Comments of the united states on the answers of brazil to further questions from the panel to the parties following the second panel meeting
The Panel Should Draw Adverse Inferences from the US Failure to Cooperate
Download 2.58 Mb.
|
- Navigate this page:
- 7. The US Refusal to Provide the Information Requested Renders Brazil’s 14/16 th
- 8. Brazil’s Intended Methodology For Allocating Contract Payments as Support to Upland Cotton
- 9. Using the Problematic and Incomplete US 18 / 19 December 2003 Aggregate Data for Allocating Contract Payments as Support to Upland Cotton Does Not Contradict Brazil’s 14/16th Methodology
6. The Panel Should Draw Adverse Inferences from the US Failure to Cooperate 57. In light of the US failure, in response to the Panel’s four requests for information on the amount of contract payments provided to current producers of upland cotton, to provide information within its exclusive control, Brazil requests the Panel to draw the following adverse inferences: That the application of Brazil’s methodology272 for allocating PFC, MLA, CCP, and DP payments to current producer of upland cotton in each of the marketing years 1999-2002 using data exclusively within the control of the United States would have resulted in payments that were higher than those estimated by Brazil’s 14/16th methodology. That the application of the US methodology for allocating PFC, MLA, CCP, and DP payments to current producer of upland cotton in each of the marketing years 1999-2002 using data exclusively within the control of the United States would have resulted in payments that were as high or higher than those estimated by Brazil’s 14/16th methodology. That the information withheld by the United States would have been detrimental to its arguments that PFC, MLA, DP, and CCP payments are not support to upland cotton, within the meaning of Article 13(b)(ii) of the Agreement on Agriculture, or alternatively, “non-product-specific” support. 58. The legal basis for the Panel’s drawing adverse inferences is found in the Appellate Body’s decision in Canada – Aircraft, where it found that “the authority to draw adverse inferences from a Member’s refusal to provide information belongs a fortiori also to panels examining claims of prohibited export subsidies. Indeed, that authority seems to us an ordinary aspect of the task of all panels to determine the relevant facts of any dispute involving any covered agreement: a view supported by the general practice usage of international tribunals.”273 The Appellate Body in Canada – Aircraft stated that “if we had been deciding the issue that confronted the panel [when referring to the drawing of adverse inferences] we might have concluded that the facts of the record did warrant the inference that the information Canada withheld included information prejudicial to Canada.”274 The Appellate Body based its “adverse inferences” holding on an interpretation of the DSU and the SCM Agreement, as well as on support from the following international law jurisprudence: In The Corfu Channel Case, where the International Court of Justice stated that “… the victim of a breach of international law is often unable to furnish direct proof of facts giving rise to responsibility. Such a State should be allowed a more liberal recourse to inferences of fact and circumstantial evidence. This indirect evidence is admitted in all systems of law, and its use is recognized by international decisions.”275 Case Concerning Military and Paramilitary Activities In and Against Nicaragua, where on the basis of the facts before it, the International Court of Justice found that it could reasonably infer that certain aid had been provided from Nicaraguan territory.276 Case Concerning the Barcelona Traction, Light and Power Company Limited, where Judge Jessup, in his separate opinion, opined that “… if a party fails to produce on demand a relevant document which is in its possession, there may be an inference that the document if brought, would have exposed facts unfavourable to the party… .”277 In William A. Parker (USA) v. United Mexican States, the Mexican-United States General Claims Commissions stated that “in any case where evidence which would probably influence its decisions is peculiarly within the knowledge of the claimant or of the respondent Government, the failure to produce it, unexplained, may be taken into account by the Commission in reaching a decision.”278 59. In US – Wheat Gluten, the Appellate Body provided additional guidance on the drawing of adverse inferences, noting that the complaining party must identify which facts support a particular inference and what inferences the Panel should have drawn from those facts.279 60. In view of these legal standards, Brazil sets forth and references the facts which support the drawing of these three adverse inferences it has requested. 61. First, at the time it refused to produce the requested information on 20 January 2004, the United States was aware of the following key facts: (1) it knew Brazil’s latest estimates using the 14/16th methodology,280 (2) it knew the results of the EWG database tabulations,281 and (3) it had access to all farm-specific contract, yield, payment and planted acreage data in a centralized database that would have permitted a calculation of the total amount of support under a variety of methodologies, including that advocated by the United States. These facts support the drawing of the first two (adverse) inferences requested by Brazil. 