COMMITTEE ON COMMODITY PROBLEMS
Rome, Italy, 11-13 April 2005
IMPACT OF OECD AGRICULTURAL AND TRADE POLICIES ON DEVELOPING COUNTRIES: EXPLORING ALTERNATIVE DEGREES OF DECOUPLING OF DOMESTIC PAYMENTS
1. The 64th Session of the CCP recommended that the Secretariat undertake further work on the impact of domestic and trade policies. Since that Session FAO has increased its capacity to examine commodity policies in two related activities. In one, it has upgraded its commodity policy modelling activity to quantify the impacts of specific policies on developing countries in order to respond to specific requests made by the Committee. In another activity, following from its mandate to monitor the implementation of commodity policies in accordance with agreed guidelines, it has joined a consortium of international agencies to construct a comprehensive set of agricultural policy indicators (API) for developing countries. The information generated by these activities is intended to enhance understanding of policy effectiveness, as well as to build capacity in developing countries to undertake these tasks on their own1.
2. Furthermore, FAO has undertaken a major study examining the linkages between trade reforms and food security at the conceptual and regional level, followed by detailed case studies for 15 countries on the linkages between trade-related economic policy reforms and food security.2 Studies have also been undertaken on the impact of import surges on developing countries. More studies on this subject are underway.
3.Various model-based studies on the likely impact of further trade liberalization, including the Doha Round agreement, on trade, welfare and income distribution in developing countries show, on the whole, that the gains from agricultural trade reform are large. Market access invariably appears to be the dominant channel through which the gains are realized, but also cuts in domestic subsidies – in conjunction with tariff cuts – contribute markedly to the expansion of market access and trade. Also importantly, there are significant gains to be made from South-South trade, following reforms. Two key channels through which the developing countries are shown to lose are negative terms of trade effects and loss of tariff preferences. With global reforms, world market prices of food products are expected to rise, leading to terms of trade losses for food importing countries and gains for net exporters. The analyses indicate that when the losses in trade preferences are taken into account, welfare impacts turn out to be negative for preference receiving countries.. This negative impact is small overall, but large for some of these countries. A side event to this 65th CCP Session also focuses on these issues, providing Members with detailed current analysis of these issues3.
4.In the context of assessing the impact of policy reforms, an emerging issue is how new domestic payment programmes are being implemented, and specifically how they affect production and trade. There is a marked tendency by many countries to adopt such policies that purport to be decoupled or minimally trade distorting. As this is an area that has not as yet been studied and discussed adequately amongst Members, this document makes an initial and limited assessment of how important the degree of decoupling of domestic payments may be for world commodity prices and trade. The analysis of this paper attempts to put some bounds on the effects of the degree of decoupling of such programmes, and to trace their potential impact on developing and least developed countries.
5. The Uruguay Round Agreement on Agriculture (URAA) made major progress in establishing effective rules and disciplines in three basic pillars of government intervention in agriculture: domestic support, market access, and export competition. On-going discussion, research and negotiation have continued on the basis of these pillars. There is commitment to reduce export subsidies with a view to phasing them out. While no such end objective has been established for the other two areas, there is commitment to increase market access and to decrease trade distorting domestic support significantly over time. However, in the case of domestic support, minimal or non trade distorting domestic payments remain an area exempt from reductions.
6. Prima facie, the concepts of decoupling, non-trade distorting, or minimal trade distorting domestic support appear to be attractive ones. First, programmes with these characteristics are said to provide an efficient means to help achieve income distributional goals4. Second, in theory or at least by definition, they have no or minimal spillover effects on trade and international prices paid or received by others. For these basic reasons, almost all new programmes and reforms of existing programmes in developed countries in the past decade claim at least some non-trade distorting features.
7. However, against the apparent positive attributes, there are some critical issues. First, there is an issue of asymmetry in the ability of trading countries to implement decoupled policies. Many importing developing countries have limited fiscal capacity to implement such programmes. Policy reforms which reduce tariffs (increase market access) imply lower domestic prices and hence reduce rural incomes and increase adjustment pressures. They also reduce fiscal capacity by lowering tariff revenues. Hence with trade liberalization, these countries have less capacity to use exempted decoupled instruments to achieve income distributional objectives5. Developed countries may not have this problem.
