The Burden of Proof (the “Peace Clause”) In the case in point, Brazil asserts in its first written submission that Article 13 is by nature an “affirmative defence” or “exception” and “not itself a positive obligation”, therefore the United States carries the burden of proof on whether its subsidies are in conformity with Article 13.
Our view, in summary, is that it is inappropriate to label Article 13 as an “affirmative defence” or “exception”. Indeed this would mean the Article having much less than its originally intended effect. Article 13 in itself confers rights and imposes positive obligations on Members. It is not there simply for the convenience of resolving the question of the burden of proof. The right that it confers of entitlement to being “exempt from actions”, for example, would be rendered pointless if the burden of proof were on the respondent. It is surely for Brazil, therefore, as a complainant, to prove a breach of a positive obligation by demonstrating non-conformity, rather than for the United States to bear the burden of proof.
In our written submission, we suggest that in arriving at a proper interpretation of the burden of proof in Article 13, it might also be helpful to make some comparisons with the different types of exceptions, exemptions and defences that exist in other Articles of WTO Agreements.
We mention, for example, disputes arising in connection with agreements not covered by the DSU, where the complaining party would bear the burden to prove that the issue in dispute falls within the purview of the DSB.
Also, where a matter is specifically excluded from the dispute settlement procedures by certain relevant agreements – such as Article 6 of the TRIPS agreement - the provision concerned allows the Member applying it to prevent dispute settlement procedures and the burden of proof falls on the complaining party.
And by way of further comparison, we refer to other cases where exceptions or exemptions are granted under relevant agreements providing specific obligations.
While Article 13 of the AOA is clearly in this case not dealing with a matter under a non-covered agreement or a matter that is specifically excluded from the dispute settlement procedures as in Article 6 of TRIPS, it is also not typical of the type of exception contained in a number of the GATT Articles. By its singular nature, Article 13, in our view, falls between these examples, therefore the procedures for applying its provision should be interpreted separately and differently.
And finally, as far as the burden of proof is concerned, we submit that requiring the respondent to prove that the subsidy measure in question is in conformity with the Agreement on Agriculture will, to a certain extent, offset the respondent’s right to claim for the exceptions provided by the Article 13 provisions, which is surely contrary to the drafters’ intent.
REBUTTAL SUBMISSIONS OF PARTIES
Annex D-1 Executive Summary of Brazil's Rebuttal Submission (22 August 2003)
Annex D-2 Executive Summary of the United States Rebuttal Submission (22 August 2003)
Annex D-3 Brazil's Comments on U.S. Rebuttal Submission (27 August 2003)
Annex D-4 Comments of the United States on New Material in Brazil's Rebuttal Filings and Answer of the United States to the additional Question from the Panel (27 August 2003)
BRAZIL’S REBUTTAL SUBMISSION TO THE PANEL
REGARDING THE “PEACE CLAUSE”
AND NON-PEACE CLAUSE RELATED CLAIMS
TABLE OF CONTENTS
Page 1. The United States Has No Peace Clause Protection for Non-Green Box
Domestic Support Measures to Upland Cotton for MY 1999-2002 3
1.1. Production Flexibility Contracts and Direct Payments Are Non-Green Box
Domestic Support 3
1.1.1. The Amounts of PFC and Direct Payments Depend on the “Type” of Production 3
1.1.2. Direct Payments Are Not Green-Box Because the Base Periods for
Determining Eligibility Have Been Updated in the 2002 FSRI Act 3
1.2. PFC, Market Loss Assistance, Direct and Counter-Cyclical Payments (CCP) and Crop Insurance Subsidies Are “Support To” Upland Cotton 4
1.2.1. Production Flexibility Contract Payments 4
1.2.2. Market Loss Assistance Payments 5
1.2.3. Direct Payments 5
1.2.5. Counter-Cyclical Payments 6
1.2.6. Crop Insurance Payments 6
1.3. The US Support to Upland Cotton in MY 1999-2002 Exceeded the Support
Decided in MY 1992 6
1.4. Challenges to Actionable Subsidies under Article 5 and 6 of the SCM Agreement
and Article XVI of GATT 1994 Are Not Limited to the Marketing Year in
which a WTO Panel Is Established 8
2. The GSM 102, GSM 103 and SCGP Export Credit Guarantee Programmes
Constitute Export Subsidies in Violation of Articles 10.1 and 8 of the Agreement
on Agriculture, and Item (j) and Articles 3.1(a) and 3.2 of the SCM Agreement 8
2.1. CCC Export Credit Guarantees Constitute Export Subsidies under Articles 1 and
3.1(a) of the SCM Agreement 9
2.2. The CCC Export Credit Guarantee Programmes Constitute Export Subsidies
