Comments of the united states on the answers of brazil to further questions from the panel to the parties following the second panel meeting



Download 2.58 Mb.
Page9/46
Date11.05.2018
Size2.58 Mb.
#48562
1   ...   5   6   7   8   9   10   11   12   ...   46

1. Introduction and Summary
1. Brazil and the Panel have repeatedly requested information from the United States regarding the amount of four different types of contract payments to upland cotton producers. The Panel requested this information five months ago, in August 2003, and again in October 2003, in December 2003 and, finally, in January 2004. Brazil first requested this information in November 2002. This information is relevant to the “peace clause” portion of the dispute. It would have permitted the Panel to determine with considerable accuracy the amount of support provided to upland cotton from PFC, market loss assistance, direct and counter-cyclical payment subsidies. After repeatedly denying for over thirteen months that it even collected or maintained the payment information, the United States finally acknowledged in its 18/19 December 2003 and 20 January 2004 Letters to the Panel that USDA collected, compiled, and organized all data that would permit an allocation of the amount of contract payments received by producers of upland cotton for MY 1999-2002. Regrettably, even after admitting it has this highly relevant information, the United States refuses to produce the data. Accordingly, Brazil has no choice but to request the Panel to draw adverse inferences from the refusal of the United States to produce this data.
2. In this document, Brazil first describes in Section 2 the factual background and the purposes behind its various requests for contract payment information. In Section 3, Brazil details exactly what the United States produced (and did not produce) on 18/19 December 2003 when it “scrambled” farm-specific data and refused to produce other types of data. Brazil then addresses in Section 4 the post hoc and invalid US arguments that the farm-specific information is “confidential” under US law. In Section 5, Brazil sets forth the relevant WTO jurisprudence interpreting Article 13.1 of the DSU, which requires Members to produce even confidential information when so requested by panels. In Sections 6-8, Brazil then makes its request for the Panel to draw adverse inferences that the withheld information would show contract payments higher than those estimated in Brazil’s 14/16th methodology, and that, therefore, the Panel should find that the United States has no peace clause protection. In Sections 9 and 10, Brazil attempts to use the summary information provided by the United States on 18/19 December 2003, together with a number of assumptions, to apply Brazil’s methodology for calculating the amount of the contract payments, as well as to perform the methodology advanced by the United States. The results of this analysis each support the amount of allocated payments, as calculated using Brazil’s 14/16th methodology. Finally, in Section 11, Brazil demonstrates that the Japan – Agricultural Products decision relied on by the United States is inapplicable.
2. Brazil’s Request for Information as Posed by the Panel on 8 December 2003 and 12 January 2004
3. Brazil’s request for information is set out in Exhibit Bra-369. Brazil requested “farm-specific data to determine the amount of cotton planted on contract base acreage during marketing year 1999-2002.” The purpose of Brazil’s request for contract acreage and planted acreage data for each farm producing upland cotton was to obtain actual data to permit as exact an allocation as possible of contract payments (from both upland cotton as well as other programme crop base acreage) to current producers of upland cotton. This farm-specific base and planting information (together with requested yield base information) would have permitted the calculation of the amount of support to upland cotton from contract payments by combining two different components of support: (a) upland cotton contract payments to current producers of upland cotton, and (b) other crop contract payments to current producers of upland cotton. These payments would then be allocated to the current plantings of upland cotton on the farm and, finally, aggregated.164 This would help the Panel in determining the amount of contract payments that constitute “support to” upland cotton, within the meaning of Article 13(b)(ii) of the Agreement on Agriculture, and to decide on the peace clause. Finally, to a large extent, the requested information would have also permitted the Panel to apply the (incorrect) US-proposed methodology for calculating the amount of “benefit” from these subsidies to current upland cotton producers.
4. Brazil first requested this information in November 2002.165 The Panel requested it in August, October, and December 2003, as well as in January 2004.166 Faced with repeated US denials that it “maintained” such information, Brazil offered its so-called “14/16th” methodology during the peace clause phase of this dispute.167 Because Brazil conclusively learned in late November 2003 (inter alia via the rice FOIA request168) that the United States had falsely stated that it did not maintain contract and planted acreage information for each farm, it sought the information detailed in Exhibit Bra-369 to confirm further its 14/16th methodology, which was already supported by a large amount of circumstantial evidence. In effect, Brazil sought to replace the 14/16th methodology with a methodology using actual farm-specific data that would provide precise information that the Panel could rely on in making its determination of the amount of support to upland cotton.169 There is no question that the information withheld by the United States would have been the best information for allocating contract payments that constitute support to upland cotton.
