Paragraphs 114-117 10. The United States argues for the first time in paragraphs 114-117 of its Rebuttal Submission that using the “price-gap” methodology is the appropriate way to calculate the portion of upland cotton AMS that stems from marketing loan gains, certificate exchange gains, and loan deficiency payments (collectively known as “marketing loan payments”). The effect of applying the price gap methodology would be to transform the $2.5 billion in budgetary expenditures for marketing loan payments in MY 2001 into a “negative” amount for purposes of total current AMS.474 The United States bases its new argument on an alleged statement by Brazil and “agrees” that “Brazil is correct when it states that a non-exempt direct payment dependent on a price gap may be calculated using a price gap methodology, rather than budgetary outlays. . ..”475 11. The United States refers to Brazil’s 11 August Answer to Question 67, paragraph 130, as the basis for its assertion. Brazil’s statement cited by the United States reads as follows:
Brazil notes that the United States has notified the deficiency payments using the price gap methodology provided for in Annex 3. [footnote citing Exhibit Bra-150 (G/AG//N/USA/10)] Brazil considers it appropriate to follow the US decision and will therefore, calculate the amount of support to upland cotton provided by the deficiency payment programme by using the “price gap” approach detailed in Annex 3 paragraph 10 and 11.
Contrary to the US interpretation of this statement, Brazil’s point was that any calculation of AMS for deficiency payments (and for the other programmes that require such calculation) must be consistent with the actual choice of methodology originally made by the United States for calculating its domestic support reduction commitments as well as its yearly current AMS notifications. Indeed, the United States’ entire argument in paragraphs 114-117 of its Rebuttal Submission is based on the alleged obligation for Brazil “to be consistent.”476 As demonstrated below, it is Brazil’s AMS calculation, not that of the United States, that is “consistent.”
12. Members are required to notify annually their current total AMS to provide other Members the opportunity to review the consistency with their domestic support reduction commitments pursuant to Article 6.3 of the Agreement on Agriculture.477 The total AMS reduction commitments were negotiated during the Uruguay Round based on a calculation of “total AMS” provided in marketing years 1986-1988. The initial AMS calculation used for the purposes of the reduction commitments was performed pursuant to Annex 3 of the Agreement on Agriculture with Members choosing either budgetary expenditures or a “price-gap” methodology expressed in total monetary terms. Like Article 13(b)(ii), the comparison between current total AMS and the AMS ceiling, i.e., the reduction commitment, must allow for an “apples to apples” annual comparison in accounting for the same measures. It follows that once a Member uses a budgetary approach for one measure to establish the AMS ceiling, it cannot use a price gap approach for that same measure in calculating total current AMS. Instead, a Member is required to report current total AMS consistently with the choice it made for that particular type of support in its original total AMS calculation.
13. This interpretation of Annex 3 and Article 6.3 of the Agreement on Agriculture is consistent with its context and object and purpose. Opting for a methodology that would permit Members to change their original methodology (i.e., from budgetary to price-gap) could sanction what the United States proposes – the covering up of billions of dollars of marketing loan payments (originally calculated on a budgetary basis) and turns them into “negative support” by using a “price-gap” formula. This would be inconsistent with the entire reason for the reduction commitments of Article 6 of the Agreement on Agriculture.
14. Brazil’s calculation of AMS for, inter alia, marketing loan payments followed the actualdecision of the United States during the Uruguay Round478 as reflected in its notifications.479 During the Uruguay Round, the United States calculated the upland cotton portion of what would eventually become its domestic support reduction commitment by using the following methodologies: it used the price gap formula for upland cotton deficiency payments480 and used budgetary outlays for all other domestic support measures.481 In its MY 1995 notification to the Committee on Agriculture the United States similarly notified deficiency payments using the price-gap formula and using budgetary outlays for all other domestic support measures subject to reduction commitments482 consistent with its AMS calculation during the Uruguay Round. After the termination of the deficiency payment programme in 1996, all later domestic support (current total AMS) notifications of the United States for upland cotton only use budgetary outlays. Thus, Brazil’s approach to calculating upland cotton AMS for MY 1992 and 1999-2002 is entirely consistent with the US approach as evidenced in its domestic support notifications483 and with the US obligations under the Agreement on Agriculture.