under Item (j) of the Illustrative List of Export Subsidies 9
2.3. The CCC Export Credit Guarantee Programmes Threaten to Circumvent US
Export Subsidy Reduction Commitments 10
2.4. The GSM 102, GSM 103 and SCGP Export Credit Guarantee Programmes
Constitute Export Subsidies in Violation of Item (j) and Articles 3.1(a) and 3.2
of the SCM Agreement 10
3. The Step 2 Export and Domestic Subsidies Are Prohibited Subsidies in Violation
of Articles 3.1(a) and 3.1(b) of the SCM Agreement 10
4. The ETI Act Subsidies Violate Articles 10.1 and 8 of the Agreement on Agriculture
and Are Prohibited by Articles 3.1(a) and 3.2 of the SCM Agreement 11 1. The United States Has No Peace Clause Protection for Non-Green Box Domestic Support Measures to Upland Cotton for MY 1999-2002 1.1. Production Flexibility Contracts and Direct Payments Are Non-Green Box Domestic Support 1.1.1. The Amounts of PFC and Direct Payments Depend on the “Type” of Production
1. PFC payments and direct payments are non-green box support because both limit the “amount“ of payment based on the “type” of production inconsistent with the requirements of Annex 2, paragraph 6(b) of the Agreement on Agriculture. The relevant text of paragraph 6(b) prohibits any linkage of the “amount of payments” to any “type of production” of an agricultural product. The “amount” of payments under the PFC and direct payment programmes falls when base acres are used to produce fruits, vegetables and wild rice. Thus, the undisputed evidence demonstrates that PFC and direct payments do not meet the policy-specific criteria for “de-coupled income support” in Annex 2, paragraph 6(b).
2. Prohibiting payments if certain types of crops are produced while at the same time permitting payments if other types of crops are produced violates Annex 2, paragraph 6(b). Contrary to the US argument, requiring no production, i.e., prohibiting production, does not relate the amount of payments to the “type” of production, as no individual “type” of production would be eligible to payments. The notion of “type of production” in paragraph 6(b) is necessarily linked to the amount of payment to some “type” of commodity that is “produced” and not to a production requirement itself.
3. In addition, Brazil also presented evidence that the US restrictions on fruits, vegetables and wild rice prevent producers with PFC and direct payment base acreage from growing these alternative crops. This restriction, therefore, channels production into particular “types of production” by prohibiting other “types of production” and, therefore, violates Annex 2, paragraph 6(b).
1.1.2. Direct Payments Are Not Green-Box Because the Base Periods for Determining Eligibility Have Been Updated in the 2002 FSRI Act
4. Direct payments are also not properly in the green box because the amount of payments are based on an updated “base period” and not on a “fixed” base period as required by Annex 2, paragraphs 6(a) and (b). Paragraphs 6(a) and (b) require a fixed and, therefore, unchanging base period for de-coupled domestic support measures with the same structure, design, and eligibility criteria. The evidence demonstrates that there are no significant changes in the payment eligibility criteria between the PFC programme and its direct successor, the direct payment program. Indeed, PFC payments made during 2002 were deducted from the amount of direct payments due in 2002.
5. Further, the updating permitted under the 2002 FSRI Act for direct payments was significant – one-third of eligible farms updated their PFC base acreage as of June 2003 in order to increase the base acreage – and payments – under the direct payment programme. This updating creates production-distorting effects because it creates expectations of future updates and will incite farmers to produce more of the programme crops that qualify for support.
6. The United States interprets the word “fixed” in Annex 2, paragraph 6(a) and (b) as being “fixed” only for the life of a particular legal measure. A Member could change a measure every year, update the “base period” to reflect the prior year’s acreage, increase current payments to reflect the updated (and increased) “historical” acreage, and label it differently under a new law. Thus, the US interpretation would permit payments to be completely “coupled” to production, just with a one-year time lag. It would render any disciplines reflected in the use of the term “a” and “fixed” “base period” in Annex 2, paragraph 6(a) a nullity. This is contrary not only to the ordinary meaning of the term “fixed” but also to the object and purpose of Annex 2, paragraph 6(a) to not permit Members to increase payments over time in a manner linked to increases in production over time. The re-linkage of payments to production is also inconsistent with the “fundamental requirement” in Annex 2, paragraph 1.
