Groundnut Policies, Global Trade Dynamics and the Impact of Trade Liberalization Ndiame Diop, John Beghin, and Mirvat Sewadeh* World Bank World Bank Policy Research Working Paper 3226, March 2004 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. *N. Diop is an economist with the Trade Department, J. Beghin is a professor of economics at Iowa State University, and M. Sewadeh is a consultant with the Trade Department. The authors thank Ataman Aksoy, John Baffes, Harry de Gorter, Eric Dohlman, Fred Gale, Cheng Fang, Ashok Gulati, Dinghuan Hu, Steve Jaffee, Holger Matthey, Don Mitchell, Gary Pursell, and Funing Zhong for comments and information. 1 Abstract: Groundnut products are of central economic importance to millions of smallholders in Africa, India and Southern China. The products generate 60 percent of rural cash income and account for about 70 percent of the rural labor force in Senegal and Gambia. Groundnut trade remains, however, heavily distorted, and this has affected the competitive position of various players in world markets. Using a new partial-equilibrium, multi-market, international model, we analyze the trade and welfare impacts of several groundnut trade liberalization scenarios compared with the recent historical baseline. Net welfare is evaluated as the sum of consumers' equivalent variation, quasi-profits in farming, quasi-profits in crushing, and taxpayers' revenues and outlays implied by distortions. We find that trade liberalization in groundnut markets has a strong South-South dimension with policies in India, and to a lesser extent China, heavily depressing the world prices of groundnuts at the expense of smaller developing countries mainly located in Africa. Under free trade, African exporters would gain because they are net sellers of groundnut products. In India, consumers would be better off with lower consumer prices resulting from the removal of prohibitive tariffs and large imports of groundnut products. The cost of adjustment would fall on Indian farmers and crushers. In China, crush margins would improve because of the large terms of trade effects in the oil market relative to the seed market. China's groundnut product exports would expand dramatically. Net buyers of groundnut products in OECD countries would be worse off. We draw implications for the Doha negotiations. Keywords: Doha, groundnut, peanut, oil, trade liberalization, protection, distortion, negotiations 2 Table of Contents I. INTRODUCTION............................................................................................................................................. 4 II. GROUNDNUTPRODUCTIONANDTRADE.................................................................................................... 5 Groundnut Production, Cost, and Utilization..........................................................................................................................5 TABLE2. ECONOMICCOSTSANDRETURNSINGROUNDNUTS, CHINAANDTHEU.S.................................... 7 Global Trade and Market Shares Dynamics...........................................................................................................................8 Groundnuts International Prices............................................................................................................................................ 13 III. DOMESTICGROUNDNUT POLICIESOFMAJORCOUNTRIESINWORLDMARKETS ...............................14 Groundnut Policies in the United States............................................................................................................................... 14 Groundnut Policies in India, China, and Argentina............................................................................................................. 17 Groundnut Policies of Key African Exporters..................................................................................................................... 19 Groundnut Trade Policies of High-income Importers......................................................................................................... 21 IV. IMPACT OF GROUNDNUT PRODUCT POLICY REFORMS ON WORLD PRICES, TRADE FLOWS AND WELFARE.......................................................................................................................................................22 Policy Reform Scenarios....................................................................................................................................................... 22 Results..................................................................................................................................................................................... 22 V. POLICYIMPLICATIONS ANDCONCLUSIONS.............................................................................................31 REFERENCES..................................................................................................................................................33 ANNEX. ..........................................................................................................................................................34 I- Unit Root and Co-Integration Tests for Edible and Groundnut Oil International Prices.............................................. 34 II- Granger-Causality Test..................................................................................................................................................... 35 Figure A1. World Groundnut Oil Prices versus other Vegetable oils................................................................................ 35 *N. Diop is an economist with the Trade Department, J beghin is a professor of economics at Iowa State University, and M. Sewadeh is a consultant with the Trade Department. The authors thank Ataman Aksoy, John Baffes, Harry de Gorter, Eric Dohlman, Fred Gale, Cheng Fang, Ashok Gulati, Dinghuan Hu, Steve Jaffee, Holger Matthey, Don Mitchell, Gary Pursell, and Funing Zhong for comments and information. 3 I. INTRODUCTION World production of groundnuts stood at 34 million tons in 2001, accounting for 10 percent of the world's total oilseed production of 324 million tons1. China is the world's largest groundnut producer, with 40 percent of world's production, followed by India (23 percent), a group of Sub-Saharan African (SSA) countries (8.4 percent) and the United States (5.6 percent). Groundnuts provide a valuable source of protein, fats, energy, and minerals, and generate cash income to many poor farmers in the developing world, especially in SSA and Asia. In Senegal for instance, groundnut account for 70 percent of the rural labor force and 60 percent of households' agricultural income. Groundnut production and processing represent about 2 percent of GDP and 9 percent of exports in that country. Due to generally high level of self-consumption, international trade in groundnuts is thin: only 5 percent of world production is sold in the international markets. Of the three major groundnut products traded internationally (edible groundnuts, groundnut oil and groundnut meal), edible groundnuts are the most traded, with a volume of 1.2 million tons in 2001, against 250,000 tons for groundnut oil. Further, the different groundnut products face different export dynamics in international markets. While global export of edible groundnuts increased by 2.2 percent over the last 20 years, exports of groundnut oil and meal declined by 1 and 2.5 percent annually despite growing global consumption of these two products. China is the world largest exporter of groundnut, with 32 percent of world edible groundnut exports, followed by the US (19 percent) and Argentina (10.5 percent). Sub-Saharan Africa (Senegal, Gambia, Nigeria, Malawi, South Africa and Sudan) has lost most ground in world edible groundnut markets, and collectively accounts for only 5 percent of the world market. In the groundnut oil market segment however, Senegal is the largest supplier, but this market has become all the more thinner as other vegetable oils are increasingly used as substitutes of groundnut oil. Since the mid 1990s, all major exporters have been gradually liberalizing their groundnut sectors, in part to fulfill their commitments under WTO agreements. However, the results are mixed and trade in groundnut products remains heavily distorted. Both China and India have removed some import restrictions and have allowed wider private sector participation in importing groundnuts. Still, tariffs on groundnut products remain very high in India and to a lesser extent in China. The market size of both countries exacerbates these distortions and their effects on the world market. In the US, the 2002 Farm Bill has suppressed many unsustainable features of the previous groundnut policies (e.g., the high support price and production quotas) but has introduced new distortions, such as counter-cyclical payments and the floor price mechanism. These policies would subsidize U.S. groundnut exports in situations of low world prices with a potential to depress world market prices. Argentina still selectively subsidizes some processed groundnut products and exports and applies moderate export taxes on groundnuts. In SSA the heavy producer taxation of earlier government intervention has ended, and unilateral liberalization efforts are continuing. Protection of processing remains significant however. 1Groundnuts are also known as peanuts, earthnuts, goobers, pinders, and Manila nuts. The groundnut plant is a hairy, tap rooted, annual legume that measures 1 to 1.5 feet in height. Shelled groundnuts are basically used as seed, consumed as raw edible groundnuts or after transformation into "prepared" groundnuts (roasted, salted, flavored, etc.) or into groundnut butter/paste. They can also be crushed for oil and a by-product, groundnut meal (animal feed). Groundnut oil is an excellent quality cooking oil with a high smoke point (440 degree F) and neutral flavor and odor. It allows food to cook quickly with a crisp aspect and with little absorption. 