5. Based on its experience working with the data produced in the rice FOIA request, Brazil knew how easy it is to use the computerized farm-specific data to perform the necessary calculations and present the evidence in a summary form. As the Panel could see from the computerized rice data displayed during the meeting on 3 December 2003, this farm-specific data permits the ready calculation of the amount of contract payment support provided to upland cotton producers. Brazil had an entire team ready on 18 December 2003 (and on 20 January 2004) to perform these operations quickly. Brazil had intended to present the evidence in a summary form (obviously not in a farm-specific form) for its Answers to Questions on 22 December 2003. But as described below, unlike USDA’s response to the rice FOIA request, the United States intentionally “scrambled” the farm-specific information to make it useless for purposes of the comparison sought by the Panel and Brazil, i.e., for calculating the allocated contract payment figures (which the Panel had requested the United States to provide, inter alia, in response to Question 67bis on 27 August 2003).
6. On 12 January 2004, the Panel again requested the United States to produce the data, and offered the United States the option to (1) produce the data using “substitute farm numbers which still permit data-matching,” and (2) to arrange for special confidentiality procedures.170 In its 20 January 2004 Letter to the Panel, the United States refused again to provide the information.171
3. The United States 18/19 December 2003 and 20 January 2004 Responses to the Panel’s and Brazil’s Request for Information172
7. The first and fundamental conclusion from the US 18 December 2003 Letter and the accompanying data is that the United States admits, for the first time, that it collects, maintains, and can readily analyze whether every farm planting upland cotton is doing so on contract base acreage. It admits that it knows how many farms planted upland cotton in MY 1999-2002.173 It admits that it knows the amount of acreage for each farm that planted upland cotton.174 It admits that it knows the number of upland cotton contract acres planted to upland cotton.175 It admits that it knows the amount of non-upland cotton contract acreage planted to upland cotton.176 And it admits that it knows the yield information for each farm with contract acreage.177 Indeed, it admits that it knows the amount of contract payments received by each current producer of upland cotton.178
8. In other words, the United States knows today, and it knew on 18 December 2003 and on 20 January 2004, exactly how many dollars of contract payments were paid to producers of upland cotton during MY 1999-2002. Moreover, the United States, at a minimum, had the information necessary to make all payment calculations for MY 1999-2001 in November 2002 when Brazil first requested it.179 And the United States had this information for MY 1999-2002 in August 2003, in October 2003, in December 2003, and in January 2004 when the Panel requested its production.180
9. The Panel can readily see from the US three-page description of the data it presented on 18 December 2003 that the data provided prevents any farm-specific analysis that would permit the calculation under either of the allocation methodologies advanced by the parties in this dispute. On page 2, the United States states that it is producing “second, a farm-by-farm file (with particular farm identification information) with all the requested data (plus additional data regarding payment quantities) but not including planted acres and cropland.”181 Further, the United States states that it is producing “[t]hird, farm-by-farm files for planted acres . . . with the order of the farms scrambled in order to prevent any matching of farms.182 In short, the United States admits that it does not “provide planted acreage information associated with an individual farm.”183 Translated, this means that neither the Panel nor Brazil can use the provided information to calculate the amount of contract payments made to all farms producing upland cotton and allocate the support to upland cotton. Brazil has explained this briefly in its 22 December 2003 Answers to Questions.184
10. While the United States provided a considerable amount of farm-specific data, the fact that it scrambled the data on plantings made any matching of contract base and current planting data impossible, and renders all data useless for purposes of producing any conclusive figures under any legitimate allocation methodology. Admitting this, the United States then provides aggregate data for three groups of farms: (1) those planting upland cotton and having at least some upland cotton base acreage, (2) those not planting upland cotton but having at least some upland cotton base acreage, and (3) those planting upland cotton and not holding upland cotton base.185 The United States appears to consider that this summary data could serve as a substitute for the farm-specific information.186 This is fundamentally wrong.
11. The United States states as follows in its 20 January 2004 Letter to the Panel:
[T]he information relevant to the Panel’s assessment would not be farm-specific data but rather some aggregation of data to permit this “assessment of . . . total expenditures.” As noted above, the United States has provided both farm-specific and aggregated contract data that would permit an assessment of total expenditures of decoupled payments to farms planting upland cotton. 187

