WPS5403 Policy Research Working Paper 5403 How Do Governments Respond to Food Price Spikes? Lessons from the Past Kym Anderson Signe Nelgen The World Bank Development Research Group Agriculture and Rural Development Team August 2010 Policy Research Working Paper 5403 Abstract Food prices in international markets spiked upward in changes were much less for developing than high-income 2008, doubling or more in a matter of months. Evidence countries, suggesting the dispersion of distortion rates is still being compiled on policy responses over the among farm industries decreased in developing countries. following two years, but lessons can be learned from the The adjustments were virtually all due to suspension and price spike in 1973, the magnitude and speed of which then reinstatement of import restrictions, with changes were similar to those experienced around the 2008 in export taxation by developing countries playing spike. In developing countries, policy responses to the an additional (but minor) role during 1972­74. This earlier spike lowered the (negative) nominal assistance beggar-thy-neighbor dimension of each government's coefficient for agriculture by one-third between 1972 and food policies is worrying because it reduces the role that 1974 before it was returned to the same level by 1976. trade between nations can play in bringing stability to the That was twice the extent of the fall and recovery of the world's food markets. More effort appears to be needed (positive) nominal assistance coefficient for high-income before a multilateral agreement to desist can be reached. countries. However, the trade and welfare effects of those This paper--a product of the Agriculture and Rural Development Team, Development Research Group--is part of a larger effort in the department to understand better the ways in which governments respond to economic shocks. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at kym. anderson@adelaide.edu.au or viaWmartin1@worldbank.org. 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 views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team How do governments respond to food price spikes? Lessons from the past Kym Anderson University of Adelaide and CEPR kym.anderson@delaide.edu.au and Signe Nelgen University of Adelaide signe.nelgen@googlemail.com Keywords: Commodity price stabilization policies, Domestic market insulation, Distorted incentives, Agricultural and trade policies, Trade restrictiveness indexes JEL codes: F14, Q17, Q18 This is a product of a World Bank research project on Distortions to Agricultural Incentives. The authors are grateful for helpful comments from workshop participants, and for funding from World Bank Trust Funds provided by the governments of the Netherlands (BNPP) and the United Kingdom (DfID) and from the Australian Research Council. The views expressed are the authors' alone and not necessarily those of the World Bank and its Executive Directors, nor the countries they represent, nor of the institutions funding this research. Forthcoming in the Journal of International Commerce, Economics and Policy,1(2), Fall 2010. Corresponding author: Kym Anderson, School of Economics, University of Adelaide, Adelaide, SA 5005, Australia; Phone +61 8 8303 4712, Fax +61 8 8223 1460, email: kym.anderson@adelaide.edu.au 2 How do governments respond to food price spikes? Lessons from the past Kym Anderson and Signe Nelgen 1. Introduction Food prices in international markets spiked upward in 2008, doubling or, in the case of rice, trebling in a matter of months. The magnitudes of the price rises, and the speed of their subsequent fall back to trend, were similar to those experienced in 1974. Figure 1 shows real international prices for the five years around the earlier and most-recent spikes for three key grains and for groundnut oil (which also is an important food staple in numerous low-income countries). On both occasions, the rising price generated panic buying by individual households, especially of rice in Asia, which exacerbated the international price spike. Both episodes also lead to urban food riots in numerous low-income countries. [Insert Figure 1 about here] Many scholars have studied the responses of market participants leading up to and during periods of upward and downward commodity price spikes, seeking among other things to explain the spikes (see, e.g., Deaton and Laroque 1992, 1996). This paper is not about actions of private agents, but rather of governments. It is understandable that in such periods governments feel the need to be seen to be doing something to lessen the impact on those adversely affected in their country. How and to what extent do their policy actions alter domestic producer and consumer prices of the commodities concerned, relative to the changes in international prices? Evidence is still being compiled on policy responses in 2008-10, but how have governments responded in the past, and in particular how did policies react to that mid-1970s price spike? The next section of the paper briefly reviews the domestic market insulation impacts of agricultural policies used in the past by governments attempting to stabilize their domestic food markets. Those impacts include protecting producers in times of downward international price spikes (as in 1986) as well as protecting consumers in periods of upward price spikes ­ in each case at the expense of the other domestic group who otherwise would have benefitted from the price movement. The main section then examines in detail the policy responses as reflected in various annual indicators of distortions to producer and 3 consumer incentives during the years 1972 to 1976. The final section of the paper draws together some policy lessons from this evidence of past government responses to food price spikes. In particular, it draws attention to the fact that when many countries seek to reduce gyrations in their domestic food markets by altering trade restrictions at their national border, such actions are collectively self-defeating: they reduce the role that global trade can play in dampening fluctuations in international prices, and they prolong the adjustment process. 