4 The current situation raises many important questions about the future of the sector and prospects for various players. How will multilateral groundnut trade liberalization affect the competitive positions of the various players? Which countries are likely to gain and capture larger market shares and which ones will lose? How will small SSA producers be affected? We address these questions in this paper. The next section examines the dynamics of global production, trade and prices of groundnut products. The third section reviews the groundnut policies of the major players in the groundnut market, with an emphasis on the most distortive policies. In the fourth section, we use a multi-market, partial-equilibrium model to assess the impact of removing major policy distortions on world prices, trade flows, and welfare for major producing and trading countries involved in the groundnut market. We analyze multiple scenarios. First, we consider full multilateral trade liberalization for groundnut products with and without the removal of the U.S. groundnut program. Then we consider multilateral groundnut trade liberalization, again with and without the removal of the U.S. farm groundnut program. Then we consider full trade liberalization in the two largest and most distorted groundnut markets, India and China. The last section derives the major policy implications of the study and concludes. II. GROUNDNUT PRODUCTION AND TRADE Groundnut Production, Cost, and Utilization Groundnuts belong to the oilseeds category and account for 16 percent of world oilseed harvested area. The major producers are China, with 40 percent of world's production, followed by India (23 percent), Sub-Saharan Africa (8.4 percent), the United States (5.6 percent) and, to a lesser degree, Japan (Table 1). Table 1. Groundnut production and utilization, average 1996-01 Area Dom estic Net Country harvested Yield Production Utilization Food use Crushed Feed/seed Export (1000 ha) (m t/ha) (1000 m t) (1000 m t) (1000 m t) (1000 m t) (1000 m t) (1000 m t) W orld 21452 1.4 29997 29896 12416 14590 2891 169 M ain producers and exporters China 4234 2.9 12204 11777 4753 6140 884 427 India 7902 0.9 7176 7082 534 5581 967 94 USA 569 3.0 1701 1428 978 280 170 220 Argentina 280 1.5 403 191 21 155 15 213 M ain producers in Africa Nigeria 1187 1.1 1340 1340 636 427 278 0 Senegal 690 1.1 722 730 317 304 109 -6 South Africa 98 1.7 161 123 72 32 19 33 Malawi 117 0.9 103 101 78 18 5 2 Gam bia 89 1.0 95 80 26 54 0 15 M ain importers EU 1 1.0 1 454 433 17 5 -449 Canada 0 0.0 0 115 115 0 0 -115 Japan 12 2.3 28 129 121 2 6 -103 south Korea 7 2.2 15 30 30 0 0 -15 Source: USDA Note: The difference between production + net exports and dom estic utilization reflects stock variation, not shown here Ending stocks are negligible for all country except the US fo which it stood at 28 percent of total production during 1996-01 Groundnut production conditions vary considerably across countries, reflecting 5 differences in technological development, access to modern inputs and irrigation, and farm management practices. Table 1 shows large differences in yields across major producing countries. Groundnut yields are highest in the United States and China, and lowest in SSA (except South Africa) and India. Yet between the US and China, important differences exist in farm practices. While in the U.S. production is highly mechanized, the high yields in China reflect high labor-intensive farming practices in small plots of land and intensive use of draft animal (Colby et al. 1992). The low yields in Africa and India are the result of limited use of modern inputs including high-yielding seed varieties and high dependence of production upon rainfall. Driven by a tremendous growth in China, global production of groundnuts exhibited a strong expansion between 1981 and 2000. Shelled groundnut production grew by 34 percent between 1981-85 and 1996-00. Growth in production has however been uneven across countries. As Figure 1 shows, production growth in China has been impressive, following Chinese 1978 market reforms2. This country doubled its production between 1992 and 2000. Groundnut production in India grew up from 1987 to 1998, before dropping to its 1970s level of 6 million tons in recent years. Indian production exhibited however important fluctuations throughout this whole period. Production in SSA picked up in the early 1990s, after a long period of decline. Groundnut production in the United States and Argentina stagnated around 2 million and 300,000 metric tons respectively over the last three decades. Figure 1. Trends in Groundnut Production Among the World's Main Porducers (million metric tons) 16 14 12 10 8 6 4 2 0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 China SSA Argentina India USA Source: Oil world. Table 2 compares the economic cost per acre between the U.S. and China, the two countries of comparable yields and that account for about half of the world groundnut production 2 China took also advantage of increased use of high yield varieties (HYV) and agricultural inputs (fertilizers, pesticides, insecticides, mechanization, and irrigation) to boost yields and production (Colby et al. 1992). 6 and 50 percent of world's edible groundnut export in 2001. Due to lack of more recent data, the comparison is done for the years 1992 and 1993. As Table 2 shows, economic costs per acre were more than three times higher in the US than in China in the early 1990s. Cost per acre in groundnut production stood at $694.03 on average for the United States, more than three times higher than the average cost for China, $164.45 per acre, in 1993. The larger economic costs per acre for U.S. groundnuts were chiefly attributed to production quota rent, land value, and "other expenses" (i.e., costs of using and maintaining farm equipment, cost of fuel, electricity, repair, capital replacement). Quota rent and land value are not an effective cost for farmers in China, since there is no production quota as such in China, and land is considered as a public property, belonging to local communities organized in groups of 30-40 households (Chen et al. 1997)3. Net returns between the U.S. and China were not significantly different if one excludes quota rent (irrelevant since the 2002 Farm Bill) in US production costs. US cost disadvantage is however compensated by higher producer prices brought about by the groundnut program and higher US quality groundnuts. The elimination of quota production (thus quota rent) in the 2002 Farm Bill reduces further US production costs. This development as well as the reputation of high quality groundnut producer warranting a high price premium in international market may well uphold US future competitiveness vis-à-vis China4. Table 2. Economic costs and returns in groundnuts, China and the U.S. (US$/acre, 1992 and 1993) 1992 1993 Item US China US China Variable costs: Seed 70.32 43.83 71.18 45.96 Fertilizer 43.27 25.03 42.40 26.13 Chem icals 89.70 3.40 92.57 3.68 Labor 89.14 71.51 86.17 75.86 Other expenses 212.84 41.43 188.54 12.82 Subtotal 505.27 185.20 480.86 164.45 Fixed costs: Land value 92.58 97.77 Quota rent 113.38 115.40 Total Costs: 711.23 185.20 694.03 164.45 Yield (lb/acre) 2576 2520 1940 2135 Revenue (producer price*yield) 753.66 323.69 570.58 280.83 Source: Chen et al. (1997) There has been a gradual increase in the use of groundnuts for food purposes, and a resulting decline in the share of groundnut production crushed for oil and meal over the last 20 years (Figure 2). The largest increase in food use occurred in developed countries and fast 3Any growers in the group are eligible to farm a certain number of acres of land. Farmers who use the land are obligated, however, to pay agricultural tax in kind and sell a certain amount of their products to the state government at regulated prices (Chen et. 1997). 4Export markets display relatively high quality premiums/discounts depending upon quality reputation of exporters. Prices of groundnuts originated from the US set in fact a ceiling for international prices because the US export high quality groundnuts which command a relative high price premium in international markets. In 2000, edible U.S. groundnuts commanded a 40 percent premium on world markets compared to shelled Chinese groundnuts, according to FAO trade data. 7 growing economies in Asia5. In contrast, producers in India and most of SSA still devote more than 60 percent of their production to crushing for oil and meal. The overall declining use of groundnut for oil and meal reflects the increasing availability of cheaper, low-fat vegetable oils such as soybeans and rapeseeds, as substitutes. Similarly groundnut meal competes with meal from other oilseeds and with cereal-based products such as maize gluten. Figure 2. Utilisation of Groundnut (Percentage of World Production) 70 60 50 entcreP40 Food Oil 30 Feed 20 10 0 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 Source: Oil World. Global Trade and Market Shares Dynamics Among the different groundnut products, shelled edible groundnuts, prepared groundnuts and groundnut oil are the most traded products in international markets. Table 3 shows the trade balance of each of these products for the major countries involved in groundnut trading in groundnut markets in 1996-2000. Countries with net exports over US$100 million are China, Argentina, and the U.S. The bulk of these countries' exports is constituted of edible groundnuts. India's trade balance reached US$94 million while Senegal, the largest African net exporter in the period registered a US$53 million groundnut trade surplus. Ninety two percent of Senegalese net export was groundnut oil. The EU, Canada and Japan are the major edible groundnut markets. The EU is the world's largest groundnuts market, accounting for 43 percent of world groundnut imports. EU's total value of net groundnut imports was just below US$500 million in 1996-2000. The other important groundnuts markets include Canada (9 percent of world imports) and Japan (8.2 percent). 5 8 Table 3. Value of net exports, by groundnut product, 1996-2000 Value of Net Exports in Million US $ Edible Groundnut Groundnut Oil Prep.Groundnut Total China 193.79 2.82 111.06 307.68 India 86.85 -0.13 7.27 93.99 USA 126.43 -12.77 28.26 141.92 Argentina 160.98 51.52 25.82 238.32 Nigeria -3.29 4.64 0.00 1.35 Senegal 3.34 48.99 0.60 52.92 South Africa 16.01 4.68 0.27 20.95 Malawi 0.77 0.00 0.00 0.77 Gambia 4.49 1.09 0.05 5.63 European Union (15) -378.47 -115.12 -4.54 -498.13 Canada -76.67 -1.19 -3.31 -81.18 Japan -44.00 -1.85 -71.46 -117.31 Korea, Republic of -4.55 0.01 -14.31 -18.86 Source: FAOSTA, 2002 Note: Prepared groundnut = roasted, salted, flavored groundnuts. Peanut butter not included. Figure 3 and 4 show the structure and trend of EU's edible groundnut (Figure 3) and groundnut oil (Figure 4) imports. Competition among exporters in this market has increased. The US has lost share in the EU edible groundnuts market between 1996 and 2001 to Argentina, which has become the top exporter of that particular product in that market (Figure 3). Figure 3. Competition in the EU edible groundnut market EU Imports of Edible Groundnut by EU Imports of Edible Groundnut by Origin in Origin in 1996 2001 ROW ROW USA 14% 15% 18% SSA USA SSA 2% 28% 5% India India 5% 3% China Argentina 26% China 32% Argentina 27% 25% Source: Oil World Imports of groundnut oil are much more concentrated than that of edible groundnut, but here also, the EU represents a sizable block, with more than 60 percent of world imports for the 1996-2000 period. Demand from the EU has declined overtime as cheaper vegetable oils were increasingly substituted for groundnut oil. As Figure 6 shows, even as EU's demand for groundnut oils has increased substantially, groundnut oil lagged behind that of palm and rapeseed. 