Further, the United States has provided both farm-specific and aggregated contract data that would permit the Panel to make the assessment it identifies, that is, an assessment of total expenditures of decoupled payments to farms planting upland cotton. 188

12. Both the Panel’s 12 January 2004 Communication and Exhibit Bra-369 are clear that the United States was required to produce farm-specific information, on the basis of which the allocation of contract payments as support to upland cotton could be performed. The Panel’s 12 January 2004 Communication states:
[T]he Panel requests the United States to provide the same data that it agreed to provide in its letters dated 18 and 22 December 2003 but in a format which permits matching of farm-specific information on contract payments with farm-specific information on plantings. The Panel considers it both necessary and appropriate to seek this information in a suitable format in order to undertake its mandate to assist the DSB in discharging its responsibilities under the DSU and the covered agreements.189

13. There is a fundamental reason why the US summary (non-farm specific) data does not allow for an allocation of support to upland cotton in any given marketing year. To obtain undistorted results, any allocation calculation has to be done on the basis of individual farms. Aggregating farm-specific data does not generate the correct amount of contract payments to be allocated to upland cotton. On the contrary, using the aggregate data provided by the United States to allocate payments will inevitably trigger distortions of the results due to an aggregation problem. This problem is illustrated by considering two cotton farms with the following combination of upland cotton base and current upland cotton plantings.







Farm 1

Farm 2

Cotton Base

99 acres

1 acre

Cotton Plantings

1 acre

99 acres

Aggregate Cotton Base

100 acres

Aggregate Cotton Plantings

100 acres

Suggested Aggregate Cotton Plantings on Cotton Base

100 acres

Actual Cotton Plantings On Cotton Base

1 acre

1 acre

Real Aggregate Cotton Plantings on Cotton Base Acreage

2 acres

Farm 1 plants 99 acres of upland cotton and has 1 upland cotton base acre. Farm 2 plants 1 acre of upland cotton and has 99 upland cotton base acres. In the aggregate, there are 100 acres of upland cotton planted and 100 upland cotton base acreage – a perfect match suggesting that 100 per cent of the upland cotton is planted on upland cotton base. The underlying farm-specific data reveals, however, that this is not the case. Both farms only plant 1 upland cotton acre on upland cotton base. Thus, the real aggregate upland cotton planting on upland cotton base is only 2 acres – not 100 acres. The remaining upland cotton acreage may be planted on some other crop base acreage – as the information requested by the Panel and Brazil would have established.