1.2. PFC, Market Loss Assistance, Direct and Counter-Cyclical Payments (CCP) and Crop Insurance Subsidies Are “Support To” Upland Cotton
7. The narrow US specificity test of “tied to production” seeks to impose a “form” of specificity on the text of Article 13(b)(ii) that is not there. It further contradicts the only analogous criteria to Article 13(b)(ii) for calculating annual levels of support – the AMS calculation criteria of Annex 3. In addition, it contradicts the broad definition of “in favour of” in defining AMS in Article 1(a) of the Agreement on Agriculture, and the “in general” language of the same provision. The “substance” the United States seeks to avoid with this unjustified interpretation is the $12.9 billion dollars in payments for the production of upland cotton from MY 1999-present.
8. Applying its narrow specificity criteria, the United States argues that PFC, market loss assistance, direct and counter-cyclical payments as well as crop insurance subsidies are not “support to” upland cotton. Brazil presents evidence that all five domestic support measures provide “support to” the production of upland cotton between MY 1999-2002.
1.2.1. Production Flexibility Contract Payments
9. Brazil has presented considerable evidence demonstrating that PFC payments to holders of upland cotton base acreage in MY 1999-2001 are support to upland cotton. The 1996 FAIR Act established a specific payment formula permitting those upland cotton farmers who had traditionally farmed upland cotton to continue to receive payments following the elimination of the deficiency payment program. The 1996 FAIR Act singled out upland cotton and only six other crops for such PFC payments. Recipients were “producers” who “shared in the risk of producing a crop”, and who farmed one of the seven crops in the three immediate years prior to the 1996 FAIR Act (MY 1993-95). Only a small minority of the producers of crops in the United States received PFC (and market loss assistance) payments. Brazil has demonstrated that between MY 1999-2001, the seven types of programme crops receiving PFC represented on average between MY 1999-2001 only 14.19 per cent of total US farm revenue. In addition, the total acreage of the seven PFC and market loss assistance crops in MY 2001 represented only 22 per cent of total US farmland. Thus, PFC payments were not provided to US producers in general.
10. The best available evidence demonstrates that upland cotton producers during MY 1999-2001 received PFC (and market loss assistance) payments. USDA reported that 97 per cent of farms producing upland cotton representing 99 per cent of upland cotton acreage from MY 1993-95 signed up to receive upland cotton PFC payments for MY 1996-2001. Upland cotton base acreage under the PFC (and market loss assistance) programme was 16.2 million acres. Between MY 1999-2001, the average acreage planted to upland cotton was 15.24 million acres. In addition, USDA reported that 95.7 per cent of the 16.2 million US upland cotton base acreage was planted to PFC programme crops in MY 2001 – a higher percentage than for any of the other 6 types of PFC programme crops. Thus, the evidence suggests that upland cotton producers in MY 1999-2001 were receiving PFC (and market loss assistance) payments.
11. Brazil has presented evidence demonstrating that PFC payments have production and trade distorting effects that arise from the prohibition on planting fruits, vegetables, and wild rice, as well as from the various “wealth effects” that result from the size of the subsidy averaging more than 15 per cent of the market value of upland cotton between MY 1999-2001. These effects provide further confirming evidence that the selected, targeted PFC (and market loss assistance) payments are support to upland cotton.
1.2.2. Market Loss Assistance Payments 12. The evidence provided by Brazil with respect to PFC payments is also relevant to market loss assistance payments because these payments were made only to farmers with PFC contracts for the seven PFC crops, and additionally to soybeans. Thus, historic upland cotton producers (producing upland cotton in MY 1993-1995) received “upland cotton-specific” market loss assistance payments in MY 1998-2001. Even with the addition of soybeans, these 8 crops only represented on average 20.75 per cent of total US farm revenue in MY 1999-2001. PFC crop base acreage and soybean acreage in MY 2001 represented only 29 per cent of total US farmland. Thus, as with PFC payments, market loss assistance payments were not paid to US agricultural producers in general but rather to only a select group of US producers.