9 Figure 4. EU's import of groundnut oil versus palm and rapeseed oils 4000 3500 Palm oil Groundnut oil Rapeseed oil 3000 2500 2000 1500 1000 500 0 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 Figures 5a to 5f depict the trends in exports and market shares of raw edible groundnuts prepared groundnuts and groundnut oil since 1976. Consistent with growth in world consumption of raw edible and prepared groundnuts, export of these two products expanded rapidly since the mid-1980s. Exports of edible groundnut increased by 8 percent between the first and second half of 1990s. This growth followed a dramatic increase of over 20 percent over the 1980s. The pattern of growth in prepared groundnut export broadly mirrors that of edible groundnuts signaling the highly integrated nature of these markets as shown in Figure 5c. China has been the major beneficiary in the expansion of the groundnut markets while the U.S. role has decreased significantly. From barely 1 percent in 1976, China's market share in the world edible groundnut market increased dramatically to 32 percent in 2001. During the same period, US' market share dropped from 32 percent to 19 percent (Figure 5b). The emergence of China as a leading groundnut exporter is even more impressive in the prepared groundnut market. Other producers have increased their share from about 20 percent to 35 percent (Figure 5d), also contributing to the erosion of the U.S. export market share. While the international edible groundnut market has become more concentrated with 61 percent of exports controlled by China, the US and Argentina in 2001, the market for prepared groundnuts has become more fragmented. The concentration of the edible market partially reflects the significant decrease in SSA's share, from 17 percent to 5 percent. African shares have been quite volatile, as several African countries (including Nigeria, Malawi, and the Gambia) enter intermittently the edible export market depending on their crop quality and world market demand, and are not dependable suppliers in this market. Over the last 25 years, many countries have exited the groundnut oil market (e.g., Brazil) or have chosen to enter it only when the quality of groundnut harvested cannot be sold in the edible market (e.g., United States). In this rather stagnant groundnut oil world market, Argentina and Senegal remain the world's leading exporters but the market has significantly fragmented. In 1976, Senegal, Argentina, Brazil, and the United States supplied 85 percent of total exports while in 2001, these four countries represented only 52 percent of total exports. China and Brazil experienced sharp decreases in shares as they elected to exit the groundnut oil market and shift emphasis on edible groundnut and other vegetable oils. 10 Figure 5. The world groundnut markets: exports by product and market shares Figure 5a. World export and "consumption" of raw edible Figure 5b. Market shares in world raw edible groundnuts groundnuts (1000 metric tons) World Consumption World Export1800 0.90 0.80 1600 0.70 36000 0.60 1400 0.50 Consumption 32000 0.40 1200 0.30 28000 Export 1000 0.20 24000 0.10 800 0.00 20000 16000 76 78 80 82 84 86 88 90 92 94 96 98 00 02 Argentina China USA Sub Sah. Africa World Edible Export World Groundnut "Consumption" Figure 5c. World export of prepared groundnuts (tons) Figure 5d. Market shares in world prepared groundnut 350000 1.00 0.90 300000 0.80 250000 0.70 0.60 200000 0.50 0.40 150000 0.30 0.20 100000 0.10 0.00 50000 0 USA China Netherland 76 78 80 82 84 86 88 90 92 94 96 98 00 02 World Prepared Peanut (roasted, salted, etc) Exports 11 Figure 5e. World export of groundnut oil (1000 metric Figure 5f. Market shares in world groundnut oil tons) 550 0.90 Post NAFTA and URAA 0.80 500 0.70 450 0.60 0.50 400 0.40 350 0.30 0.20 300 0.10 250 0.00 200 Senegal Argentina USA Brazil 150 76 78 80 82 84 86 88 90 92 94 96 98 00 02 World Groundnut Oil Export The decline in African country shares in world groundnut markets has significantly reduced the contribution of groundnut products in export earnings of many countries. As Table 4 shows, the importance of groundnut products as a source of export earnings has declined dramatically in Senegal and South Africa and in Malawi since the early 1980's. It has only increased in Gambia significantly in Gambia where it accounted for 84 percent of total good exports in 2000-02. Table 4 also depicts the changes in the volume and value of groundnut exports, for shelled (raw edible) groundnuts and groundnut oil. The volume exported of raw edible groundnuts decreased significantly in Malawi, Nigeria and South Africa and quasi-stagnated in Senegal and Gambia. As a result of declining and quasi-stagnant volumes, shelled groundnut export earnings dwindled, but the extent of the decline suggests that unit values have decreased for Malawi, Gambia and Senegal. This is confirmed in Figure 6. The export performance of groundnut oil is more contrasting. Senegal and Nigeria have increased both their volume and value of exports while South Africa significantly exited the groundnut oil market. Table 4. Share of groundnut products in total merchandise exports Nigeria South Africa Senegal Gambia Malawi Shelled gr. export volume 1980-82 1,026 41,333 2,725 27,333 14,867 (metric ton) 2000-02 412 34,830 2,915 27,939 662 Shelled groundnut exports 1980-82 400 29,730 2,145 11,743 12,333 (1000 US$) 2000-02 204 22,875 1,371 5,763 436 Gr. oil export volume 1980-82 0 22,667 82,693 7,651 0 (metric ton) 2000-02 1,287 1,519 98,879 8,633 0 Groundnut oil exports 1980-82 0 14,071 60,285 6,400 0 (1000 US$) 2000-02 797 1,053 63,007 6,333 0 Share of gr. In total exports 1980-82 0.003 0.21 16.17 59.62 4.65 (%) 2000-02 0.006 0.08 8.16 84.64 0.10 Source: Production and groundnut exports data, FAOSTAT; Total good exports, WDI Note: The share of groundnuts in total good exports inlude export of groundnut meal 12 The evolution of Africa's edible groundnut exports prices has not been uniform across major African exporters. As Figure 6 shows, despite a decline in recent years, prices for South African groundnuts have held up on average at a high level since the mid-1980's. In contrast, prices sharply declined for Gambia and, to a lesser degree, Senegal. The discount on the latter two countries' prices reflect their lower groundnuts quality and stricter EU quality and technical standards. The EU has indeed become more demanding, both from a public health perspective (permissible level of aflatoxin)6, and from a technical standpoint (size, uniformity). Figure 6. Unit value of raw edible groundnut of African origin (US$/ton) 1200 Gambia 1000 Senegal South Africa 800 600 400 200 0 Groundnuts International Prices International prices of edible groundnuts and groundnut oil in Rotterdam market (the reference for groundnut trade) exhibit two distinct patterns since 1970 (Figure 7). First, during 1970-81, both prices were non-stationary, drifting away, and following an increasing trend. Co- integration tests show no co-integration between edible groundnut and groundnut oil during this period (Annex Tables 1 to 3). Prices were high and the world market was dominated by the US and SSA which supplied respectively 45 and 18 percent of world exports. China exported no edible groundnut or groundnut oil. Second, in sharp contrast with the 1970s, groundnut prices over the last 20 years have been stationary, constantly reverting to their mean values following shocks (Annex Tables 4&5). Price fluctuations were thus important but two sub-periods can be distinguished. Prior to 1994, prices of groundnuts displayed a higher level of volatility. The coefficient of variation of prices stood at 20 percent between 1980 and 1994 against only 7 percent between 1995 and 2001. This shift in world price variability warrants at least two questions. First, what are the main causes? Second, is the change in price variability permanent? 6 Aflatoxin is a cancer-causing chemical produced by species of aspergillus moulds that can contaminate groundnuts. The spores of these moulds, present anywhere in the air and the soil, require specific temperature, moisture and nutrient substrates to germinate. Aflatoxin contamination of groundnuts can occur during its cultivation in the field, as well as during harvesting, post-harvesting, storage and processing. While aflatoxin disappears with crushing, they tend to be present in groundnut meal and edible groundnuts. 13 Revoredo and Fletcher (2002) recently addressed theses specific questions. They analyze both production instability (originated in producer-exporting countries) and consumption instability (originated in consumer-importing countries) and find that steady expansion of Chinese exports, which are negatively correlated with exports from US and Argentina, have been a stabilizing force in the second half of the 1990s. This occurs in spite of the fact that India, Argentina, and South Africa now transmit a higher proportion of their supply shocks to the world market (Revoredo and Fletcher 2002). It seems therefore that substitution between Chinese and US groundnuts has increased in recent years, but detailed data on substitution in world markets to substantiate this inference is as yet inconclusive. Regarding the groundnut oil market, the influence of Senegal on world prices remains significant. In 2000/01, Senegal exported about 100,000 metric tons of groundnut oil, representing one-third of the world's export and more than 60 percent of demand from EU, Senegal's main export market. While 2000/01 was an exceptional production year, Granger causality tests strongly indicate that variations in Senegal's exports Granger-caused international prices and the reverse was not true (Table A6). Figure 7. International prices of groundnuts, Rotterdam, CIF, $/metric ton 1400 Prices are Non Stationary Period of Stationary Prices and Not Integrated (tendency to revert to their mean values) 1200 Edible groundnut 1000 800 Groundnut oil 600 400 200 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 Edible Groundnut, US Shelled Runner 40/50 (CIF Rotterdam) Groundnut Oil, any origin (CIF Rotterdam) Finally as Figure A1 (Annex) shows, groundnut oil markets are broadly integrated with that of other vegetable oils (soy oil, rapeseed oil, palm oil, and sunflower oil), however, the integration between oilseeds markets other than groundnuts seems to be much stronger. III. DOMESTIC GROUNDNUT POLICIES OF MAJOR COUNTRIES IN WORLD MARKETS This section reviews the groundnut policies of the US, China, India, Argentina and SSA. Since domestic producer support/taxation and trade policies essentially determine excess supply and trade flows, it is necessary to examine them in some detail to anticipate the potential implications of policy changes on the distribution of gains/losses across countries. Groundnut Policies in the United States Groundnut products are a minor sector nationally, but they are a key component of agriculture and rural development in the Southern part of the United States. Based on USDA 14 Census of Agriculture, many counties in the South derive 50-70 percent of their agriculture income from groundnut. The first level of value-added activities, such as shelling, are performed locally as are many groundnut products manufacturing (Fletcher 2001). Groundnut policies have played a major role in maintaining rural income in these regions of the US. The foundation of U.S. groundnut policy is the U.S. peanut program, which traces its roots to the 1930s. Until the signature of the New Farm Bill in May 2002, the pillars of the system were production regulation through quotas, high producer support prices, and import control. The groundnut support program existed as a two-tier price support program. The support price for edible groundnuts was $610 per short ton paid for production under quota. Other groundnuts ("additionals") could be either exported at world prices or sold to the