8. A second issue is one of extent: what is “minimal trade effect” and how can it be measured? The degree of trade impact depends on programme nature and size. Given that many developing countries do not have the fiscal capacity to use exempted domestic support programmes, how much should they be concerned about their use by others, even if these programmes are claimed to be minimally distorting, or fit criteria negotiated in the WTO? Even if it can be shown that such programmes have less effect on markets than coupled domestic programmes, or restrictive market access measures, or export subsidy policies, what is the potential for the increasing trend for large domestic payments to affect world agricultural markets in ways that reduce the market opportunities of developing countries?
9. Since trade negotiations started to recognize the links between domestic policy and trade, many countries have undertaken policy reforms that have lowered market price support and direct price intervention and/or output based payments and have increased direct payments to producers, with varying degrees of linkages to prices and production. The trend is illustrated in Chart 1, where it can be seen that in OECD countries, market price support has fallen from 77 percent of the producer support estimate in 1986, to 62 percent in 20036. Market price support is determined fundamentally by market access measures (tariffs). Chart 2 illustrates the changing composition of domestic payments. Output and input based domestic payments have been relatively constant since 1986, while other forms of payments, largely based on fixed base production or area levels, historical entitlements, or income, have increased by some US$41 billion in the OECD area over this period. Increasing faster have been the payments based on historical entitlement. Such payments have also increased substantially following the European Union (EU) reform of its Common Agricultural Policy in 2003; starting in 2004, the EU Single Farm Payment for both crops and animals, which is replacing a considerable part of previous area payments and per head livestock payments, has been implemented.
10. The trend to decoupling of support/payments from production levels has also been evident in developing countries. Some examples of decoupling in policy reform include Mexico’s PROCAMPO reforms of 1994 in which guaranteed prices for cereals, oilseeds and protein crops were replaced with direct decoupled payments. In Turkey, various border measures, administered prices and output and input subsidies were replaced by a direct income support payment, starting in 2001. Other recent examples include China, where direct payments have been instituted for producers in the grain sector7.
11. The wording of the URAA provides ex-ante, legal criteria for programmes which have no or minimal production or trade-distorting effects that are exempt from reduction commitments in the Green Box. Decoupled income support programmes are defined by specific criteria on an individual basis. The agreement also provides criteria for support-limited programmes listed in the Blue Box which have also been exempt from reduction commitments. WTO negotiations are continuing to discuss the criteria for these boxes, with wide ranging views to improve specificity, flexibility, and to limit trade distortions8. Proposals include options to eliminate the Blue box, as well as to cap all payments, including those of the Green Box. Such proposals reflect concerns by Members about the potential impact of such payments.
12. In assessing domestic support programmes, economists have mainly used definitions of decoupling that concern the impact of programmes on different aspects of the market. These refer to how programmes affect both the position and the shape of a sector’s supply and demand curves with and without the programmes in place. Existing analyses (OECD (2001), Cahill (1997)) define “full-decoupling” as not only the situation where equilibrium production, trade and price remain unchanged, but also where the shape of the supply and demand curves remains unaffected by the programmes. The term “effective decoupling” is less stringent, and refers to a situation where equilibrium production and trade remain unchanged, even if the shapes of the supply and demand curves are altered (e.g. by production quota, price sensitive insurance schemes etc.). The difference between these two concepts largely concerns how short term domestic market outcomes are affected by demand and supply shocks. For example, a production limiting programme set at a historical production base level, may not affect the equilibrium production level of a country, but it may affect the supply response to short term price increases due, for instance, to high international prices.
13. It is increasingly acknowledged that neither fully decoupled nor effectively decoupled programmes actually can exist in practice9. Theoretical and applied analyses have examined how such programmes affect market agents. In this sense, the decoupling of programme payments from production decisions has come to mean the reduction of the degree of impact on production/price in a continuum, from fully coupled, to partially decoupled, to completely decoupled. There is concern that criteria which were thought to be sufficient to contain or minimize production and trade impact, may actually fail to do so as expected.