14. This example demonstrates that using aggregate data to allocate contract payments distorts the results, possibly to a considerable extent. Without non-scrambled farm-specific data (exclusively controlled by the United States), it is impossible to assess how severe this problem is. More importantly, the same aggregation problem arises when other crop plantings and crop base acreages are considered. This increases the possible distortion that stems from the analysis of aggregate data and makes any judgement as to the direction of the distortion (over- or under-stating the results) impossible.
15. It follows from the discussions above that it is also necessary to have farm-specific data for any allocation exercise. It is only these farm-specific data that allow for farm-specific payment allocations, which, in turn, can be aggregated to “total expenditures.”190 Contrary to the US assertions,191 aggregate data will not suffice. In sum, despite its numerous assertions on 18 December 2003 and 20 January 2004 to the contrary, the United States failed to provide the requested data.
16. In addition to not providing usable farm-specific planted acreage data, the United States also refused to provide other information requested by the Panel and Brazil.
17. First, the Panel and Brazil also requested the United States to produce farm-specific data on the amount of other contract crop base acreage on farms producing upland cotton with no upland cotton base acreage. But the United States refused to provide this information – even in summary form.192 Although the United States claims otherwise in its 20 January 2004 Letter,193 the summary files, in fact, do not contain data on contract payments to farms that plant upland cotton but do not have upland cotton base.194 The fact that farms do not have any contract base for upland cotton cannot mean that no such farms have any base acreage whatsoever, which the US-produced data appears to suggest. To the contrary, at least some of these farms are likely to have contract base for other crops, such as rice, corn, and wheat, among others. Because the United States withheld that information, Brazil and the Panel would need to make assumptions on the amount of contract payments received by farms in this third category. Any such assumptions necessarily distort the resulting allocation of contract payments that constitute support to upland cotton.
18. It appears that the reason for the US refusal to provide even summary data on non-upland cotton contract base acreage for these farms stems from the erroneous US argument that Brazil has only challenged upland cotton contract payments made to current producers of upland cotton.195 The United States concludes, therefore, that it is inappropriate for Brazil to allocate contract payments made to producers of upland cotton producing on other types of base acreage.196 This refusal is based on an entirely flawed legal argument. In fact, the Panel’s terms of reference include the following broad references to contract payments:
Subsidies and domestic support provided under [the 2002 FSRI Act] … related to … direct payments, counter-cyclical payments … that provide direct or indirect support to the US upland cotton industry.197

Subsidies and domestic support provided under [the 1996 FAIR Act] … relating to … production flexibility contract payments … providing direct or indirect support to the US upland cotton industry.198

Subsidies provided under the [1998-2001 Appropriation Acts, i.e., market loss assistance subsidies]199

19. These terms of reference in Brazil’s request for the establishment of a panel are not limited in any way to contract payments made on upland cotton base acres. Brazil listed contract payments (unqualified) providing support to the US upland cotton industry, not upland cotton contract payments providing support to the US upland cotton industry. Indeed, the reference to “indirect” subsidies is more than broad enough to encompass any type of payment, including payments made from non-upland cotton base acreage. In addition, Brazil’s allocation methodology always has included payments made to producers of upland cotton on non-upland cotton base acreage.200 In short, the US argument ignores the Panel’s terms of reference, evidences a misunderstanding of Brazil’s allocation methodology, and a misunderstanding of the Panel’s and Brazil’s requests for information.201 (As with the other data the United States refuses to produce, Brazil discusses, in Section 5 below, how this constitutes a violation of the US obligations under Article 13.1 of the DSU to cooperate in a panel proceeding.202)


20. Second, the United States has not provided the requested data for market loss assistance payments received by the farms listed for MY 1999-2001. Therefore, the Panel and Brazil would have to make assumptions about the amount of market loss assistance payments that constitute support to upland cotton, based on the information about PFC payments that constitute support to upland cotton. This might distort the results. Further, since market loss assistance payments were also made for soybean base (that otherwise was not eligible to receive PFC payments), this allocation methodology may significantly underestimate the amount of market loss assistance payments that constitute support to upland cotton.203 In its 20 January 2004 Letter, the United States appears to recognize that market loss assistance payments were within the scope of the Panel’s request.204 It also implies that it has provided the information – at the very least on an aggregate basis.205 The fact, however, remains that it has not done so.206
21. Third, Brazil notes that the 19 December 2003 US data concerning MY 2002 does not contain any information regarding the amount of direct and counter-cyclical payments made on peanut base.207 As with the missing information on market loss assistance payments (specifically the soybean payment information), the missing data on peanut contract payments for MY 2002 would cause lower aggregate payments allocated as support to upland cotton.208 It follows that any such figures would lead to understating the real support to upland cotton by an amount unknown to Brazil.
22. The United States asserts that the summary data it provided allows for the “easy calculation” of “total expenditures of decoupled payments to farms planting upland cotton.”209 This is simply not true. The table below presents Brazil’s calculations of total contract payments to farms that actually produce upland cotton in as far as they were “easy” to perform.210 The table also reflects the holes in the US summary data described above. These holes prevent any “easy,” or even “difficult,” calculation of the amount of contract payments to upland cotton farms. Only by making a number of assumptions, as detailed in Sections 9 and 10 below, can the incomplete summary data be used.