domestic crush industry and were eligible for a lower support price ($132 in 2001). The quota farmgate price tended to be higher than the prevailing export prices as shown in Table 6. Table 6. Producer support prices and international prices Prices faced by peanut producers under former peanut programs (US cents/pound) Item 1993/94 1994/95 1995/96 1996/97 1997/98 1998/99 Quota price 38.8 33.9 33.9 30.5 30.5 30.5 Additional price 6.6 6.6 6.6 6.6 6.6 6.6 Average farm price 30.4 28.9 29.3 28.1 28.3 28 CCC export price 20 20 20 20 20 20 Export unit value 33 28.6 29.2 32.2 32.7 32.8 Rotterdam c.i.f. Price 37.1 29.2 33.6 31.6 36 29 Source: Skinner (1999) The domestic support as measured by the aggregate measure of support (AMS) for U.S. groundnuts was estimated at around $330 million during that year or $31 million more than the average AMS for 1996-2001. The average cost of aggregate support in 1996-01 stood at $206 per metric ton of groundnut produced in the U.S. (Table 7). Table 7. U.S. aggregate measurement of support (million dollars) 1986-88 1995-96 1996-97 1997-98 1999-2000 2000-01 AMS Million US$ 347.2 414.6 299 305.76 300 330 AMS/ Metric Ton of US Production 203.3 264.1 180.0 190.5 172.7 222.8 Source: Skinner (1999), Hart and Babcock (2002) for 2000-01 AMS and USDA database for production data With the 2002 farm bill, groundnut production quotas were eliminated (with a quota buyout) and the groundnut price support program converted to a system of direct and counter- cyclical payments, and a price floor cum production subsidy (non-recourse loans with marketing loan provisions). The key features of the new program are detailed as follows: (i) All current groundnut producers have equal access to a marketing loan program, under which producers can pledge their crops as collateral to obtain a marketing loan rate equal to $355 per short ton. Producers may repay the loan at a rate that is the lesser of USDA-set repayment rate plus interest or the marketing loan rate plus interest, or they can forfeit the loan. (ii) For producers with a history of groundnut production, a new direct and fixed payment of 15 $36 per ton is available. Historic producers are those involved in groundnut production during the period 1998-2001. Eligible production would equal the product of average yield in the base- period and 85 percent of base-period acres. These payments are made regardless of current prices or the actual crop planted, so long as the farm remains in approved agricultural uses. (iii) Producers with a history of groundnut production are also eligible for a new counter- cyclical payment when market prices are below an established target price of $495 per ton minus the $36 per ton direct payment. The payment rate is the difference between the target price ($495 per ton) minus the direct fixed payment ($36 per ton), and the higher of the 12-month national average market price for the marketing year for groundnut or the marketing assistance loan rate ($355 per short ton). Total counter-cyclical payment to each eligible producer is calculated as the product of the payment acres (85 percent of base acres), the base-year average yield and the payment rate. (iv) Owners of groundnut quota under the previous legislation receive compensation payments for the loss of quota asset value. Payments may be made in five annual installments of $220 per short ton during fiscal years 2002-06, or the quota owner may opt to take the outstanding payment due in a lump sum. These payments are based on the quota owner's 2001 quota, so long as the person owned a farm eligible groundnut quota (Wescott, Young, and Price 2002). Beginning in 1994, under the URAA and NAFTA, the U.S. opened its market to gradually increasing quantities of groundnut imports through a tariff-rate quota (TRQ) system. For edible groundnuts, the total TRQ in 2001 was 57,059 metric tons or 4 percent of domestic consumption, allocated to historical importers (e.g., Argentina 77 percent, Mexico 7.4 percent) and then on a first-come, first-serve basis (table 8). In-quota tariffs for edible and prepared groundnuts range between 6.6 to 9.35 cents per kilogram while out-of-the quota tariffs are very high, ranging between 131.8 percent and 163 percent under URAA (Table 9)7. Table.8. US edible groundnut TRQ allocation Table 9. US over-the-quota tariffs Year Argentina Mexico Other Total NAFTA URAA URAA TRQ NAFTA TRQ URAA TRQ NAFTA + URAA Edible & Year Prepared Prepared 1995 26,341 3,478 4,052 33,871 Groundnut & Peanut Butter 1996 29,853 3,582 5,043 38,478 Base 123.1 155 1997 33,364 3,690 6,034 43,088 1994 120 1998 36,877 3,801 7,024 47,702 1995 116.9 151.1 1996 113.9 147.3 1999 40,388 3,915 8,015 52,318 1997 110.8 143.4 2000 43,901 4,032 9,005 56,938 1998 107.7 139.5 2001 43,901 4,153 9,005 57,059 1999 104.6 135.7 2002 43,901 4,278 9,005 57,184 2000 93 131.8 2001 81.4 131.8 -- 43,901 4,278 9,005 -- 2002 69.8 131.8 2008 43,901 unrestricted 9,005 -- 2008 0 Source: Uruguay Round Agricultural Negotiation, USA, Source: Uruguay Round Agricultural Negotiation, USA, Revised Country Schedule and NAFTA Revised Country Schedule and NAFTA Note: Prepared groundnut = roasted, salted, flavored groun 7The levels of quota and tariff for the period post-2003 are currently under negotiation. 16 Phase out of groundnut trade barriers under NAFTA and URAA is limited in scope, but it continues to have a dramatic impact on U.S. imports. From virtually no imports to the U.S. prior to 1994, edible groundnut imports have increased dramatically (Table 10). Argentina and Mexico averaged a fill rate of 87 and 77 percent respectively, but over-of-the quota imports were quite important, averaging 25,000 tons over the last six years. Edible groundnut imports comprised 6 percent of U.S. total domestic consumption of groundnuts in 2001. Table 10. US Imports of edible groundnuts Argentina Mexico Total Total Over-the-Quota year Import Quota Import Quota Import Quota Imports 1996 38270 29853 4710 3583 57000 38478 18522 1997 40622 33365 6148 3690 64000 43088 20912 1998 34465 36875 4834 3801 70000 47702 22298 1999 39494 40388 4916 3915 82000 52318 29682 2000 72230 43901 4864 4032 97000 56938 40062 2001 37557 43901 3611 4153 81000 57059 23941 2002 29927 43901 4406 4278 46795 57184 Not filled 2003 4692 43901 292 4278 5698 57184 Not filled Source: USDA The initial impacts of the new Farm bill are also visible here with the collapse of imports in 2003 (Table 10). The elimination of production quotas decreased the price paid by US processors and thus, increased domestic use of groundnuts (Figure 8). It also removed the incentive to import edible groundnuts. Figure 8. The decline in US domestic groundnut prices after the Farm Bill 2001 1100 1000 Rotterdamprice 900 800 700 600 US domestic price 500 400 New Farm Bill 300 93 94 95 96 97 98 99 00 01 02 03 Groundnut Policies in India, China, and Argentina For India, China and Argentina, a general trend since the mid-1990s is a gradual reduction of potentially market-distorting direct government intervention in production and marketing of groundnut products. However, in response to declines in groundnut product prices, 17 India has intensified its use of trade policy measures to protect its producers and processors. India is now the largest source of distortions in these product markets, but China is also a significant source of distortions in the groundnut oil market, because of its size relative to other countries present in these markets. India removed most restrictions on domestic trade (inter-state), storage, and export of groundnuts by 1998, and permitted trading in groundnut futures. However, import tariff levels remain very high for all the three groundnut products considered here. As Table 12 shows, tariffs on edible groundnut and groundnut meal stood at 45 percent, while that of groundnut oil was 35 percent in 2001. Since 2002, India has reversed its trade liberalization course on vegetable oil and has increased its applied tariffs on groundnut oil to 65-75 percent in 2002/3 and 85 percent for 2003/4 (Gulati, Pursell, and Mullen; and Pursell). The bound tariff is 100 percent. Furthermore, additional regulatory burdens increase domestic costs and prices. One example is the obligation to sell and purchase groundnut only in the "Agricultural Produce Wholesale Market."8. Another example of costly legislation is the "small-scale reservation" policy in groundnut processing, which sets limits on fixed assets in plant and machinery and thus prevents the domestic processing industry from realizing economies of scale. As in India, China liberalized groundnut trade to some degree in recent years. Imports of groundnut are now opened to private firms while prior to 1999, only six public agencies imported groundnut products. However, while the government has committed to cap and reduce trade-distorting domestic subsidies as part of its WTO accession commitments, guaranteed prices and government procurement schemes remained in place9. Furthermore, groundnut border protection remains high in China for processed groundnut (30%). The tariff on raw groundnut was only 15 percent in 2001 and many regions of China are natural exporters of groundnuts making the tariff redundant. In-quota tariffs on groundnut oil and groundnut meal were much lower, at 10 and 5.0 percent, respectively. One has however to factor in the issue of reported uneven application of the Chinese value-added tax (VAT) on imported and domestic products (USDA Attaché reports). The VAT is significant, ranging between 13 to 17 percent depending on the product; there is ample room for tax evasion (USDA FAS [a]; USDA FAS [b]), and the non-uniformity in application prevents a more accurate measure of the impact. Our policy analysis considers several cases with and without the VAT included in the trade barriers. China's State trading imposes quantitative restrictions through quotas and licenses on groundnut oil imports and imposes tariff barriers on seeds, meal, and oil. These barriers create a wedge between domestic and world market prices. Domestic prices of most oils including groundnut oil are significantly higher than international market prices. Tariffs and rents on import licenses explain price differentials between domestic and CIF prices. For example, the international price of groundnut oil in Hong Kong for 1998 was U.S. $728 per ton while the corresponding wholesale price in China was 67.8 percent higher (Fang and Beghin 2002). 8This legislation is costly to both farmers and processors because even if they are located very close to each other geographically, they all have to travel to the wholesale market, pay an "agent commission" and other marketing fees before the transaction is processed. 9According to FAO, these policies provide little incentive to expand production due to unattractive administrative price levels and greater involvement of private sector in marketing operations. Data on the size of domestic support is not available. 