14. Assessing how programmes affect production is at least as complicated as the programmes themselves. They each have different attributes, from simple unconditional, lump sum transfers to others which may be market-state dependent, such as countercyclical programmes which make varying lump sum payments (on base area and yield) as price conditions vary. Accordingly, they have different effects on producers. The literature has variously attempted to categorize these effects. Generally, programmes have direct output effects by influencing the “supply inducing” price or return to the producer. These direct effects also may apply to commodities which are complements and substitutes in the production process. Programmes also may have indirect output effects by their impact on levels of income and wealth and hence on farm household labour allocation and on farm investment. The size and direction of these effects depend on the nature of national labour markets (mobility of labour from farm to non-farm, rural to urban), and capital markets (ability and cost to source investment funds). Both direct and indirect effects may be viewed in dimensions that refer to static (current situation), dynamic (expectations), and risk (insurance) related factors that affect production decisions. Producers who are risk averse can be shown to increase production if payments can be expected to offset market income losses, not only currently, but in future years. Such expectations not only depend on current programme design, but also on past experiences of how governments have changed domestic policy in response to market pressures of the day. Even theoretically decoupled programmes may become coupled if they adapt to new circumstances, in response to domestic pressures for support under adverse market conditions.
15. The economic literature on this topic, both theoretical and applied, is still relatively new10. Empirical understanding of new programmes is underdeveloped, since it takes time and experience to examine them using statistical methodologies which are required to test how recipients respond to payments. In the absence of such definitive analysis, most assessments rely on models that can accommodate measures of both the direct and indirect effects of programmes. In such approaches, extensive use is made of both simple assumptions regarding impacts, as well as sensitivity analysis around those assumptions. For example, a detailed simulation framework such as OECD’s Policy Evaluation Matrix (PEM) system11 specifies market supply and demand in a complete framework that includes detail on factor markets that enables a good understanding of both the direct and indirect effects of payments on commodity production and trade. Sensitivity analysis, moreover, looks at a range of key parameters in that system and provides corresponding ranges of the likely impact of programmes.
16. Some assessments have used decoupling assumptions applied to existing market models. These assumptions imply “price proportionality adjustments” that are applied to the supply functions for the various products that are eligible for payments. For example, in the supply function for a particular crop the price effect includes the direct payment for that crop multiplied by a proportionality constant. This constant may range from a value of 1 for a fully coupled payment to 0 for a fully decoupled payment. Similarly, a risk premium, which is subtracted from the price or revenue facing a producer, may also be applied. This risk premium would be zero or near zero for a situation where changes in risk exposure of farmers from programme implementation is small, but would increase if perceived risk increases, for instance as exposure to market price variations increases. Expression (1) below describes this specification more concisely, where S is supply, P is market price, DP is direct payment, and R is the risk premium. The coefficient n is a proportionality constant, with an assumed value between 0 and 1, according to its effect relative to market price, and the parameters a0 and a1, characterize the supply function.
S = a0 + a1(P + nDP – R) (1)
By specifying the proportionality constant and the risk premium in such a framework, parameters that are presumably functions of the specific domestic programme, one can calculate the effect on production, trade and hence on international markets. A similar approach has been followed by the OECD (2002) in examining the impact of the United States Farm Security and Rural Investment Act.
17. Todate, studies of the impact of decoupled and minimally distorting direct payments on markets indicate small effects on production and trade and international market prices. For example, a study published by the United States Department of Agriculture (2001) concluded (based on 1998 data) that removal of all domestic support, with direct payments mostly decoupled, would have increased world agricultural prices by +3.6 percent; assuming full coupling the price increase would have been +4.8 percent, a difference of 1.2 percent12. Furthermore, as noted in Dewbre (2001), fixed base area payments are assessed as the least trade distorting of domestic support instruments. These studies have made assumptions as to the supply response to these direct payment programmes. However there are doubts about the conclusion that these programmes have small effects on markets, which arise from the presumption that both the direct and indirect effects of these programmes are larger than these studies assume. Furthermore, over time direct payments are becoming more substantial. Indeed, as explained in Beghin et al (2004), there is considerable doubt concerning the decoupled nature of new programmes, given their size, design, and frequency of change.
18. For this paper FAO’s new agricultural commodity policy model has been used to explore, very tentatively, the impacts of alternative decoupling assumptions. This model (Commodity Simulation Model or “Cosimo”) has been developed in collaboration with OECD by using its Aglink model components for OECD countries, and by specifying 29 additional developing countries and regions, in order to provide greater capacity to look at policy impacts on these countries. The basis for analysis is a ten year projection period (2004-2013), compared to which impacts are assessed by changing assumptions and examining differences in a “shock minus control” method. The main purpose of analysis undertaken was to gauge the possible range of impacts on international markets that direct payment programmes might have according to alternative extreme decoupling assumptions. It is important to note that currently this model has not fully integrated the markets for sugar or cotton, and results do not, therefore, apply for these commodities.
