137 Further Submission of the United States, paragraph 94.
138 The Panel stated that: "[W]e must assess the 'effect of the subsidies' on the interests of another Member to determine whether serious prejudice exists, not the effect of 'subsidy programmes'. We note that at any given moment in time some payments of subsidies have occurred in the past while others have yet to occur in the future. If we were to consider that past subsidies were not relevant to our serious prejudice analysis as they were 'expired measures' while future measures could not yet have caused actual serious prejudice, it is hard to imagine any situation where a panel would be able to determine the existence of actual serious prejudice." Panel Report on Indonesia – Automobiles, paragraph 14.206.
139 See WT/DS219/AB/R, paragraph 80: " … we understand a POI to provide data collected over a sustained period of time, which period can allow the investigating authority to make a dumping determination that is less likely to be subject to market fluctuations or other vagaries that may distort a proper evaluation".
140 Further Submission of the United States, paragraph 97.
141 See also Further Third-Party Submission of New Zealand, paragraph 2.34.
142 Idem, paragraph 101.
143 See Annexes BRA-7 (ERS Data: Commodity Costs and Returns); BRA-257 ("Cost of Farm Production Up in 2003", USDA, 6 May 2003) and BRA-82 (USDA Agricultural Baseline Projections until 2012, USDA, February 2003, p.48).
149 First, Article 18.104.22.168 of the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994 ("Anti-Dumping Agreement"), which deals with the calculation of cost of production, singles out "non recurring items of cost which benefit future and/or current production" (emphasis added). Second, the Appellate Body has acknowledged that non recurring subsidies may be allocated over time. In US Lead Bar II, the Appellate Body found that it was permissible for an investigating authority in a countervailing duty proceeding to rely on a rebuttable presumption "that a ‘benefit’ continues to flow from an untied, non recurring ‘financial contribution’" (emphasis added). Third, the Report by the Informal Group of Experts to the Committee on Subsidies and Countervailing Measures, G/SCM/W/415/Rev.2 (15 May 1998), recommends that certain subsidies be expensed to the year of receipt and that the benefits from other subsidies be allocated over time.
150 For example, Brazil includes payments made to recipients that do not produce upland cotton and fails to allocate non product specific payments across the total value of the recipient firm’s sales. Brazil has not reduced the value of decoupled income support payments to account for the capture by landowners of those payments made to farms on which cotton cropland is rented (65 per cent of total cotton cropland). Further, Brazil has Brazil has not identified the value of the cotton export credit guarantees under the GSM 102 programme, conceding that it "is not in a position to quantify the benefit to the recipients that has arisen from the application of the GSM 102 export credit guarantee programme to exports of US upland cotton between MY 1999 2002".
151 First, footnote 17 to Article 6.3(d) provides an exception to the provision where "[o]ther multilaterally agreed specific rules apply to the trade in the product or commodity in question". This exception applies only to "trade" because "multilaterally agreed specific rules" would be unlikely to apply exclusively to domestic consumption; however, the use of the world "trade" in the footnote to Article 6.3(d) but not in the text of the Article itself suggests that "world market share" does not merely encompass shares in world "trade". Second, Article 27.6 speaks of a developing country Member reaching export competitiveness when its "share . . . in world trade of that product" reaches a certain level. This use of "world trade" stands in contrast to the phrase "world market share" in Article 6.3(d). Third, GATT 1994 Article XVI:3 uses the phrase "world export trade", which also stands in contrast to the phrase "world market share".
152 While US share of world consumption in MY2002 was projected to be higher than the preceding three year average, that increase has not followed "a consistent trend over a period when subsidies have been granted" – in this case, for purposes of argument, since the 1996 Act came into effect. Reversing direction every year since marketing year 1996 cannot constitute "a consistent trend".
153 First, to use either marketing years 1998 or 2001 as one end of a period for comparison contradicts Brazil’s own argument that the "period of investigation" should be marketing years 1999 2002. Second, marketing year 1998 was a year in which US harvested acreage and production were severely impacted by weather conditions, in particular, drought. The record shows record abandonment during that year (that is, the difference between planted acres and harvested acres). Thus, to begin a comparison of harvested acreage or production with marketing year 1998 will overstate any resulting increase. Third, marketing year 2001 was a year in which US production increased, primarily because of record yields (as Brazil has acknowledged). That is, while planted acreage increased over marketing year 2000 in large part due to the decline in expected returns from competing crops, production increased by a much greater percentage because of uncommonly favourable weather conditions. Thus, to end any comparison of production with marketing year 2001 will overstate any resulting increase.