18 Table 13. Groundnut trade policies distortions in India, China and Argentina (%) Country Product Description 1999/2002 Edible groundnuts Import tariff 45 India Groundnut oil Import tariff 35'1 Groundnut meal Import tariff 45 Raw edible groundnuts Import tariff 15 China Processed edible groundnut Import tariff 30 Groundnut oil Import tariff 10 Groundnut meal Import tariff 5 VAT on edible groundnuts and gr. oil VAT 17 Edible groundnuts Import tariff 5 Groundnut oil Import tariff 13 Argentina Groundnut meal Import tariff 8 Edible groundnuts Export tax 3.5 Groundnut meal Export rebate 3.2 groundnut oil (refined) Export rebate 2.3 Source: WTO, WITS. USDA GAINS Report 1. Note: (1) India raised its tariff on oil to 65% in 2002 and 85% in 2003 Raw edib. gr. =raw, not roasted or cooked, in shell or shelled gr. Processed groundnuts = Bleached, preserved or otherwise prepared gr., incl. Roasted, salted. gr. butter Argentina's groundnut trade policy contrasts sharply with that of India and China, as almost all the distortions are associated with exports, with a 3.5 percent tax on exported raw groundnuts. With the peso devaluation of 2001, export retention on groundnut increased to 20 percent. This export tax may countervail the positive signal sent to groundnut exporters through the peso devaluation. Argentina maintains import tariffs on groundnut products, which exhibit some escalation (5, 8 and 13 percent on edible groundnut, meal, and oil, respectively). These tariffs are frequently redundant since the country is a net exporter of groundnut products. Groundnut Policies of Key African Exporters After decades of extensive intervention in the groundnut sector, African countries have, to a varying degree, undergone market reforms in the 1980s under structural adjustment plans (SAP). One of the main objectives of market reforms was to eliminate direct and indirect taxation of farmers that had undermined production incentives in the 1970s and early 1980s, and that led to underutilized processing capacities in many groundnut producing countries (Badiane and Kinteh 1994). Reforms have been piecemeal and partial. Governments have generally withdrawn from input markets, leading to difficult access to inputs (chiefly certified seeds and fertilizer) where important market failures (e.g., in the credit market) and high transaction costs prevailed, as in Senegal and Gambia (Akobundu 1998). Governments have however been reluctant to liberalize groundnut processing, for which privatization efforts started only recently (Senegal, Gambia). In Senegal and Gambia, producer prices are still set by governments. African governments have traditionally used pricing policies as levers to conveniently tax or subsidize farmers based on countries' industrial policy and political circumstances10. Taxation of groundnut farmers was high in the 1970's but has been reversed since the early 1990's in most countries while real world prices have trended downward (Badiane and Kinteh 1994). In Senegal 10Taxation of producers was direct, i.e., when marketing boards or similar agencies captured the rent equals to the difference between net world price and producer price, or indirect, via real exchange rate appreciation. This taxation was generally mitigated by input subsidies and border protection. 19 and Gambia, the main rationale for state intervention in the groundnut sector has been to safeguard the viability of state-owned processing mills. Consequently, the share of the export price to groundnut farmers has consistently been less than 60 percent in these two countries. This policy has been counter-productive for the state-owned enterprises, since farmers have bypassed large public processing companies, creating excess capacities and financial difficulties. With regard to trade policies, there are wide differences among African traditional groundnut exporters. Senegal and Malawi apply high tariffs on processed groundnuts, to encourage in-country processing (oil production in the case of Senegal) (Table 14). In contrast, Gambia and Nigeria have a liberal trade policy, with no export taxes or import tariffs. South Africa's tariff structure exhibits a slight escalation, with processed groundnuts subject to a tariff of 6 percent while unprocessed groundnuts enter duty free. In Senegal and Gambia however, unofficial cross-border trade is significant, with farmers frequently crossing the border to and from Senegal, depending mainly on respective producer prices and domestic supply levels. Oil imports face a 20 percent tariff in Senegal, South Africa and Malawi. Table 14. Tariffs on groundnut products in major African producers (%) Country Product 1999/2002 Raw edible groundnut 5 Senegal Processed edible groundnut 20 Groundnut meal 0 groundnut oil (refined) 20 Edible groundnuts 0 Nigeria Groundnut meal 0 groundnut oil (refined) 0 Edible groundnuts 0 The Gambia Groundnut meal 0 groundnut oil (refined) 0 Raw edible groundnut 0 South Africa Processed edible groundnut 6 Groundnut meal 0 groundnut oil (refined) 20 Raw edible groundnut 5 Malawi Processed edible groundnut 25 Groundnut meal 0 groundnut oil (refined) 20 Source: WTO Database. Note: Raw edib. gr. =raw, not roasted or cooked, in shell or shelled gr. Processed groundnuts = Bleached, preserved or otherwise prepared gr., incl. Roasted, salted. gr. butter The EU has harmonized country regulations of maximum permissible level of aflatoxin in 1998, and has fixed maximum levels of aflatoxin to the lowest possible level (e.g., 0.002- milligram for B1 type aflatoxin for edible groundnuts). Technical processes exist to reduce aflatoxin contamination (e.g., through ammoniac as used in Senegal for groundnut meal) but the best method is to improve farm practices through use of the best quality and resistant seeds, proper management of farms, and appropriate storage to avoid exposure to high temperature and humidity. African countries also face difficulties in meeting stricter product and quality standards. In Senegal and Gambia, groundnut varieties were originally selected for oil production, which can accommodate lower quality seeds and raw groundnuts. A seed variety in Malawi proved successful in better yields, but lacked commercially viable characteristics. These problems transpire in the international prices of African origin, which are discounted compared 20 to high quality groundnut sold in the EU market (Figure 9). Figure 9. Prices of African Shelled Groundnut Exports Compared to other Developing Countries (US$/ ton) 2500 2000 1500 1000 500 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 SSA Argentina China India Source: Oil World Shifting out of groundnut oil and upgrading the quality of edible groundnut may be an option for African producers. Unlike demand for oil and meal, demand for confectionary groundnut (the higher quality edible groundnut) has been rising and is expected to continue to increase in the medium term. At the same time, confectionary groundnuts receive a price premium that can be as high as 100 percent compared to grades used for oil and meal. In Senegal, one ton of first grade confectionery groundnut sells for $800 to $900, which is equivalent to the price of groundnut oil. It takes, however, three tons of unshelled groundnuts to produce one ton of oil. Similarly, prices of Gambian groundnuts are about $300, $450 and $600 FOB for (crushing), bird food and edible respectively. Were Gambia able to upgrade 50 percent of its 10,000 tons of exports from crushing to edible, the increased revenue would be $1.5 million. Groundnut Trade Policies of High-income Importers In spite of a general pattern of tariff escalation, barriers to groundnut are not a major obstacle in high-income major importers: the two largest groundnut importers in this category, the EU and Canada, have a zero tariff for unprocessed groundnut and low processed groundnut tariffs for GSP and LDC countries (Table 15). Assessment of market access in these countries should however take into account the strict quality standards. In contrast to the EU and Canada, Japan and Korea have a high tariff regime for groundnuts. Japan applies a high tariff on processed groundnuts and it offers a very limited preference margin of 4 percent for groundnut exports from LDC. Korea has very high tariffs on both raw and processed groundnuts, with tariffs on the former standing at more than 200 percent. This high tariff on raw groundnut may reflect the government desire to stimulate production, which has plummeted over the past 15 years. In contrast to edible groundnuts, groundnut oils and meal enter into all these high-income countries duty free. 21 Table 15. Simple average tariffs on edible groundnuts, unprocessed and processed MFN Tariffs GSP Tariffs LDC Tariffs Unprocessed Processed Unprocesse Processed Unprocesse Processed EU 0% 13% 0% 9% 0% 0% Japan 3.7% 19% 3.7% 19% 3.7% 15% Canada 0% 4% 0% 4% 0% 3.2% Korea 243% 65% 243% 65% 243% 65% Source: WTO IV. IMPACT OF GROUNDNUT PRODUCT POLICY REFORMS ON WORLD PRICES, TRADE FLOWS AND WELFARE We summarize key findings of our quantitative analysis of distortions in groundnut markets (see Beghin, Diop, Matthey, and Sewadeh 2003). A full description of the model and further details can be found in the latter reference. We first report results obtained under the most plausible assumptions underlying the model. We later discuss sensitivity analysis which concerns US policy and the uncertainty on China's protection (VAT and protection of processed groundnuts). Policy Reform Scenarios We analyze multiple scenarios. First, we consider full multilateral trade liberalization for groundnuts, meal, and oil, with and without the removal of the U.S. peanut program. We call these two scenarios FMTL&US, and FMTL. Then we consider multilateral groundnut trade liberalization, again with and without the removal of the U.S. farm peanut program (GMTL&US, and GMTL scenarios). We then consider full trade liberalization in the two largest and most distorted groundnut markets, China and India (CIFTL scenario). We report results on these key five scenarios in Tables 16 to 21. All results regarding changes in price and physical flows are reported in percent change from the baseline, Tables 16-20. Changes in welfare are reported in 1995 PPP U.S. dollars (purchase power parity holding in 1995) in Table 21. The baseline and simulations were run for 3 years (1999-2001), and averaged. We report the 3-year average impacts in the last column of the tables. Results In countries characterized by high groundnut protection, the combined effect of the world price increase and removal of their own protection is beneficial to final users of groundnuts, other things being equal. For countries with moderate or no protection prior to reform, the net impact (tariff removal and terms of trade) is an increase in domestic groundnut prices, handicapping groundnut users (final consumers, crushers), other things being equal. A similar logic and dichotomy of cases carries through for groundnut oil and meal for which the combined effect of world price and tariff removal has to be assessed. These substantial terms-of-trade effects have a large impact on trade and welfare. Allocative efficiency gains in domestic markets can be offset by large price increases originating in post-reform world markets. Further, in countries with high protection of the oil and/or meal sectors (e.g., India), the oil and cake tariff removal, net of the world price hike, induces lower domestic prices for these two products and reduces crush margins. As a result, the domestic excess supply of groundnut crushed into