields grew from 1.2 to 2.1 mt/ha, with most of the increase occurring since 1960. The principal factors responsible for these increases are well-known-- increased chemical and fertilizer use (nitrogen production, for example, increased from less than five to 60 million nt), expansian of irrigated area and the development and dissemination of fertilizer-responsive seed varieties. However, the global average conceals a substantial diversity in experience, both among regions and among particular cereal crops. Table 2 provides regional yield data for wheat, maize and rice. The developed economies (both market and CPEs) achieved yield increases equal to or greater than global averages for each of the grains. The developing CPEs (primarily China), have also achieved high growth rates, particularly in wheat and rice. The experience of the developing market economies provides a stark contrast, as yield increases were below global average for all three crops. This generalization extends to individual regions as well, with the exception of rice in the Near East and wheat in the Far East. Wheat and rice yields increased by about 65 percent, while maize yields increased by about 45 percent. These figures highlight the rather limited impact of the Green Revolution on global food production. 3y 1980, high-yielding varieties of wheat and rice had been adopted on less than 60 million ha, or about 40 percent of rice and wheat area in the developing market economies. Maize was largely unaffected, and the share of improved varieties in total cereal production of the developing market economies is only about 20 percent. 6A TABLE 2. CROP YIELDS, 1948/52 - 1979/80. Percent 1948/52 1961/65 1969/71 1979/80 Change mt/ha WHEAT WORLD 1.01 1.21 1.58 1.87 85 DEVELOPED Market 1.30 1.74 2.13 2.34 80 CPEs 0.84a 1.06 1.55 1.78 87 DEVELOPING Market 0.87 0.98 1.14 1.43 64 Africa 0.60 0.70 0.79 0.85 42 Latin America 1.05 1.43 1.40 1.47 40 Near East 0.91 1.00. 1.07 1.43 57 Far East 0.79 0.84 120 1.51 91 CPEs 0.69 0.88 1.48 2.02 193 HAIZE WORLD 1.58 2.17 2.54 3.16 100 DEVELOPED Market 2.31 3.50 4.35 5.48 137 CPEs 1.30a 2.15 2.81 3.71 185 DEVELOPING Market 0.97 1.13 1.28 1.39 43 Africa 0.76 0.91 1.01 0.90 18 Latin America 1.06 1.23 1.43 1.68 58 Near East 1.52 1.90 2.33 2.65 74 Far East 0.77 1.00 1.11 1.21 57 CPEs 2.43 2.57 2.98 23 RICE (paddy) WORLD 1.61 2.04 2.38 2.71 68 DEVELOPED Market 3.27 4.91 5.39 5.40 65 CPEs 1.45a 2.48 3.43 3.98 174 DEVELOPING Market 1.42 1.62 1.84 2q10 48 Africa 0.97 1.28 1.35 1.35 39 Latin America 1.68 1.73 1.72 1.95 16 Near East 2.37 3.41 3.71 4.36 84 Far East 1.40 1.60 1.84 2.12 51 CPEs 2.17 2.61 '3.Z9 3.84 77 Source: Food and Agriculture Organization, Production Yearbook, 1963, 1976, 1980. a USSR only. 7 The diversity in production growth rates among countries provides a supply-side explanation for the growth in international trade. Table 3 provides regional data for cereal production, imports and exports in per capita terms. Relatively high rates of yield increase and low rates of population growth in the developed countries resulted in substantial increases in per capita production and potential exportable surpluses. In contrast, low rates of yield increase and high rates of population growth resulted in virtual stagnation or declines in per capita production in the developing market economies, particularly since 1960. This effect is particularly significant in Africa and Latin America, where per capita production declined during the last decade by 30 and 10 kilograms (kgs), respectively; net incorts per capita increased by 21 and 36 kgs. The regional data of Table 3 conceals even more diverse experiences among countries. Bachman and Paulino (1979), for example, found that per capita staple food production declined in 53 of 93 developing market economies during the 1961-76 period. Increased imports were thus necessary to maintain per capita grain availabilities in many developing market economies. Estimates of cereal production in individual countries and population growth can be used to provide rough approximations of the contribution of -these two factors'to the growth in import demand for cereals over the 1950-80 period. The contribution of production declines to increased imports can be -estimated as 1980 1950 1980 SPOP. (0 - 0 1 1 si si where the subscript i refers to importing countries which experienced declines in per capita cereal production during the period, POP refers to population and Os represents per capita production. Appendix Table 2 provides data on TABLE 3. CEWR.L POfXETION AN) 'WE, 1948/52 - 1979/80. ITREAL PRCI fION GRO0SS IMPOfS ROSS 1Y0MPTS NE '1W~ (= exports, + = iports) 1948 1960 1969 1979 1948 1960 1969 1979 1948 1960 1969 1979 1948 1960 1969 1979 -52 -61 -71 -80 -52 -61 -71 -80 -52 -61 -71 -80 -51 -61 -71 -80 (K-s/capita) (K;/capita) (Ts/capith) (Ygs/capita) D)EVE[IPED Dkarket Econoies 503 548 582 692 59 69 80 95 64 90 110 229 -5 -21 -30- 1 34 CPFs 407 579 665 684 ia 26 32 123 2 26 30 16 na 0 2 107 tarket Econoies 178 199 212 207 11 14 19 32 8 8 11 11 3 6 8 21 Africa 144 143 153 123 6 11 15 34 8 7 3 1 -2 4 12 33 IatinAnerica 192 218 246 236 20 22 29 60 26 25 41 36 -6 -3 -12 24. ikar East 223 241 257 262 16 30 38 .87 11 5 6 6 5 25 32 81 Far East 177 204 212 215 8 12 14 14 5 5 5 7 3 7 9 7 CIFs 270 233 257 302 na 9a 8 19 - 1 2 2 na 8 6 17 I-K)U) 286 308 339 357 16 22 30 48 Sources: Appendix Table 1; Food and Agrialture Organization, Trade Yearbook 1959, 1963, 1972, 1975, 1978, 1980. Rone, Italv. Tang and Stone, 1980, Food Production in the People's Republic of China, International Food Policy Research Institute Research Peport N. 15, Washingtön, D.C. a 'Fstimtes for 1961 (Tang and Stone, 1980, p. 31). per capita cereal production. Fifty-one countries demonstrated declines during the 1950-80 period. Africa accounts for 24 of the countries with - declining per capita production.Only two developed countries (Japan and Portugal) and one CPE (Kampuchea) experienced declines in per caoita cereal production over the 30 year period. The impact of population growth can be estimated in a similar fashion, as 1950 1980 1950 11 M. (POP. - POP. ) , where Mj represents gross imports in the jth region. Sources of growth estimates are dependent on the selection of time frame, and a decade-by-decade perspective would undoubtedly produce some difference in estimates of the importance of particular factors. The countries with declining production identified by Bachman and Paulino for the 1961-76 period, for example, differ slightly from those which experienced declines in the 1950-80 period. The ceteris paribus assumption should also be emphasized. The impact on cereal imports of production declines in non-cereal staples are ignored, for example, and thus the effect of declines in per capita prod,iction will be understated. This source of bias is probably small, however, as total cultivated area of roots and tubers in LDCs was only about 35 million has in 1980. In spite of these shortcomings, some useful generalizations emerge from -the*results, presented in Table 4. Declines in per capita production can account for a substantial proportion of the growth in imports in Africa and the Far East. For Latin America and the Near East, however,.these factors cannot explain very much. In the developing market economies as a whole, maintenance of per capita cereal availabilities at their 1948/52 levels would have required only 30 million mt of increased imports, or about half of the TABLE 4. SOURCES OF GROWTH IN IMPORT DEMAND FOR CEREALS, 1948/52 - 1979/81. Effects on Import Demand of Total Declines in Population Region Growth Per Capita Productinol, Growth Residnal (million mti5 DEVELOPED Market Economies 46.02 9.33 17.83 18.86 CPEs 46.30 0 1.74a 44.56 DEVELOPING Market Economies 58.34 19.02 10.80 28.52 Africa (11.50) (9.66) (1.11) (0.73) Latin America (18.59) (1.90) (4.04) (12.65) Near East (!7.04) (2.34) (1.80) (12.90) Far East (11.21) (5.12) (3.83) (2.26) CPEs 19.58 1.38 3.14a 15.06 TOTAL 170.24 29.73 33.51 107.00 Sources: See Appendix Tables 1,2. Notes: a Calculations for CPE's are made for the 1960-80 period, due to lack of data on intra-CPE trade for the 1948/52 period. Trade between market economies and CPE's was extremely small during this period, but increased substantially during the 1950's. If per capita imports by CPE's during the 1948/52 period were assumed equal to those of 1960, the effects of population growth on trade would increase to 2.77 and 4.15 million mt for the developed and developing CPE's, respectively. These calculations yield overestimates, and do not alter the conclusions presented in the text. 03 observed change in import demands. For the CPEs, these factors account for only ten percent of the growth in imports. In total, 63 percent of the increase in import demand remains as an unexplained residual. The residual column of Table 4 comprises a potpourri of effects-- increases in per capita production, and consunotion effects induced by changes in real prices and incomes. Most importing countries demonstrated increases in per capita cereal production, and these production increases would reduce import requirements. Since the residuals are positive in all regions, however, this category of production effects must he dominated by demand forces. While real grain prices declined during the 1950's, most of the increases in trade volume occurred after 1960, a period in which real grain prices were constant (see below). Thus relative price changes for consumers are not likely to provide a significant factor in the explanation of the residual of Table 4.4/ Income growth emerges as a principal force in increased import demand. Two empirical generalizations--Engel's Law and Bennett's Law--are useful to elucidate the role of income growth in increased demand and world trade. Engel observed that the income elasticity of demand for food is less than one and declines as incomes increase. A key implication of this result is that income growth in L.DCs will have a larger proportional impact on food demand than an equivalent growth rate among consumers in developed countries. In the 4/ - Section four demonstrates that world prices have no necessary implications for domestic consumers, due to government monopolization of international trade. Thus a necessary condition for the elimination of real nrice change as a determinant of increased trade is that the rate of protection of domestic consumers is constant or declining over time. Such a presumption is probably reasonable in the aggregate, but is clearly not accurate for all countries. In the USSR, for example, nominal prices of flour were held constant throughout most of the 1960-80 period. 