154 Brazil has alleged that increased income can induce producers to take riskier choices, thus potentially increasing production and distorting markets. The economic literature suggests any such effects are empirically trivial. Recipients of decoupled payments use many market mechanisms to reduce their risk exposure in their farm operation. These strategies to manage risk reduce the extent to which changes in risk attitude due to decoupled payments, if any, will be evidenced in their production levels or demand for inputs.
155 Comparing marketing year 1999 planted acreage to base acreage, the ratio of planted to enrolled acreage, by region, in 1999 ranged from only 51% in the West to 141.25% in the Southeast. In the Southeastern United States (Alabama, Florida, Georgia, North Carolina, South Carolina and Virginia), for example, upland cotton planted acreage exceeded base acreage by over 1 million acres. In each of the other three regions, planted acreage was between 879,000 and 1 million acres less than base acreage. The variations by State are even more extreme.
156 (a) Specifically, several of these papers simply remove the full outlay of the marketing loan program. This implies that farmers at the time of planting knew what actual prices would be at harvest time. Brazil’s own expert recognizes that it is producers’ expectations of harvest season prices that drive planting decisions. Thus, using the full outlays will overstate the influence of the marketing loan programme on the planting/production decision when actual prices turn out to be below the expected prices at the time of the planting/production decision.
157 In three different submissions, Brazil presents three different per pound revenue figures derived from market revenue and US support programmes, and purports to represent this figure as average revenue received by upland cotton farmers in that year for every pound of cotton produced. This combined per pound figure in no way represents what a cotton farmer would have received – or even could have expected to receive – in the specific year in the way of government support. In addition, Brazil’s measure of revenue for upland cotton producers – revenue from sales of cotton lint and cottonseed – is incomplete. Revenue from all sources – commodity sales, contracts in futures markets, off farm employment, investment income – are needed to put the costs into perspective.
158 The marketing year 2000 harvest season futures price at planting time was 61.31 cents per pound, suggesting that the market expected prices in marketing year 2000 to recover from the previous year’s levels. For marketing year 2001, the harvest season futures price at planting time was 58.63 cents per pound (nearly the same as futures in marketing years 1999 and 2000), once again indicting that market participants expected prices in marketing year 2001 to recover from their marketing year 2000 levels. It is only in marketing year 2002 that persistent lower than expected farm prices translated into a lower harvest season futures price at planting. For marketing year 2002, the February average futures price for December delivery fell to 42.18 cents per pound. However, even in marketing year 2002, market participants expected prices to recover and run higher than the "lagged price" of 29.80 cents per pound suggested.
159 We also recall that Brazil failed to properly analyze marketing loan payments through its use of "lagged prices" instead of futures prices. During marketing years 2000 2003, lagged prices significantly understate the harvest season prices expected by producers, thereby inflating the expected effect of the marketing loan rate.
160 Brazil’s evidence under Article 6.3(c) must establish the volume of subsidized US upland cotton that is "in the same market" as Brazilian upland cotton, the extent of subsidization, and the prices of those respective products sufficient to establish its claim of "significant price suppression". Brazil has not even shown that for each foreign market, there have been any US exports of upland cotton.
161 Even if one were to look to the period since the 1996 Act when different subsidies were in place, there is no consistent trend over a period when those subsidies have been granted. The facts demonstrate that since marketing year 1996, US world market share has increased and decreased in alternating years, and US world market share in marketing year 2002 is lower than in marketing years 1996 1997. These data cannot support a finding of a consistent trend. Brazil seeks to evade these facts by ignoring the change in subsidies over the years and by interpreting "world market share" contrary to the ordinary meaning of those terms.
