VI. Payments With Respect to Base Period Production of Certain Commodities But Not Others Are Not Inherently Product-Specific Support
29. Brazil has argued that production flexibility contract payments, market loss assistance payments, direct payments, and counter-cyclical payments are product-specific support. The United States had addressed some infirmities in Brazil’s approach in response to Additional Question 67bis from the Panel and in previous submissions.614 The United States now briefly addresses two arguments presented by Brazil.
30. First, Brazil argues that each of these payments is product-specific because base acreage is defined as acreage on which only some commodities were historically produced during a defined and fixed base period. This argument, again, rests on an “obsolete” definition of “in general” (in the definition of non-product-specific support) as “universal” or “without exception” and a determined refusal to quote accurately and interpret the definition of product-specific support in Article 1(a).615 That these payments are made with respect to base acreage for only some commodities is not relevant to the question whether they are support “provided for an agricultural product in favour of the producers of the basic agricultural product.”616 None of these payments satisfies either part of this definition: they are neither provided “for an agricultural product” (rather, they are made with respect to historic production of several products) nor “in favour of the producers of the basic agricultural product” (no production is necessary for payments to be made).
31. Brazil also appears to now argue that the requirement under paragraph 6(a) of Annex 2 that eligibility for payments under a decoupled income support measure “shall be determined by clearly defined criteria such as income, status as a producer or landowner, factor use or production level in a defined and fixed base period” requires that these payments be made to all producers for all commodities. This approach would seriously limit the ability of Members to move to decoupled income support. It is not clear that any Member would be willing to switch to decoupled income support if it required expanding support to whole new classes of producers or commodities. We can find no basis for this approach in the text of paragraph 6(a). This definition does not require comprehensive coverage of all or nearly all production in “a defined and fixed base period”; it merely requires “clearly-defined criteria.” Thus, under Brazil’s reading, a measure could satisfy the requirement of Annex 2, paragraph 6(a), and yet qualify as product-specific support under Article 1(a).
32. Second, Brazil again selectively quotes the statutory definition of “producers” to suggest that recipients of these payments had to be growers who “shared in the risk of producing a crop.”617 As the United States has previously noted, the statute defines “producers” (those eligible in the first instance to receive payment) as persons who “would have shared had the crop been produced.”618 Thus, both the 2002 and 1996 Acts make clear that a payment recipient need not produce any crop (including upland cotton) to receive payment. It is thus a serious error to imply that a payment recipient is necessarily a “producer” in the Agreement on Agriculture rather than a “producer” (meaning “recipient”) in the statutory sense.
33. Nowhere in Brazil’s submission is there any suggestion of how its approach can be found in the Agreement on Agriculture. It does not make sense of the definitions of product-specific support and non-product-specific support in Article 1(a), which Brazil has recognized guide the interpretation of the phrase “support to a specific commodity” in the Peace Clause. In sum, Brazil’s argument provides ample evidence that the phrase “support to a specific commodity” in the Peace Clause must be interpreted in the context provided by the Agreement on Agriculture. To divorce it from that context may result in an unworkable and illogical interpretation along the lines suggested by Brazil.
VII. Brazil’s New Arguments Relating to Crop Insurance Do Not Demonstrate that Crop Insurance Payments Are Product-Specific Support
34. Brazil presents a number of arguments claiming that crop insurance payments are “support to a specific commodity.” In part, this argument relies on the notion that such payments are not support provided to agricultural producers “in general” and, hence, not non-product-specific support. We note, however, that in making these arguments Brazil avoids any reference to the definition of product-specific support in Article 1(a). This is a fundamental interpretive error: Brazil cannot claim that payments are not support to “agricultural producers in general” under Article 1(a) without providing an interpretation of the other component of support in Article 1(a), namely, product-specific support (support “provided for an agricultural product in favour of the producers of the basic agricultural product”). In fact, given that crop insurance support is available to approximately 100 agricultural commodities, representing approximately 80 per cent of US area planted and greater than 85 per cent of the value of all US crops, crop insurance payments are not support “provided for an agricultural product.”619 The support to these approximately 100 commodities is the same: that is, the crop insurance premium subsidies do not vary by commodity or plan of insurance.