62. Second, the United States was fully capable of calculating the amount of payments allocable to current US producers of upland cotton. As an initial matter, the data would allow for an exact payment total of the amount of upland cotton contract payments received by upland cotton farmers. Thus, without using any allocation methodology for non-upland cotton payments, the United States knows the exact amount of all these cotton-specific payments. Further, the United States was fully instructed on how Brazil calculated the support attributable to rice in Exhibit Bra-368, Tables 2-3 and accompanying data. Further, throughout the briefings on the different allocation methodologies, the United States left no doubt that it was fully aware of how to calculate the amount of payments using a number of different methodologies, including Brazil’s 14/16th methodology, as well as the methodology advocated by the United States itself.282 Given this knowledge, opportunity, and capability, the Panel may infer that the United States knew the amount of payments resulting from the application of both the Brazilian as well as their own allocation methodologies. These facts support the drawing of the first two (adverse) inferences. 63. Third, the consistently misleading information provided by the United States concerning its possession of data regarding acreage and payment information for contract payments is another fact supporting the drawing of all three adverse inferences. Brazil first requested this information in November 2002, and then repeatedly through December 2003. The Panel made similar requests. It is uncontested that the United States government, through USDA’s Kansas City Administrative Office and its database, collected, organized and maintained in a centralized database all of the data that would be responsive to Brazil’s and the Panel’s requests. The United States now argues it did not “maintain” information on the amount of contract payments paid to current producers of upland cotton.283 The ordinary meaning of the word “maintain” is “practice habitually,” “observe,” “cause to continue (a state of affairs, a condition, an activity).”284 The rapid response of USDA’s Kansas City office to the rice FOIA request provides compelling evidence of habitual practice of the US government in “maintaining” both contract and planted acreage information.285 In sum, the pattern of misrepresentations supports the finding that the United States sought to hide this information and to mislead Brazil and the Panel, because it knew that the requested information would be harmful to its defence. 64. Fourth, a major issue in this dispute has been whether the US contract payments are support to upland cotton, or alternatively, as the United States argued, whether they are “non-product-specific support.” As the United States knew on 20 January 2004, a key element of Brazil’s proof of support to upland cotton is demonstrating the extent to which the allegedly “decoupled” contract payments are actually paid to current producers of upland cotton. Brazil has provided extensive circumstantial evidence demonstrating this link.286 The refusal of the United States to provide the information provides the basis for the Panel to infer that the United States knew this information would be detrimental to its argument that these payments were not linked to current cotton production. 65. In deciding whether to draw any adverse inferences, the Panel may wish to consider the precedential impact of the US refusal to cooperate. If a Member can easily block a panel’s request for information without any consequences, it will provide a clear roadmap for avoiding subsidy and other WTO disciplines in the future. It will also encourage parties to a future dispute to obfuscate and refuse to provide requested evidence. The WTO dispute settlement system will simply cease to function over the long run unless there are consequences that follow from a Member’s non-cooperation. In short, the US lack of cooperation, in the words of the Appellate Body in US – Wheat Gluten, is to be “deplored,” rather than rewarded.287 66. Finally, any adverse inferences drawn by the Panel become part of the evidence on the basis of which the Panel must make an objective assessment of the facts under DSU Article 11. Among the best available “facts” are the adverse inferences themselves. The Appellate Body in Canada – Aircraft held that “a panel must draw inferences on the basis of all of the facts of record relevant to the particular determination to be made,”288 and that “[w]here a party refuses to provide information requested by a panel, that refusal will be one of the relevant facts on record, and indeed an important fact, to be taken into account in determining the appropriate inference to be drawn.”289 7. The US Refusal to Provide the Information Requested Renders Brazil’s 14/16th Methodology the Most Accurate Information on Record Concerning the Amount of Support to Upland Cotton from Contract Payments 67. Since the United States has not provided the data, the best available information before the Panel is Brazil’s so-called “14/16th” methodology. This information is the most accurate proxy for the amount of contract payments that constitute support to upland cotton. By this methodology, Brazil has made the assumption that all upland cotton is planted on upland cotton base. Or, put differently, for each acre planted to upland cotton an average contract payment in the amount of an upland cotton contract payment is received. Accordingly, Brazil has adjusted the amount of upland cotton contract payments made in any marketing year by the ratio of the acreage actually planted to upland cotton and the upland cotton contract payment base acreage in that marketing year.290 For MY 2002, the original ratio was about 