15. The United States accounted for the marketing loan payments in the same manner as in its notifications when it answered Question 67 in its 11 August submission.484 This is, furthermore, the methodology the Panel indicated the United States should use – referring to the US notification of MY 1999 domestic support in G/AG/N/USA/43, in which the United States – in line with its obligation – used budgetary outlays.485
16. In sum, like many other US arguments in this phase of the dispute, this US argument is designed to cover-up expenditures and support to upland cotton that increased significantly since the US Uruguay Round commitments came into effect. Therefore, the Panel should reject it.
Paragraphs 124-127 and Exhibit US-24 17. The United States presents an additional critique of Professor Sumner’s analysis in Exhibit US-24 prepared by Dr. Joseph W. Glauber, the Deputy Chief Economists of USDA, as well as in paragraphs 124-127 of its Rebuttal Submission.
Marketing Loan Benefits 18. Dr. Glauber notes that the analysis of Professor Sumner did not include in the upland cotton acres eligible for marketing loans 447,164 acres of upland cotton planted on Flex acres from other programme crops.486 Dr. Glauber is correct that any such acreage would be eligible for receiving marketing loan payments and therefore should be included in the calculation.487 Dr. Glauber refers to “Acreage Reduction compliance reports” as his source of this number. Dr. Glauber does not provide a citation for these compliance reports and the United States has not made them available to Brazil. Therefore, Brazil cannot confirm the actual number of acres. Brazil also notes that the number listed is planted acres not harvested acres.488
19. Dr. Glauber further rests his finding that 100 per cent of US upland cotton production in MY 1992 benefited from marketing loan payments on his statement that upland cotton farmers “often” report land that had been planted and abandoned as land left idle and therefore never planted. No citation, authority or reference is provided for this assertion. The assumption in this assertion is that farmers report one thing to the US Federal Government, yet actually do something else. Thus, Dr. Glauber’s presumption appears to be that farmers engage in what would appear to be widespread misrepresentation. Brazil does not know if such assumed large-scale misrepresentations were legal under the 1992 US programme, but it certainly contradicts “programme” expectations.
20. Professor Sumner concluded that 1.99 million acres used to produce upland cotton in MY 1992 were not eligible for the marketing loan payments because they did not participate in the deficiency programme. Dr. Glauber confirms Professor Sumner’s general approach on non-participating acreage in footnote 1 on page 2 where he acknowledges that “some base building occurred during the early 1990’s.” What this means is that a substantial amount of upland cotton must have been planted outside the programme to accommodate expansion of upland cotton base by 200,000 acres in 1993, as identified by Dr. Glauber. This acknowledgement supports Professor Sumner’s analysis that a significant amount of upland cotton must have been grown outside the deficiency programme. The basis for this analysis is as follows:
21. Under the rules existing in MY 1992, in order to “build” base a farm was required to plant all of its upland cotton outside the programme.489 The expansion of upland cotton base is equal to one-third of the amount of additional cotton acres planted in each of the previous three years. An acre of base is added if an additional acre of cotton is planted for three consecutive years.490 In order to plant more than the current base, a farm was required to leave the programme altogether so that current base acres would also be planted outside the programme. For example, assume a farm had 1000 acres of upland cotton base and wanted to add – over the two-year period 1991-1992 – 200 acres of base by 1993. That farm would withdraw from the programme for those two years and plant 1300 acres of cotton (300 acres more than the previous base) in 1991 and 1992. The 1993 upland cotton base would than be calculated as follows: (1000 acres + 1300 acres + 1300 acres) / 3 equalling 1200 acres, thus 200 acres more than previously. Therefore, to add the 200,000 acres of base in 1993 (which Dr. Glauber stated were actually added) a much larger amount of upland cotton would have been required to be planted outside the programme in 1992. In addition to building of additional base, farmers planted upland cotton outside the programme because they did not comply with payment limit rules and for some more idiosyncratic reasons.