35. Brazil’s specific arguments fail to address the definition of product-specific support in Article 1(a); thus, each fails to demonstrate that crop insurance payments are “support to a specific commodity” rather than “support to several commodities.”
36. First, Brazil argues that certain policies (and accompanying premiums) on irrigation failures are available only to upland cotton and a few other commodities. The United States has previously addressed this argument and directs the Panel’s attention to that argument.620 37. Second, Brazil argues that a larger pool of types of insurance policies are offered to upland cotton than most other crops. Brazil has not explained how the types of crop insurance policies offered by private companies621 can affect whether US crop insurance payments (premium subsidies that do not vary by commodity or insurance plan) are product-specific or not. Brazil’s “facts” are also misleading in some instances and erroneous in others. For example, Brazil suggests that “in many instances, the policies available for cotton enterprises are not available for other crops.”622 However, we note that commodities other than upland cotton can be insured under the same types of policies as upland cotton.623 38. Third, Brazil argues that are there specific upland cotton provisions in certain policies.624 This is true – an insurance product offered by a private company must be tailored for the situation and desires of the insurance purchasers – but also irrelevant as the policies are generally similar in underwriting rules and share the same subsidy schedule.
39. Fourth, Brazil argues that upland cotton producer participation rates in the crop insurance programme “is much higher that for other crops.”625 We first note that Brazil neglects to mention that participation rates for the major field crops are generally quite high (over 75 per cent of insurable acres). Any producer who received disaster assistance was required to purchase federal crop insurance in the following year; cotton participation may be slightly higher because of droughts that have hit cotton regions in recent years. More importantly, that cotton producers may choose to take up crop insurance more than producers of other commodities might is irrelevant to whether the payments are provided “for an agricultural product.” Again, the crop insurance premium subsidy is identical for all commodities and for each plan of insurance.
40. Fifth, Brazil argues that tracking the cost of reinsurance provided to private companies is “further evidence that USDA treats crop insurance for upland cotton separately from crop insurance provided to other crops.”626 Of course, the way the US Department of Agriculture “tracks cost[s]” is irrelevant to the analysis of whether crop insurance payments provide support “for an agricultural product.” Brazil also misinterprets the Standard Reinsurance Agreement between the US Government and private insurers. Under that Agreement, net underwriting gains and losses for each insurer are calculated at the state level over all crops, not separately for individual crops (such as upland cotton).627 Thus, Brazil errs when it claims that reinsurance provides evidence that crop insurance for upland cotton is treated separately from crop insurance provided to other crops.
41. Sixth, Brazil claims that “the 2000 ARP Act denies subsidies to producers of other agricultural products.”628 It is true that there are certain products for which policies have not been developed. However, development of new policies is ongoing; for example, provisions of the Agricultural Risk Protection Act of 2000 allow for the development of livestock insurance products. A number of livestock products are currently available on a pilot basis, including price insurance for hogs and feeder cattle and gross margin insurance. We also note that producers may currently insure livestock and dairy revenue as part of whole farm insurance offered through the Adjusted Gross Revenue Insurance.629 Finally, Brazil’s argument here again reads domestic producers “in general’ to mean “universally” or “without exception”; as noted above, that definition is now considered obsolete.
42. Finally, with respect to Brazil’s references to the literature on the effects of crop insurance on production,630 the findings are (contrary to what Brazil has claimed) mixed. While several studies (such as those cited by Brazil) have suggested crop insurance payments may have a slight effect on acreage, the effects on production are less clear.631 If crop insurance encourages moral hazard problems (as claimed by Brazil), crop yields will be adversely affected as producers attempt to increase crop insurance indemnities. If moral hazard and adverse selection problems are severe, crop insurance support could potentially have a negative effect on production.632 The potential production effects of crop insurance payments, moreover, goes to whether such payments are “amber box” support but does not figure in the question whether such payments (which are offered at the same rate across commodities and policies) can be support “for an agricultural product.”