2. Past domestic market stabilization efforts Governments of almost all countries deliberately seek to reduce fluctuations in domestic food prices and in the quantities available for local consumption. There is a huge analytical literature on the economics of such price stabilization efforts. Its connection with trade policy was highlighted by Johnson (1975) immediately following the upward spike in world food prices in 1973-74. His analysis of grain prices suggested that if free trade in grain had been in place in 1975, prices would have been so much less variable ­ because trade could mitigate local supply variability ­ that only negligible quantities of carryover/storage would have been profitable. A subsequent study of global food trade provided complementary results: using a stochastic model of world markets for grains, livestock products and sugar, Tyers and Anderson (1992, Table 6.14) found that instability of international food prices in the 1980s was three times greater than it would have been under free trade in those products. Such government intervention is in response to lobbying efforts from and society's concern for groups destined to otherwise lose from exogenous shocks (Thompson et al. 2004, Freund and Ozden 2008) ­ although it needs to be kept in mind that stabilizing prices is not the same as stabilizing real incomes of the target households. An additional justification sometimes given for such intervention in poor countries is that credit markets are underdeveloped, or inefficient because of local monopoly lenders, so low-income consumers and producers have difficulty smoothing their consumption over time as prices fluctuate. However, it is difficult for governments to stabilize even prices, let alone incomes for target households. Indeed, more than 60 years ago Hayek (1945) warned that such intervention is likely to lead to government failure that could reduce welfare more than the cost of the market failure it seeks to overcome, given the high cost of the information needed to do it well. The extensiveness of that required information is made clear in the seminal theoretical study by Newbery and Stiglitz (1981), the econometric work by Deaton and 4 Laroque (1992) and by the studies of storage in particular by Williams and Wright (1991) and Deaton and Laroque (1996). One way countries try to achieve their stabilization objectives is by varying the restrictions on their international trade in food according to seasonal conditions domestically and changes in prices internationally. Effectively this involves exporting domestic instability and not importing instability from abroad. A simulation exercise by Tyers (1991) suggests that between three-fifths and three-quarters of the global cost of OECD agricultural protection is due to the insulating component of high-income countries' policies. That practice has continued unabated despite the signing of numerous multilateral and preferential trade-liberalizing agreements over the past two decades. Another indication of how much governments are intervening for stabilization purposes is provided by annual estimates of national nominal rates of assistance (NRAs). A NRA captures the extent to which the domestic producer price differs from the most comparable international price at a country's border. NRAs have been measured for all major farm products in more than 70 countries from the mid-1950s to 2007 in a recently completed study summarized in Anderson (2009). That study provides an ideal annual database for analysing government responses to food price spikes, which is fully detailed in Anderson and Valenzuela (2008). As a prelude to looking at those distortion estimates, it is illustrative to compare the movements in their two components, namely domestic and border prices. Pursell, Gulati and Gupta (2009), for example, report pertinent prices for rice in India which reveal that the Indian government has been able to maintain an almost-constant real domestic rice price for decades despite huge fluctuations in the international price of rice. Similar if less-complete attempts have been made by most governments of South and Southeast Asia and in parts of Sub-Saharan Africa where rice is also a key food staple (Figure 2). When averaged across all developing countries, Figure 3 reveals that that insulating behavior of governments hugely reduced the rise in domestic prices relative to the spike in international prices for rice (and also for wheat). As a result, especially since Asia produces and consumes four-fifths of the world's rice (compared with about one-third of the world's wheat and maize), this market- insulating behavior of Asian and to a lesser extent African policy makers means that very little rice production has been traded internationally: less than 7 percent in 2000-04 (and less than 5 percent pre-1990), compared with 14 and 24 percent for maize and wheat. This insulating behavior of governments also means international prices are much more volatile for rice than for those other grains, as is evident for the two periods shown in Figure 1. 