10 case of cereals, this generalization extends to absolute quantities as well, since income elasticities of direct demand are zero or negative in the DCs. Mellor (1960) and Heady (1962) provide estimates of -0.2 to -0.5 for DC consumers for the 1950s and 60s. At higher income levels, elasticities probably-return towards zero. US expenditure data for the 1965-1981 period, for example, show no significant change in the per capita consumption of cereals (USDA, 1981). Income elasticities of direct demand appear low in most developing economies as well, although significantly greater than zero. Studies by Mellor (1960), Timmer (1971) and Timmer and Alderman (1979), for example, suggest a range of elasticities for cereal demand between 0.1 and 0.7. These elasticities usually are based on expenditure data, and quantity elasticities are probably somewhat lower. But even with low elasticities, the income growth of the developing world during the post-war period was substantial enough to cause major increases in per capita grain demand. Real incomes roughly doubled during the 1950-75 period (see Morawetz, 1977) and even if the average quan ity elasticity was as low as 0.2, per capita grain consumption would increase by about 20 percent (at constant real prices). The data of Table 2 indicate an increase in per capita consumption of about this magnitude. Engel's Law is of little use in explaining increased demand in the developed countries, however, as income elasticities are near zero or negative. This group of countries accounted for about 90 of the 170 million mt increase in imports between 1950-80. Bennett's Law identifies the linkages between income and increased grain demand in these countries. Bennett's Law states that as incomes increase the share of the food budget devoted to starchy staples will decline. As a result, the budget shares devoted to animal products will increase. Thus total grain demand may increase indirectly via the use of grains in .the production of meat, eggs, milk and milk products. Income elasticities of demand for these products appear significantly positive at all income levels. Mellor (1966) provides data for 1960 which show meat demand elasticities in developed countries ranging from 1.7 to 0.4, generally varying inversely with respect to income level. Since grain input animai production ratios are well above unity for most animal. products, these income elasticities will exert a magnified impact on grain demand.- Cross-sectional data for feed grain consumption in 1973 are used to estimate a relationship between income and grain for feed. Cross-sectional estimates are complicated by a number of factors. Government policies influence frequently both prices and availabilities of animal feeds. Pasture may substitute for grains in the production of most meat products. Meats vary substantially in the efficiency of grain conversion, and cross-country differences in preferences for types of meat can create substantial variation in feed grain consumption. Relative prices and availabilities of animal products vary substantially across countries. Finally, observations at the national level reflect a particular income distribution within the country. While these differences are probably not systematically correlated with income level, the resulting left-out variable problem causes high standard errors and reduces the probability of statistically-significant estimates. In spite of these drawbacks, significant relationships appear between income level and indirect grain use. Equation (1) presents the results of a double logarithmic estimation of feedgrain consumption in 1978 in 36 countries. The data are presented in Appendix 3. 12 In CFeed -6.57 + 1.43 InY (1) (-5.96) (8.91) R= 0.69 F = 79.,42 D = 1.83 (1,34) As expected, the intercept is negative, as feedgrain consumption is zero at the lowest income levels. The demand for feedgrains is income-elastic, a result of the magnified effect of income elasticities for meat demand. The results of equation (1) are presented graphically in Figure 1, together with data on direct grain consumption. Feedgrain consumption becomes positive at per capita iacomes of $100 (although this effect is undoubtedly due to intra- country income distribution), and remains fairly small in absolute quantities until per capita incomes surpass the $2000 range. At incomes of about S3,PO0, indirect demand for grain reaches 185 kgs, roughly equivalent to direct consumption, and increases rapidly thereafter. At the highest income level ($9,500), indirect grain consumption is nearly four times as large as direct grain consumption. Thus income provides thi key to understanding increases in grain imports by the developed as well as the developing economies. In spite of major increases in per capita cereal production in nearly all developed economies, .consumptioa and imports increased even more rapidly. Net imports in the developed CPEs, for example, rose from zero to 107 kgs per capita between 1960 -and-1980, in spite of per capita production increases of 105 kgs per capita during the same period. In sum, the data suggest that the growth of international grain trade served primarily as a means to augment per capita consumption levels rather than as a substitute for lagging domestic production in particular countries. This result does not imply that the growth of trade was the only avenue to increased consumption. As the followina section Figure 1. Income and Grain Consumption,.1978. 900 800 700 Total Grain Consumption 600 500 400 j 300 Direct Grain Consumption 200 100 Income 0 750 1500 3000 4500 6000 7500 9000 (US $/cap 13 demonstrates, government price policies for consumers and producers are largely responsible for the prominent role of the international market.in facilitating consumption growth. IV Government Poliey, and International Trade One indication of the importance of government policy fir the growth and price behavior of the international grain market is provided by a comparison of domestic market prices with border prices. With free trade, domestic prices would equal border prices, and differences between the two thus indicate the presence of market imperfections or government-induced distortions. Since large numbers of firms are involved in production and processing in most countries, price diveruonces invariably reflect the presence of government policies. Table 5 and Appendix Table 4 list estimates of domestic nominal protection coefficients (the ratio of the domestic consumer price to the cif or fob price) for a number of countries in the late 1970s, The list is drawn from a number of sources and comprises about 95 percent of world wheat and maize consumption and 92 percent of world rice consumption. When multiple estimates of coefficients are available, the extreme values of the coefficients are indicated. In only six cases do studies disagree over the issue of subsidization or taxation of consumers--wheat in Australia, maize in Thailand, Malawi and South Africa, and rice in Senegal and Argentina. Rather, differences occur mainly in the estimation of the magnitude of subsidization or taxation. This variation is expected given the differences in choice of base year and hence world prices. Official exchange rates of many countries have also changed significantly during the 1970s. Some of the estimates utilize shadow rather than official rates of exchange, as currencies TALE 5. (XISLYBPTION AN) NF' PYUECf[ON UMFICU-US RMI fl.S, 1975-80. WAIlENT MAI7X, RIfY 0.95 < 0.45 < 0.5 < Region NKC < 0.95 NPC < 1.05 NIC > 1.05 NW < 0.95 NIC < 1.05 NU > 1.05 NPC < 0.95 NIC < 1.05 NIC > 1.05 Share of Vrld Constuption (percent) DEVEJ)PED ~t~art,et Economies 1.72 5.87 13.48 0 36.29 11.79 0.07 0.75 4.41 CPls 31.76 0 0 10.51 0 2.12 0.81 0 0 Kirket Economies Afrca 1.55 0.34 0.44 1.29 0.03 0.81 0.25 0.16 0.79 IAtin Amrica 3.73 0.22 0.74 9.84 0.37 1.24 0.18 3.15 0 Ncar East 5.42 1.59 0.11 0.99 0.16 0.39 0.56 0.04 0.10 Fär East 11.01 0.16 0.86 3.41) 0.46 1.32 39.20 2.55 2.95 CPEs (China only) 16.00 0 0 14.16 0 0 35.98 0 0 a 71.19 8.18 15.68 40.19 37.31 17.67 80.02 3.68 8.25 Sources: &e Appendix Table 4. Notes: a Cnefficient values are nnt knm for 4.95, 4.83 and 8.05 percent of Jheat, maize and rice constnptIton, respecttvely. 14 in many developing and centrally-planned economies are believed overvalued. In this case, nominal protection coefficients will he smaller when estimated at the shadow rate of exchange than when measured at the official rate. Universal application of shadow prices of foreign exchange will thus shift the observed distribution of net protection coefficients toward values less than one.5/ Finally, errors may result if estimates do not compare identical qualities of grain. vlhite corn, for example, commands a 20 percent premium to yellow corn on international markets, and comparison of domestic white corn prices to international prices for yellow corn could suggest taxation of consumers where none exists. Most information sources anpear well aware of this potential problem, however. The results suggest major differences between grain nolicies of the developed market economies and those of the rest of the world. In the former group, only eight percent of wheat consunption and one percent of rice consumption takes place under conditions in which domestic prices are less than world prices. The remainder of developed country consumers pay domestic prices for the major grains which are equal to or greater than world prices. . No theoretical considerations mandate the use of shadow rather than official rates of exchange. First, foreign exchange accounts are not out of balance in most countries, as foreigners are unwilling to accumulate the relevant domestic currency. Thus the official exchange rate reflects an equilibrium in foreign balances. Second, foreign exchange rates do not have to be altered to reflect the opportunity costs of comnmodities and resources. The only requirement of equilibrium is that the relative prices of nontraded and traded commodities reflecz a particular ratio. This equilibrium can be achieved at fixed exchange rates by alteration of the prices of doestic factors of production. Factor price changes may occur in addition to, or in complete substitution for, changes in the domestic currency prices of tradable goods. Changes in factor prices will affect factor incomes and the quantity consumed of tradable goods, but given the small shares of most countries in total consumption of cereals, these income-induced changes in demand will have no effect on world prices or the net protection coefficient. See Bruno (1976) and Dornbusch (1973). Among the remaining country categories, consumers are generally subsid.ized. In only a few countries--Ghana, Nigeria, Zaire, Colombia, Guatemala, Taiw'an and Korea--do consumers face grain prices which are above world price