VIII. Brazil May Not Act Unilaterally on Procedural Matters
43. The United States takes note of Brazil’s statement in its 25 August 2003 letter to the Panel633 that, concerning paragraph 20 of the Panel’s determination of 20 June 2003, “Brazil interpreted this ruling as permitting it to provide no later than 22 August all of its evidence and argument that had not already been provided in its earlier submissions. This is the manner in which it treated the new evidence and arguments presented by the United States in its Rebuttal Submissions.” The United States is unable to reconcile Brazil’s position concerning its own ability to provide evidence and arguments at any time up through August 22 with Brazil’s repeated assertions that the United States “should have” provided particular material in its replies to the Panel’s questions.634 There is of course no basis for Brazil’s assertions that particular material “should have been” provided in replies to questions rather than in a rebuttal submission. There is no basis for Brazil to dictate to another Member what it may or may not include in its rebuttal submission. Brazil is fabricating an obligation and attempting to impose it on the United States at the same time that it exempts itself from this obligation. In this, Brazil’s approach is similar to its repeated attempts in this dispute to add to the obligations in the Agreement on Agriculture and the Subsidies Agreement.
IX. Conclusion 44. The United States has demonstrated that using any measurement that reflects the support “decided” by the United States rather than factors (such as market prices) beyond the United States’ control, US support to upland cotton in marketing years 1999-2002 has not exceeded the 1992 marketing year level. The question then is whether the Panel will find that the United States has breached the Peace Clause simply because market prices were lower in some recent years than they were in 1992.
45. The United States submits that the Peace Clause must be interpreted in a way that permits Members to comply in good faith – that is, Members must be able to tell if they will breach the Peace Clause or not. Brazil’s budgetary outlays approach does not do that. Brazil’s approach would mean that Members could not know if they had complied with the Peace Clause until it was too late to do anything about it. The best way to interpret the Peace Clause in a way that allows Members to comply is to use the “support” as “decided” by a Member during the 1992 marketing year as the basis for comparison. Recognizing, as the United States believes is required by the Peace Clause text, that “decided” and “grant” cover only those parameters over which Members exercise control would also be consistent with this approach of allowing “good faith” compliance since it would permit Members to control whether their measures conformed to their obligations.
46. The United States has disciplined itself to grant support not in excess of that decided during the 1992 marketing year. Therefore, we are entitled to the protection of the Peace Clause, and we respectfully request the Panel to find that Brazil may not maintain this action challenging these conforming US measures.
2. US Department of Agriculture, Compliance Report for 1992 Acreage Reduction Programme
SUBMISSIONS OF PARTIES AND THIRD PARTIES FOR THE
RESUMED SESSION OF THE FIRST SUBSTANTIVE MEETING
Annex E-1 Executive Summary of Brazil's Further Submission
Annex E-2 Executive Summary of United States Further Submission
Annex E-3 Third Party Further Submission of Argentina
Annex E-4 Third Party Further Submission of Benin and Chad
Annex E-5 Third Party Further Submission of Canada
Annex E-6 Third Party Further Submission of the European Communities
Annex E-7 Third Party Further Submission of New Zealand
Annex E-8 Third Party Further Submission of Paraguay
ANNEX E-1 Brazil’s Further Submission to the Panel
Summary Of The Argument Regarding Brazil’s Further Claims 1. Brazil demonstrates in its Further Submission that US subsidies from MY 1999-2007 supporting the production, use and export of US upland cotton cause or threaten to cause serious prejudice to the interests of Brazil within the meaning of Article 5(c), 6.3(c) and 6.3(d) of the SCM Agreement as well as violate GATT Article XVI.
2. The measures challenged by Brazil comprise domestic support subsidies including the marketing loan programme635, crop insurance subsidies, market loss assistance payments and their successor counter-cyclical payments, production flexibility contract payments and their successor direct payments, cottonseed payments and “other payments”.636 The measures also include prohibited export and local content subsidies including Step 2 export and domestic payments, and the subsidies provided by the US GSM 102 export credit guarantee programme. These collective subsidies are referred to as “the US subsidies”.