22. In summary, the evidence of an expanding base is consistent with the assessment of Professor Sumner that a substantial amount of acreage was planted to upland cotton outside the programme. This evidence is not consistent with Dr. Glauber’s undocumented or unsubstantiated claim that all upland cotton harvested was eligible for marketing loans.
23. To summarize, if the Panel were to accept the undocumented assertion by Dr. Glauber that 447,164 acres of cotton were planted on flex acres from other programme crop base acreage, then there would be 1.54 million acres (1.99 million acres – 0.45 million acres) that were planted to upland cotton but were not eligible for marketing loan payments. This 1.54 million represents 12 per cent of planted acreage.491 Thus, adjusting 52.35 cents per pound by 0.88 results in a support from the marketing loan programme of 46.1 cents per pound. This is an increase of 1.76 cents over the marketing loan level of support set forth in Appendix Table 1 to Professor Sumner’s 22 July 2003 Statement.
Deficiency Payments 24. Brazil has already rebutted the US argument that it is inappropriate to adjust the support provided by the deficiency payment programme by non-participation and the resulting non-eligibility to receive payments.492 25. The various “decisions” in MY 1992 with respect to the deficiency payment programme were calculated to establish rules that encouraged some producers to forego eligibility of the programme. Furthermore, the record establishes that US policy makers had relatively precise prior knowledge of how many acres would remain out of the programme based on their policy choices on required land idling and loan rates.493
26. Dr. Glauber criticizes Professor Sumner for relying on a programme yield of 531 pounds per acre to calculate the ratio of payment yield to expected yield and states that the true programme yield in MY 1992 was 602 pounds per acre.494 Dr. Glauber references a USDA press release that is not available to those outside the US government as his source of the payment yield information.495 Assuming that the figure of 602 pounds per acre used by Dr. Glauber is correct, he incorrectly continues to rely on Professor Sumner’s calculation of the expected yield by stating that the “expected yield based on an average of the upland cotton yields over the five preceding crop years is 601 pounds per acre.” Professor Sumner estimated the payment yield based on actual yields per planted acre, whereas the payment yields that Dr. Glauber cites appears to be the approximate average yields per harvested acre. To achieve an apples-to-apples comparison, Brazil has re-calculated the expected yield for MY 1992 as the average yield per harvested acre during the 1990 to 1994 based on USDA’s “Fact Sheet: Upland Cotton” (656.4 pounds per acre).496 Thus, the relevant adjustment factor is not 1.002 as suggested by Dr. Glauber, but 0.917 (603 / 656.4).
27. Dr. Glauber offers no critique of Professor Sumner’s analysis of the mandatory land idling cost component in the calculation of deficiency payment support. However, Dr. Glauber neglected to include these costs associated with the participation in the programme. As Professor Sumner explained, such costs are properly subtracted from the gross benefits of the cotton deficiency payment programme.
28. Brazil provides a revised calculation below, taking account of the revised yield adjustment factor of 0.917 – reflecting the deficiency payment yield as provided by Dr. Glauber’s and the expected comparable yield for MY 1992 that Dr. Glauber erroneously did not correct. In addition, Brazil continues to deduct the cost figure calculated by Professor Sumner from the deficiency payment programme. The revised formula is as follows:
Deficiency payment support = 20.55 cents per pound * 0.75 * 0.917 – 0.84 cents per pound = 13.29 cents per pound
Using the new payment yield and the new expected yield that is comparable to it, results in a 0.04 cents per pound upward adjustment to the 13.25 cents per pound presented by Professor Sumner in his 22 July Statement.