levels. Particular grains may be taxed in some countries--wheat in Thailand and the Philippines, for example--but these cases do not usually involve the primary staple of the particular country. In all develoning country regions except Africa, the vast majority of countries subsidize consumption of the primary staple. The aggregate results are presented in the final row of Table 5, and indicate the dominance of subsidy policies in world grain consumption. The results are particularly striking for wheat and rice, as more than 70 percent of wheat and 80 percent of rice consumption takes place at prices below those of the world market. In contrast, only 40 percent of maize consumption appears subsidized, reflecting the much larger shares of the developed market economies in world consumption (48 percent), and its widespread use as an animal feed rather than as a staple for human consumption. Only 16 percent of wheat, 18 percent of maize and eight percent of rice consumption is taxed. The developed market economies account for more than two-thirds of the wheat and maize totals and half of taxed rice consumption. The patterns of net protection coefficients and trade policies reflect substantial differences in domestic objectives across countries. In the ceveLoped market economies, high consumer prices have been utilized as a means of effecting income transfers to domestic producers and attaining a substantial degree of price stability for both groups (Johnson, 1973; Josling, 1974). A reliance on trade policy to realize transfers reduces the treasury burden of income-support programs. The variable levy is particularly well- suited to the stability objective in this circumstance, as world price fluctuations result only in fluctuations in tariff revenues. Further, these policies are recognized by some as a means of exerting monopoly power in trade (Carter and Schmitz, 1979;-Sarris and Schmitz, 1981), as the protectionist policies of Western EurGpe and Japan are seen capable of changing the terrms of trade with grain exporting countries by decreasing world market prices. The data of Table 5 show, however, that protectionist policies are far from dominant in world markets. Low consumer prices have been the principal objective in most countries. In both the developing and centrally-planned economies, price subsidies are critical policies to maintain low-cost wage goods in the industrial sector and LO ensure nutritional adequacy for low income consumers (see, for example, Timmer and Falcon, 1971; Chibber, 1979; Parker and Coyle, 1981). Unless production is equally subsidized, reliance on international markets is larger than under free trade and world market prices are artificially high. Production subsidy programs are not much more common than consumer tax programs, as government budget constraints usually dictate trade monopoly policies which force domestic producers to provide part of the subsidy to consumers. In a study of rice i:: nine Asian countries, for example, Timmer and Falcon (1975) find an average ratio of producer prices of rice to fertilizer of about 0.5, while world price incentives would dictate a ratio two or three.times as large (Falcon and Honke, 1979-80). A cross- section time-series production function, estimated in the former study, suggested that changes of this magnitude in prices would alter the net trade position of the sample of countries from zero to exports of ten-fifteen million nt. These amounts are roughtly equal to total world exports of rice. A second circumstance in which world market prices might not increase under free trade arises if rationing constrains consumption. Perhaps the mos.t significant programs in this regard include grain consumption in the developed 17 CPEs, where meat rations may artificially constrain feedgrain demands. These constraints have become less important in recent years, however, as increases in meat supplies have been a principal objective of consumer policies in these countries (Poleman, 1975). In developing countries, such as China, Egypt and India, direct grain concumption does not appear constrained in most years, and both free market and ration shop prices are below world market prices (Timner, 1976; C1-2ber, 1979; Parker and Coyle, 1981). ?rice stability is also an objective of consumer-oriented policies, and government monopolies on foreign trade create the potential for maintenance of domestic stability in the face of world price fluctuations. In this case, however, increases in world prices or shortfalls in domestic production increase government expenditures rather than reduce government revenues as in the case of variable levies, and budget constraints may occasionally limit the extent to which imports can be increased. In addition, the government must predict quancities of domestic production and required consumption, an empirical task avoided by the use of a variable levy. As a result, stability under sul-sidized price levels has proven a more difficult task than stability under high price regimes. Four alternative policy interventions are possible in the maintenance of divergences between domestic consumer prices and world market prices. Government monopolies could directly handle all grain imports and exports. Quantitative restrictions could be imposed, in,which the private sector handles trade subject to quota or licensing arrangements. A third category involves trade tax/subsidy policies, such as tariffs or variables levies. A final alternative policy comprises free trade, in which divergences between domestic and world prices are maintained by the provision of direct taxes/subsidies to domestic consuners and producers. 12 Table 6 and Appendix*Table 5 categorize the trade policies of countries in which domestic prices deviate substantially from world market orices. Government monopolies are the most common policy, accounting for 75 percent of distorted wheat consumption, 86 percent of distorted rice consumption and 63 percent of distorted maize consumption. These ratios reflect the importance of the C2Es, as foreign trade in almost all products is government- controlled. China and the USSR alone account for 40, 20, and 37 percent of world wheat, maize and rice consumption, respectively. In the developing market economies, government monopoly policies again predominate. Trade taxes and quantitative restrictions are the dominant policies only in the developed market economies, with the variable levy systems of the EEC and other European countries of partitcular importance. In almost all countries where consumer prices differ from world prices, free trade is not allowed. Saudi Arabia and Guyana represent the only exceptions. Three factors help to explain the reliance on trade policies rather than direct interventions in producer and consumer markets. First, alternative policies differ in their impact on the government budget. The theory of optimal taxes and subsidies (see, for example, Corden, 1974) suggests that trade interventions are never optimal ways to distort world prices, as trade interventions impose unnecessary efficiency losses on consumers or producers in the process of providing the desirable tax/subsidy to the remaining group. From the perspective of the government .treasury, however, trade interventions will always be preferrable methods to distort world prices. In two circumstances, the use of trade policy generates government revenues. Exporting countries who desire low consumer prices can collect the revenues from export taxes while importing countries who desire high producer prices can collect import tariffs. The alternative policy (consumer or producer TABLE 6. GRAIN TRADE POLICIES. a Covern-ent Trade Ouotas and Trade Taxes/ Free Monopoly Licenses Subsidies Trade Shares of World Consumption (pert) W1EAT - Developed Market Economies 1.00 1.71 12.90 0 Developed CPEs 28.01 2.40 1.10 0 Developing Market Economies 20.53 1.39 1.80 0.30 Developing CPEs 16.00 0 0 0 TOTALS 65.55 5.58 15.23 0.30 IAIZE Developed Market Economies 2.43 0.37 9.36 0 Developed CPEs 8.61 3.71 3.71 0 Developing Narket Economies 13.87 3.13 2.93 0 Developing CPEs 14.16 0 0 0 TOTALS 39.07 7.21 16.00 0 RICE Developed Market Economies 4.06 0 0.35 0 Developed CPEs 0.78 0 00 Developing Market Economies 37.54 8.75 3.29 0 Developing CPEs 35.94 0 0 0 TOTALS 78.36 8.75 3.64 0 Sources:. See Apppendix Table 5. Note: a Only those countries wiht net protection coefficients different from unity are Included in this Table. 0 19 subsidies) represents a burden on the treasury. In the remaining two cases, the treasury burden is less for a trade intervention than for a direct subsidy. An import subsidy in an importing country which desires low consumer prices and an export subsidy in an exporting country which desires high producer prices-allow domestic producers and consumers to pay some of the costs of distortions. A second reason for the reliance on trade controls involves the implementation costs of domestic tax/subsidy programs. These administrative costs can be substantial, especially in developing countries, where the dominance of grain production and consumption in aggregate output and expenditure, limitations on the revenue base and an underdeveloped infrastructure make the establishment of universal subsidy programs costly and often impossible to finance. Trade policy, in contrast, requires minimal administrative manpower, and operates usually through well-established marketing channels in the major port areas. Finally, and related to the previous points, trade controls are usually a necessary component of rationing programs. Consumer prices Thich are held below world prices do not necessarily imply that consumption levels are larger than they would be under free trade, as quantitative limits may constrain individual demands. The maintenance of such restrictions, however, requires the prohibition of access to international markets by domestic producers, as higher world prices would attract all of domestic production. Logistical considerations may also mandate the use of imports to supply ration shops, particularly when the urban-oriented programs are located nearer to port facilities than to rural production areas. In China, for example, wheat imports are utilized primarily to supply urban areas on the cqxt (International Wheat Council, 1976-77). 