162 The effect of such higher market prices is vividly suggested by Brazil’s use of the January 2003 FAPRI baseline versus the November 2002 preliminary FAPRI baseline in Dr. Sumner’s new model. We, of course, strongly disagree with what we understand to have been the way in which Dr. Sumner has most recently modeled all of the US payments at issue, but we note that a mere change in baselines that increased the baseline A index price by an average of 4.24 cents per pound per year over MY 2003 2007 reduced the estimated impact of removal of all US subsidies on A index prices by nearly one third. Price movements since January 2003 would suggest that Dr. Sumner’s estimated impacts using more current data would be smaller still. For example, the January 2003 FAPRI baseline projected a 2003 marketing year A index price of 58.40 cents per pound while the year to date A index price has been 68.73 cents per pound, an increase of more than 10 cents per pound over the January baseline.
163 For example, Brazil has repeatedly argued that the challenged US subsidies provided $12.9 billion in support over marketing years 1999 2002; this figure was based on payments made under specific programmes, including decoupled income support with respect to upland cotton base acres only. Brazil also has argued that decoupled payments for upland cotton base acres (net of base acres not "planted to cotton") are all support to upland cotton irrespective of what is planted on the land now.
164 See Brazil’s Further Rebuttal Submission, para. 208 n. 344 ("Brazil agrees that the recurring subsidies at issue would be allocated to the year in which they are paid for purposes of a CVD analysis . . . .").
165 For example, the marketing year 2002 base acreage increase means that, on average over marketing years 1998 2001, 2.6 million acres of upland cotton were planted on farms without upland cotton base acreage or in excess of those farms’ upland cotton base acreage, suggesting that Brazil’s theory that upland cotton must be planted on upland cotton base acreage is not supported by the facts.
166 A recent study concluded that production flexibility payments had "no effects on agricultural production in either the short run or the long run". USDA, ERS, Decoupled Payments: Household Income Transfers in Contemporary US Agriculture, M.E. Burfisher and J. Hopkins, Eds. (February 2003), at 23. (See Exhibit US 53). Other studies cited in Exhibit US 23 and discussed in the US rebuttal and further rebuttal submissions suggest that the effects of decoupled payments on planted area are less than 0.5 per cent.
167 Under Articles 5 and 6 a Member cannot claim threat of serious prejudice using the "more than equitable share" standard because that standard is not enumerated in SCM Agreement Article 6.3(c). Therefore, under Brazil’s interpretation, a Member could show a threat of "serious prejudice" (under GATT 1994 Article XVI:1) by showing a threat of something that is not "serious prejudice" within the meaning of Articles 5 and 6.
168 The payments Brazil identifies as "mandatory" are "mandatory" only if price conditions are fulfilled. Thus, the likelihood that price conditions will be satisfied must be taken into account. The payments Brazil identifies are also not "unlimited". For decoupled payments, the payments are set by multiplying fixed base acres times fixed base yields times the fixed or statutory maximum payment rate. The challenged payments are also not unlimited because a "circuit breaker" in the 2002 Act could result in these "mandatory" payments not being made.
169 For example, if a recipient of decoupled income support can choose to produce cotton, something else, or nothing at all, the payment is not tied to production of a particular product. There is nothing in the Agreement on Agriculture to suggest that support may be at one and the same time "product specific support" and "non product specific support". Thus, in attributing part of the decoupled payments on upland cotton base acres to producers and part to non producers, Brazil concedes that such payments are non product specific support.
170 Brazil’s Further Submission, para. 432 ("When US upland cotton farmers plant their crop in spring, farmers expect a certain price level. But, by no means is it ensured that this price level will be accomplished. However, given the US subsidies, that is irrelevant. . . . . The single fact that these programmes exist ensures a guaranteed revenue amount from the production of upland cotton.") (italics added).
171 PC/IPL/12, circulated 2 December 1994 (exhibit US 99).
172 See, e.g., Exhibit US 99, paras. 1(c), 1(e), 1(i), 2; Table ES:1 and Supporting Tables ES:1 and ES:2.
173 See, e.g., Article 18.5, 18.6, and 18.7.
174 Notification, G/AG/N/USA/47, circulated 6 June 2003 (exhibit US 100).
175 Statement of Brazil Second Panel Meeting (2 December 2003), para. 81.
176 Statement of Brazil Second Panel Meeting (2 December 2003), para. 84.
177 Compare Statement of Brazil Second Panel Meeting (2 December 2003), para. 91, with Answer of Brazil to Panel Question 142 (October 27, 1993) paras. 95, 100.
178 US Further Rebuttal Submission (18 November 2003), para. 201.