19. Chosen for this illustrative analysis are two specific direct payment programmes in the United States and European Union. These were chosen because they are among the largest direct payment programmes which have various characteristics of decoupling. Furthermore, as of November 2004, none of these programmes in their current form had been notified to the WTO, and hence their formal classification into the Green or Blue Box categories has not been made.
20. In the United States the programmes examined are the fixed direct area payments (FDP) and countercyclical payments (CCP) for crops. The direct fixed payment is provided on a per unit area basis on a base area for all eligible crops, and was initially offered under the FAIR Act of 1996. The base area and yield data were updatable in the FSRI Act of 2002. The fixed payment amounts to some US$4.4 billion or about US$54 per base hectare. A countercyclical payment is provided to a producer on the basis of 85 percent of a fixed base area for each eligible crop, and given differences in a target price (excluding the fixed payment) and farm price for each crop. The primary aspect of these payments is that given base areas and yields by crop, payments are independent of current production. However, depending on expectations that base yields and areas may be updated in the future, this latter claim would be tenuous. Payments, nevertheless, do vary with market prices. If prices for eligible crops are relatively high, and assuming no upward adjustment to target prices, countercyclical payments will be low; if market prices are low payments may increase dramatically. Payments are not crop specific, but paid on a producer’s base area and yield. However, at average base yields the payment amounts could be as high as US$58/ha, US$113/ha, and US$30/ha for wheat, maize and soybeans respectively, given base yield, target price, fixed direct payment and loan rate levels. Payments do not depend on current production levels, but they are coupled to current prices over a specific range of values.13 In this sense, the CCP assures a more stable distribution of net farm revenues for programme crops.
21. In the European Union, the analysis includes the single farm payments for both crops and livestock. These were part of the reform of the Common Agricultural Policy of 2003, and have been implemented in many EU countries in 2004. These payments are determined on the basis of previous per hectare and per head programmes (2000 – 2002 average) under the CAP’s previous legislation. They amount to about €250/hectare for crops and about €560 per head for livestock14. While single farm payments have not yet been fully implemented across the European Union, they are in principle unrelated to current production and prices.
22. As the degree of decoupling of these programmes is difficult to ascertain, the model has been used to generate bounds of effects for these major programmes in order to illustrate where concerns may lie. Two scenarios were undertaken with the model to explore the possible range of impacts. These are compared to a base projection in which programmes are assumed to be fully decoupled in the sense of having the same market outcomes with or without payments.
23. The first scenario was constructed to examine the impact of introducing the programmes under the assumption that they are partially coupled. For this scenario it was assumed that in the United States domestic payments have a 9 percent equivalent effect on producer prices. In the EU, the effects were assumed to be 11 percent for crops and 6 percent for livestock. In both countries appropriate risk premia were assumed to be 2 percent to reflect lower risk that may prevail under the programmes. These values for n and for Rt in equation (1) above have been derived from work by OECD15, and are similar to those made currently in the literature which investigates the decoupling issue, and reflect background research used in the model. The scenario may be viewed as setting a plausible estimate on the impacts of these programmes given the existing literature on the subject. Table 1 provides detail on the nature of impacts on international prices production and trade for important groups of countries.
24. The results show that compared to a base of full decoupling, or no programmes, the institution of domestic payment programmes under an assumption of partial coupling suggested by existing research would have made small differences to market prices by year 2010. The maximum price effect is for wheat, where in 2010, world reference prices would be 1 percent lower than a projection where one assumed programs were fully decoupled. Note however, that while price effects are minor, production and trade effects are higher, indicating that even a modestly low price effect may cause displacement of production and trade on international markets. For instance in all basic foods and oilseeds the least developed countries (LDCs) are net importers (negative net exports) in the base simulation. The introduction of partially decoupled domestic payments by the United States and European Union, of the type discussed above, would expand the imports of wheat for this group by 1.2 percent and those of coarse grains by 15.8 percent over the base projections, while they would decrease the imports of rice by 1.3 percent, and oilseeds by 0.4 percent. The effects are more pronounced for the group of all the other developing countries (that is, excluding the LDCs), which includes many of the higher income developing countries. For this group imports of wheat, maize and oilseeds would increase by 1.0, 2.0, and 4.3 percent respectively, while their exports of rice would decrease by 3.4 percent.