20 The data of Tables 5 and 6 reinforce the view of the international grain markets as a.means to augment domestic consumption levels. The maintenance of low consumer prices, and shifts in the demand curve induced by population and income growth were riot matched by compensatory shifts in the domestic supply curve. Whatever the reasons for lagging production response--low producer prices, lagging rates of technological change, constraints on potential grain area--government trade policies have translated net demand increas,. into international market demands. The prominence of consumer subsidy and producer tax policies across countries suggests that importer reliance on the international market is larger than under free trade, and that world market prices are above their free-trade level. V. Price Behavior on the International Market The dominance of government monopolies and quantitative restrictions imply that movements in world prices have only indirect relevance for most consumers and producers. This result is particularly significant for rice and wheat, where world prices equal domestic market prices for only four and eight percent of world consumption of the two grains, respectively. Even in countries where domestic prices are held above world prices, trade taxes and subsidies are used only rarely. Further, most trade taxes involve variable levy policies, and domestic consumers and producers again have latitude to alter behavior without regard to world prices. Thus international market behavior becomes a reflection of the impact of government policies aggregated across countries rather than a direct consequence of the actions of consumers and producers. The low price elasticity of demand for grains suggests that prices provide a more sensitive barometer of changes in international market 21 conditions than quantities. Figure 2 illustrates the movement of nominal international prices of wheat, maize and rice for the 1950-80 period. Data are provided in Appendix Table 6. The prices represent widely-traded varieties, ahd thus may not be representative of changes in the markets for special varieties of these grains. Prices of round-grain rice (Petzel and Monke, 1979-80), durum wheat and white corn, for-example, may represent commodity markets which perform independently of the markets for more widely traded varieties. Such varieties, however, do not account for a substantial share of cereal production or trade. Nominal grain prices demonstrated little change throughout the 1950s and 60s, indicative of market conditions.in which available export supplies were in excess of import demand. The price movements of individual commodities show some differences, roughly synchronized with the relative growth rates of production in the exporting countries. Coarse grain prices trended downward in the 1950s, caused initially by lack of purchasing power among importers in the immediate post-war period as absolute trade volumes fell below pre-war levels. Subsequent price declines refl-)cted major increases in coarse grain production, due largely to the spread of hybrid corn in the US and Canada. By the early 1960s, these two countries accounted for more than half of world corn production and three-fourths of world trade. Similar conditions prevailed in the wheat market, as developed country yields increased throughout the 1950-70 period. Rice prices showed somewhat greater fluctuations, reflecting the concentration of 90 percent of world production within a single climatic zone, the monsoon region of Asia. In addition, the small size of world trade relative to world production made the international rice market more sensitive than wheat and corn to production variations. ItUS $/mt 21A 550 . 500 450 Rice 400 350 300 250 200 Wheat 150 Maize. 100 50 1950 1960 1970 1980 Figure 2. Nominal Grain Prices, 1950-80 22 Producer price policies in the US and Canada were of particular importance to the stability of world prices during the 1950s and 1960s. Domestic prices remained consistently above world prices, and the inability to restrain production via acreage controls resulted in rapid increases in government-owned stocks of wheat and corn. As a result, stocks were well in excess of world trade during most of the 1950-70 period. Wheat stocks in the major exporting countries, for example, increased from 18 to 55 million mt between 1951 and 1957, with four-fifths of the total held by the US and Canadian governments. By 1960, total wheat trade was only 40 million mt. Under these conditions, declines in world prices were not needed to encourage increases in annual carryover, and increases in world prices were not needed to encourage a reduction in annual carryover. Rather, the difference between domestic and world prices, and governmental willingness to subsidize exports appeared as critical determinants of stock changes. With stocks well in excess of trade volumes during most of the 1950-70 period, government actions maintained an effective ceiling on world prices. Major increases in import demands, such as USSR and South Asian wheat ir.ports in the mid-1960s, were met without substantial changes in price. Government stock accumulation also encouraged the growth of concessional exports. The growth of maize imports by developed countries and wheat imports by LDCs represent the most significant categories of increased trade in individual grains during the 1950-70 period. Almost all of the increase in LDC wheat imports (from 8 to 23 million mt) is accounted for by concessional exports. Table 7 provides data on the growth of concessional trade, and shows the continual upward trend in aid shipments until the mid-1960s. The US dominated the trade until this period, and wheat was the principal commodity traded, accounting for 75 to 85 percent of total cereal aid during the 22A TABLE 7. CONCESSIONAL EXPORTS OF CEREALS, 1955-80. Share of Yeara US Canada EEC Total orld Trade (million metric tons) 1955 0.30 56 3.22 57 7.14 53 6.22 59 7.89 60 10.09 61 11.12 0.19 62 12.18 0.05 63 12.84 0.07 64 12.32 0.26 65 13.16 0.40 66 12.26 1.17 67 9.50 0.80 63 7.81 0.78 69 10.01 0.66 1.36 12.77 13.2 70 9.10 1.61 0.98 12.83 11.4 71 9.26 1.09 0.98 12.56 10.6 72 7.02 0.89 0.99 10.11 7.7 73 3.20 0.49 1.21 7.85c 5,0 74 4.71 0.59 1.41 8.38 5.6 75 4.28 1.03 0.93 6.86 4.3 76 6.15 1.18 1.13 9.11 5.3 77 5.99 0.88 1.49 9.34 5.8 78 6.24 0.74 1.24 9.57 5.1 79 5.44 0.55 1.29 9.00 4.5 80b 5.73 0.60 1.65 9.39 4.2 a July/June shipments before 1979. For 1979/80 and 1980/81, data relate to budget year of each country. b Commitments or allocations. c Includes shipments by USSR of 2.2 million at to India and Banglidesh. Sources: 1969-80 FAO, Food Aid Bulletin, Rome 1955-68 USDA, ERS, 1974, US Agricultural Exports under Public Law 480 ERS-Foreign 395, Washington, D.C. 1961-68 Canada FAQ, National Grain Policies, 1972. 23 1960s. After 1966, US sales terms were made increasingly strict. Payments were Made in US dollars (.7ith varying credit terms), rather than in local currency, and availabilities were conditioned on recipient country efforts to increase food production. By the late 1960s, aid levels had peaked at about 13 million mt. Declines in US availabilities were particularly significant after 1972, and more than offset subsequent increases in aid from other areas, such as the EEC. Aid shipments ranged between seven and nine million mt during the 1973-80 period, or four to six percent of world trade.6/ Two exceptions to the scenario of stable or declining prices in the 1950- 70 period are of particular historical interest. The first case involves the increase in wheat prices in 1964-65, as the USSR attempted to offset partially a 21 million mt production decline with nine million mt of increased imports. This magnitude represented about 20 percent of world trade volume, and was the first major attempt by Soviet planners to use the international market as a way to offset variability in domestic production. The second event concerns the increases in rice prices in 1966 and 1967, a consequence of consecutive years of unfavorable weather and political turmoil in Asia. Similar events were to recur in 1972-73, with far more serious consequences for prices. In the 1970s,,world prices became much more sensitive to variations in world production, as new stock and production policies of the developed 6/ Johnson (1973, Chapter 8) provides a concise summary of the possible effects of these aid programs on the developing countries. Whether the aid programs served to support international commercial market prices is not clear. Certainly if commercial export availabilities had increased by ten- fifteen percent, world market prices would have fallen. But continued retention of stocks or destruction of the commodities may have been preferrable ways to manage domestic production surpluses. 24 countries reduced the capacity of supply-side adjustments to limit market price increases. Particularly important was the reorientation of US and Canadian pLoduction towards world market incentives. Export subsidies from the US govertment were suspended in late 1972, and government grain stocks weZe largely eliminated. Wheat stocks held by the Comodity Credit Corporation, for example, declined by 19 million mt in 1972/73. In addition, producer price policy was altered. The Agricultural and Consumer Protection Act of 1973 replaced cost-of-production parity prices with a target orice system. These prices were set well below world prices, and for most of the 1970s target prices served as floor prices rather than market prices. While the bill maintained governmental authority to adjust production, set-asides were given reduced priority and not utilized after January, 1973. This policy shift increased potential crop area by six million ha, and reduced the potential of acreage adjustments to limit market price increases. The most dramatic indication of this new constraint appeared in 1972/73, as bad weather in North America, the USSR and Asia caused declines in the production of ell cereal grains. Import demands increased as production declined in exporting countries, resulting in unprecedented increases in grain prices. World cereal production declined by three percent, and was no larger .than six percent among the individual grains. International market prices for wheat, maize and rice increased in 1973 and 1974 by 158, 136 and 270 percent, respectively, reflecting the inelastic demand of government importers. Trade volume increased by 30 million nit between 1972 and 1973, and price responsive storage behavior in the US caused export shares to climb to 70 percent in coarse grains and 50 percent in wheat. A similar production scenario recurred in 1974/75, as world production of wheat, coarse grains and rice fell by four, three and one percent, respectively. In this case, however, income declines 25 in many importing countries and decreases in feedgrain use in the US and Canada prevented a repetition of the 1)73 price increases (Johnson, 1977). Nominal prices remained at a relatively constant level through the latter half of the 1970s, although price variability remained substantially greater than in the earlier decades. As in the 1950-70 period, rice prices demonstrated the greatest