|
|
|
espa�ol - fran�ais - portugu�s |
Search
|
|
|
|
espa�ol - fran�ais - portugu�s |
|
Search
|
UNITED STATES � COUNTERVAILING DUTY INVESTIGATION ON DYNAMIC
Report of the Appellate Body
120. We turn to Korea's cross-appeal, which challenges the Panel's finding that certain evidence referred to by the USDOC was "sufficient for an objective and impartial investigating authority to properly find government entrustment or direction in respect of KFB".199 This finding was made in the context of the Panel's examination of the USDOC's reference to alleged threats by the GOK against KFB and another two Korean banks.200 121. Korea pointed out to the Panel that "ultimately KFB declined to participate in the Fast Track [Debenture] Programme, and in fact did not participate in the Fast Track [Debenture] Programme, and exercised its appraisal rights in the October restructuring."201 "KFB's actions", according to Korea, "are hardly consistent with the [United States'] theory of coercion from the [GOK]."202 The Panel did not find Korea's argument relevant because "[its] analysis at [that] stage [was] concerned first and foremost with the acts of the GOK, rather than private entities' reaction to those acts."203 122. On appeal, Korea asserts that the Panel's conclusion regarding entrustment or direction of KFB rests on an incorrect interpretation of Article 1.1(a)(1)(iv) of the SCM Agreement. Korea explains that a finding of entrustment or direction under Article 1.1(a)(1)(iv) requires that the private body carry out one of the functions listed in that provision.204 "Mere direction without action", Korea submits, is not sufficient.205 Korea adds that, in this case, KFB did not carry out the action it was allegedly entrusted or directed to carry out and, therefore, the Panel's finding of entrustment or direction in respect of KFB is incorrect.206 123. The United States responds to Korea's appeal by explaining that the Panel's finding in respect of KFB does not relate specifically to KFB's participation in the Fast Track Debenture Programme.207 It states that "[t]he fact that KFB did not participate in the Fast Track [Debenture] Program was never in dispute."208 The United States further explains that "the GOK's threats and coercive behavior occurred because of 'KFB's failure to participate in the Fast Track [Debenture] Programme.'"209 The Panel record confirms that KFB participated in the financial transactions that preceded and followed the Fast Track Debenture Programme.210 We do not read the Panel's finding that the evidence was sufficient to demonstrate GOK entrustment or direction of KFB as necessarily related exclusively to the Fast Track Debenture Programme. Thus, even assuming arguendo Korea is correct that a finding of entrustment or direction requires that the function so entrusted or directed be carried out, we are not persuaded that the basis for the Panel's finding is as narrow as that alleged by Korea. 124. In any event, a finding of entrustment or direction, by itself, does not establish the existence of a financial contribution. Where a government entrusts or directs a private body�by giving responsibility to or exercising its authority over the private body�it is likely that the function that is allegedly entrusted or directed will indeed be carried out. The private body's refusal to carry out the function may be evidence that the government did not give it responsibility for such function, or that the government did not exercise the requisite authority over it such that the private body did not heed the government.211 It does not, however, on its own, mean that the private body was not entrusted or directed. Depending on the circumstances, a private body may decide not to carry out a function with which it was so entrusted or directed, despite the possible negative consequences that may follow. 125. Still, this does not mean that it is possible to make a finding of a financial contribution under Article 1.1(a)(1)(iv) where a private body does not carry out the function allegedly entrusted or directed to it. Failure by the private body to carry out one of the functions of the types listed in paragraphs (i) through (iii) means that nothing of economic value has been transferred from the grantor to the recipient.212 Simply put, if the private body has not carried out the function allegedly entrusted or directed to it, nothing will have changed hands. Therefore, there is no financial contribution and, consequently, there would be no right to apply countervailing measures. 126. For these reasons, we uphold the Panel's finding, in paragraph 7.117 of the Panel Report, that the evidence was "sufficient for an objective and impartial investigating authority to properly find government entrustment or direction in respect of KFB". VI. The Panel's Review of the USDOC's Evidence
127. We consider next the Panel's examination of the evidence underlying the USDOC's finding of entrustment or direction. After providing a general interpretation of Article 1.1(a)(1)(iv) of the SCM Agreement, the Panel turned to the evidence relied upon by the USDOC in order to determine whether it was sufficient to support the USDOC's finding of entrustment or direction. Based on its review of the evidence, the Panel concluded that the USDOC "could not properly have found that there was sufficient evidence to support a generalized finding of entrustment or direction with respect to private bodies spanning multiple creditors and multiple transactions over the period of investigation."213 Accordingly, the Panel concluded that the USDOC's subsidy determination is inconsistent with Article 1.1(a)(1)(iv) of the SCM Agreement.214 128. The United States alleges multiple errors in the Panel's application