25. A second scenario was constructed to examine the impact of assuming that the payments analysed were fully coupled: that is where the effect of per unit payment is equivalent to increasing market price and where risk premia are completely removed. Specifically the proportionality constants were assumed to be 1.0 rather than 0, and the risk premia were also assumed to be 0. These extreme assumptions are used to obtain an upper estimate of the effects, and not to suggest that such assumptions are appropriate or likely. They go beyond those found in the existing literature studying the issue.
26. The results of the second scenario are provided in Table 2. They indicate that with the extreme assumption that these programmes are fully coupled, the impact on international prices is considerably greater than the first scenario. The maximum effects are on wheat and coarse grain prices, at -6.8 percent and -6.1 percent respectively. It must be noted here that these effects could be larger in a projection in which baseline world crop prices were lower. In that case, the United States countercyclical payments would have been higher, and the assumption of full coupling to these larger payments would mean correspondingly larger effects.
27. In this extreme scenario under a full coupling assumption, the production effects are also more significant, although still relatively low in all countries, including the United States and the EU, given low supply responses to incentive prices that are commonly estimated in these countries. However, the trade effects are estimated to be quite large in many cases, particularly in countries/regions where base net trade levels are low compared to production/consumption. As Table 2 shows, developed country net-exports increase considerably (in this case due to United States and European Union increased net exports), and these result in large percentage reductions in exports of other economic groups. For LDCs the impact of fully coupled domestic payments in the United States and European Union is to increase imports of wheat and coarse grains by 7.8 and 81.7 respectively, while decreasing the imports of oilseeds by 10.9 percent. The same large effects on imports and exports also obtain for the other developing countries. The estimated response for each country or region depend not only on the extent of international price change, but also how open their markets are to such changes, and their own supply response characteristics for the crops affected the most. For example, in the wheat and coarse grain markets, a country or region’s response depends on the importance of these crops in land use, and on the type of domestic and trade policies in place. In the model underlying the scenarios shown, it has been estimated that the supply response is higher in least developed countries than in other developing countries. This is due largely to the presence of very large countries in the latter group, such as India and China, whose markets do not react as much to changing external prices. This implies that trade responses are larger for these large countries, and hence that their trade is affected more by changes in world prices.
28. There has been a significant trend toward the decoupling of domestic support, particularly in developed countries. As market price support measures decline due to increased market access (i.e. tariff reductions), the domestic pressures for direct payment support may increase. Trade negotiations continue to discuss the means to enable governments to make direct payments to producers, subject to specific criteria, which may be exempted from reduction commitments.
29. The objective of the analysis undertaken in this document was to examine the importance of decoupling assumptions in assessing the market impacts of domestic payment programmes. As much remains uncertain about how to assess the degree of decoupling, the paper examines the bounds of the impacts on markets from different degrees of decoupling of domestic payment programmes. As examples, specific direct payment programmes in the United States and European Union were analysed using FAO’s new world policy commodity model. The results of the analysis indicate that the estimated effects on international prices, determined by extreme assumptions of coupling or decoupling, may lie generally within a reasonably small range. They also indicate that depending on the degree of coupling and on the size and nature of the programmes involved, significant distortions to trade may be involved despite low international price effects. The implication is that assessing the degree of coupling of new types of reform programmes may be important.
30. In this study only specific programmes in two major producing jurisdictions are included in the analysis. While those analysed are among the largest such programmes in terms of expenditure, a greater movement on the part of many countries who can afford to re-instrument their support toward decoupled or minimally trade distorting programmes may lead, in sum, to a more significant impact on trade. This trend may leave those who cannot afford such programmes in a disadvantaged position as far as support to their farmers is concerned.
31. In light of the evidence provided, delegates are invited to address, inter alia, the following questions:
Aksoy, M., and J. Beghin (2005), Global Agricultural Trade and Developing Countries. World Bank. Washington.
Andersson, F. (2004) Decoupling: The concept and past experiences, Working Paper 1., IDEMA project, Swedish Institute for Food and Agricultural Economics.
Anton, J. and C. LeMouel. (2003), Do counter-cyclical payments in the FSRI Act create incentives to produce? International Conference of Agricultural Economists, Durban, August.