3. Table 1 summarizes the amounts of US subsidies in terms of US dollar and as a percentage of subsidization in terms of the market value of US upland cotton:
4. Brazil’s actionable subsidy claims fall into two basic temporal and legal categories: first, claims of present serious prejudice resulting from subsidies provided in MY 1999-2002; second, claims of threat of serious prejudice from subsidies that are required to be paid by USDA to the US upland cotton industry during MY 2003-2007 under the Farm Security and Rural Investment Act of 2002 (2002 FSRI Act) and the Agricultural Risk Protection Act of 2000 (2000 ARP Act).
5. Because Brazil’s claims involve the adverse effects of “subsidies”, Brazil first establishes that each of the US domestic and export programmes is a “subsidy” within the meaning of Article 1.1 of the SCM Agreement. As Table 1 above indicates, each provides a “financial contribution” which confers a “benefit”.
6. Each of the subsidies is “specific” within the meaning of Article 2.1(a) and/or (c) of the SCM Agreement. The United States and Brazil agree that the specificity test of Article 2.1 was not intended to function as a loophole through which narrowly focused subsidies provided to or used by discrete segments of an economy could escape review. The four direct subsidies – the PFC, direct payment, market loss assistance, and CCP payments are de jure specific because the 1996 FAIR Act and the 2002 FSRI Act explicitly limit access of payments to holders of upland cotton base acreage. The per acre upland cotton base acreage payments of the four programmes are single specific subsidies because the payments are based on individual cotton-based criteria and are significantly higher than payments to most other base acres in each program. Alternatively, even if the total payments to all base acres for each of the four direct subsidies are treated as single subsidies, they are specific because the payments are excluded from the significant majority of US farmland and the value of the crops produced with such payments is less than one-quarter the value of total US commodities.
7. Crop insurance subsidies are specific because there are specific policies and groups of policies available only for upland cotton (or a few other crops) and not for the majority of crops in the programme. Alternatively, crop insurance is not specific because the 2000 ARP Act denies benefits to commodities representing more than half of the value of US agriculture. Further, US crops represent only 0.8 per cent of total US GDP.
8. Within each of the present serious prejudice and threat of serious prejudice claims, Brazil has asserted three different claims regarding the application of these actionable subsidies.
9. Present Significant Price Suppression – Article 6.3(c) of the SCM Agreement:The US subsidies provided during MY 1999-2002 cause present significant price suppression in the world and Brazilian market, as well as in markets where Brazilian producers export. Brazil first establishes that Brazilian upland cotton is “like” US upland cotton based on common tariff classification, USDA’s designation of upland cotton as a separate commodity, and the interchangeability and treatment by the markets of US and Brazilian cotton as like products.
10. The bulk of Brazil’s price suppression analysis involves establishing the causal link between the US subsidies and suppressed prices in the US, world, Brazilian, and other markets. Most of this evidence is also relevant for Brazil’s other present and threat of serious prejudice claims. The enormous size of the MY 1999-2002 US subsidies in terms of amount ($12.9 billion) and percentage as market value (95 per cent) coupled with the dominating US world market share of 41.6 per cent of a fungible commodity create a de facto presumption of production, export, and price-suppressing effects. The effects of the subsidies are also seen in the significant increase of US production, exports, and world market share in MY 1999-2002 while US, A-Index, and Brazilian prices fell to record lows and remained suppressed. The causal link is further confirmed by the fact the average total US upland cotton costs of production was 77 per cent higher than market prices received by US farmers in MY 1999-2001, at the same time that US production and exports increased remained at high levels. Another demonstration of the causal link is the 15 per cent increase in the value of the US dollar between MY 1999-2001 at the same time that US export market share increased 68 per cent and A-Index prices declined by 21 per cent.
11. The link between US subsidies and suppressed A-Index, Brazilian, and other third country upland cotton market prices is further confirmed by the fact that the nature, size and global impact of the US market permits it to drive and suppress world prices. The large size of the US world market share of 41.6 per cent that is generated and sustained by the US subsidies suppresses world market prices. Production and export developments in the US market are widely publicized and impact the New York Cotton Exchange’s futures market, which in turn directly influence and impact A-Index and Brazilian prices.