5 [Insert Figures 2 and 3 about here] To get a sense of how much this practice varies across products, whether it is mirrored by the opposite adjustment in times of downward international price spikes such as in the mid-1980s, and whether it has changed much since policy reforms began around the mid- 1980s, Table 1 reports the average across countries of the percentage point deviation each year of national NRAs for 12 key farm products around their trend value for the sub-periods before and following each of those two spike periods (1972-76 and 1984-88). For the majority of those products that indicator is higher for the period since the late 1980s than in the earlier two non-spike periods. Those annual percentage point deviations since 1988 have been several times the average percentage NRA for the world as a whole (shown in parentheses in the first column of Table 1). During the upward price spike years 1972-76, the NRAs tend to be below the estimates for the periods either side of the spike, whereas in the mid-1980s when prices spiked downwards they tend to be above the NRAs in the more normal periods either side of that spike. [Insert Table 1 about here] That NRAs tend to be above trend in years of low international prices and conversely in years when international prices are high is also clear from Table 2, which shows the extent of the negative correlation between the NRAs for various products and their international price. That coefficient globally and in high-income countries is negative for all but beef, and even in the various developing country regions it is negative in all but one-quarter of the cases. For almost all of those 12 products the (negative) regional correlation is highest for the South Asian region. Among the developing countries it is again rice, sugar and milk that have the highest correlation coefficients. [Insert Table 2 about here] The proportional deviation of the international price of a farm product from its long- run trend value each year turns out to be highly significant when added to equations aiming at explaining NRA differences in the Anderson and Valenzuela (2008) database for those 12 products over the full time series from 1955 to 2007. The other prospective explanatory variables are per capita income, an index of agricultural comparative advantage (farm land per capita) and a dummy for distinguishing export industries. Anderson et al. (2010) show that the price deviation variable is highly significant for the three grains, sugar, cotton and coffee, but is not significantly different from zero for non-staple livestock products and soybean. 6 Yet another way of capturing this insulation phenomenon statistically is to estimate the elasticity of transmission of the international product price to the domestic market. Following Tyers and Anderson (1992, pp. 65-75), a geometric lag formulation is used by Anderson et al. (2010) to estimate elasticities for each product for all focus countries for the period 1985 to 2007. The average of estimates for the short run elasticity range from a low of 0.3 for sugar and milk to 0.5 for rice, wheat and pigmeat, 0.6 for cotton, cocoa, maize and poultry, and 0.7 for beef, soybean and coffee. The unweighted average across all of those 12 key products is 0.54, suggesting that within the first year little more than half the movement in international prices is transmitted domestically. Even the long run elasticity appears well short of unity after full adjustment: the average of the elasticities for those 12 products across the focus countries is just 0.69. 3. Changes in the extent of distortions during the food price spike of the mid-1970s We turn now to the responses to the most extreme food price spike prior to 2008, focusing on the five years around 1974. The effects of government actions are reflected in the movement of domestic relative to international prices and thus in estimated NRAs. These are reported by country in Anderson and Nelgen (2010, Appendix Table B) for the three key grains, for the important food staple of groundnut oil, and for all farm products for which estimates are available in Anderson and Valenzuela (2008). In the case of rice, for example, the NRA for developing countries as a group was 7 percent in 1972, but it fell to -27 percent in 1973 and - 55 percent in 1974 before recovering slowly to -24 percent in 1975 and -8 percent in 1976. 3.1 Nominal assistance coefficients To compare the extent of the fall and rise of food prices in developing countries with that in high-income countries (where assistance rates are much higher), it is easier if the NRA is converted to an NAC (nominal assistance coefficient, defined as 1+ NRA/100). The rice NAC for developing countries fell from 1.07 in 1972 to 0.45 in 1974, which is similar proportionally to the fall for high-income countries, whose NAC also more than halved over that short period, from 2.65 to 1.24 (Table 3). That means rice was still protected in high- income countries even in 1974 (NAC>1), whereas in developing countries its domestic price averaged less than half the international price in that year. [Insert Table 3 about here] 7 The NAC falls for wheat were not as severe as for rice, but were still substantial: from 1.24 in 1972 to 0.71 in 1973 for developing countries, and from 1.05 to 0.88 for high-income countries over the same 12-month period. For