oil and cake decreases, drawing imports. By contrast, countries with moderate or no 22 protection in their oil and cake markets face a net price increase for oil and cake after full trade liberalization. Their final consumption of these value-added products decreases, and crushing increases as their crush margin improves with the reform. Their excess supply of these products increase and they exhibit larger exports. The two full trade liberalizations scenarios with and without the removal of U.S. farm policy, FMTL&US and FMTL, bring strong price increases for all three products, 10 percent for groundnuts, 18 percent for groundnut cake, and 27 percent for groundnut oil, as shown in Tables 16 and 17. The welfare impact of the FMTL&US and FMTL reforms is influenced by the change in the groundnut oil price, which affects the crush margin. Specifically, crush margins deteriorate in the EU and India, decreasing supply. However, margins improve in China, Gambia, Nigeria, Senegal, South Africa and the USA. As shown in Tables 16 and 17, trade patterns change dramatically. China expands its exports of the three products. The high increase in the price of oil improves the crush margin and stimulates crush in China. Higher prices for groundnut oil in the world markets translates into larger exports. In India the lower crush margin reduces oil and meal production; lower consumer prices for all groundnut products stimulate oil and feed demand and eventually imports to meet the need associated with the liberalized price level. African producers expand their exports of value-added products. Senegal and Malawi decrease their exports of groundnuts because of larger domestic use, India experiences trade reversal becoming a large importer of groundnut oil and cake. Accounting for the trade reversals in Table 16 and 17, aggregate trade in groundnuts increases by 16 percent, and trade in value-added products more than doubles. The first two columns of Table 21 show the welfare impact of these two reforms. The aggregate net welfare effects of FMTL&US and FMTL amount to about 791 million dollars at 1995 prices in each scenario, respectively. China and India experience the largest welfare gains, not surprisingly since they have the two largest and most distorted groundnut product markets. China's welfare gains are about 666 million dollars, whereas India's' gains are about 213 million dollars. The "moderate" world welfare effect first comes from offsets--some countries gain in aggregate whereas some others lose, chiefly the EU-15. Further, for many countries other than China and India, individual net gains/losses are moderate, mostly because of the small size of the three groundnut markets and their price-inelastic nature, which bring large transfers but small deadweight losses. Indeed, substantial transfers occur between consumers, crushers, and producers. These transfers offset each other. Price effects induced by the reforms have a similar impact (large welfare rectangles and small triangles), including in countries with undistorted markets. For example in Nigeria, following FMTL, groundnut producers gain 34 million dollars of quasi-rents; consumers experience welfare losses of 65 million dollars because of higher oil and processed groundnut prices; crushers gain 51 million dollars; meal users (feed users) lose about 3 million dollars. The country in aggregate is better off by 16 million dollars. Under multilateral trade liberalization for all three products, the removal of the U.S. program has impact on trade flows, terms of trade or welfare. The strong price effects of trade liberalization invalidate the price floor established by the U.S. loan rate. The only remaining production-distorting element is the fixed payment (fully coupled in our model), which is small. Results under both scenarios (full trade liberalization, FMTL) and FMTL&US) are qualitatively identical, except for the USA, which experience additional welfare gains of 3.5 million dollars for the removal of its domestic distortions (gains to U.S. taxpayers net of losses of U.S. 23 producers). The world price impacts of the FMTL scenario are identical to those of FMTL&US (10%, 18% and 27% respectively for groundnuts, cake and oil). Similarly trade flows are barely affected by the removal of the US domestic program under free trade. U.S. peanut exports are about 15,000 mt lower in the FMTL&US scenario as compared to their level in the FMTL scenario. Given that our parameterization of US farm policy assumes full coupling of payments received by producers to production, our assessment provides an upper bound on the effect of the current U.S. peanut program.11 Many debates of the Doha round of the WTO revolve around narrow agricultural negotiations of substantial importance to developing countries. Hence, it is useful to assess what a narrow agricultural liberalization would achieve relative to a full trade liberalization encompassing the value-added products of groundnut oil and cake. The GMTL&US, and GMTL scenarios consider these reforms and their impacts, with results shown in Tables 18 and 19 and in the third and fourth columns of Table 21. Much is achieved by groundnut trade liberalization alone, but with a large second-best component since distortions are present in the value-added markets. In these groundnut liberalization scenarios, the price of cake and oil is little affected and crush margins are primarily affected by changes in groundnut prices. Margins improve in India, but deteriorate in countries with limited groundnut distortions. Consumer welfare implications are also different in these trade scenarios. In highly protected groundnut oil markets, prices are higher under the groundnut trade scenarios (GMTL scenarios) than they are under all-product trade liberalization (FMTL scenarios). In countries with no oil distortions, prices roughly remain as their baseline level and consumers do better under the groundnut trade liberalization than under FMTL scenarios. For the latter reason, the rest of the world fares much better under GMTL scenarios than under the FMTL scenarios. By contrast African economies do much better with the FMTL scenarios than with groundnut trade liberalization reforms. The potential Africa-5 welfare gains nearly double moving from GMTL scenarios to FMTL ones.12 If China and India liberalized alone (CIFTL scenario), the qualitative results of the FMTL scenarios would hold. What is striking in this last scenario is the importance of India's and to a lesser extent China's distortions and market size in the welfare, trade, and price effects. As suggested by Table 18 and the last column of Table 20, FMTL really hinges on the removal of distortions in China and India. With the implementation of CIFTL, world price increases for the three products would be substantial: 8 percent for groundnuts, 18 percent for meal, and 26 percent for oil. The major welfare differences occur in the Rest of the World where consumers do worse than they would under the GMTL, since oil prices are higher. Africa-5 improves its lot in aggregate but not as well as it would under the FMTL scenario, since groundnut prices are not as high and because Africa-5 own distortions are still in place. 11We also ran a U.S. distortion removal scenario under existing trade distortions. We obtain a 0.13% increase in the world price of peanuts and virtually no increase in world cake and oil prices. U.S. peanut exports decrease by 10 percent or about 20,000 mt. Hence unlike in the case of some other commodities subsidized by U.S. taxpayers and consumers (e.g., rice, cotton, sugar), the impact of the current U.S. farm program on peanut world price and trade is nearly negligible. 12Africa-5 denotes our aggregate of the Gambia, Malawi, Nigeria, Senegal, and South Africa. 24 Table 16. Full trade liberalization and removal of US farmpolicy (FMTL&US scenario) new levels after reform baseline levels average change 99/00 00/01 01/02 99/00 00/01 01/02 for 3 years Peanuts Trade (1000mt) Net Exporters Argentina 241 190 196 226 177 185 7% China 699 659 687 540 450 525 36% Gambia 9 12 17 8 11 15 11% India 89 -1 33 100 100 125 -62% Malawi 0 1 1 2 3 3 -80% Nigeria 30 39 42 0 0 0 3667% Senegal -5 -10 -3 2 4 5 -287% South Africa 26 20 38 20 16 35 22% USA 272 162 234 255 141 231 8% Total Net Exports 1361 1072 1245 1153 902 1124 16% Net Importers Canada 111 102 105 116 107 110 -5% European Union 441 428 448 457 441 463 -3% Mexico 94 65 69 101 72 75 -8% Rest of the World 525 467 563 290 272 415 63% Residual 189 10 61 189 10 61 0% Total Net Imports 1361 1072 1245 1153 902 1124 16% Peanuts Price US Run. 40/50 CIF Rotterd 896 972 779 820 888 700 10% Peanut Meal Trade (1000 mt) Net Exporters Argentina 73 54 63 67 50 52 13% China 111 119 124 9 15 25 741% Gambia 7 12 11 5 10 10 22% India -311 -297 -212 10 20 100 -1702% Malawi 0 0 0 0 0 0 9% Nigeria 26 26 34 0 0 0 2867% Senegal 137 151 145 130 144 140 5% South Africa -4 2 1 -5 0 0 95% USA 33 29 31 6 5 5 484% Rest of the World 134 136 103 8 14 -12 499% Total Net Exports 206 233 300 230 258 320 -9% Net Importers European Union 162 169 158 186 194 178 -12% Residual 44 64 142 44 64 142 0% Total Net Imports 206 233 300 230 258 320 -9% Meal Price 48/50%CIF Rotterdam$/mt 144 159 147 122 134 125 18% Peanut Oil Trade (1000 mt) Net Exporters Argentina 49 44 50 46 41 42 11% China 55 64 76 0 5 2 3469% Gambia 5 6 6 0 0 0 589% India -238 -225 -266 0 0 0 -24288% Malawi 0 0 1 0 0 0 43% Nigeria 72 72 77 35 35 30 123% Senegal 102 108 114 98 102 109 5% South Africa 0 1 1 0 0 0 49% USA 23 -18 6 2 -30 -10 288% Rest of the World 114 103 105 18 11 8 861% Total Net Exports 185 155 170 199 164 181 -6% Net Importers European Union 136 101 109 150 110 120 -9% Residual 49 54 61 49 54 61 0% Total Net Imports 185 155 170 199 164 181 -6% Peanut Oil Price CIF Rotterdam$/mt 933 866 851 744 685 659 27% welfare(million dollars) 690 920 763 791 25 Table 17. Full trade liberalization (FMTLscenario) newlevels after reform baseline levels 99/00 00/01 01/02 99/00 00/01 01/02 average change for 3 years (%) Peanuts Trade (1000 mt) Net Exporters Argentina 241 190 195 226 177 185 6% China 693 655 678 540 450 525 34% Gambia 9 12 17 8 11 15 11% India 87 -1 29 100 100 125 -64% Malawi 0 1 1 2 3 3 -82% Nigeria 29 38 40 0 0 0 3564% Senegal -6 -10 -4 2 4 5 -298% South Africa 26 20 38 20 16 35 22% USA 287 169 259 255 141 231 15% Total Net Exports 1367 1075 1254 1153 902 1124 16% Net Importers Canada 111 102 105 116 107 110 -5% European Union 442 428 448 457 441 463 -3% Mexico 94 65 69 101 72 75 -8% Rest of the World 531 470 571 290 272 415 65% Residual 189 10 61 189 10 61 0% Total Net Imports 1367 1075 1254 1153 902 1124 16% Peanuts Price US Run. 40/50, CIF Ro 895 972 778 820 888 700 10% Peanut Meal Trade (1000 mt) Net Exporters Argentina 73 54 63 67 50 52 13% China 111 119 123 9 15 25 739% Gambia 7 12 11 5 10 10 21% India -311 -297 -212 10 20 100 -1703% Malawi 0 0 0 0 0 0 8% Nigeria 26 26 34 0 0 0 2862% Senegal 137 151 145 130 144 140 5% South Africa -4 2 1 -5 0 0 95% USA 33 29 32 6 5 5 487% Rest of the World 135 137 103 8 14 -12 499% Total Net Exports 206 233 300 230 258 320 -9% Net Importers European Union 162 169 158 186 194 178 -12% Residual 44 64 142 44 64 142 0% Total Net Imports 206 233 300 230 258 320 -9% Meal Price 48/50%CIFRotterdam$/ 