12. The link between the US subsidies and significant price suppression is also confirmed by USDA and other economists’ findings of the individual production, export and price-suppressing effects of each of the US subsidies have . USDA and other economists have identified the US marketing loan payments as having the greatest production, exports and A-Index price-suppressing effects. Crop insurance subsidies, like marketing loan payments, are directly tied to production, and have been found to create similar types of effects by USDA and other economists, as they eliminate risk and induce farmers to put marginal land into production. CCP payments (and to a lesser extent market loss assistance, PFC and direct payments) have production impacts as a result of additional income and wealth effects that keep land in production and maintain base acres in production in anticipation of future base updating like that permitted under the 2002 FSRI Act. Step 2 export payments directly stimulate US exports and permit US exporters to export high-cost US upland cotton with the effect of suppressing A-Index prices. The GSM-102 export credit guarantee programme facilitated the export of more than $1 billion worth of US upland cotton between FY 1999 and the present, thereby increasing US exports and suppressing world prices.
13. A number of econometric studies found that the US subsidies collectively have the effect of increasing US production, exports and suppressing world prices. The ICAC found that world prices would increase by 11 cents per pound price from the removal of US subsidies in MY 2001 and 6 cents in MY 2000. The University of Tennessee found an average 11.4 per cent price suppressive effects from removing US subsidies during MY 2003-2007. The IMF estimated world prices would increase by 4 per cent based on removing $1 billion of US subsidies in MY 1998. The World Bank/IMF found a 25-30 per cent increase in world prices from elimination of US upland cotton subsidies in MY 2003-2007. The Centre for International Economics found an increase of world price by 13.4 per cent by eliminating US and Chinese subsidies in MY 1998. Professor Daniel Sumner determined that removing the US subsidies in MY 1999-2002 results in an average increase in the A-Index price of 12.6 per cent.
14. Further, Brazilian prices for domestic Brazilian sales as well as Brazilian export sales are suppressed by the effects of the US subsidies. The small size of the Brazilian market and low applied tariffs mean that Brazilian producers are price takers, not price makers. Negotiations to determine Brazilian domestic and export prices are heavily influenced by New York Cotton Exchange’s futures prices and A-Index prices. The movements in prices between the Brazilian prices and prices in these international markets are closely tied. Prices in other third country markets also show a close linkage with Brazilian, New York futures, US spot, US A-Index and other prices included in the A-Index calculation.
15. Finally, the amount of price suppression between MY 1999-2002 as reflected in various econometric studies of world prices varies from 6 cents per pound to 11 cents per pound. Professor Sumner found the A-Index price suppression to be 6.5 cents per pound between MY 1999-2002. This estimated worldwide price suppression is “significant” because it materially affects producers in Brazil and throughout the world. Total income loss from the price suppression is $3.587 billion and Brazilian producers lost an estimated $478 million from suppressed prices.
16. Increasing World Market Share in MY 2001 – Article 6.3(d) of the SCM Agreement:The US subsidies for upland cotton contributed significantly to the production and export of large quantities of upland cotton in violation of Article 6.3(d) of the SCM Agreement. The three-year average US world market share in MY 1998-2000 was 22.3 per cent. In MY 2001, the subsidy-enhanced US world market share increased to 38.3 per cent. This MY 2001 increase follows a consistent trend since the 1996 FAIR Act was enacted with US world market share increasing from 25 per cent to 38.1 per cent. The evidence linking US subsidies and increasing production and exports in the price suppression analysis is also relevant for the Article 6.3(d) claim. Professor Sumner finds that but for the US subsidies, US exports between MY 1999-2001 would have declined on average by 39 per cent.