maize the NACs fell even less than for wheat, bottoming out a year later at 30 percent lower for developing countries and one-eighth lower for high-income countries. And the fall was smaller again for groundnut oil. Considering all covered farm products, the NAC for developing countries fell by exactly one-third in the first two years before rising by almost the same amount in the subsequent two years. This was a little more than twice the extent of the fall and recovery for high-income countries, and is due mainly to Asia's developing countries (see bottom segment of Table 3). What is remarkable is that these U-shaped paths over that 5-year period for this large sample of 75 countries are the mirror image of the inverted U-shaped paths of the international prices shown in Figure 1. This is so even if the depths of the former are not quite as big as the heights of the latter in proportional terms ­ confirming that there is at least some transmission of the international price changes to domestic markets. Table 4 reports for comparison the regional NRAs and regional consumer tax equivalents (CTEs) of agricultural policies. These two indicators are almost identical for high-income countries, reflecting the fact that most of the farm price distortions are coming from border measures in those countries. For developing countries, however, the CTEs are one-tenth lower initially than the NRAs (or rather their coefficient counterparts are), and they are negative on average, meaning consumers were paying less than they would under free markets. However, note that the CTEs for developing countries decline less than do their NRAs by 1974, before rising back to slightly above their pre-shock level by 1976. [Insert Table 4 about here] 3.2 Welfare and trade reduction indexes The cost of government policy distortions to incentives in terms of inefficiency of resource use and consumer spending tends to be greater the greater the variation of NRAs and CTEs across industries/products within the agricultural sector. It is helpful to have a single indicator of the overall welfare effect of each country's regime of agricultural price distortions in place at any time, so as to be able to trace its path over time and make cross-country comparisons. To that end, the family of indexes first developed by Anderson and Neary (2005) under the catch-all name of trade restrictiveness indexes has been drawn upon to generate indicators of distortions imposed by each country's agricultural policies on its economic welfare, and also 8 on its agricultural trade. Lloyd, Croser and Anderson (2010) define and estimate a Welfare Reduction Index (WRI) and a Trade Reduction Index (TRI) for the same 75 countries as included in this paper, taking into account the fact that for some covered products the NRA and CTE differ. As their names suggest, each of these two new indexes captures in a single partial equilibrium indicator the direct welfare- or trade-reducing effects of distortions to consumer and producer prices of covered farm products from all agricultural and food price and trade policy measures in place. Specifically, the WRI (or TRI) is that ad valorem trade tax rate which, if applied uniformly to all farm commodities in a country that year would generate the same reduction in economic welfare (or trade) as the actual cross-commodity structure of agricultural NRAs and CTEs for that country, other things equal. The WRI measure reflects the partial equilibrium welfare cost of agricultural price- distorting policies better than the NRA because it recognizes that the welfare cost of a government-imposed price distortion is related to the square of the price wedge. It thus captures the disproportionately higher welfare costs of peak levels of assistance or taxation within the farm sector, and is larger than the mean NRA/CTE and is positive regardless of whether the government's agricultural policy is favoring or hurting farmers. In this way the WRI and TRI go somewhat closer to what a computable general equilibrium can provide in the way of estimates of the welfare and trade (and other) effects of the price distortions captured by the product NRA and CTE estimates ­ and they have the advantage over a computable general equilibrium model of providing an annual time series. Summaries of those two indexes are provided for the major regions in the lower half of Table 4. In 1972 the WRI was almost 50 percent for developing countries and almost 60 percent for high-income countries. Over the subsequent two years, that index went down by nearly one-third in high-income countries but hardly changed for developing countries ­ even though the NAC for developing countries fell by one-third. Almost the same is true for the TRI: no change for developing countries, but a big drop (a halving) for high-income countries. This suggests for developing countries that the dispersion of NRAs among farm industries increased in individual developing countries, for example because NRAs for export industries became more negative to a greater extent than NRAs for import-competing farm industries became less positive. For high-income countries, if most of the assistance for farmers is provided by import restrictions such as variable levies to stabilize domestic prices, one would expect those levies to be reduced as international prices rise. 9 3.3 Contribution of different policy instruments to the aggregate indicators To test the possibility in the previous sentence, it is necessary to attribute each industry's NRA and CTE to a particular