14/16th,291 therefore the name of the methodology. In fact, Brazil has used a different adjustment factor for each marketing year.292 68. Brazil has demonstrated with circumstantial evidence that current upland cotton producers need contract payments to generate sufficient returns to remain economically viable.293 Brazil has also demonstrated that current producers of upland cotton need high per-acre contract payments (such as those provided for upland cotton, rice, peanut or corn base) to cover the cost of upland cotton production.294 Thus, in the absence of the withheld data, it is fair to assume that some upland cotton was planted on base acreage with higher payments than upland cotton base (e.g., rice295), and some was planted on base acreage yielding lower contract payments than upland cotton base (e.g., corn). These phenomena would, on average, cancel each other out, so that one can reasonably assume that the average planted upland cotton acre received a contract payment in the amount of an upland cotton contract payment.296 69. In addition, the incomplete EWG data supports the conclusion that most US upland cotton is planted on upland cotton base.297 Three quarters of all upland cotton base payments are paid to producers of upland cotton.298 The EWG data further demonstrates that the great majority of both contract payments and marketing loan payments received by upland cotton producers are for upland cotton.299 Thus, while part of the contract payment support to upland cotton comes from non-upland cotton contract payments, any over- or under-counting resulting from Brazil’s “14/16th” methodology would be minimal. 70. Finally, in making this factual finding, the Panel should also take into account, as a fact of record, the adverse inferences outlined in Section 6 above, which fully support a finding that Brazil’s 14/16th methodology is correct. In addition, the Panel should take into account the adverse inference that the US allocation methodology would have resulted in contract payments at least as high as those in Brazil’s 14/16th methodology. However, in making its objective assessment of the facts, the Panel should reject the partial information proffered by the United States in its 18 and 19 December 2003 Letter and reflected in various submissions.300 As in Canada – Aircraft, this heavily redacted information, which excludes the most relevant information requested, should be ignored. 8. Brazil’s Intended Methodology For Allocating Contract Payments as Support to Upland Cotton 71. Brazil intended to use the data requested from the United States to calculate the amount of contract payments that constitute “support to” upland cotton,301 within the meaning of Article 13(b)(ii) of the Agreement on Agriculture.302 In its 20 January 2004 Answers to Additional Questions, Brazil provided more detailed information concerning its allocation methodology.303 Brazil emphasizes that it intended to apply this methodology since early on in this dispute, but was prevented from doing so due to (1) the US denial that the data existed and (2) the US refusal to provide the data requested by the Panel and Brazil.304 Only because of the US argument that it did not have the information requested and – after demonstrating the incorrectness of that argument – because of the US refusal to provide the information, Brazil suggested to apply its so-called “14/16th” methodology as a proxy. 9. Using the Problematic and Incomplete US 18 / 19 December 2003 Aggregate Data for Allocating Contract Payments as Support to Upland Cotton Does Not Contradict Brazil’s 14/16th Methodology 72. In this Section, Brazil presents its results from applying a simplified version of its methodology applied to the US summary data. Brazil incorporates all of its reservations it has expressed regarding this data.305 Brazil further recalls its various arguments why contract payments constitute support to specific commodities.306 Therefore, contract payments are principally allocated to the programme crops covered, as suggested by Brazil’s methodology presented in its 20 January 2004 Answers to Additional Questions.307 73. As the record demonstrates, none of four types of contract payments308 is truly “decoupled,” given the production of programme crops by the farms holding contract payment base. To the contrary, they are intended to and, in fact, do provide support for the production of programme crops. This is particularly true in the case of high per-acre payment crops, such as upland cotton and rice.309 Therefore, Brazil allocates contract payments to the programme crops covered. This approach is reasonable, since contract payments are eliminated, for instance, if fruits and vegetables are grown. In sum, Brazil maintains its position – supported by all third parties310 – that the support from contract payments that can be allocated to upland cotton is product-specific support within the meaning of the Agreement on Agriculture.311 74. Although Brazil could not use the data provided by the United States for its intended purpose, Brazil has attempted to apply a similar but less complex allocation methodology312 to the aggregate data provided by the United States.313 The summary data provided by the United States groups farms in three different categories and provides aggregate data for these categories: (1) those farms planting upland cotton and holding upland cotton contract base acreage, (2) those farms not planting upland cotton but holding upland cotton contract base acreage, and (3) those farms planting upland cotton and not holding upland cotton contract base acreage.314 For purposes of this analysis, only the aggregate data concerning farms in category (1) and (3) are of interest, as only those farms actually plant upland cotton. Brazil discusses below how it has calculated the contract payment support to upland cotton for each of these categories. 