13. The evidence presented by Brazil indicates that while producers holding PFC/market loss assistance base acreage had the legal “freedom to farm” different crops, if they produced upland cotton, they would suffer adverse financial consequences unless they produced upland cotton on upland cotton, corn or rice base acres. The evidence highlights the practical impossibility of growing upland cotton without any type of PFC and market loss assistance payment in MY 2001. This evidence confirms NCC statements and supports a conclusion that any upland cotton produced in MY 1999-2001 – as a matter of economic reality and viability – needed and received PFC and market loss assistance payments to meet the high cost of production.
14. Further evidence that market loss assistance payments are support to upland cotton stems from the fact that the United States notified these subsidies as trade and production distorting amber box support. The evidence demonstrates that the targeted market loss assistance payments triggered by market price declines have even more trade and production-distorting effects than PFC payments. Further, as with PFC payments, production and trade distortions occurred because of the prohibition or restriction on receiving such payments based on growing fruits, vegetables, or wild rice. The production and trade-distorting effects on upland cotton are further confirmed by the fact that market loss assistance payments represented on average 17.87 per cent of the market value of upland cotton between MY 1999-2001. Thus, even though upland cotton producers were not required to produce upland cotton to receive market loss assistance payments, the record demonstrates that they continued to produce upland cotton between MY 1999-2001, and they continued to benefit from the 17.87 per cent subsidies represented by these payments.
1.2.3. Direct Payments
15. Direct payments are targeted support to “producers” farming, inter alia, on upland cotton base acreage. The eligible upland cotton producers who grew upland cotton in MY 1998-2001 (or in MY 1993-95) – together with eligible producers of only nine other crops – are a select group, who grew crops representing only 23.49 per cent of total farm cash receipts and 30 per cent of total US farm acreage. Thus, direct payments are not available to the great majority of US producers of agricultural commodities, i.e., they are not provided to US agricultural producers in general.
16. The United States argues that direct payments and CCP payments are not “support to upland cotton” because there is no legal requirement under the 2002 FSRI Act for holders of upland cotton base acreage to grow upland cotton. However, Brazil has demonstrated that the theoretical legal planting flexibility in the 2002 FSRI Act is not reflected in the economic reality of growing high-cost crops like upland cotton. Farmers who did plant the 14.2 million acres of upland cotton for MY 2002 could only have covered their costs by receiving upland cotton, rice or peanut direct payments and counter-cyclical payments. This evidence strongly confirms what the NCC officials have stated repeatedly, that their members need, rely on, and receive direct payment and counter-cyclical payment support. And this evidence refutes the United States argument that the legal flexibility to grow other crops – or not produce at all – is the single relevant fact justifying a finding that direct payments and CCP payments did not support upland cotton in MY 2002.
17. Further evidence that direct payments are support to upland cotton is derived from the effects on upland cotton production caused by the updating of the base acreage between the PFC and the direct payment programmes. Brazil has presented evidence indicating that this updating creates a re-linkage between production and the direct (and counter-cyclical) payments. Production effects are also caused by channeling the payments into crops other than fruits, vegetables, and wild rice. Further, the size of the subsidy (over 15 per cent of the current upland cotton market value) also contributes to wealth creation that has production effects. These production effects demonstrate that the direct payments (and CCP payments) are not de-linked from production – as argued by the United States – and support a conclusion that they are support to upland cotton.
1.2.5. Counter-Cyclical Payments 18. The United States argues that because producers receiving CCP payments are not required to produce upland cotton to receive payments, these payments cannot, as a matter of law, be considered support to upland cotton within the meaning of Article 13(b)(ii). Nevertheless, the evidence provided by Brazil demonstrates that CCP funds in MY 2002 paid to “historic” (i.e., 1998-2001 or 1993-1995) upland cotton producers are paid to a tiny fraction of total US producers of agricultural commodities and not to US producers of agricultural products “in general”. Further, the evidence supports the conclusion that the recipients of these payments in MY 2002 needed these payments to continue producing upland cotton. They constitute “support to upland cotton”.