set of policy instruments. This is done in Table 5(a) for developing countries and Table 5(b) for high-income countries. The latter shows that in the mid-1970s import restrictions were indeed the dominant form of assistance to farmers in high-income countries, with domestic and export subsidies accounting for less than one-sixth of the total NRA. Between 1972 and 1974 it was both import restrictions and export subsidies that contributed to the fall in the NRA for high-income countries, as they did to its subsequent rise. [Insert Table 5 about here] For developing countries, by contrast, domestic price-depressing policies ­ especially export taxes ­ were the dominant policy instrument in 1972, although import restrictions and domestic producer subsidies played non-trivial roles as well. By 1974, though, import restrictions and domestic subsidies had been mostly suspended while the absolute contribution to the total NRA from export taxes and import subsidies hardly changed on average. As a result, the NRA for developing countries fell from 3 percent to -29 percent over those two years, before rising back to an average of almost zero again by 1976. The contribution to the CTEs by different instruments was fairly similar to that for NRAs and so is not shown, but they are almost as important as the NRAs in terms of their contribution to the WRI and TRI. To attribute those index values to different policy instruments requires doing it separately for the producer and consumer sides of the market: each instrument share is computed by first converting the index percentage to constant 2000 $US billions by multiplying it by the average value of production or consumption for that instrument group at the country level, and then dividing the dollar amount for each instrument by the country average value of production or consumption. The weighted average of the production and consumption components could give an overall attribution, but for the sake of transparency they are shown separately in Tables 6 and 7. [Insert Tables 6 and 7 about here] For high-income countries, their WRI fall of one-third and the halving of their TRI by 1974 are virtually all due to import measure changes, and the contribution from the consumption side is only a little greater than that from the production side. By 1976, however, both the WRI and TRI had risen by more than they had fallen in the previous two 10 years, and both production and consumption of farm products in high-income countries were more distorted immediately after than before the price spike. For developing countries, the situation summarized in Tables 6 and 7 is more complex. First, because the export tax instrument dominated in 1972 for developing countries, the contribution to both indexes from the production side is greater than from the consumption side. Second, even though their average WRI and TRI did not increase by 1974, each of the two component parts did and especially on the production side. And third, those increases are all due to increases in export restrictions and some import subsidies and are despite substantial falls in the contributions from import restrictions. 4. Lessons from the past Clearly a lot happened to global agricultural market distortions during the brief spike in international prices of farm products in the mid-1970s. By way of summary, the key points that emerged from the above analysis are the following: The magnitudes of the international price rises in 1973-74, and the speed of their subsequent fall, were very similar to those experienced in 2006-10; National NRAs and CTEs tend to be above trend in years of low international prices and conversely in years when international prices are high, reflecting the fact that trade restrictions adjust to reduce the fluctuations in domestic prices that would otherwise occur; The agricultural sector's NAC for developing countries fell by exactly one-third between 1972 and 1974 before rising by almost the same amount in the subsequent two years, which was slightly more than twice the extent of the NAC fall and recovery for high-income countries; By contrast, the WRI (TRI) for covered agricultural products went down by almost one-third (nearly halved) in high-income countries but hardly changed for developing countries between 1972 and 1974, even though the NAC for developing countries fell by one-third, suggesting that the dispersion of NRAs among farm industries decreased in individual developing countries; 11 Between 1972 and 1974 it was both import restrictions and export subsidies that contributed to the fall in the NRA for high-income countries, as they did to its subsequent rise; For developing countries, where export taxes were the dominant policy instrument in 1972 (although import restrictions and domestic producer subsidies played non-trivial roles as well), their import restrictions and domestic subsidies were mostly suspended by 1974 while the absolute contribution to the total NRA from export taxes and import subsidies hardly changed over those two years; For high-income countries, their WRI fall of one-third and the halving of their TRI by 1974 are virtually all due to import measure changes, but by 1976 both the WRI and TRI had risen by more than they had fallen in the previous two years, indicating that agriculture in high-income countries was more distorted immediately after than before that mid-1970s price spike. These findings for the mid-1970s, together with the broader findings in Section 2 of a tendency for governments to use border measures to reduce