variability. Trade in rice remained between three and four percent of world production, and the production of major exporting and importing countries continued to demonstrate positive correlation. In addition, rice stocks comprised less than ten percent of consumption in most years, and the dominance of human consumption in total utilization gave the rice market particularly inelastic price behavior. Coarse grain stock levels were comparable to rice, ranging from nine to fourteen percent of utilization. Prices, hourever, show more stability than wheat and rice prices due to the greater price elasticity of meat demand and the presence of non- grain substitutes for feed, such as pasture and starches. Figure 3 summarizes the institutional influences on international market behavior over the 1950-80 period. Panels a and d provide a diagrammatic representation of international market conditions during most of the 1950s and 60s. In this case, exporters desire high producer prices, creating a surplus which is potentially available to international markets. Importers desire subsidies to consumers, resulting in a potential import demand. High producer prices and changes in production practices caused the supply curve in developed market economies to shift outward, while demand remained relatively stagnant. At the same time, demands were expanding in the importing countries due to income and population growth. Thus high producer prices created surpluses which effectively damped exogenous demand shocks from importing countries, and prices remained stable from year to year. 25A Figure 3. Trade Policy and Domestic Objectives Government Revenue \ Government Subsidy on Traded Quantities on Traded Quantities P P D D \ /Q ' S PP P SS D S a. Importer, Low Consumer Pricos b. Exporter, Low Consumer Prices' pP D S D S SD D c. Importer, High Producer Prices d. Exporter, High Producer Prices '24 The 1970s represented a marked chanke in policies of the exporting countries, as the programs for high producer prices were largely eliminated. The result of this policy change was that exogenous shocks to production or consumption could no longer be accomodated without price adjustments on the international market. Grain exporters who maintain low consumer prices (panel h) will attempt to compensate for shortfalls in domestic production with exnort restrictions, while importers with a similar objective (panel a) will attempt to increase imports. The result is the creatibn of excess demand without compensatory availabilities of excess supplies, and prices must adjust to resolve the temporary conflicts between the trade pilicies of different countries. The magnitude of price adjustment will depend on the size of the international market relative to total production, since government trade monopolies can attempt to transmit the entire deviation in production or consumption to international market.7/ The emergence of incompatibilities among international market policies provides only a part of the explanation of market behavior during the 1970s. -'he international market remained dominated by government policy, but the behavior of the principal actors became increasingly erratic. Particularly important were the policies of the CPEs, who accounted for nearly half of the 100 mt increase in.cereal trade during the decade. In the USSR and Eastern Europe, limited data indicate increases of 15-30 percent in numbers of cattle, hogs-and poultry (International Wheat Council,,1978/79). Increased numbers were largely supported by wheat and maize imports, as domestic cereal 7/ Panels c and d also depict an incompatible policy set, as exporting countries will attempt to dispense with excess production at the same time as importers are attempting to decrease imports. Prices would be expected to move downward in this circumstance. 27 production has increased by only 10 percent during this period. This increase in trade participation rates has been accompanied by increased variability, reflecting the,commitment to maintain herd sizes in the face of domestic production shortfalls (Poleman, 1975). Grain imports of the USSR, for example, increased by 16 million mt in 1972/73 ar. by 20 million mt in 1975/76, relative to the previous year. The PRC has followed similar policies, although of a lesser magnitude. Cereal imports increased by about 12 million mt during the 1970s, becoming increasingly variable toward the end of the decade. Wheat imports, for example, increased from 3.2 to 8.6 million mt between 1976/77 and 1977/78, and from 8.8 to 12 million mt between 1979/80 and 1980/81. This change was accompanied in each case by an increase of only 250,000 - 300,000 mt in rice exports. Two estimates of real prices for each of the grains are provided in Figure 4. One deflator is the World Bank index of cif unit values of developing country imports of manufactured goods, and provides an indication of the relative cost of grain imports in LDCs. The second deflator is the FAO index of fob unit values of developing market economy exports of ag-icultural products, which provides an indication of purchasing power of LOC grain importers. The latter index is not seriously biased by the inclusion of ,cereal exports, as cereals accounted for only a small percentage of the value of agricultural exports by market LDCs (6.7 percent in 1980, for example). Doth-indexes suggest similar patterns of real price movements. The real price estimates provide two additional insights into the international grain economy. First, increases in per capita cereal imports of the CPEs and LDCs have been met at constant or declining real costs. While prices became more variable in the 1970s, the overall trend in real prices was negative (but not significantly different from zero at the 95 percent Figure 4. Grain Prices, 1960-80. Wheat, 1975 LS $/mt --- cif index 200 L fob index WIIEAT 150 100 50 _._..................---~. --_- -- ~ - 1960 1965 1970 - :1975 1980 Maize 1975 US $/mt MAIZE 150 100 50 1960 1965 1970 1975 1980 t Rice RICE 1975 US $/mt 600! 5001 4001 300 200 1960 1965 1970 1975 1980 1 4i confidence level).S/ This result is surprising, given the major increases in the volume of trade in all three grains since 1960 and the sharp reductions in developed country surplus disposal during the 1970s. Second,,international prices for the three grains move largely in concert, pointing to the importance of linkages between the three markets. Table S provides correlation coefficients for the various price series. All the coefficients Are significantly different from zero. Substitution possibilities in production are fairly limited, at least in the short run, and price linkages are thus due to adjustments in consumption or stocks. Relationships among the grains are most significant for -heat and rice and wheat and corn, and suggest that wheat plays the central role in the linkage of the various grain markets. The linkage between wheat and corn reflects the use of both grains as animal feeds. The corn trade is directed primarily toward animal feed uses, and competed traditionally with barley and other coarse grains. Wheat has become increasingly important as a feedgrain, however. Wheat provided between 1.4 and 5 percent of total feedgrain use in the US during the 1970s, for example, and use of wheat as a feed by major wheat exporters increased from 10 to 20 million over the last two decades. While this volume is small relative to total wheat consumption and total fe-3dgrain use, it represents twenty percent of world wheat trade, and thus provides substantial latitude for adjustment should maize prices deviate significantly relative to wheat. Regressions of real price- (FAO index) against a time trend for 1960-80 indicated average annual price declines of 0.7 percent, 1.0 percent and 0.4 percent for wheat, maize, and rice, respectively. Regressions based on the World Bank index indicated average annual declines of 0.9, 1.2, and 0.5 percent. 28A TABLE 8. CROSS-PRICE CORRELATION COEFFICIENTS FOR GRAINS, 1960-80. a. Grain prices deflated by World Bank import index a Wheat Maize Rice Uheat 1 Naize 0.84 1 Rice 0.83 0.71 1 b. Grain prices deflated by FAO export index a Wheat Maize Rice Wheat 1 Maize 0.81 1 Rice 0.79 0.65 1 Source: See Appendix Table 6. totes: a/ Correlation coefficients greater than 0.44 are significantly different from zero at the 95 percent confidence level. 29 Importing countries may also engage in substitution, particularly when the domestic wheat crop is of poor quality for human consumption. This circumstance will result in increased use of wheat as feed, decreases in corn imports and increases in wheat imports for bread. Eastern Europe appears the most significant region to arbitrage the markets in this fashion (International Wheat Council, 1980/81).9/ Critically important to the wheat-rice linkages is the willingness of rice consumers to substitute wheat, partictlarly in Asia. Popular beliefs that rice consumers were intransigent with respect to staple food consumption are belied by both econometric demand analyses and trade statistics. Timer (1971) and Timitrer and Alderman (1979), for example, provide evidence of high cross-price elasticities for Bangladesh and Indonesia. Elasticities appear particularly high among low income consumers, and in the Indonesian case, substitution extends to starchy staples (cassava) as well. The expression of consumer preferences depends to a large degree on the international buying habits of government marketing agencies. While these agencies may bring quite inelastic demands to the market in terms of total ctreal requirements, trade data suggest that governments are willing to substitute among grains. Examples of this behavior include increases in wheat and wheat flour imports among traditional rice consuming areas, particularly during periods of high rice prices. In 1974, for example, Indian imports of wheat increased by two million mt while rice imports declined by 200,QOO mt. _. Johnson (1977) points out that the growth of feedgrain use has increased the capacity of the international market to meet global food shortages even more rapidly than is suggested by changes in the share of trade in world production. 