19. Moreover, CCP payments create additional production effects due to the “base-update” permitted under the 2002 FSRI Act for both base yields and base acreage compared to market loss assistance payments. Further, the fruits, vegetables, and wild rice prohibitions or restrictions channel production into upland cotton. This evidence collectively supports a conclusion that CCP payments are “support to upland cotton”.
1.2.6. Crop Insurance Payments 20. Brazil has demonstrated that upland cotton farmer benefit from specialized and specific crop insurance policies provided under the 2000 Agricultural Risk protection Act. Premium subsidies are directly tied to the amount of acreage planted by an upland cotton farmer. Also the participation rate, the share of policies at higher buy-up levels and the crop insurance loss ratio are higher for upland cotton than for other crops. This is confirmed by USDA’s own economists, who have found that crop insurance subsidies cause much higher production and export effects for upland cotton than for other crops. In sum, crop insurance subsidies tied directly to the production of upland cotton are “support to a specific commodity” for the purposes of Article 13(b)(ii).
1.3. The US Support to Upland Cotton in MY 1999-2002 Exceeded the Support Decided in MY 1992
21. The United States has raised a number of post hoc arguments related to a supposed “rate of support” decision it alleges to have made during MY 1992. In the SAA, the United States stated that Members would have peace clause protection from adverse effects and serious prejudice challenges in the WTO “unless the AMS for the particular commodity exceeds the level decided in the 1992 marketing year”. The phrase “AMS for the particular commodity” is an explicit recognition by the United States of the test in Annex 3, paragraph 6 which states: “For each basic agricultural product, a specific AMS shall be established expressed in total monetary terms.” The US “rate of support” methodology is not an expression in “total monetary terms,” nor does it permit such a calculation. There are only two types of methodologies that would allow an expression in monetary terms of a decision (or decisions) taken by the United States in MY 1992 regarding its level of support to upland cotton: “using budgetary outlays” or the “gap between the fixed reference price and the applied administered price multiplied by the quantity of production eligible to receive the administered price”.
22. Brazil disagrees with the United States’ assertion that it did not “decide” on budgetary outlays. The alleged US decision to provide a rate of support must necessarily be accompanied by a decision to authorize whatever budgetary outlays would be necessary to meet the rate of support. The United States took specific administrative decisions which meant that the United States decided on the payment rates that resulted from the “rate of support” and, therefore, on the amount of budgetary outlays it would use from its unlimited spending authority. For the United States to argue, post hoc, that these decisions did not also include expenditures is inconsistent with its SAA interpretation of the peace clause that the 1992 decision must be expressed in “total monetary terms.”
23. Brazil has demonstrated that expenditures for MY 1992 are lower than they are for any of the marketing years from 1999-2002. Therefore, under this methodology, the United States has no peace clause exemption for MY 1999-2002. While Brazil does not believe that calculating the upland cotton AMS based on the AMS methodology in Annex 3 is the appropriate methodology – based on the absence of the terms “AMS”, “product-specific” and “non-product-specific” in Article 13(b)(ii) – Brazil has provided evidence that by using this methodology the United States support to the basic agricultural commodity “upland cotton” exceeded the support decided during the 1992 marketing year in all marketing years from 1999-2002.
24. In the event the Panel decides not to use a “total monetary value” methodology, then there are two “rate of support” methodologies: (1) budgetary outlays per pound of support, and (2) the expected guaranteed income rate of support set out in Professor Sumner’s analysis. Brazil has provided extensive analysis of each of these two methodologies. However, Brazil does not endorse either methodology.
25. Brazil has demonstrated that the preferable methodology would be to rely on budgetary outlays per pound of upland cotton production. Professor Sumner’s approach should be used only as an alternative to the simplistic US “72.9 methodology” because it is much more accurate than the United States approach accounting for eligibility criteria, effective programme limitations and costs that the US ignores. In any event, Brazil has demonstrated that also under both rate of support methodologies the US support in MY 1999-2002 exceeds the support decided during MY 1992
26. Any methodology that does not account for eligibility and effective participation criteria is inconsistent with Article 13(b)(ii). It is also inconsistent with the context of Article 13(b)(ii) which includes Annex 3, paragraphs 8 and 10 requiring calculation of the monetary value of support by factoring in “production eligible to receive the administered price.” And it is also inconsistent with object and purpose of the Agreement on Agriculture, which is – after all – “correcting and preventing restrictions and distortions in world agricultural markets.”