the variability of domestic food prices in general, are not dissimilar to what appeared to be the case in the most recent food price spike ­ although the data are not yet available to confirm that anecdotal impression. That beggar-thy-neighbor dimension of each government's food policies ought to be of concern. It is worrying because it reduces the role that trade between nations can play in bringing stability to the world's food markets: the more countries insulate their domestic markets, the more other countries perceive a need to do likewise, exacerbating the effect on world prices such that even greater changes in each nation's NRAs are desired. By increasing the volatility of world markets as they seek to reduce domestic volatility, such actions by national governments are collectively self-defeating. Clearly there is scope for multilateral agreement to desist. The World Trade Organization (WTO) is the most obvious place to seek restraints on varying trade restrictions (exports as well as imports), but the initial responses to proposals along these lines in the WTO's current Doha round of trade negotiations has been at best cool to date. It remains to be seen whether, over time, responses become warmer or even cooler with the expected increase in volatility of international food markets as climate change adds to the frequency of extreme weather events around the world. 12 References Anderson, J.E. and J.P. Neary (2005), Measuring the Restrictiveness of International Trade Policy, Cambridge MA: MIT Press. Anderson, K. (ed.) (2009), Distortions to Agricultural Incentives: A Global Perspective, 1955-2007, London: Palgrave Macmillan and Washington DC: World Bank. Anderson, K. and J.L. Croser (2009), National and Global Agricultural Trade and Welfare Reduction Indexes, 1955 to 2007, database at www.worldbank.org/agdistortions Anderson, K., J.L. Croser, D. Sandri and E. Valenzuela (2010), "Agricultural Distortion Patterns Since the 1950s: What Needs Explaining", Ch. 2 in Anderson, K. (ed.) (2010), The Political Economy of Agricultural Price Distortions, Cambridge and New York: Cambridge University Press. Anderson, K. and S. Nelgen (2010), "How do Governments Respond to Food Price Spikes? Lessons from the Past", Discussion Paper 1015, Centre for International Economic Studies, University of Adelaide, May. Anderson, K. and E. Valenzuela (2008), Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007, data spreadsheets at www.worldbank.org/agdistortions Deaton, A. and G. Laroque (1992), "On the Behavior of Commodity Prices", Review of Economic Studies 59(198): 1-23, January. Deaton, A. and G. Laroque (1996), "Competitive Storage and Commodity Price Dynamics", Journal of Political Economy 104(58): 896-923, October. Freund, C. and C. Ozden (2008), "Trade Policy and Loss Aversion", American Economic Review 98(4): 1675-1691, September. Hayek, (1945), `On the Use of Information in Society', American Economic Review 35(4): 519-30, September. Johnson, D.G. (1975), "World Agriculture, Commodity Policy, and Price Variability", American Journal of Agricultural Economics 57(5): 823-28, December. Lloyd, P.J., J.L. Croser and K. Anderson (2010), "Global Distortions to Agricultural Markets: New Indicators of Trade and Welfare Impacts, 1960 to 2007", Review of Development Economics 14(2): 141-60, May. Newbery, D.M.G. and J.E. Stiglitz (1981), The Theory of Commodity Price Stabilization: A Study in the Economics of Risk, London: Oxford University Press. 13 Pursell, G., A. Gulati and K. Gupta (2009), "India", Ch. 10 in Anderson, K. and W. Martin (eds.) (2009), Distortions to Agricultural Incentives in Asia, Washington DC: World Bank. Thompson, S.R., P.M. Schmitz, N. Iwai and B.K. Goodwin (2004), `The Real Rate of Protection: The Income Insurance Effects of Agricultural Policy", Applied Economics 36: 1-8. Tyers, R. (1991), "On the Neglect of Dynamics, Risk and Market Insulation in the Analysis of Uruguay Round Food Trade Reforms", Australian Journal of Agricultural Economics 35(3): 295-313, December. Tyers, R. and K. Anderson (1992), Disarray in World Food Markets: A Quantitative Assessment, Cambridge and New York: Cambridge University Press. Williams, J.C. and B.D. Wright (1991), Storage and Commodity Markets, Cambridge and New York: Cambridge University Press. 14 Figure 1: Indexes of real international prices of rice, wheat, maize and groundnut oil, 1972- 76 (1972 = 100) and 2006-10 (2006 = 100) (a) Rice 400 350 19721976 300 20062010 250 200 150 100 (b) Wheat 400 350 19721976 300 20062010 250 200 150 100 15 Figure 1 (continued): Indexes of real international prices of rice, wheat, maize and groundnut oil, 1972-76 (1972 = 100) and 2006-10 (2006 = 100) (c) Maize 400 350 19721976 300 20062010 250 200 150 100 (d) Groundnut oil 400 350 19721976 300 20062010 250 200 150 100 Source: Authors' derivation drawn from World Bank Pink Sheets of nominal prices deflated by the Unites States GDP implicit price deflator (see Anderson and Nelgen 2010, Appendix Table A). 16 Figure 2: Rice NRAs and international rice price, South and Southeast Asia, 1970 to 2005 (left axis is int'l price in current US dollars, right axis is weighted average NRA in percent) (a) South Asia 600 30 20 500 10 400 0 Intern. price (USD) -10 300 -20 NRA (%) -30 200 -40 100 Intern. Price in USD -50 NRA S Asia -60 0 -70 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Correlation coefficient is -0.79 Note: Countries included are Bangladesh (except for 1970-73), India, Pakistan, Sri Lanka. (b) Southeast Asia 600 30 20 500 10 400 0 Intern. price (USD) -10 300 -20 NRA (%) -30 200 -40 100 -50 Intern. Price in USD -60 0 -70 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Correlation coefficient is -0.56 Note: Countries included are Indonesia (except for 1970-74), Malaysia, Philippines, Thailand and Vietnam (except for 1970-85). 