Other Payments 29. Brazil has already responded at length to Dr. Glauber’s claims endorsing the arguments of the United States that PFC, market loss assistance, direct and CCP payments, as well as crop insurance payments are not “support to” upland cotton. Dr. Glauber’s statement simply restates assertions in the legal briefs of the United States and offers no economic analysis to support his assertions.
30. Dr. Glauber asserts that it is relevant that Step 2 payments are not paid directly to producers.497 As Brazil explains in its 11 August Answer to Question 18, a basic principle is that the effect of a subsidy is independent of who initially receives the subsidy.498 That is the economic common sense behind the United States notifying Step 2 payments as product-specific support to upland cotton. And it is the basis for including such payments as “support to a specific commodity” “decided” by a Member with respect to MY 1992. The text of Article 13(b)(ii) requires, under any methodology, the calculation of a level of support to upland cotton, not to producers of upland cotton.
31. Dr. Glauber’s assertions about “double counting” are also incorrect. All the support programmes Dr. Glauber discusses (marketing loan payments, CCP payments and deficiency payments) have production and trade effects largely independent of Step 2 payments. But the purpose of calculating a rate of support under Article 13(b)(ii) is not to assess the amount of production, export, and price effects of the simultaneous application of all measures of support. Brazil will present an equilibrium analysis of the full economic effects of these support programmes simultaneously for Brazil’s “Further Submission.” Such an analysis is, however, not required for the purposes of calculating the rate of support under Article 13(b)(ii). The fact that the United States notified Step 2 as “product-specific” support indicates its position that the Step 2 programme provides additional support to upland cotton. This has certainly been the strongly held view of the US National Cotton Council.499 Thus, it was appropriate for Professor Sumner to include this production and trade-distorting subsidy in the total rate of support.
32. Brazil notes that Dr. Glauber does not criticize any other calculation made by Professor Sumner. For the convenience of the Panel, Brazil reproduces the chart containing Professor Sumner’s calculation as amended following the detailed US critique of Professor Sumner’s methodology. As the Panel will note, the results do not materially change. The support granted by the United States in MY 1999-2002 exceeds the support decided in MY 1992. Thus, even under this methodology, the United States does not enjoy peace clause exemption from actions based on Articles 5 and 6 of the SCM Agreement or Article XVI:1 of GATT 1994.
Paragraphs 135-146, and Exhibits US-25 through US-29 33. Brazil has demonstrated that under the ordinary meaning of Article 10.2 of the Agreement on Agriculture, in its context and according to the object and purpose of Article 10 and the Agreement on Agriculture overall, export credit guarantees are subject to the general export subsidy disciplines contained in that Agreement.500 Article 10.2 announces Members’ intent to work toward negotiations on specific disciplines for export credits. In the meantime, the general disciplines on export subsidies included in the Agreement on Agriculture apply to export credits, if those export credits constitute export subsidies.501
34. The United States asserts that Article 10.2 carves out export credit guarantees from the general export subsidy disciplines of the Agreement on Agriculture, including the anti-circumvention provisions of Article 10.1. The Appellate Body has, however, concluded that to exempt or carve-out particular categories of measures from general obligations such as the export subsidy obligations in the Agreement on Agriculture, the exemption or carve-out must be explicit in the text of an agreement.502 Article 10.2 includes no such explicit carve-out or exemption. The negotiators knew how to make such an exemption or carve-out explicit, as evidenced by, for example, Article 13 of the Agreement on Agriculture, footnote 15 to Article 6.1(a) of the SCM Agreement, and the second paragraph of item (k) of the Illustrative List of Export Subsidies.503
35. In support of its interpretation, the United States appeals to “subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation,” within the meaning of Article 31(3)(b) of the Vienna Convention.504 According to the United States, negotiations on agricultural export credit issues that have taken place in the OECD subsequent to the effective date of the WTO Agreement, and a statement by the OECD Secretariat, constitute “subsequent practice” establishing the agreement of WTO Members that Article 10.2 of the Agreement on Agriculture exempts export credits from any and all disciplines.