30 A second way in which government policy links international prices of rice and wheat involves the trade policy of the PRC. Chinese rice exports and wheat imports are related to relative international prices of the grains rather than changes in domestic production. Rice exports of China have consistencly remained between 0.8 and one mmt during the last two decades, with the exception of 1966-67 (1.24 mit) and 1973-75 (I.Q mnt). In both periods, international rice prices were high relative to wheat prices. Such arbitrage maximizes net foreign exchange earning from agriculture while maintaining per capita calorie availabilities for consumers. In addition, the policy appears logistically feasible, as most wheat imports are directed to urban coastal areas, which are also relatively large rice producing areas (Timer, 1976; International Ldheat Council, 1976/77). Chinn (1979) provides statistical confirmation of this relationship. VI. Policies and Instzbility in Tradel0/ On average, governm-nts have been able to utilize international markets at constant or declining real costs throughout the 1960s and 1970s. But stagnation in the availabilities of concessional aid and the increased instability in the international markets have spawned new criticisms of the grain trade. Two problems have been mentioned most frequently. The first difficulty involves the availability of foreign exchange among LDC importers. The foreign exchange requirements for food imports are believed large, and export earnings are seen as constraints on cereal imports and 10/ Much of the data presented in this section is taken from Salah Abdelsalam, 1982, Food Self-Sufficiency and International Grain Trade, unpublished U.S. Thesis, Department of Agricultural Economics, University of Arizona. 31 consequently on the Lbility to maintain desired consumption levels (Abbott, 1979; Siamwalla and Valdes, 1980). A second problam, somewhat related to the first, involves.instahility on the international markets. Even if countries are able to 'pay for grain imports at average prices, the extreme fluctuations in the international market create temporary foreign exchange constraints. As a result, desired levels of domestic price stability cannot be attained. Table 9 provides data on the costs of cereal imports relative to commodity export earnings for fifteen of the sixteen largest importers (data for China were not available). A more comprehensive list of 65 LDCs is presented in Appendix Table 7. The results suggest that, on average, LDC importers do not spend a particularly large share of commodity export earnings on cereals. While petroleun imports in the late 1970s accounted for about 30 percent of export earnings of LDCs (excluding petroleum exporters), Ban7ladesh and Egypt are the only major cereal importers who spent a similar share of export earnings on cereals. No other countries spent more than fifteen percent of earnings on cereals during the 1970-80 period. Similar results hold for the 65 country sample, as 56 countries spent less than fifteen percent of export earnings. Forty-seven countries spent less than ten percent, and 29 countries spent less than five percent of export earnings. The list is further reduced when other sources of foreign exchange are considered. Repatriated earnings are particularly important sources of foreign exchange in Egypt, Upper Volta, Jordan.and the Yemen Arab Republic. In sum, only five countries--Senegal, Somalia, Haiti, Bangladesh and Sri Lanka--devoted substantial shares of foreign exchange earnings to cereal imports during the 1970s. The choice of fifteen percent as a 'large' number is arbitrary, as there is no way to determine a critical value for the ratios presented in Table 9 TABLE 9. RATIO OF 'lUUAL yFAL IMPO ortant in the maintenance of constant nominal prices, as increases in the demands of importing countries were met by drawdowns in government stocks. Elimination of these stocks in the 1970s brought substantial changes in the behavior of the international grain markets, as government policies in the importing countries now dictated world price behavior. Importing countries were forced to compete for supplies, with price serving as the allocative mechanism. Changes in international market behavior and increased instability in prices have created new problems for participants, particularly the LDCs. The magnitude of these difficulties should not be exaggerated, as foreign exchange requirements remained a relatively small proportion of availabilities throughout the 1970s. But perhaps the greatest irony surrounding the instability issue is that the policies of the LOCs themselves are at least partly responsible for the magnitude of price instability. Clearly, shifts in policies toVard freer trade would substantially increase-price stability in the international markets, as global production declines of a few percent could no longer generate increases of 100-200 percent in world prices. Yet as long as countries continue to subsidize domestic consumption, international market controls are likely to remain as key components of grain policy. The treasury costs and logistical difficulties of simultaneous producer and consumer subsidies are critical constraints for the relaxation of trade controls. How to cut the Gordian knot? Return to subsidized stock programs? Develop new ways to reach low income consumers in LCDs and producers in DCs? Rely on economic growth to shift government objectives? Resolution of this issue will continue to occupy center stage in future discussions of the world grain economy. 39 REFERENCES Abbot, Philip C., 1979, Modeling International Grain trade with Government- Controlled Markets, American Journal of Agricultural Economics 61(1): 22-31. Bachman, Kenneth L. and L.A. Paulino, 1979, "Rapid Food Production Growth in Selected Developing Countries: A Comparative Analysis of Underlying Trends, 1961-76", International Food Policy Research In-titute Research Report No. L1, Washington, D.C. Bruno, Michael, 1976, "The Two-Sector Open Economy and the Real Exchange Rate," American Economic Review, 66: 566-577. Carter, Colin and Andrew Schmitz, 1979, "Import Tariffs and Price Formation in the World Wheat Market," American Journal of Agricultural Economics, 61: 517-522. Chibber, Ajay, 1979, "Optimal Public Policy and Decision Analysis: A Case Study of the Indian Public Distribution System," Working Paper, Food Research Institute, Stanford University. Chinn, Dennis L., 1979, "A Calorie-Arbitrage Model of Chinese Grain Trade," Working Paper, Food Research Institute, Stanford University. Corden, W.H., 1974, Trade Policy and Economic (7elfare, (Oxford: Clarendon Press). Dornbusch, Rudiger, 1973, "Devaluation, Money and Nontraded Goods," American Economic Review, 63: 871-880. Falcon, Walter P. and Eric A. Monke, 1979-80, "International Trade in Rice," Food Research Institute Studies 17(3): 279-306. Heady, Earl 0'., 1962, Agricultural Policy Under Economic Development, (Ames, Iowa: University of Iowa Press). Hillman, Jimmye S., 1981, "The Role of Export Cropping in Less Developed Countries," American Journal of Agricultural Economics 63(2): 375-383. International 1heat Council, various years, Review of the World Wheat - 'Situation, London. Johnson, D. Gale, 1973, World Agriculture in Disarray, (London: Fontana/Collins. Johnson, D. Gale, 1977, "The Soviet Impact on World Grain Trade," British- North American Committee, Washington, D.C. Josling, Tim, 1974, "Agricultural Policies in Developed Countries: A Review," Journal of Agricultural Economics, 25: 229-264. 40 Joslincl, Tim, 1977, "Government Price Policies and the Structure of International Agricultural Trade, Journal of Aaricultural Economics, 28: 261-277. Knudsen, Odin K. and P.L. Scandizzo, 1982, "The Demand for Calories in Developing Countries," American Journal of Agricultural Economics 64: 79- 36. Lappe, Frances Moore and Joseph Collins, 1977, Food First, (Boston: Houghton Iifflin). ellor, John W., 1966, The Economics of Agricultural Development (New York: Cornell University Press). Morawetz, David, 1977, Twenty-five Years of Economic Development, 1950 to 1975 (Washington, D.C.: The World Bank). Paarlberg, Robert L., 1980, "Lessons of the Grain Embargo," Foreign Affairs, 59(1): 144-162. Parker, John B. and J.R. Coyle, 1981, "Urbanization and Agricultural Policy in Egypt," United States Deoartment of Agriculture, Economic Research Service, Foreign Agricultural Economic Report No. 169, Washington, D.C. Petzel, Todd E. and Eric A. Monke, 1979-80, "The Integration of the International Rice Market," Food Research Institute Studies 17(3); 307- 326. Poleman, Thomas T., 1975, "World Food: A Perspective," in Philip H. Abelson, ed., Food: Politics, Economics; Nutrition and Research, (Washington, D.C.: American Association for the Advancement of Science), 8-16. Sarris, Alexander H. and Andrew 3chmitz, 1981, "Toward a U.S. Agricultural Export Policy for the 1980s," American Journal of Agricultural Economics, 63(5): 832-839. Siamwalla, Ammar and Alberto Valdes, 1980, "Food Insecurity in Developing Countries, Food Policy. Tang, Anthony M. and Bruce Stone, 1980, "Food and Agriculture in China," International Food Policy Research Institute Research Report No. 15, Washington, D.C. Timmer, C. Peter, 1971, Estimating Rice Consumption, Bulletin of Indonesian Economic Studies VII(2): 70-88. Timmer, C. Peter, 1971, "Wheat Flour Consumption," Bulletin of Indonesian Economic Studies, 7: 83-95. Timmer, C. Peter, 1976, "Food Policy in China," Food Research Institute Studies, 15(1): 53-69. 41 Timer, C. Peter and Henry Alderman, 1979, "Estimating Consumption Parameters for Food Policy Analysis," American Journal of Agricitural Economrics, 61: 932-987. Timer, C. Peter and Ualter P. Falcon, 1975, "The Political 7conomy of Rice Production and Trade in Asia," in Lloyd G. Rleynolds, ed., Agriculture in Developbent Theory, (New Haven: Yale University Press), 373-408. United States Department of Agriculture, 1982, "Handbook of Agricultural Charts," Agriculture H-andbook N7o. 609, Washington, D.C. Valdes, Alberto and Barbara [Ruddleston, 1977, "Potential of Agricultural Exports to Finance Increased Food Imports in Selected Developing Countries," Occasional Paper No. 2, International Food Policy Research Institude, Washington, D.C. 42 APPENDICES APPENDIX TABlY 1. ESTIMATES OF POPLATION (miilons) AND IUMA OEAL PRACTII (million rt). 1948-52 1960-61 1969-71 1979-0 Prouction Population Prodiction Population Production Poptdation Production Popilation IEVEIDPED Market 244.31 480.71 299.62 546.49 422.48 725.72 542.26 782.94 CPKs 109.90 269.77 179.17 309.52 231.52 347.93 257.84 376.46 DEVEIDPDE Market 213.59 1202.67 299.13 1,503.34 363.15 1,713.17 452.40 2,185.02 Africa 26.96 186.93 33.78 236.23 43.26 282.73 45.84 371.32 Iatin America 30.96 161.58 46.11 211.85 69.51 282.70 85.68 363.63 tar F.ast 22.80 102.22 31.44 130.30 42.61 165.89 56.15 214.71 Far East 132.87 751.95 187.80 921.74 207.74 977.80 264.68 1,230.29 CPEs 54.4a 569.9 159.0 632.0 228.59 889.88 310.98 1,030.76 1-DUD 'DE 722.20b 2,523.05 936.92 3,038.51 1,245.73 3,676.69 1,563.4( 4,375.13 Soirces: Food aid Agriatiture Organizatton Prcxaction Yearbook, 1963, 1972, 1976, 1979; 1980. (Iinese data taken frcm Anthony Mf. Tai and Bnce Stone, Food Production in the lople 's Republi.c of China, International Food Pol.cy Iesearch institute Research Peport #1511 KMay, 1980. a Ita are for China ari for 1952 only. b (1dna and USSR are excltyled frcn the FAO estimates for 1948-52. 