Subsidy Programme

Farms Planting Cotton and Holding Cotton Base

Farms Planting Cotton and Not Holding Cotton Base1

Total Payments

Marketing Year 1999

PFC Payments
MLA Payments2

$695,912,510211
?

?
?

?
?

Marketing Year 2000

PFC Payments
MLA Payments2

$650,579,667212
?

?
?

?
?

Marketing Year 2001

PFC Payments
MLA Payments2

$520,230,908213
?

?
?

?
?

Marketing Year 2002

Direct Payments3
CCP Payments3

$619,049,305214
$1,062,580,729215

?
?

?
?

1 The United States has not provided any information on contract payments to farms that plant upland cotton, but do not hold any upland cotton base in its summary files.
2 The United States has not provided specific information on market loss assistance payments. In particular, any information on soybean payments to upland cotton farms is missing.
3 Brazil recalls that the United States withheld information on peanut base payment, thus, these figures are in all likelihood understated.
23. As the Panel can readily see, the only “easy to calculate” payment figures are the PFC, DP and CCP payments figures to farms planting upland cotton and holding upland cotton base (with the DP and CCP payment figures being understated due to the peanut base issue). The other payment figures relating to all market loss assistance payments, and to farms that plant upland cotton but hold only non-upland cotton base cannot be calculated, because the United States produced no data.
24. These calculations are, however, only one step to allocating the amount of contract payments that are actually support to “upland cotton.” As Brazil has detailed above, any allocation of “support to” upland cotton has to be performed on a farm-specific basis, to avoid aggregation problems. Any calculation using the US summary data will further distort the results by due to its incompleteness, as demonstrated by the table above.
25. In sum, the fundamental shortcomings in the scrambled and incomplete data provided by the United States on 18 and 19 December 2003 render it unusable for purposes of Brazil’s allocation methodology, discussed below,216 and, for that matter, any allocation methodology.
26. While the United States argued in its 20 January 2004 Letter that the summary data it provided was useful for calculating the amount of contract acreage that constitutes support to upland cotton217, the Panel must realize that this is – at best – “second-best” evidence, which, due to all the shortcomings just discussed, only allows for distorted calculation results, heavily depending on assumptions.
27. Nevertheless, because the redacted, scrambled and incomplete US data is what was produced, Brazil has attempted to use it.218 In particular, in Section 9, Brazil has attempted to use the summary data provided by the United States on 18 and 19 December 2003 to allocate contract payments that constitute support to upland cotton, within the meaning of Article 13(b)(ii) of the Agreement on Agriculture.219 Due to the limitations of the data discussed above, Brazil was required to make a number of critical assumptions. None of Brazil’s assumptions would have been necessary had the United States produced the requested complete farm-specific information within its exclusive control. Therefore, Brazil cautions against the use of these results. However, Brazil offers these calculations as further circumstantial evidence that its 14/16th methodology is reasonable.
28. In Section 10, Brazil has also attempted to use the US data to perform the US-proposed methodology of allocating contract payments over the total value of the crops produced on the farms producing upland cotton.220 Brazil strongly disagrees with the US position that such an allocation is required under Part III of the SCM Agreement and GATT Article XVI.221 Similarly, although Brazil cautions against relying on Brazil’s results from applying the US-proposed methodology222 due to the limitations of the data and the resulting assumptions it had to make, the results would indicate that Brazil’s 14/16th methodology generates fairly similar results.

Download 2.58 Mb.

Share with your friends:
1   ...   5   6   7   8   9   10   11   12   ...   46




The database is protected by copyright ©ininet.org 2024
send message

    Main page