36. The United States is wrong. The United States has not established “a ‘concordant, common and consistent’ sequence of acts or pronouncements which is sufficient to establish a discernible pattern implying the agreement of the parties [to a treaty] regarding its interpretation,” which is the standard adopted by the Appellate Body to establish “subsequent practice” under Article 31(3)(b) of the Vienna Convention.505 It is evident from the positions taken by Canada, the European Communities and New Zealand in this dispute that not even those WTO Members that participated in the OECD negotiations agree with the United States’ interpretation of Article 10.2.506 Nor is there any evidence of “subsequent practice” signifying agreement on the United States’ interpretation amongst the 136 WTO Members that did not participate in the OECD negotiations.
37. Brazil notes, moreover, that the WTO Secretariat, which is in a better position to address interpretations of the covered agreements than is the OECD Secretariat, does not appear to agree that agricultural export credits are exempt from the general export subsidy disciplines of the Agreement on Agriculture by virtue of Article 10.2.507
38. The United States also argues that the negotiating history of Articles 9.1 and 10.2 of the Agreement on Agriculture supports its argument that export credit guarantees are exempt from the general export subsidy disciplines of the Agreement on Agriculture. The United States raises three arguments in this regard.
39. First, the United States addresses the negotiating history of Article 9.1.508 40. In the DeZeeuw framework agreement, the United States points to paragraph 20(e), which contemplated Members providing “data on financial outlays or revenue foregone . . . in respect of export credits provided by governments or their agencies on less than fully commercial terms.”509 The United States apparently considers that since paragraph 20(e) was not carried over into the Agreement on Agriculture, export credits are not subject to the export subsidy disciplines in the Agreement.
41. Brazil notes, however, that paragraph 20(g) addressed “export performance-related taxation concessions or incentives.” This provision was also not carried over into the Agreement on Agriculture. Nonetheless, panels and the Appellate Body have ruled that export performance-related taxation concessions or incentives like the United States’ FSC and ETI measures are subject to the general export subsidy disciplines of the Agreement on Agriculture, including Article 10.1.
42. Similarly, the United States points to Addendum 10 of Chairman Dunkel’s Note on Options, which includes an illustrative list of export subsidy practices.510 A number of the items on that illustrative list were eventually included, with modifications, in Article 9.1 of the Agreement on Agriculture.511 Others were not, including item (h), which refers to “[e]xport credits provided by governments or their agencies on less than fully commercial terms,” and item (i), which refers to “[s]ubsidized export credit guarantees or insurance programmes.” The United States apparently considers that since items (h) and (i) were not carried over into the Agreement on Agriculture, export credits, including export credit guarantees and insurance programmes, are not subject to the export subsidy disciplines in the Agreement.
43. Brazil notes, however, that item (g) of Chairman Dunkel’s illustrative list refers to “[e]xport performance-related taxation concessions or incentives other than the remission of indirect taxes.” This provision was also not carried over in Article 9.1 of the Agreement on Agriculture. Nonetheless, as Brazil has already noted, panels and the Appellate Body have ruled that export performance-related taxation concessions or incentives like the United States’ FSC and ETI measures are subject to the general export subsidy disciplines of the Agreement on Agriculture, including Article 10.1.
44. Brazil assumes that the United States simply overlooked paragraph 20(g) of the DeZeeuw framework agreement and item (g) from Chairman Dunkel’s illustrative list when it states, in paragraph 143 of its Rebuttal Submission, that
the negotiating history reveals that the Members very early specifically included export credits and export credit guarantees as a subject for negotiation and specifically elected not to include such practices among export subsidies. In contrast, the negotiating history reveals no comparable discussion involving FSC.
45. In light of these facts, it is evident that the negotiating history of Article 9.1 does not offer support for the United States’ argument that export credit guarantees are exempt from the general export subsidy disciplines of the Agreement on Agriculture.