44 APPENDIX TABLE 2. DECLINES IN CEREAL PRODUCTION PER CAPITA, 1948/52 - 1979/31. Aggregate Per Capita Production Production Population Decline 1943/52 1979/81 1979/81 ('000 t)(kgs) ('000) AFRICA Algeria 2,081 221 111 18,919 Angola 248 86 51 7,078 Benin 85 148 124 3,530 Chad 468 254 149 4,455 Ethiopia 1,133 174 138 31,468 Gambia 20 160 126 603 Ghana 105 69 60 11,679 Guinea 341 162 94 5,014 Guinea-Bissau 49 162 77 573 Lesotho 202 313 162 1,341 Liberia 153 198 120 1,967 Mali 666 259 163 6,940 Mauritania 47 60 31 1,634 Mauritius 4 5 1 959 Morocco 2,517 299 175 20,216 Mozambique 335 78 46 10,473 Reunion 3 33 27 525 Senegal 108 151 132 5,661 Sierra Leone 38 156 145 3,474 Togo 24 123 114 2,625 Uganda 264 117 97 13,201 Upper Volta 83 183 171 6,908 Zaire 538 47 28 28,291 Zambia 150 189 163 5,766 t =b LATIN AMERICA Barbados 0.3 5 4 263 Chile 433 196 157 11,104 Ecuador 64 91 83 8,021 Guatemala 7 162 161 7,262 Haiti 314 122 68 5,809 Honduras 299 195 114 3,691 Jamaica 22 14 4 2,188 Nicaragua 131 172 124 2,733 Panama 110 180 123 1,927 Peru 483 109 81 17,625 Puerto Rico 29 9 1 3,675 Trinidad and Tobago 10 35 26 1,168 1,903 45 Aggregate Per Capita Production Production Population Decline 1948/52 1979/31 1979/31 ( t000 mt) (kgs) ('000) NEAR EAST Cyprus 6 199 184 620 Egypt 420 200 190 41,963 Iraq 1,307 272 172 13,072 Jordan 402 154 30 3,244 Lebanon 133 67 17 2,658 Saudi Arabia 36 36 32 8,960 Syria 36 348 344 8,977 2,340 FAR EAST Bangladesh 1,146 251 238 88,164 Ilong Kong 92 18 0 5,106 Japan 3,875 199 123 116,782 Kampuchea 1,376 366 162 6,747 Korea 577 223 208 38,455 Nepal 2,866 452 250 14,288 Yemen AR 413 208 137 5,812 15,366 WESTERN EUROPE Portugal 452 169 123 9,8336 :977-22 Sources: Food and Agriculture Organization, Production Yearbook, v,-rious years. APPENDIX TABLE 3. GRAINS: DIRECT AND FEED CONSUMPTION, 1978. Grain. ConsumlptIn' (MMt) GNP per Population U[heat & Country Capita ($US) (millions) Rice Coarse Gratns Fecd MARKET ECONOMIES Bangladesh 90 84.7 13.158 1-.622 0 Mall 120 6.3 0.126 1.115 0 Somalia 130 3.7 0.024 0.365 0.290 India 180 643.9 50.389 62.890 2.280 Sri Lanka 190 14.3 1.458 0.722 0 Sierra Leone 210 3.3 0.330 0.036 Niger 220 5.0 1.502 0 Pakistan 230 77.3 2.118 11.524 0 Tanzania 230 16.9 0.199 2.314 0.076 Sudan 320 17.4 3.037 0 Kenya 330 14.7 0.004 2.233 0.085 Senegal 340* 5.4 0.316 0.940 0.025 Indonesia 360 136.0 18.714 4.557 0.645 Egypt 390 39.9 1.377 9.868 1.656 Thailand 490 44.5 8.304 0.432 0.790 Phlilippines 510 45.6 4.280 2.690 1.350 Colonhia 850 25.6 1.112 1.542 0.608 Paraguay 850 2.9 0.045 0.546 0 Syria 930 8.1 0.087 1.989 0.909 Tunisia 950 6.0 1.523 0.270 Halaysla 1,090 13.3 1.457 0.405 0.551 Turkey 1,200 43.1 0.200 13.702 5.373 Nexico 1,290 65.4 0.360 14.702 4.955 Argentina 1,910 26.4 0.120 4.862 6.840 Venezuela 2,910 14.0 0.339 1.818 1.235 Spain 3,470 37.1 0.223 5.691 14.200 UK 5,030 55.8 0.151 10.032 12.243 Japan 7,280 114.9 10.299 8.426 15.805 Australia 7,990 14.2 0.060 2.817 2.900 Canada 9,180 23.5 0.089 5.245 17.325 Uis 9,590 221.9 1.686 38.345 141.673 APPENDIX TABLE 3 (Cont). Grain Constirption" (mit) CNP per Population Uheat & Country Capita ($US) (millions) Rice Coarse Grains - Feed CENTRALLY PLANNED ECONOMIES China 230 952.2 92.218 144.246 0 Cuba 810 9.7 0.499 1.460 0 USSR 3,700 261.0 1.981 100.520 119.177 Czechoslovakia 4,720 15.1 0.070 4.220 7.596 GDR 5,711) 16.7 0.040 4.630 7.672 Sources: Population and income data taken from World Bank, 1980. World Development Report (Baltimore: Johns Hopkins). Grain consumption data are taken from US Department of Agriculture, 1981, Reference Tables on Wheat, Corn and Total Coarse Grains Supply--Distribution for Individual Countries, Foreign Agricilture Circular FG-4-81, and US Department of Agriclture, 1980, Reference Tables on Rice Supply--Utilization for Individual Countries, Foreign Agriculture Circuilar FG-38-80. Notes: a Consumption data for feed includes wheat and coarse grain use only. Rice is not widely used as a livestock feed. Direct grain consumption ---stimates include use for seed and industrial purposes, as well as spoilage and losses. APPENDIX TABLE 4. ESTIMATES OF Wfif PRGflMECN G I FFICIEFfS R CRFAlS, 1975-8° Reg!on/Ckountry Ne--t Protection Coefficients Percentage Sliares of Sources Ikrl d Constrptton, 198 Uieat aize Rice Ieat ttaze Rice - DIWEI )ED tARKMIr £Ux»4ES Austtralia 0.91-1.13 1-1.07 0.65 0.03 4, 5, 9, 10 histria 1.15-1.48 1.37-1.82 0.18 0.24 4, 9, 11 Canada 0.77-0.84 1.03 1.0 1.24 1.45 0.03 4, 5, 1Ø EEC2' l.15-l.)7 1.15-1.72 1.5-1.86 9.97 7.33 0.34 2, 5, 10, 12 Finland 1.28-1.95 0.09 4, 11 Israel 0.80c 1.0 0.15 0.06 9, 10, 11 Japan 1.18-2.90 1.06 2.03-3.70 1.39 3.34 4.06 2, 3, 5, 10, 12 New Zealard 1.0 1.0 0.08 0.04 4, 11 2.65 0.08 0.02 4, 10 Portugal 0.83-0.84 1.02-1.2 0.6-0.7 0.25 0.75 0.07 6, 8, 9, 12 South Africa 1.22-1.58 0.86-1.43 1.0 0.43 1.66 0.05 5, 9, 11 Spain 1.34 1.67 1.12 1.65 5 Suxden 1.38 1.18 1.18 0.16 0.01 0.01 4, 10 Switze:rland 2.69 3.03 0.14 0.08 10 U.S. 1.0 1.0 1.0 5.14 31.37 0.67 10 Suhtotal ti < 0.95 1.72 0 0.07 0.95 < NIC < 1.05 5.87 36.29 0.75 Nt < 1.05 13.48 11.79 4.41 DEVE11NI) CPEs Czechoslovakia 0.6 <1 0.6 1.34 0.31 0.03 9, 10 Ge-nin Wm. P\ep. 0.67 0.74 0.86 0.58 9, 10 Ik ngary 0.50 1.13 1.10 1.43 4, 9, 10 loland 0.65c 1.70 1.73 0.69 10, il Romnia 0.50 0.56 0.64 1.33 2.74 0.03 4, 4 USSR 0.88 0.81 <1 24.1 4.60 0.75 9, 10 Yugoslavia 0.54 0.71 1.30 2.28 APPENDL iÅBJE 4 (Cont). Region/Country Net Protection Coefficients Percentage Shares of Surces SbrId (onstuption, 1980 U ieat Maize Ri ce ljleat tilze Ri.ce Subtotal NRC < 0.95 31.76 10.51 0.81 0.95 < r1C < 1.05 0 0 0 NPC > 1.05 0 2.12 0 DEVE[PM, IVARKET Eakf fITS Afrtca A1lgeria 0.86c 0.71 4 Caxwroon 0.5 0.03 12 (Iana 1.27 1.50 1.13 0.03 0.10 0.03 4 Ivory Coast 0.8 1.0-1.38 0.07 0.20 7, 12 Kenya 1.12-1.29 0.81-0.96 1.30 . 0.06 0.47 0.01 6, 10, 11, 12 LTeria 1.29 0.09 7 Malari 0.51-1.5 0.03 4,12 Hali 0.61 0.05 7 tH-uritius <1 0.03 4 trocco 0.35 1.17 0.77 0.11 11 Nigeria 2.26 1.4-1.6 2.5-2.69 0.32 0.46 0.39 10, 11, 12 Senegal 0.7-1.25 0.16 7, 12 Sierra Leone 1.36 0.14 7 Tanzania 0.72 0.73 0.3 0.02 0.24 0.09 4, 12 'jnisia 1.0 0.41 0.31 0.04 9, 12 Zaire 1.0 1.46 6.69 0.03 0.14 0.07 9 Zambia 1.99 0.6-0.75 0.03 0.14 9, 12 Zhh4x- 0.89 0.55 0.05 0.33 9 Subtotal NP9C < 0.95 1.55 1.29 0.25 0.95 < NIC < 1.05 0.34 0.03 0.16 C M3> 1.05 0.44 0.81 0.79 APPENCL' TABLE 4 (Cont). Region/Country Net Protection Cocfficicnts Percentage Shares of Sources -krld Cnnsturption, 198 lieat Maize Rice eat Wttze Rice lAtin h¥rica Argentina 0.54-0.94 0.49-0.94 0.81-1.1 0.90 0.82 0.05 2, 5, 6 12 lblivia <1 0.07 9 Brazil 0.58-0.64 0.76-0.92 0.6 1.69 5.57 2.56 5, 10, 12 Oile <1.12 1.0-1.1 0.44 0.18 9, 10 colonbla 1.37 1.34 0.9 0.11 0.25 0.45 10, 12 Costa Rca 1.0 1.64-1.73 0.56 0.02 0.02 0.03 9, 10, 11 Ikxinican Repiblic 1.07 1.20 1.0 0.05 0.05 0.10 9, 10 Ecuador 0.68 1.0 0.81 0.07 0.04 0.06 9, 1i El Salvador 1.0 Q 0.12 10 Giatennla 1.45 1.33 0.03 0.30 10 Ouyana 0.84 1.0 0.01 0.004 11 Ibnduras 1.15 0.81 1.0 0.02 0.10 0.02 9 Mexico 0.79 0.86 0.79 3.29 5, 10, 11 Nicaragua 0.88 0.06 10 Panama 1.52 0.72 0.03 0.05 9 Pni 0.46c 1.4 0.20 0.22 9, 10, 11 Uiruguay 1.3 1.0 1.0 0.09 0.03 0.01 9, 1<), l1 Venezuela 1.0 1.14 0.20 0.37 10, 11 Subtotal NPC < 0.95 3.73 9. 4 0.18 0.95 < NiC < 1.05 0.22 0.37 3.15 NPC > 1.05 0.74 1.24 0 Ln APPFNDIX TABIE 4 (Cont). Region/tbuntry rt Protection Coefficients Percentage Shares of 1-hrld Consturption, 1980 1hat thize Rice Meheat titze Rice ikar East Egypt 0.2C-0.57 0.52-0.7 0.37 1.68 0.99 0.56 2, 5, 6, 12 Tran 0.98 1.03 1.46 0.12 10 Jordan 0.70 0.08 4 oenon 0.59 1.0 0.09 0.04 10 Satuli Arabia <1 0.29 4 Syria 0.44c 1.13 1.0 0.47 0.08 0.04 9, 10, 11 Sdahn 1.0 0.13 12 'Ibrkey 0.71--0.9 1.2-1.36 1.5 2.81 0.31 0.10 10, 12 Yeren Ar. Rep. 1.2 0.11 12 Subtotal N[C < 0.95 5.42 0.99 0.56 0.95 < tPC < 1.05 1.59 0.16 0.04 N[C < 1.05 0.11 0.39 0.10 Far East Bangladesh 0.6 0.64-0.7 0.50 5.16 3, 12 Buraa 0.13 2.32 3 ling Kong 1.0 1.0 0.04 0.14 10 India 0.84 0.81 0.52-0.85 7.66 1.43 19.31 3, 5, 10, 12 Indonesia 0.73-0.87 0.84 0.80-0.83 0.31 0.88 7.84 3, 5, 10, 11 Korea (Eep). 1.05-1.8 1.10-1.67 1.75-2.30 0.54 0.67 2.21 1, 3, 5, 12 Malaysia 1.03 1.02 0.10 0.15 10, 11 Pakistan 0.5--0.84 0.94 0.40-0.97 2.54 0.19 0.81 29 3, 5, 6, 10, 12 lld.ippirnes 1.43 0.7-1.0 0.7-1.16 0.16 0.8W . 1.78 3, 10, 12 Singaxore 1.0 1.0 0.02 0.12 11 Sri Lanka 0.98 0.63 3 Taiwan 1.11-1.18 1.06-1.43 1.29 0.16 0.65 0.74 3, 5, 1ø 'i.1ar1 1.22 0.8-1.02. 0.63-0.76 0.05 0.28 3.26 2, 3, 5, 9, 10, 11, 12 APPIEN)IX TABIE 4 (Cont). Iagon/Country Net Protection Cœfficients Percentagt± Shares of Scirces Ibrid Consurption, 190 14Mat Maize Rlce t Maitze Rtce Sub)total NR, < 0.95 11.01 3.40 39.20 0.95 < MIC < 1.05 0.16 0.46 2.55 NIC > 1.05 0.86 1.32 2.95 EVEIOG CPEs ciina 0.72 0.73 0.56 16.00 14.16 35.98 3, 11 IbrId Totals 14-*C < 0.95 71.19 40.19 80.02? 0.95 < NPC < 1.05 8.18 37.31 3.68 ItPC > 1.05 15.68 17.67 8.25 Unwyn4.95 4.83 8.05 APPENDIX TAIFlE 4. (Cont.) Sources: Cnsuntlon shares are estimated from 1980 data provided in United States Departr.yýnt of Agrilture, Forelyn Agriciilture Service, 1980, Reference Tables on Rice Supply-UtilizatIon for Inlivtdual Countrtes, F1-38-80, an 1981, Reference Tables on lAat, Ckorn, and Total Coarse Gratin Supply-Distribition for Individtial- Contries, FG-4-81. Net protection coefficients are drawn from the follUwing sources: 1. Ahn, In-Oin an~ Kym Anderson, 1982, Korean Conparative Advantage, Agriciltural Protection ani DIsti Resouirce Cost of ice Production, 1965 to 1980, mnscrlpt,- Department of Economcs, Research School of Pactfic Stulles, Astralian National University. 2. Bale, Malcolm D. and Ernst bitz, 1979, Price Distortions In Agriailture aMd Thelr Effects: An International C~parison, 1.brll Ikuk Staff lbrking Paper No. 359, Wshington, D.C. 3. Falcon, 1Llter P. and Eric A. 1bnke, 1979-80, International Trade in Ricn, Food Research Institute Studies XVI1(3):279-306. 4. Food andi Agrlctilture Organization, 1975, National Crain Ilblicies, Pane, Italy. 5. Jahara, Cathv L., 1981, Trade Restrictions in International Grain and Otlseæd rarkets: A Conparative Cintry Analysis, Jnited States Departmniit of Agriculture, Economics and Statistics Service, Foreign Agric,-. tural Econonic Report No. 162. 6. lutz, Ernst ail Pasquale Scandizzo, 1980, Price Distortions in Developing Cxntrfes: A Bias Against Agricilture, FAtrmean Neview of Agricultural Econmiics 7(1):5-27. 7. Pearson, Scott R., J.D. Stryker, C.P. 11nphreys and others, 1981, Rice in 11-st Africa: Policy anis Ecojnocs (Stanford, CA: Stanford University Press). 8. Policy and Econnte Studies Team of PROCMÆER, 1982, Comparattve Advantage of Policy (lioices in Portugiese Agrictiltuire, mio, Delpartint of Aricultiral Econonics, thversity of Arizona, 'icson, Arizona. 