75. Due to the summary nature of this data, Brazil had to make several critical assumptions that would not have been necessary had the United States provided usable non-scrambled farm-specific data rather than summary data and “scrambled” farm-specific data. These assumptions are set out below. 76. First, Brazil had to assume that no “aggregation problem” exists with the US summary data. Brazil’s example in Section 3 above demonstrates that basing calculations on aggregate rather than farm-specific data can distort the results to a considerable extent. Without usable non-scrambled farm-specific data, it is impossible to avoid this problem – although Brazil does not know how severe the problem is in the case of upland cotton contract payments. Using the US aggregate base acreage data and aggregate planting data to allocate contract payments will most likely trigger distortions of the results due to the allocation problem. 77. It bears repeating that a proper allocation calculation has to be done on the basis of individual farms to obtain undistorted results. Only aggregating farm-specific allocations of contract payments as support to upland cotton generates the correct total amount of contract payments to be allocated to upland cotton. 78. Further critical assumptions are discussed in the course of the description of the allocation methodology used, as set out below. 79. First, Brazil has considered the data for upland cotton planted on farms that also have upland cotton base. As indicated in Brazil’s allocation approach,315 as a first step in these calculations, Brazil has assigned all aggregate upland cotton base payments received by farms producing upland cotton and holding upland cotton base as support to upland cotton – yet only up to the share of upland cotton base acreage that was actually planted to upland cotton.316 The monetary value of contract payments to these farms was calculated by multiplying the payment units for a crop base (as provided by the United States) by the payment rate for that crop in the marketing year in question.317 For MY 1999-2001, all upland cotton PFC payments to farms producing upland cotton and holding upland cotton base were allocated as support to upland cotton, because the aggregate acreage planted to upland cotton by that group of farms exceeded their aggregate upland cotton base acreage.318 For MY 2002, only a percentage of total upland cotton direct and counter-cyclical payments was allocated as support to upland cotton, because upland cotton acres planted by farms holding upland cotton base were below their updated upland cotton base for that marketing year.319 The percentage of payments allocated corresponds to the ratio of aggregated upland cotton planted acreage to aggregated upland cotton base acreage for the group of farms producing upland cotton and holding upland cotton base.320 It follows that in MY 2002 for this group of farms no further direct or counter-cyclical payments from other crops were allocated. 80. Since for MY 1999-2001 upland cotton acreage planted by farms that also had upland cotton base exceeded that base, additional PFC payments paid on other crop base were allocated as support to upland cotton. As discussed in Brazil’s methodology,321 PFC payments for other crops were primarily allocated as support for these crops – up to the share of contract acreage planted to the respective crop – in a manner identical to the above described first step for upland cotton. Any further payments stemming from contract acreage not planted to the respective base crop were pooled and allocated as additional support to those contract crops whose aggregate planting exceeded their aggregated base acreage.322 81. Second, Brazil has analyzed the US summary data concerning upland cotton that has been planted on farms without upland cotton base. Unfortunately, the United States did not provide the requested information (even in summary form) regarding the amount of contract acreage that existed on those farms.323 The second critical assumption Brazil, therefore, must make is that, for each upland cotton acre planted on a farm without upland cotton base, contract payments were received in an amount identical to the average per-acre payment allocated to upland cotton planted by the first group of farms. This assumption triggers further uncertainty about the reliability of the results, because it assumes that this group of farms receives contract payments in an equal manner as compared to group planting upland cotton and holding upland cotton base.324 82. Third, since the United States has not provided any information on the amount of market loss assistance payments received by any farm producing upland cotton – the United States did not even provide aggregate information – Brazil has assumed325 that any allocation of market loss assistance payment as support to upland cotton would have to be made in the same manner as the allocation of PFC payments in the marketing year in question.326 83. Combining the resulting PFC, market loss assistance, direct and counter-cyclical payments from these calculations yields the following amount of contract payments that would be considered support to upland cotton.
Directory: english -> tratop e -> dispu e dispu e -> Annex g list of exhibits submitted by the parties dispu e -> World Trade Organization dispu e -> World Trade Organization Download 2.58 Mb. Share with your friends:
|