144 159 147 122 134 125 18% Peanut Oil Trade (1000 mt) Net Exporters Argentina 49 44 50 46 41 42 11% China 55 64 76 0 5 2 3459% Gambia 5 6 6 0 0 0 587% India -238 -225 -266 0 0 0 -24304% Malawi 0 0 1 0 0 0 43% Nigeria 72 72 77 35 35 30 123% Senegal 102 108 114 98 102 109 5% South Africa 0 1 1 0 0 0 49% USA 24 -17 6 2 -30 -10 290% Rest of the World 115 103 105 18 11 8 864% Total Net Exports 185 155 170 199 164 181 -6% Net Importers European Union 136 101 109 150 110 120 -9% Residual 49 54 61 49 54 61 0% Total Net Imports 185 155 170 199 164 181 -6% Peanut Oil Price CIFRotterdam$/mt 933 866 851 744 685 659 27% welfare(milliondollars) 691 924 757 791 26 Table 18. Peanut trade liberalizationand removal of US peanut program(GMTL&US scenario) newlevels after reform baseline levels 99/00 00/01 01/02 99/00 00/01 01/02 average change for 3 years Peanuts Trade (1000 mt) Net Exporters Argentina 260 208 249 226 177 185 22% China 748 703 688 540 450 525 42% Gambia 11 15 18 8 11 15 31% India -482 -553 -415 100 100 125 -556% Malawi -1 1 1 2 3 3 -93% Nigeria 68 75 82 0 0 0 7470% Senegal 4 2 2 2 4 5 -8% South Africa 25 20 38 20 16 35 20% USA 355 244 301 255 141 231 48% Total Net Exports 989 714 962 1153 902 1124 -17% Net Importers Canada 112 102 106 116 107 110 -4% European Union 440 426 449 457 441 463 -3% Mexico 95 66 70 101 72 75 -7% Rest of the World 153 109 275 290 272 415 -47% Residual 189 10 61 189 10 61 0% Total Net Imports 989 714 962 1153 902 1124 -17% Peanuts Price US Run. 40/50, CIF Rotterdam 884 960 759 820 888 700 8% Peanut Meal Trade (1000 mt) Net Exporters Argentina 64 47 47 67 50 52 -6% China -11 -9 12 9 15 25 -144% Gambia 5 10 10 5 10 10 -2% India 71 95 151 10 20 100 344% Malawi 0 0 0 0 0 0 1% Nigeria -2 -2 -2 0 0 0 -193% Senegal 129 143 140 130 144 140 -1% South Africa -5 0 0 -5 0 0 -7% USA -14 -18 -13 6 5 5 -380% Rest of the World -7 -5 -25 8 14 -12 -70% Total Net Exports 231 260 320 230 258 320 0% Net Importers European Union 187 196 178 186 194 178 1% Residual 44 64 142 44 64 142 0% Total Net Imports 231 260 320 230 258 320 0% Peanut Meal Price 48/50%CIF Rotterdam 122 133 125 122 134 125 0% Peanut Oil Trade (1000 mt) Net Exporters Argentina 44 39 38 46 41 42 -6% China -13 -11 -7 0 5 2 -705% Gambia 0 0 0 0 0 0 -5% India 45 55 38 0 0 0 4591% Malawi 0 0 0 0 0 0 0% Nigeria 33 32 28 35 35 30 -6% Senegal 97 101 109 98 102 109 -1% South Africa 0 0 0 0 0 0 -4% USA -14 -48 -24 2 -30 -10 -194% Rest of the World 7 -3 -2 18 11 8 -103% Total Net Exports 200 165 181 199 164 181 0% Net Importers European Union 151 111 120 150 110 120 0% Residual 49 54 61 49 54 61 0% Total Net Imports 200 165 181 199 164 181 0% Peanut Oil Price CIF Rotterdam$/mt 747 686 664 744 685 659 0% welfare (million dollars) 782 1024 799 868 27 Table19.Impactofpeanuttradeliberalization(GMTLscenario) newlevelsafterreform baselinelevels averagechange 99/00 00/01 01/02 99/00 00/01 01/02 for3years PeanutsTrade Net Exporters Argentina 260 208 248 226 177 185 22% China 742 699 678 540 450 525 41% Gambia 11 15 18 8 11 15 30% India -483 -554 -419 100 100 125 -557% Malawi -0.6 0.7 0.6 2 3 3 -95% Nigeria 67 74 80 0 0 0 7358% Senegal 3 2 2 2 4 5 -20% SouthAfrica 25 20 37 20 16 35 20% USA 371 253 327 255 141 231 55% Total Net Exports 995 717 972 1,153 902 1,124 -16% Net Importers Canada 112 102 106 116 107 110 -4% EuropeanUnion 440 426 449 457 441 463 -3% Mexico 95 66 70 101 72 75 -7% Rest oftheWorld 159 112 285 290 272 415 -45% Residual 189 10 61 189 10 61 0% Total Net Imports 995 717 972 1,153 902 1,124 -16% PeanutsPrice:USRunners40/50, CIFRotterdam 883 959 758 820 888 700 8% PeanutMealTrade Net Exporters Argentina 64.42 46.98 47.23 67.00 50.00 52.00 -6% China -10.71 -9.49 11.42 9.00 15.00 25.00 -146% Gambia 4.87 9.81 9.92 5.00 10.00 10.00 -2% India 70.15 94.53 150.54 10.00 20.00 100.00 342% Malawi 0.00 -0.03 0.06 0.00 0.00 0.00 1% Nigeria -1.89 -2.37 -1.64 0.00 0.00 0.00 -196% Senegal 129.18 142.68 139.58 130.00 144.00 140.00 -1% SouthAfrica -5.09 -0.19 -0.05 -5.00 0.00 0.00 -7% USA -13.70 -17.66 -12.35 6.00 5.00 5.00 -376% Rest oftheWorld -6.06 -4.64 -24.20 8.00 14.00 -12.00 -69% Total Net Exports 231.18 259.63 320.50 230.00 258.00 320.00 0.4% Net Importers EuropeanUnion 187 196 179 186 194 178 1% Residual 44 64 142 44 64 142 0% Total Net Imports 231 260 321 230 258 320 0% PeanutMealPrice:48/50%CIFRotterdam 122 133 125 122 134 125 0% PeanutOilTrade Net Exporters Argentina 44 39 38 46 41 42 -6% China -13 -12 -7 0 5 2 -713% Gambia -0.07 -0.16 0.04 0 0 0 -6% India 44 55 38 0 0 0 4558% Malawi -0.006 -0.027 0.023 0.000 0.000 0.000 0% Nigeria 33 32 28 35 35 30 -6% Senegal 97 101 109 98 102 109 -1% SouthAfrica 0 0 0 0 0 0 -4% USA -13 -48 -23 2 -30 -10 -192% Rest oftheWorld 8 -3 -1 18 11 8 -99% Total Net Exports 200 165 181 199 164 181 0% Net Importers EuropeanUnion 151 111 120 150 110 120 0% Residual 49 54 61 49 54 61 0% Total Net Imports 200 165 181 199 164 181 0% PeanutOilPrice:CIFRotterdam 746 686 664 744 685 659 0% Welfare(milliondollars) 783 1,028 795 869 28 Table20.ImpactofChinaandIndiafullliberalization(CIFTLscenario) Newlevelsafterreform baselinelevels averagechange 1999/2000 2000/01 2001/02 1999/2000 2000/01 2001/02 for3years(%) PeanutsTrade NetExporters Argentina 223 174 160 226 177 185 -6% China 586 531 589 540 450 525 13% Gambia 8 11 16 8 11 15 3% India 55 -38 0 100 100 125 -94% Malawi 4 5 5 2 3 3 84% Nigeria 12 19 22 0 0 0 1776% Senegal -24 -19 -9 2 4 5 -708% SouthAfrica 24 19 37 20 16 35 14% USA 258 142 238 255 141 231 2% TotalNetExports 1147 845 1058 1153 902 1124 -4% NetImporters Canada 112 103 106 116 107 110 -4% EuropeanUnion 448 433 453 457 441 463 -2% Mexico 96 67 70 101 72 75 -6% RestoftheWorld 302 232 369 290 272 415 -7% Residual 189 10 61 189 10 61 0% TotalNetImports 1147 845 1058 1153 902 1124 -4% PeanutsPriceUSRun.40/50,CIFRotterdam$/mt 877 952 763 820 888 700 8% PeanutMealTrade NetExporters Argentina 76 57 67 67 50 52 18% China 114 122 126 9 15 25 759% Gambia 7 12 11 5 10 10 22% India -309 -294 -210 10 20 100 -1690% Malawi 1 0 0 0 0 0 46% Nigeria 26 26 34 0 0 0 2867% Senegal 150 165 155 130 144 140 14% SouthAfrica -3 3 2 -5 0 0 139% USA 37 33 35 6 5 5 563% RestoftheWorld 108 108 80 8 14 -12 385% TotalNetExports 205 233 300 230 258 320 -9% NetImporters EuropeanUnion 161 169 158 186 194 178 -13% Residual 44 64 142 44 64 142 0% TotalNetImports 205 233 300 230 258 320 -9% MealPrice48/50%CIFRotterdam$/mt 144 159 147 122 134 125 18% PeanutOilTrade NetExporters Argentina 51 45 53 46 41 42 16% China 53 61 74 0 5 2 3354% Gambia 5 6 6 0 0 0 567% India -240 -226 -269 0 0 0 -24481% Malawi 2 2 2 0 0 0 185% Nigeria 71 71 76 35 35 30 120% Senegal 116 126 126 98 102 109 19% SouthAfrica 2 2 2 0 0 0 224% USA 26 -14 9 2 -30 -10 326% RestoftheWorld 92 79 85 18 11 8 665% TotalNetExports 180 151 166 199 164 181 -8% NetImporters EuropeanUnion 131 97 105 150 110 120 -12% Residual 49 54 61 49 54 61 0% TotalNetImports 180 151 166 199 164 181 -8% PeanutOilPriceCIFRotterdam$/mt 924 857 844 744 685 659 26% Welfareeffects(million$) 765 1013 815 - - - 864 29 Table21.Welfareeffectsofpolicyscenariosinmilliondollarsat1995prices (average1999-2001) Country FMTL&US FMTL GMTL&US GMTL CIFTL Argentina 16.07 15.94 9.97 9.84 12.66 EU-15 -51.83 -51.27 -34.40 -33.82 -58.87 China 666.25 668.76 650.65 653.33 716.25 India 213.27 214.11 196.57 197.79 228.59 Restoftheworld -126.69 -127.06 -4.21 -4.86 -71.06 Canada -5.94 -5.87 -4.88 -4.81 -4.59 Mexico -7.43 -7.34 -6.11 -6.01 -5.73 Senegal 41.03 40.96 21.93 21.86 21.39 Nigeria 15.93 15.77 7.22 7.07 13.45 SouthAfrica 2.30 2.28 2.19 2.17 0.53 Malawi 7.45 7.45 7.60 7.61 -1.06 Gambia 0.43 0.42 0.24 0.24 0.36 USA 20.18 16.70 21.71 18.40 12.39 Africa-5total1 67.14 66.89 39.18 38.95 34.67 Total 791.01 790.87 868.48 868.79 864.32 1.DenotestheaggregateofSenegal,Nigeria,SouthAfrica,Malawi,andtheGambia. We investigated two key assumptions in the model: the prevailing groundnut market price underlying the US market, and the level of protection of the groundnut markets in China. We calibrated the model on 2002/3 US prices ($389/mt) to see if the new US policy would have had a stronger impact on the world market under lower prevailing prices. US farm prices in 2002/3 were 25 percent lower than in 2001/2. We remove the loan rate, counter-cyclical payments, and fixed payments (recall the latter two components are assumed fully coupled in our model to provide an upper bound on the effect of the US program), while holding all other distortions in place in all other countries. The price floor provided by the loan rate is effective under the lower 2002/3 farm price. US Output decreases by 7 percent under the new prices and US exports decreases by 52%, inducing a 0.9 percent increase in the world price of groundnuts and further negligible price impacts in the other markets. The aggregate net welfare effect is negligible and negative. Higher world prices exacerbate distortions in other markets or increase import costs in net-importing countries. The US gains about 22 million dollars (program cost savings net of the producer loss). We also tested the same change but with all other distortions removed in all countries (FMTL&US scenario). In this scenario, the world price levels of groundnuts was 0.5 percent higher than under free trade cum U.S. farm bill. We concluded that removing the farm bill incentives in a free trade world would decrease US production by about 4 percent and decrease its exports by 31 percent. The aggregate welfare gains vary by less than $1 million. Hence, the conclusions that the new US groundnut policy is much more benign than its predecessor remains unaltered under much lower prices. The sensitivity analysis on China's protection structure is more pivotal to the conclusions reached, especially the protection of the groundnut sector. We consider the following assumptions changes: the protection of groundnut producers is assumed to be 15 (tariff is redundant in the original model). The Chinese farm sector is no longer assumed to be a net exporter without assistance. Under this new assumption and following full trade liberalization (FMTL&US), China becomes a net importer of groundnuts because demands for edible and crush groundnuts increase. China's welfare gains are 1,029 million dollars. Aggregate welfare gains are 1160 million dollars. World prices increase by 18, 19, 29 percent for groundnuts, cake 30 and oil, respectively. We also lower the baseline protection of processed groundnuts to 15 percent ad-valorem tariff (original tariff was 30 percent and VAT was 17 percent). Under the latter assumptions, welfare gains from FMTL in China are only 266 million dollars and aggregate gains are 388 millions. The world price of groundnuts increases by 9 percent in this modified scenario compared to a 10 percent increase under the original run). The major change in welfare occurs in China because Chinese consumers gain much less from trade liberalization compared to the initial situation with original tariffs and VAT on processed groundnuts. V. POLICY IMPLICATIONS AND CONCLUSIONS The groundnut market is historically distorted by heavy government intervention in the North and the South. In the US, the 2002 Farm Bill has suppressed some unsustainable features of the previous policies, but has introduced new distortions with some limited potential to depress world