17. Inequitable World Export Share - 1999-2002: Articles XVI:1 and 3 of GATT 1994:US subsidies provided from MY 1999-2002 contributed significantly to the United States having more than an equitable share of world export trade within the meaning of GATT Article XVI:3. The US share of world exports increased from 17.93 per cent in MY 1998 to 38.3 per cent in MY 2001 and increased further to 41.6 per cent in MY 2002. The causal link between the US subsidies and the increased US export market share is based on the evidence of production and export effects of US subsidies set out in the price suppression analysis. Professor Sumner concluded that but for the US subsidies US exports would be 41.2 per cent lower and US production would decline by 28.7 per cent on average between MY 1999-2002. The current US share of world exports of upland cotton is not “equitable” because producers from countries competing with the United States do not receive any or at most only small amounts of subsidies and have costs of productions are far lower than the United States. But for the effects of the US subsidies, these producers, including producers in Brazil, would have increased their share of the world export trade.
18. Threat of Serious Prejudice:Brazil’s second set of adverse effects claims involves the demonstration of a threat of serious prejudice under Article 5(c), 6.3(c), and 6.3(d) of the SCM Agreement and GATT Articles XVI:1 and 3. The record shows that five US subsidies – marketing loan payments, crop insurance subsidies, Step 2, CCP and direct payments – have no production or expenditure limitations and either no or at best no practical payment limitations. These unlimited subsides are required to be paid by USDA and eligible US producers, users and exporters. These eligible recipients have an enforceable entitlement to receive the payments – regardless of the effect of US subsidies on the world upland cotton market. Based on the EC – Sugar Exports GATT panel decisions and the US – FSC Appellate Body decision, these facts support a finding of a threat of serious prejudice in the form of significant price suppression, increases in US world market share, and inequitable share of world export trade.
19. A threat of serious prejudice is confirmed by fact that the level of US subsidies has increased by up to 10 cents per pound between MY 2001 and 2002 with the passage of the 2002 FSRI Act. Having established present price suppression, increased world market share and inequitable share of world export trade, this evidence also confirms the existence of a threat of serious prejudice for MY 2003-2007. A University of Tennessee study predicts removal of the US subsidies will increase US prices by 11.4 per cent between 2003-2007. The IMF predicts that removal of US subsidies would increase world market prices by 25-30 per cent in the short term. Professor Sumner predicts that removal of US subsidies would increase world A-Index prices by 5.9 cents per pound, decrease US production by 4.5 million bales, and decrease exports by 4.4 million bales.
20. A threat of serious prejudice also exists because the US costs of production will increase between MY 2003-2007 with USDA and FAPRI not expecting the large cost-market revenue gap to decline significantly. The most recent FAPRI baseline suggests that marketing loan and CCP payments will be made throughout MY 2003-2007. Additional evidence of a threat of serious prejudice exists from USDA and FAPRI baselines (reflecting mandatory payments under the 2002 FSRI Act) that project that US acreage, production, and exports will continue at existing high levels given the existence of the US subsidies. The increasing export orientation of US production, as US domestic textile production declines, also increases the threat of significant price suppression, increased world market shares, and inequitable shares of world export trade.
21. The evidence demonstrates that the threat of an increased US world market share in MY 2002 has already materialized, as the US world market share continued to increase in MY 2002 to 41.6 per cent, well above the MY 1999-2001 three-year average of 29.1 per cent. The threat also exists for MY 2003 as recent USDA projections of US exports indicate that the likely US share will be 38.8 per cent in MY 2003 – an increase over the three-year (MY 2000-2002) average of 34.9 per cent. This evidence further supports the finding of a threat that the US share of world export trade will continue to be inequitable for MY 2003-2007.
22. Per Se Challenges to 2002 FSRI Act and 2000 ARP Act:Brazil also challenges certain provisions of the 2002 FSRI Act and the 2000 ARP Act – in as far as they relate to upland cotton – as per se violations of Articles 5(c), 6.3(c), 6.3(d) of the SCM Agreement and GATT Article XVI:3. In particular, Brazil challenges the mandatory provisions requiring the executive branch of the US Government to make marketing loan, Step 2 domestic and export, crop insurance, direct and counter-cyclical payments to eligible upland cotton producers, users and exporters. There is no statutory mechanism in any of the challenged statutes or regulations to limit the guaranteed payments when these payments cause serious prejudice or a threat of serious prejudice. The absence of any statutory upland cotton circuit breaker threatens to cause serious prejudice, including price suppression, increased US export and an inequitable US share of world upland cotton exports.