46. Second, the United States argues that changes introduced to the text of Article 10.2 between the Draft Final Act and the final version of the Agreement on Agriculture mean that export credits are not subject to the export subsidy disciplines included in the Agreement.512
47. The version of Article 10.2 included by negotiators in the Draft Final Act read as follows:
Participants undertake not to provide export credits, export credit guarantees or insurance programmes otherwise than in conformity with internationally agreed disciplines.
The United States argues that this version of Article 10.2 “would clearly prohibit the use of export credit guarantees except in conformity with [internationally] agreed disciplines,” which it asserts “would include those contemplated by the SCM Agreement.”513
48. The version of Article 10.2 included in the Agreement on Agriculture reads as follows:
Members undertake to work toward the development of internationally agreed disciplines to govern the provision of export credits, export credit guarantees or insurance programmes and, after agreement on such disciplines, to provide export credits, export credit guarantees or insurance programmes only in conformity therewith.
The United States argues that having changed the draft, “[t]he Members clearly subsequently decided not to condition the use of export credit guarantees on conformity with the export subsidy disciplines of the Agreement on Agriculture or the SCM Agreement.”514
49. The United States’ interpretation of the negotiating history requires the Panel to accept that the version of Article 10.2 included in the Draft Final Act would have imposed a greater burden on Members than does the version of Article 10.2 ultimately included in the Agreement on Agriculture. In fact, however, Article 10.2 of the Draft Final Act was amended to make it clear that negotiators expected Members actually to pursue negotiations on specific disciplines. Whereas the version of Article 10.2 included in the Draft Final Act did not include an undertaking to pursue those negotiations, the final version of Article 10.2 does include such an undertaking. The amendment did not relieve the Members of any burden, but instead increased the burden.
50. At least some Members understood this to be the case, since soon after the conclusion of the Uruguay Round, they launched negotiations in the OECD on specific export credit disciplines.515 The United States implies that Brazil’s “admission” that those negotiations have not yet resulted in agreement on specific disciplines for export credits is fatal to its claims. Brazil has demonstrated elsewhere, however, that while those negotiations are pending, nothing in Article 10.2 (or Article 1(e)) exempts export credits from the general disciplines on export subsidies included in, for example, Article 10.1 of the Agreement on Agriculture. If export credits constitute export subsidies, they are subject to those disciplines. As noted above, the Appellate Body has concluded that to exempt particular categories of measures from general obligations such as Article 10.1, the exemption must be explicit.516 The negotiators knew how to make exemptions explicit, but did not do so in the case of export credits.517 51. Third, the United States argues that “Brazil’s interpretation would require export credit guarantees in agriculture to be subject to more disciplines than any other practice addressed in the Agreement on Agriculture,” since “not only would export credit guarantees constitute export subsidies and be subject to all of the export subsidy disciplines, but Member’s [sic] would also be specifically obligated to work toward and then apply additional disciplines.”518 This statement is incorrect for several reasons:
As clarified by Brazil, New Zealand and the European Communities, since export credits are not included in Article 9.1, they do not necessarily “constitute export subsidies.”519 They only constitute export subsidies if they are financial contributions that confer benefits and are contingent on export, or if they satisfy the elements of one of the items on the Illustrative List of Export Subsidies annexed to the SCM Agreement.
Export credits are only “subject to all of the export subsidy disciplines” of the Agreement on Agriculture if they lead to circumvention of a Member’s export subsidy reduction commitments.
It is not clear that any specific disciplines resulting from negotiations undertaken pursuant to Article 10.2 will be “additional” to those already included in the Agreement on Agriculture or the SCM Agreement. Those negotiations are not yet completed. Depending on the agreement negotiated, it is presumably possible that the resulting text could replace the disciplines included in the Agreement on Agriculture.
52. Therefore, the United States’ argument is inaccurate, and does not support its assertion that export credit guarantees are exempt from the general export subsidy disciplines in the Agreement on Agriculture, even if they meet the definition of “export subsidy.”