9. United S ates Departmænt of Agrioilture, Foreign Agricuiture Service, Foreign Agriciiltural Attache Reports, Washington, D.C. 10. lited States Department of Agricultire, Foreign Agriculture Service, 1979, Foreign Agriculture Circular, livat and Corn Prices for Selected Countrtes, FU-6-79, 11. ibid, 1981, FGr21-81. 12. l*)rld Bank, 1982, Wrld Developimnt Report (Naw York: Oxford University Press) APPI)lX TABIE 4. (Cont). tltes: a) Cal.culated as the ratio of dow-stic constuner price to fob export or cif inport price. Tf country-specific cif-fob prices x-re unavailable, prices wre taken from E-brld Lank, Cnmndity Trade and Price Trends, adjusted upward by 20-30 perent to approxiflnte insurance and freight costs. iThe percentage share of each a-xintry in ixrld consirnption for 1980 is providled in pareitheses. b) Protection rates vary for iniividual mimer countries, prinarily due to green rate exciange policies. c) Reflects subsidies on flour or bread. APPENDIX TABLE 5. CRAIN TMW)E FOLICLE.S IN SELECIF.CD XrRIES, 197(). Cintry Comw>dities with Trade NC Different Covernv,nt aiotas, Trade Taxes/ Free Frxn Zero linopoly Licences Subsidies Trade Sources DEFVEIWPEI) NARKßRI (IJwheat, H=j:nize, EGI1fIES R=rice) Austria 144 ll 5 Canada wh5 EEC 1kR 1, 5 Finland W l w Israel w ti Japan WRR w, 5 1nray ý11 Ifi 1, 5 Portugal. R 1-k 1, 5 South Africa 11 WI 5 Spain 1 t 1, 5 Suxden WR VPahR 1, 5 SAtzerland .I1 t4a 1, 5 % of world cons W 1.00 1.71 12.90 0 m 2.43 0.37 9.36 0 R 4.06 0 0.35 0 DEVEWPED CPEs Czechoslovakia WR na C-runn Dem. PCp. ti1 W 5 Hungary VM 11 1,,1 1 Poland lf l4 5 Ronnni.a Rli UR 1, 5 USSR tR 1IR 2, 5 Yugoslavia 1I WI 1, 5 li APPENIDIX TABIE 5 (Cont). Country Conlities ,ith Trade Poll-etes NC Different Goverrmnent Qiotas, Trade Taxes/ Free Fran Zero 1bnopoly llcences Shildies Trade % of wrld cons 14 28.02 2.40 1.10 M 8.61 3.71 3.71 R 0.78 0 0 Af rica Algeria w aroon R ia Q~iana Ut1W I Ivory Oast 3 Kenya 1*R Wn 1, 5 Liberia R R 1, 3 Malawi N(?) 14M tli R R 3 Mlauritis R R tbrocco W i, 5 Nigeria eIR l 1,5 Senegal R(?) R Sierra Leone R R 3 Tainzania NR unisia M M Zaire na Zambia U4 I Zimbalwe IfI na w 1.59 0.32 0.03 M 1.07 0.49 0.56 R 0.47 0.69 0.03 0i APPFIJOTX TAIE 5 (Cont). CouItry cblities with Trade Policies NIC Different Coverrunmt ftiotas, Trade Taxes/ Free Fro Zero Unopoly Licences Subsidles Trade latin Anerica Argentin leDl(?) laR (1974-76) lei (1976-80) 5 Blivia 1-1 Braz L1 UR 5 Gille 1 Colombia WR !rIR M Coxta RIca HR in 5 Difnican Republic M1 4 Ecuador IR W 5 (R na) auitenala INM tI 1, 5 Cuyana iF 5 · lbrxilras tll 4 ,1xico 111 W' 1, 5 Nicaragua fl nj Panam .R Peru 1*1 UI 1, 5 Urugpay W w 1, 5 Venezuela m Mb 5 % of wxrld cons W 2.88 0.37 0.99 0.01 9.80 0 1.44 0 R 0.47 0 0 0 Nar Fast Egypt I- filk 1, 2, 5 Jordan 1ebanon 5 Satki Arabia w 14 Syria II .5 'llikey 1 RR 5 Yeonn Ar. Rep. W m ul APPENDIX TABLE 5 (Cort) CrXIItty CowmxIittes wvith Trade PolCies NIC Different Government Motas, Trade Taxes/ Free Froi Zero bbnopoly Liences Suhsidlies Trada W 5.05 0.08 0 0.29 M 1.38 0 0 0 R 0.66 00 0 Far East ßangladesh ir le 1, 2, 5 Birma R R 2 India %R*R 1, 2, 5 Irdonesia IliR R 2, 5 Korea 1R pR 1, 2, 5 Pak3tan l-R R (high grades) R (lower grades) 1, 2 Philippines UkR(?)i 5 Taiwan 1R t 5 (R na) 'Ihallai l'bR I R 1, 2, 5 % of kirld cons lW 11.01 0.70 0.21 0 M 1.62 2.64 0.93 0 R 35.94 8.06 3.26 0 )EVEIlPIG CPEs China tWt 2, 5 % of irld cons W 16.00 14.16 R 35.98 'lU'LS % !are of 1brld Cons keat 65.55 5.58 15.23 0.30 laize 39.07 7.21 16.00 0 Rice 78.36 8.75 3.64 0 L-n 1 APPEFRlX TABLE 5 (Cont). Sources: 1. Food anrd Agriculture Organization, 1973, 1976, Nbtional Grain Policies 1972, 1975, Rae, Italy. 2. Food and Agriculture Organization, Intragovernmental Group on Rice, 1973, 1977, Qo-peniun of National Rice Trade Iblictes, Rome, Italy. 3. Pearson, Scott R., J.D. Strykey, C.P. iephreys and Others, 1981, Rice in I-bst Africa: Policy anrd Econordes (Stanford, CA: Stanford University Press). 4. United States Department of Agriculture, Foreign Agriculture Service, Foreign Agricultural Attache Reports, Washington, D.C0 5. United States Departuent of Agriculture, Foreign Agriculture Service, 1979, 1981, National "heat and Corn Prices anl humy of tattonal Grain Marketing Systems, Foreign Agriculture Circular, W-6-79, FG-21-81. flotes: na informition not available. a) Variable levy policy b) Government mrketing hoard acts as interediary for comercial firs. c) DXmestic producer ixopoly on wheat flour. LO flit. 60 APPENDIX TABLE 6. RAIN PRICES, 1950-80 (US S/metric ton) Price Deflators 1975=100 Rices faizeb Wheate Ad Be 1950 136.7 68.1 66.9 51 144.4 72.0 72.4 52 156.3 62.6 72.0 53 174.7 60.2 68.0 54 157.9 58.3 62.8 55 141.5 48.8 61.7 56 136.9 51.6 62.8 57 137.2 47.6 63.2 58 142.3 47.6 61.4 59 132.2 46.1 58.1 60 124.7 43.3 59.2 43.9 61 136.5 45.9 59.5 44.2 47 62 152.8 51.4 59.9 43.7 43 63 143.3 54.7 59.2 43.9 46 64 137.7 55.8 63.6 44.5 48 65 136.3 55.0 58.1 46.0 47 66 163.2 59.4 62.1 46.4 46 67 205.8 49.9 61.7 47.1 45 68 201.6 49.1 58.4 44.1 46 69 186.9 53.9 56.2 44.5 46 70 144.0 58.4 57.0 49.1 49 71 129.0 58.4 62.1 53.1 50 72 147.0 56.0 69.1 58.4 53 73 350.0 98.0 136.8 59.8 70 74 542.0 132.0 178.0 86.9 100 75 363.1 119.6 138.4 100.0 100 76 254.5 112.4 122.7 101.9 104 77 272.2 95.3 95.5 109.4 135 78 367.5 100.7 124.9 125.9 129 79 334.2 115.5 156.3 142.7 139 80 434.0 126.0 1683.0 154 Sourdes: World Bank Commodity Price Trends; FAO, Trade Yearbook. Notes: a Thai 5% brokens, fob Bangkok b US No. 2 yellow, fob Gulf ports c US No. 1, soft red winter, fob Atlantic, prior to 1974, No. 2 d Index of US Dollar unit values (cif) of manufactured exports to developing countries e Index of US dollar unit values (fob) of agricultural product exports from developing countries APPENDI-X 'IAßE 7. PAT,ýO OF C-RFA i 'RT 0TST 'I 7I'A. (1DDTY FXPRT EARNTI,S RA SEIF.0E ) iNE fm O(JN'RIES, 1970--80. Year Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 JIM8 Av(rage Africa Algeria .034 f)61 .06 .056 .071 .094 .065 .059 .069 .059 .053 .062 Cameroon .037 o037 .035 .025 .04 .036 .029 .034 .033 .027 .024 .032 Congo .058 .064 .053 .053 .062 .089 .091 .055 .09 .03 NA .065 Ethiopla .048 .032 .006 .014 .004 .J03 .012 .071 .06 .03 .162 04l .038 .020 .022 .036 .055 .025 .049 .062 .062 K NA .041 Ivory Coast .033 .024 .028 .045 .037 .012 .014 .029 0.32 .040 NA .029 Liberia .046 .,,36 .028 .042 .043 .042 .037 .051 .053 .059 .065 046 Libya .010 .011 .X)9 .011 .019 .025 .017 .013 .015 .012 i1.008 04 .adagascar .032 .064 .054 .029 .168 .117 .127 .082 .135 .135 A .014 Knurania .086 .104 .086 .077 .101 .145 .143 .155 .226 .194 .184 .136 Mautius .165 .156 .128 .115 .177 .129 .176 .095 .118 .107 .129 .136 I-brocco .060 .112 .056 .14) .109 .180 .163 .131 .158 .145 .143 .127 Nigeria .021 .020 .020 .016 .009 .013 .021 .033 .068 .024 .019 .024 Senegal .139 .252 .125 .323 .237 .113 .147 .153 .223 .163 .223 1<1 Tanzania .024 .017 040 .023 .2695 .291 .030 .064 .064 .032 .2ø. .097 'Ibnisia .162 .091 .076 .086 .059 .062 .064 .089 .094 .082 .079 06 Upper Volta .173 .341 .139 .336 .310 .149 .153 .173 .419 .186 .302 .244 Z,aire .025 .026 .024 .022 .043 .081 .096 .068 .089 .044 055 .052 Zambia .009 .049 .017 .018 .010 .026 .021 .015 .019 ,028 074 .026 A1PF-MIX TAB. 7 (Cont). Year Country 19710 1971 1972 1973 1974 1975 1976 - 1977 1971 1979 1980 Average Latin A.-rica ßolivia .082 .081 .077 .066 .072 .089 .044 .059 .063 .072 .0)m9 .073 Brazil .059 .047 .039 .051 .069 .044 .056 .026 .064 .078 .1 Y) Chile .039 .060 .061 .083 .135 .088 .106 .035 .069 .057 .066 Colombia .032 .057 .038 .078 .073 .041 .054 .G1 NA l NA 651 Costa luca .039 .049 .033 .048 .056 .050 .046 .018 .021 .025 .031) 6039 Doninican Repuiblic .016 .037 .037 .071 .110 .064 .086 .062 .062 .039 .068 .059 Ectador .035 .044 .034 .034 .030 .056 .040 .031 .030 .030 .030 037 El Salvador .017 .023 .015 .051 .037 .028 .033 .021 .029 .02) .028 .027 Cuatetikla .026 .020 .019 .034 .039 .042 .038 .015 .025 .024 .030 l028 (uyana .030 .029 .030 .055 .044 .035 .063 .040 .029 .040 .041 .0410 HIaiti .107 .107 .123 .219 .229 .239 .226 .246 .201 .204 N .190 lknduras .029 .024 .024 .023 .037 .078 .045 .023 .032 .033 .045 .036 Jamaica .076 .079 .086 .119 .131 .106 .123 .070 .0!9 .098 .106 .098 Ixico .046 .012 .048 .200 .170 .206 .036 .072 .061 .057 .072 .089 Panana .026 .079 .046 .070 .034 .043 .045 .044 .034 .045 .064 .048 Paraguay .070 .059 .043 .080 .066 .026 .051 .020 .019 .028 .036 .045 Peru .040 .061 .064 .104 .097 .141 .119 .094 .076 .058 NAi .085 Sirinanæ .023 .014 .014 .025 .023 .023 .0N9 .025 .018 .019 .015 .020 Trinidad etc. .030 .030 .029 .031 .024 .029 .024 .019 .019 .020 .015 .025 Uniuay A .039 .042 .056 .0005 .002 .004 .002 .027 .047 .007 .022 Venezuela .023 .022 .022 .021 .022 .021 .032 .032 .027 .021 .026 .024 CN AP1PFfI)LX TIAL. 7 (Cont). Year Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 Average tkbar East Cyprus .106 .064 .077 .289 .181 .135 .114 .0)2 .098 .104 .105 .124 Egypt .02 .206 .144 .153 .442 .523 .371 .399 .466 .421 .374 .326 Iran .JX2 .027 .030 .025 .024 .027 .017 .020 .028 .024 .058 .026 Iriq .009 .099 .020 .045 .040 .027 .022 .029 .039 .029 .024 .035 Jordan .455 .517 .495 .373 .243 .210 .325 .295 .250 .275 .176 .32) Kuwait .010 .010 .009 .o8 .004 .9 .007 .005 .X7 .5 7 MV07 leinon .200 .150 .091 .097 .046 A .117 NA NA NJA Un .013 .012 .031 .019 .026 .016 .007 .015 Saudi Arabia .029 .020 .015 .010 .005 .007 .A7 .06 .012 .011 010 012 Sxmia .188 .209 .254 .277 .221 .356 .232 .480 .390 .467 .576 .332 Sudan .057 .047 .045 .038 .079 .053 .064 .02() .m10 .132 .145 *079 Syria .217 .331 .100 .085 .128 .084 .051 .077 .07() .065 07 120 UA Fmirates N AN N .001 .JX2 .M2 .004 ."6 .A) .05 004 Yein AR 2.852 1.874 2.205 1.826 2.734 5.262 12.577 3.623 11.143 5.065 VA 4.898 Yemen DR .132 .081 .101 .136 .175 .138 .135 .153 .271 .158 må .148 Far Fast Afghaistan .151 .364 .165 .059 NA .062 .007 .026 .071 .052 .028 099 Aangladesh NA NA .610 .819 1.114 1.543 .671 .185 .474 .267 .815 .722 India .173 .119 .024 .176 .213 .296 .19)7 .034 .017 .020 .M4 1Ji Indonesia .169 .141 .114 .177 .062 .056 .063 .067 .059 .045 .040 04 Korea RIP .292 .283 .174 .126 .137 .135 .055 .082 .038 .05) 061 .130 PakLstan .028 .033 .079 .112 .146 .212 .155 .o56 .092 .172 .040 .102 (hAppines .043 .067 .085 .061 .052 .069 .056 .033 .02t] .025 .033 .050 Singapore .041 .032 .036 .030 .029 .027 .Oi<)5 .019 .017 .016 .014 .026 Srilanka .290 .209 .192 .28<9 .484 .559 .301 .311 .221 .187 .174 .292 'Ihatland .00X .007 .007 .007 .009 .007 .091 .004 .006 .009 .X5 .014 Sources: Dýita for the cost of cereal iqports are taken fmm Fond and A,ri.culture Organtzation Trade Yearboolc, variousYwi Condity exports earnings data are taken from International- Monetary url, ,nternational Finaniich-dtitstts Yearinok, . .