market prices and subsidize U.S. groundnut exports. In India and China, governments have succeeded in stimulating production and exports, thereby capturing a growing share of the international market. These gains in India have been artificial because the entire groundnut industry relies on heavy protection. In contrast, in Sub-Saharan Africa (SSA) and in Argentina, government intervention has negatively affected the sector. Following full trade liberalization, world market prices would increase by about 10 percent for groundnuts, 18 percent for meal, and 27 percent for oil. Trade in groundnuts would increase by 16 percent and trade in oil and meal would more than double, considerably expanding the global trade volume of these markets. The current U.S. domestic peanut program is now mostly a U.S. domestic support program with minimal distortive impact, unlike U.S. domestic policy for other products such as cotton, dairy, rice, and sugar. Based on our scenarios, developing countries would gain little by "forcing" further U.S. groundnut policy reform. These changes would prove ineffective unless groundnut prices fall to very low levels. Only then would the U.S. policy further destabilize world prices given its anti-cyclical nature, and sending the wrong market signal to U.S. producers. Under prevailing market conditions, U.S. producers would actually benefit from multilateral trade liberalization in groundnut product markets. Hence, on the instance of groundnuts, it would be rational for the U.S. to support foreign groundnut producers in their attempt to liberalize. As a bloc, the most OECD countries would experience welfare losses after trade liberalization--moderate gains in the USA offset by losses in the EU-15, Canada and Mexico. Mexico, Canada, and the EU-15 would lose from the trade liberalization, because there are few distortions in these markets, so consumers are directly penalized by price increases for the groundnut products. The removal of trade distortions by the two largest developing economies--India and to a lesser extent China-- is essential in the groundnut product markets. These two countries' policies have the largest distorting effect on world prices for the three traded commodities considered in our analysis because of their market size and because of prohibitive distortions in the case of India. Their policies substantially depress the world prices of the three traded commodities. Following the removal of these distortions, net buyers of these products will be worse off. However, as we have shown, the countries of Sub-Saharan Africa that export groundnuts would mostly gain from full trade liberalization. Although the net world welfare effects of liberalizing these three markets are moderate, 31 they remain significant for small agrarian economies in Sub-Saharan Africa. Hence in the context of poverty alleviation, liberalization creates welfare gains in countries where they are much needed. The simulations also show that beyond agricultural trade liberalization, the liberalization of the value-added markets is essential to achieve larger welfare gains in African countries. Although the bulk of the world welfare gains occur with groundnut trade liberalization, the additional removal of distortions in value-added markets doubles net welfare gains in the African region via larger profits to groundnut and groundnut oil producers and exporters. African countries modeled in our trade liberalization analysis would experience aggregate welfare gains of 67 million dollars, with Senegal and Nigeria reaping most of these gains. Groundnut and groundnut oil consumers in Africa tend to be urban whereas groundnut production generates income in rural areas as a cash crop. In that respect, African groundnut producers modeled in our analysis gain between 50 and 150 million dollars of farm income depending on assumptions underlying the model. These figures are significant in the context of small African economies, and represent a significant opportunity to expand rural development in these areas. Also, in scenarios tested, the rest of the world would fare worse under full trade liberalization because consumers are required to pay higher groundnut product prices. However, groundnut is not without substitutes. The recent changes present both challenges and opportunities to major countries in the market. The US is likely to continue to dominate the high end of the international confectionary market under the new program. The performance of China and Argentina show that both countries have established strong groundnut sectors that can compete favorably under free market conditions. Chinese exports have played a stabilizing role in world markets in the 1990s. All developing countries, except Argentina, face a quality challenge for meeting the requirements of the expanding confectionary markets. This is particularly true for African countries. We did not attempt to model this aspect of the groundnut market, however our qualitative assessment of the groundnut market made this point clearly. Currently, the opportunities and rewards induced by the expansion of the edible groundnut exports market are eluding African producers to a large extent because of these quality issues. 32 REFERENCES Adams, G. et al. (2001). "Do Decoupled Payments Affect U.S. Crop Area? Preliminary Evidence from 1997-2000", American Journal of Agricultural Economics, Vol. 83, N.% (2001): 1190-95 Akobundu E. (1998). Farm-Household Analysis of Policies Affecting Groundnut production in Senegal. MS Thesis, Virginia Polytechnic and State University. Badiane, O. and Kinteh, S. (1994). Trade Pessimism and regionalism in African Countries: The Case of Groundnut Exporters. IFPRI Research Report 97, May 1994. Badiane, O. and M. Gaye (1999). Liberalization of Groundnut Markets in Senegal: Impact on the Marketing and Processing Sector. IFPRI Research Report. Beghin, J., N. Diop, H. Matthey, and M. Sewadeh (2003). "The Impact of Groundnut Trade Liberalization. Implication for the Doha Round." Mimeo, Selected paper presented at the 2003 AAEA Annual Meetings, Montreal. Chen et al. (1997). "Competitiveness in Peanuts: US versus China". Research Bulletin n.430, July 1997. University of Georgia. Gulati, A., G. Pursell, and K. Mullen. (2003) "Indian Agriculture Since the Reforms: Performance, Policy Environment, and Incentives," Mimeo, The World Bank, May. Fang C., and J. Beghin. "Urban Demand for Edible Oils and Fats in China. Evidence from Household Survey Data," Journal of Comparative Economics, 30 (4) (2002): 732-753. Fletcher S. M. (2001). "Peanuts: Responding to Opportunities and Challenges from an Intertwined Trade and Domestic Policies", University of Georgia, mimeo. Fletcher S.M., and C.L. Revoredo (2003). "Does the US Need the Groundnut Tariff Rate Quota Under the 2002 US Farm Act?" paper presented at the International Conference on Agricultural policy and the WTO: where are we heading, Capri, Italy, June 23-26. Hart, C., and B. Babcock (2002). U.S. Farm Policy and the World Trade Organization: How Do They Match Up? Chad E. Hart and Bruce A. Babcock, Working Paper 02-WP 294, February 2002 IMF (2001). Senegal: Selected Issues. Country Report N. 01/188., October Pursell, G. (2003) Private email correspondence on India's groundnut protection, September 2003. Revoredo, C.L. and S. Fletcher (2002a). "The US 2002 Farm Act and the Effects on US Groundnut Exports". University of Georgia, mimeo. Revoredo, C.L. and S. Fletcher (2002b). "World Peanut Market: An Overview of the Last 30 Years." University of Georgia, mimeo. Skinner, R. (1999). "Issues Facing the U.S. Peanut Industry During the Seattle Round of the World Trade Organization". US Department of Agriculture, Economic Research Service. U.S. Department of Agriculture, Foreign Agriculture Service. GAIN Report (Various years a) "China, People's Republic of. Oilseeds and Products. Oilseeds Update." __________. (Various years b). "China, People's Republic of. Oilseeds and Products. Annual." 33 Westcott, P.C., C. E. Young, and J. M. Price. (2002) "The 2002 Farm Act Provisions and Implications for Commodity Markets," Agriculture Information Bulletin Number 778, USDA, ERS, Washington DC, November. ANNEX. I- Unit Root and Co-Integration Tests for Edible and Groundnut Oil International Prices Table A1. ADF Unit Root Test, Edible Groundnut Prices, Period 1970-81: ADF Test Statistic 0.314357 1% Critical Value* -4.3260 5% Critical Value -3.2195 10% Critical Value -2.7557 *MacKinnon critical values for rejection of hypothesis of a unit root. Table A2. ADF Unit Root Test, Groundnut oil Prices, 1970-81: ADF Test Statistic -2.588595 1% Critical Value* -4.3260 5% Critical Value -3.2195 10% Critical Value -2.7557 *MacKinnon critical values for rejection of hypothesis of a unit root. Table A3. Johansen Co-integration Tests, Edible Groundnut and Groundnut Oil World Prices, 1970-81 Date: 12/01/02 Time: 12:33 Sample(adjusted): 1970 1981 Included observations: 12 after adjusting endpoints Trend assumption: Linear deterministic trend (restricted) Series: EDIBLEGR GROIL Lags interval (in first differences): 1 to 1 Unrestricted Cointegration Rank Test Hypothesized Trace 5 Percent 1 Percent No. of CE(s) Eigenvalue Statistic Critical Value Critical Value None 0.673526 16.94914 25.32 30.45 At most 1 0.437582 5.755101 12.25 16.26 *(**) denotes rejection of the hypothesis at the 5%(1%) level Trace test indicates no cointegration at both 5% and 1% levels Hypothesized Max-Eigen 5 Percent 1 Percent No. of CE(s) Eigenvalue Statistic Critical Value Critical Value None 0.673526 11.19404 18.96 23.65 At most 1 0.437582 5.755101 12.25 16.26 *(**) denotes rejection of the hypothesis at the 5%(1%) level Max-eigenvalue test indicates no cointegration at both 5% and 1% levels Table A4. Unit Root Test, Edible Groundnut Prices, Period 1981-2000 ADF Test Statistic -3.220154 1% Critical Value* -3.8067 5% Critical Value -3.0199 10% Critical Value -2.6502 *MacKinnon critical values for rejection of hypothesis of a unit root. 34 Table A5. Unit Root Test, Groundnut Oil Prices, Period 1981-2000 ADF Test Statistic -4.142213 1% Critical Value* -3.8067 5% Critical Value -3.0199 10% Critical Value -2.6502 *MacKinnon critical values for rejection of hypothesis of a unit root. II- Granger-Causality Test Do Senegalese Exports Granger-Cause Groundnut World Oil Price? DLOGEXP = First Difference of the Log of Senegalese Exports DLOGINTPR = First Difference of the Log of International Groundnut Oil Prices The test strongly indicates that Senegal's groundnut oil export granger-cause international price while the reverse is not true. Table A6. Granger-Causality Test Between Senegalese Groundnut Oil Exports and World Prices of Groundnut oil Pairwise Granger Causality Tests Date: 02/27/03 Time: 17:00 Sample: 1961 2001 Lags: 2 Null Hypothesis: Obs F-Statistic Probability DLOGINTPR does not Granger Cause DLOGEXP 37 0.02753 0.97287 DLOGEXP does not Granger Cause DLOGINTPR 7.43861 0.00222 Figure A1. World Groundnut Oil Prices versus other Vegetable oils 1200 1000 Groundnut Oil 800 600 400 200 0 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 Groundnut Oil Soybean Oil Palm Oil Sunflower Oil Rapeseed Oil 35 wb201257 C:\NDiame\Groundnuts\PRWPnumber2803 Groundnut Trade....doc February 11, 2004 3:57 PM 36