17 Figure 2 (continued): Rice NRAs and international rice price, South and Southeast Asia, 1970 to 2005 (left axis is int'l price in current US dollars, right axis is weighted average NRA in percent) (c) Sub Saharan Africa 600 50 500 30 400 10 Intern. price (USD) 300 -10 NRA (%) 200 -30 100 Intern. Price in -50 USD 0 -70 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Correlation coefficient is -0.72 Note: Countries included are Cote d'Ivoire, Ghana (except for 1983), Madagascar, Mozambique (except for 1970-75, 1997-98, 2002-04), Nigeria, Senegal, Tanzania (except for 1970-75), Uganda, Zambia. Source: Authors' compilation based on data in Anderson and Valenzuela (2008). 18 Figure 3: Indexes of real international prices and real producer prices of rice and wheat, developing countries' unweighted average, 1972-76 (1972 = 100). (a) Rice 400 International Price 350 Producer Price 300 250 200 150 100 50 0 (b) Wheat 400 International Price 350 Producer Price 300 250 200 150 100 50 0 Note: Countries included are: Argentina, Bangladesh, Brazil, Chile, Colombia, Cote d'Ivoire, Dominican Republic, Ecuador, Ghana, India, Indonesia, Kenya, Korea, Madagascar, Malaysia, Mozambique, Nigeria, Pakistan, Philippines, Senegal, Sri Lanka, Sudan, Taiwan, Tanzania, Thailand, Uganda, Zambia, and Zimbabwe. Source: Authors' compilation based on data in Anderson and Valenzuela (2008). 18 Table 1: Deviation of national NRA around its trend value,a 12 key covered farm products,b developing and high-income countries, 1965-72, 1972-76, 1976-84, 1984-88 and 1988-2004 (NRA percentage points) Developing countries High-income countries 1965- 1972- 1976- 1984- 1988- 1965- 1972- 1976- 1984- 1988- 1972 1976 1984 1988 2004 1972 1976 1984 1988 2004 Grains, oils, sugar Rice (28%) 24 32 36 62 64 54 48 89 222 227 Wheat (18%) 30 29 37 45 46 44 18 67 119 84 Maize (7%) 38 34 37 53 28 42 28 43 76 53 Soybean (3%) 36 30 56 75 124 10 11 111 142 41 Sugar (42%) 70 38 46 95 57 216 45 133 230 160 Tropical cash crops Cotton (-5%) 30 39 42 38 32 63 42 11 21 29 Coconut (-21%) 23 17 22 15 22 na na na na na Coffee (-12%) 31 37 50 44 25 na na na na na Livestock products Milk (88%) 68 61 87 101 61 187 188 281 327 156 Beef (43%) 32 38 59 63 51 54 57 103 167 117 Pigmeat (3%) 96 84 73 49 61 75 61 75 98 74 Poultry (21%) 111 90 112 63 76 44 46 135 221 192 a Deviation is computed as the absolute value of (residual ­ trend NRA) where trend NRA in each of the two sub-periods is obtained by regressing NRA on time. The global NRA average for each product for 1985-2004 is shown after the product name in column 1. b Unweighted average of national deviations. Source: Anderson et al. (2010). 19 Table 2: Coefficient of correlation between regional NRA and international price, 12 key covered farm products,a various regions, 1965 to 2007 South East High- South Asia and Latin income All focus Africa Asia China America countries countriesb Grains, oils, sugar Rice -0.19 -0.58 -0.51 -0.52 -0.10 -0.16 (0.99) Wheat 0.01 -0.81 0.09 -0.12 -0.28 -0.41 (0.85) Maize -0.20 -0.70 -0.55 -0.04 -0.29 -0.57 (0.71) Soybean -0.15 -0.42 0.16 -0.27 -0.07 -0.18 (0.30) Sugar -0.57 -0.74 -0.57 -0.40 -0.69 -0.70 (0.99) Tropical cash crops Cotton 0.28 -0.33 -0.16 -0.29 -0.74 -0.57 (0.96) Coconut na -0.16 -0.14 na na -0.12 (0.99) Coffee -0.35 na 0.02 -0.30 na -0.28 (0.99) Livestock products Milk 0.19 -0.57 -0.70 0.33 -0.10 -0.31 (0.98) Beef 0.20 na 0.05 0.55 0.29 0.32 (0.97) Pigmeat na na -0.53 -0.47 -0.60 -0.76 (0.98) Poultry 0.59 na -0.52 -0.78 -0.22 -0.34 (0.87) a Computed using the weighted average regional NRAs and a common international reference price for each product, from World Bank (2008). b Numbers in parantheses are the coefficient of correlation between the unweighted average regional NRAs and CTEs for individual covered products. For all covered products the coefficient is 0.93. Source: Source: Anderson et al. (2010). 20 Table 3: Nominal assistance coefficientsa for rice, wheat, maize, groundnuts and all farm products, key country groups, 1972 to 1976 (percent) 1972 1973 1974 1975 1976 Rice Asia (excl. Japan) 1.07 0.72 0.43 0.76 0.92 Africa 0.98 0.61 0.36 0.60 0.82 Latin America 1.00 1.04 0.84 0.90 0.98 All developing countries 1.07 0.73 0.45 0.76 0.92 High-income countries 2.65 1.84 1.24 1.72 2.33 Wheat Asia (excl. Japan) 1.31 0.72 0.83 0.97 0.94 Africa 1.19 0.85 0.67 0.87 1.00 Latin America 0.89 0.59 0.57 0.93 1.02 All developing countries 1.24 0.71 0.75 0.95 0.96 High-income countries 1.05 0.88 0.88 0.97 0.93 Maize Asia (excl. Japan) 1.52 1.17 1.04 0.82 1.03 Africa 1.03 0.90 0.69 0.83 0.92 Latin America 1.06 0.89 0.81 0.85 0.84 All developing countries 1.18 0.98 0.83 0.83 0.92 High-income countries 1.13 1.06 1.00 1.03 1.04 Groundnut oil Asia (excl. Japan) 1.08 1.00 1.00 0.69 0.86 Africa 0.51 0.46 0.48 0.53 0.57 Latin America na na na na na All developing countries 0.92 0.88 0.78 0.63 0.75 High-income countries na na na na na All farm products Asia (excl. Japan) 1.18 0.90 0.69 0.87 1.04 Africa 0.83 0.75 0.70 0.80 0.79 Latin America 0.73 0.69 0.58 0.65 0.85 All developing countries 1.01 0.83 0.67 0.82 0.96 High-income countries 1.32 1.19 1.14 1.25 1.36 a Nominal assistance coefficient = 1 + NRA/100 Source: Authors' compilation based on data in Anderson and