Baffes, J and H. de Gorter (2003) Decoupling Support to Agriculture: An economic analysis of recent experience, paper to Annual Bank Conference on Development Economics – Europe, Paris May 15-16, 2003.
Buetre, B, R. Nair, N. Che and T. Podbury (2004), The effects of liberalization on agricultural trade on developing countries trade, incomes and activity levels, ABARE Conference Paper 04.6, ABARE, Canberra, Australia.
Bouet, A, J-C. Bureau, Y. Decreux and S. Jean (2004), Assessing the impact of multilateral agricultural liberalization: the contrasted fortunes of developing countries in the Doha Round, INRA and CEPII, Paris.
Cahill, S., (1997), Calculating the Rate of Decoupling for Crops under CAP/Oilseeds Reforms, Journal of Agricultural Economics. No. 47 p 349-78.
Conforti, P. and L. Salvatici (2004), Agricultural Trade Liberalization in the Doha Round: Alternative Scenarios and Strategic Interactions between Developed and Developing Countries, FAO Commodity and Trade Policy Research Working Paper No. 10, FAO, Rome.
Dewbre, J., J. Antón, and W. Thompson, (2001) the Transfer Efficiency and Trade Effects of Direct Payments, Journal of American Journal of Agricultural Economics. No 5. p 1204-1214.
FAO (2004), Trade Reforms and Food Security: Conceptualizing the Linkages, 2004, FAO, Rome.
FAO (2005), The Linkages between Trade-related Economic Policy Reforms and Food Security: Synthesis of Case Study Findings, forthcoming, 2005, FAO, Rome.
Hennesy, D., (1998), The Production Effects of Agricultural Income Support Policies under Uncertainty. American Journal of Agricultural Economics. No 80. p 46-57.
Poonyth, D and R. Sharma (2003), The Impact of the Doha Development Round of Trade Negotiations on Developing Countries: Results from ATPSM. In Commodity Market Review 2003-04, FAO, Rome.
OECD (2004), Synthesis of Findings from Analyses of Global Commodity Market and National Welfare Effects, Paper presented to OECD Global Forum on Agriculture. December 2004.
OECD (2003), Agricultural Outlook, 2003-2008, “The Market Implications of the 2002 US FSRI Act”. Paris. pp. 47-52.
OECD (2001), Market Effects of Crop Support Measures.
OECD (2001), Decoupling: A Conceptual Overview. OECD Papers No. 10.
Poonyth, D, A. Sarris, R. Sharma and S. Shui (2004), The Impact of Domestic and Trade Policies on the World Cotton Market, FAO Commodity and Trade Policy Research Working Paper No. 8, FAO, Rome.
Sumner, D. (2000) "Domestic Support and the WTO Negotiations." Australian Journal of Agricultural and Resource Economics. No 44, pp 457-74.
USDA (2001), Agricultural Policy Reform – The Road Ahead, Washington.
WTO (2004) Agriculture negotiations: Where we are now. Geneva. August. Website www.wto.org
1 These activities are briefly described in documents CCP 05/CRS 4 and CCP 05/CRS 5.
2 See FAO (2004) and FAO (2005), and CCP 05/11.
3 To be held April 12, “The Impact of OECD Policies on Developing Countries: What Policies Matter Most?”
4 See Dewbre et al. (2001)
5 The International Monetary Fund has examined this issue, and has proposed the Trade Integration Mechanism, but this would deal with short term measures and technical assistance and not with on-going income distributional objectives. See WTO WT/L/565, May 2004.
6 OECD (2004).
7 For 2004 the Government has allocated about US$1.2. billion for this purpose.
8 WTO (2004)
9 See J. Baffes et al., (2003).
10 For a good review of the economics of payment decoupling see F. Andersson “Decoupling: The concept and past experiences” , Swe dish Institute for Food and Agricultural Economics. 2004.
11 See Dewbre J., et al., (2001).
12 USDA (2001)
13 The range is determined by the target price levels, the fixed payment level and the loan rates of the various crops. For example, the wheat target price is US$144/tn, the fixed payment is 19/tn, and the loan rate is 101/tn.
14 Calculations based on estimated program expenditure and base area and animal numbers.
15 Estimates taken from OECD Aglink model, as used in OECD (2003), Agricultural Outlook, 2003-2008, “The Market Implications of the 2002 United States FSRI Act”. Paris. pp. 47-52.