Valenzuela (2008) 21 Table 4: NRAs, CTEs, WRIs and TRIs, all covered farm products, by region, 1972 to 1976 1972 1973 1974 1975 1976 NRA (%) Asia (excl. Japan) 18 -10 -31 -13 4 Africa -17 -25 -30 -20 -21 Latin America -27 -31 -42 -35 -15 All developing countries 1 -17 -33 -18 -4 High-income countries 32 19 14 25 36 CTE (%) Asia (excl. Japan) -5 -32 -28 -8 1 Africa -9 -18 -25 -13 -6 Latin America -19 -26 -35 -28 -9 All developing countries -9 -28 -29 -12 -2 High-income countries 32 18 14 26 36 WRI (%) Asia (excl. Japan) 47 48 46 50 41 Africa 55 52 53 48 49 Latin America 38 41 52 49 27 All developing countries 47 47 48 49 40 High-income countries 59 43 41 51 72 TRI (%) Asia (excl. Japan) 29 25 24 36 25 Africa 19 18 23 25 31 Latin America 21 25 33 33 16 All developing countries 25 24 25 34 24 High-income countries 27 18 13 24 36 Source: Authors' compilation based on data in Anderson and Valenzuela (2008) and Anderson and Croser (2009). 22 Table 5: Contributions to total agricultural NRAb from different policy instruments,a by region, 1972-76 (percent) (a) Developing countries 1972 1973 1974 1975 1976 Border measures Import tax equivalent 22 2 2 8 6 Export subsidies 4 0 0 1 1 Export tax equivalent -26 -18 -24 -22 -9 Import subsidy equivalent -6 -5 -5 -2 -1 ALL BORDER MEASURES -22 -21 -28 -16 -4 Domestic measures Production subsidies 26 7 1 1 1 Production taxes -1 0 -3 -3 0 farm input net subsidies 0 0 0 1 1 Non-product-specific (NPS) assistance except to inputs 0 0 0 0 0 ALL DOMESTIC PRODUCTION MEASURES 25 7 -1 -1 2 TOTAL NRA (including NPS and decoupled payments) 3 -14 -29 -17 -2 Producer subsidy equivalent, in real 2000 US$ billion 7 -43 -141 -72 -9 (b) High-income countries 1972 1973 1974 1975 1976 Border measures Import tax equivalent 25 18 15 21 30 Export subsidies 4 2 1 2 2 Export tax equivalent 0 -1 0 0 0 Import subsidy equivalent -1 -3 -3 -1 -1 ALL BORDER MEASURES 27 17 13 22 31 Domestic measures Production subsidies 1 1 0 1 1 Production taxes 0 0 0 0 0 farm input net subsidies 0 0 0 0 0 Non-product-specific (NPS) assistance except to inputs 1 0 1 1 1 ALL DOMESTIC PRODUCTION MEASURES 2 1 1 1 1 TOTAL NRA (including NPS and decoupled payments) 29 18 13 24 32 Producer subsidy equivalent, in real 2000 US$ billion 125 114 89 137 173 23 a In the absence of data, it is assumed the share of input tax/subsidy, domestic production tax/subsidy and border tax/subsidy payments for non- covered farm products are the same as those for covered farm products. b All entries have been generated by dividing the producer subsidy equivalent of all (including NPS and `decoupled') measures by the total agricultural sector's gross production valued at undistorted prices. c All entries have been generated by dividing the consumer tax equivalent of all measures by the total consumption value (at the farmgate level, valued at undistorted prices). Source: Authors' compilation based on data in Anderson and Valenzuela (2008) and Anderson and Croser (2009). 24 Table 6: Contributions to Welfare Reduction Index for covered products by different policy instruments, by region, 1972 to 1976 (percent) (a) Production side of economy High-income countries Developing countries 1972 1973 1974 1975 1976 1972 1973 1974 1975 1976 All measures 46 34 31 41 60 42 47 55 53 38 Border measures 46 32 31 41 58 38 44 44 43 35 Export tax 0 1 1 0 0 16 31 34 28 18 Export subsidy 6 3 1 3 3 3 1 0 1 1 Import tax 39 25 24 37 53 15 4 3 11 13 Import subsidy 1 3 5 1 2 4 8 7 3 2 Domestic taxes & subsidies 0 2 0 0 2 4 3 11 10 3 Production tax on output 0 0 0 0 0 2 1 10 8 1 Production subsidy on output 0 2 0 0 2 0 0 0 0 0 Farm input net subsidies 0 0 0 0 0 2 1 1 1 1 (b) Consumption side of economy High-income countries Developing countries 1972 1973 1974 1975 1976 1972 1973 1974 1975 1976 All measures 51 37 34 47 67 39 52 45 45 35 Border measures 51 37 34 47 67 39 51 44 43 35 Export tax 0 1 0 0 0 11 37 30 24 15 Export subsidy 5 2 1 2 3 3 1 0 1 1 Import tax 44 30 27 43 63 21 4 3 14 16 Import subsidy 2 4 6 1 2 5 9 10 4 3 Domestic taxes & subsidies 0 0 0 0 0 1 1 1 1 0 Consumption tax 0 0 0 0 0 1 1 1 1 0 Consumption subsidy 0 0 0 0 0 0 0 0 0 0 Source: Authors' compilation based on data in Anderson and Valenzuela (2008) and Anderson and Croser (2009). 25 Table 7: Contributions to Trade Reduction Index for covered products by different policy instruments, by region, 1972 to 1976 (percent) (a) Production side of economy High-income countries Developing countries 1972 1973 1974 1975 1976 1972 1973 1974 1975 1976 All measures 24 16 12 22 32 22 20 28 37 25 Border measures 24 16 12 22 32 22 24 27 33 25 Export tax 0 1 0 0 0 15 29 32 26 17 Export subsidy -4 -2 -1 -2 -2 -2 -1 0 -1 -1 Import tax 29 20 17 24 35 13 3 2 10 11 Import subsidy -1 -3 -4 -1 -2 -3 -8 -7 -3 -2 Domestic taxes & subsidies 0 0 0 0 0 0 -4 1 4 0 Production tax on output 0 0 0 0 0 0 1 1 4 0 Production subsidy on output 0 0 0 0 0 0 -5 0 0 -1 Farm input net subsidies 0 0 0 0 0 0 0 0 1 1 (b) Consumption side of economy High-income countries Developing countries 1972 1973 1974 1975 1976 1972 1973 1974 1975 1976 All measures 28 19 13 25 38 20 30 21 30 23 Border measures 28 19 13 25 38 21 30 21 30 23 Export tax 0 1 0 0 0 10 36 28 22 14 Export subsidy -3 -2 -1 -2 -1 -2 -1 0 -1 -1 Import tax 33 23 18 28 41 17 3 3 13 13 Import subsidy -1 -4 -5 -1 -2 -4 -9 -9 -4 -3 Domestic taxes & subsidies 0 0 0 0 0 0 0 0 0 0 Consumption tax 0 0 0 0 0 0 0 0 0 0 Consumption subsidy 0 0 0 0 0 0 0 0 0 0 Source: Authors' compilation based on data in Anderson and Valenzuela (2008) and Anderson and Croser (2009).