of Article 1.1(a)(1)(iv) to the facts of this case, in particular, with respect to the Panel's review of the USDOC's evidence. First, the United States claims that the Panel erred in applying an "evidentiary standard"215 that required evidence of entrustment or direction to be "probative and compelling".216 Secondly, the United States argues that the Panel erred in its approach to the examination of the USDOC's evidence under Article 1.1(a)(1)(iv), because it failed to consider the USDOC's evidence in its totality, disregarded the inferences reasonably drawn from circumstantial evidence, and, as a consequence, effectively shifted the burden of proof from Korea to the United States.217 Thirdly, the United States argues that the Panel erroneously refused to admit certain evidence submitted by the United States during the panel proceedings on the ground that arguments based on such evidence constituted "ex post rationalizations".218 129. In addition to these allegations of error in the Panel's application of Article 1.1(a)(1)(iv), the United States contends that the Panel failed to comply with its obligations under Article 11 of the DSU by relying on evidence that was not on the record of the USDOC during the investigation.219 Finally, the United States claims that the aforementioned errors, taken together, give rise to a separate and additional ground of error under Article 11 of the DSU because of the Panel's failure to apply the proper standard of review to its examination of the USDOC's finding of entrustment or direction.220 130. Before beginning our analysis, we briefly describe the USDOC's finding of entrustment or direction, as contained in its subsidy determination, to facilitate the subsequent discussion of the Panel's review of the USDOC's evidence. We then address each of the above allegations of error by the United States. Finally, we consider the implications of our analysis for the Panel's conclusion regarding the USDOC's finding of entrustment or direction.
131. In its subsidy determination, the USDOC found that numerous financial institutions, both public as well as private bodies, participated in financial transactions related to Hynix. For the purpose of this determination, the USDOC distinguished between public bodies221, government-owned and -controlled private creditors, and private creditors not owned or controlled by the GOK.222 The Panel maintained this distinction in its analysis, adopting the categorization of Group A, B, and C creditors put forward by the United States.223 Accordingly, Hynix's public body creditors were referred to as Group A creditors, and included the Korean Development Bank ("KDB"), the Industrial Bank of Korea, and other "specialized" banks.224 The GOK-owned or -controlled private creditors, which were found by the USDOC not to be public bodies225, were referred to as Group B creditors; these included the Korea Exchange Bank and KFB.226 Private entities in which the GOK had much smaller, or even non-existent shareholdings, were referred to as Group C creditors227; among these creditors were KorAm Bank, Hana Bank, and Kookmin Bank.228 We use the same categories herein. 132. In its analysis of entrustment and direction, the USDOC examined, in particular, four financial transactions. The first was an 800 billion won syndicated bank loan (the "December 2000 syndicated loan") extended to Hynix at the end of 2000 in order to finance short-term debt that was coming due in early 2001.229 The second financial transaction was the KDB Fast Track Debenture Programme, which was also designed to address the liquidity crisis caused by maturing bonds in the same time period. The USDOC's Issues and Decision Memorandum describes this programme as follows:
133. The third financial transaction was a restructuring programme agreed between Hynix and its creditors in May 2001 (the "May 2001 restructuring"). A group of 17 major creditors formed a creditors' council (the "May 2001 Creditors' Council"), based on the debt restructuring process established in June 1998 by the Corporate Restructuring Agreement (the "CRA"), which was "an informal agreement that comprised 210 [Korean] financial institutions".231 The May 2001 Creditors' Council agreed to an overall restructuring plan for Hynix, involving the rescheduling and refinancing of Hynix's debt through maturity extensions and short-term debt instruments, as well as the issuance of convertible bonds and Global Depository Shares.232 134. The fourth financial transaction identified by the USDOC was another restructuring programme, developed in view of Hynix's "continuing financial troubles" and the downturn in the DRAMS233 market, and adopted by Hynix and its creditors in October 2001 (the "October 2001 restructuring").234 This restructuring plan was formulated under the CRPA, a codification under Korean law of the corporate workout methods utilized informally under the CRA. The creditors' council governing this restructuring plan (the "October 2001 Creditors' Council") provided Hynix's creditors with three options: (i) extend new loans to Hynix; (ii) convert a certain amount of Hynix's debt into equity, with a portion of the debt being forgiven; or (iii) exercise appraisal rights235, with a portion of the debt being forgiven. As a result of this restructuring, almost 3 trillion won of Hynix's debt was converted to equity, 1.45 trillion won in debt was forgiven, and Hynix was issued new loans as well, having other loans refinanced or their terms extended.236 135. The USDOC drew three factual inferences from the evidence on the record before it: (i) the GOK maintained a policy of supporting Hynix's financial restructuring and thereby avoiding the firm's collapse237; (ii) the GOK exercised the control or influence over Hynix's creditors necessary to implement this policy238; and (iii) the GOK at times used this control/influence to "pressure" or coerce Hynix's creditors to continue supporting the financial restructuring of the firm.239 On the basis of these inferences, the USDOC arrived at a conclusion of entrustment or direction covering virtually all of Hynix's creditors and their participation in any or all of the four financial transactions examined:
136. We begin our analysis with the United States' challenge to the Panel's alleged articulation and application of an erroneous "evidentiary standard".241 Having agreed with the United States in finding that entrustment or direction need not be determined on the basis of explicit acts, the Panel nevertheless cautioned:
Subsequently, at the outset of its examination of the USDOC's evidence of entrustment or direction, the Panel reiterated its understanding that "such evidence [be] probative and compelling, in the sense that it demonstrates that each of the private creditors participating in the financial contributions was entrusted or directed to do so."243 137. The United States challenges the Panel's requirement that the evidence underlying the USDOC's determination of "entrustment" or "direction" must be "probative and compelling". The United States agrees that evidence, by its very nature, must be "probative"244, but argues that the standard of "compelling" evidence is not contained in either the SCM Agreement or the DSU. The term "compelling" is understood by the United States to refer to evidence of such weight as to require the decision-maker to arrive at one given conclusion.245 According to the United States, by applying such a standard, the Panel effectively required the USDOC to have "overwhelming" or "irrefutable" evidence that "force[d]" or "oblige[d]" it to find entrustment or direction.246 Korea submits that the term "compelling", in the context in which it was used by the Panel, is more properly understood in a sense similar to "probative", merely to describe evidence tending to persuade the decision-maker of a certain conclusion.247 Korea therefore contests the United States' characterization of the Panel's "probative and compelling" statement as a new evidentiary standard, arguing instead that it merely describes the "quality of evidence" needed to establish entrustment or direction.248 138. We agree with the participants249 that neither the SCM Agreement nor the DSU explicitly articulates a standard for the evidence required to substantiate a finding of entrustment or direction under Article 1.1(a)(1)(iv).250 Article 12 of the SCM Agreement� entitled "Evidence", specifies in paragraph 2 that a decision of the investigating authority as to the existence of a subsidy "can only be based on" evidence on the record of that agency; this applies equally to evidence used to support a finding of a financial contribution under Article 1.1(a)(1)(iv).251 Beyond this requirement, however, we see no basis in the SCM Agreement or in the DSU to impose upon an investigating authority a particular standard for the evidence supporting its finding of entrustment or direction. 139. The Panel explained that, in using the terms "probative" and "compelling", it was expressing the view that the total evidence relied upon by an agency must "demonstrate" entrustment or direction with respect to each private body in a given financial contribution.252 In so stating, the Panel, in our view, did not require that the evidence relied upon by the USDOC be "irrefutable"253, nor did it require the evidence to be of such quality or quantity so as to "force"254 the USDOC to arrive at a finding of entrustment or direction. Indeed, after reviewing the USDOC's evidence, the Panel concluded that the USDOC "could not properly have found that there was sufficient evidence to support [its] finding of entrustment or direction."255 It appears to us, on balance, that the Panel did not apply the term "compelling" in the manner suggested by the United States; had it done so, it would have erroneously imposed a qualitative standard higher than that contemplated by the SCM Agreement. Rather, the Panel properly examined whether the USDOC's evidence could support its conclusion. Thus, we do not read the Panel to have imposed an "evidentiary standard"256 beyond what we have found in the SCM Agreement.257 140. Therefore, we find that the Panel did not err in finding, in paragraphs 7.35 and 7.46 of the Panel Report, that the evidence underlying the USDOC's finding of entrustment or direction must be "probative and compelling", to the extent the Panel understood these terms to require only that the evidence demonstrate entrustment or direction.
141. We turn now to the United States' allegation that the Panel employed an erroneous approach under Article 1.1(a)(1)(iv) to its review of the USDOC's evidence. In particular, the United States alleges, first, that the Panel's approach fails to appreciate that the USDOC's conclusion rested on the totality of the evidence.258 Secondly, according to the United States, the Panel's approach impermissibly restricts the ability of the agency to draw legitimate inferences from circumstantial evidence, because it effectively requires examining whether each piece of evidence constitutes direct evidence of entrustment or direction.259 Thirdly, the United States submits that the Panel's improper approach to examining the evidence effectively shifted the burden of proof from Korea to the United States, because the Panel appeared not to have considered seriously any evidence that did not amount to a "smoking gun".260 142. At the outset of its examination, the Panel acknowledged the factual underpinnings of the USDOC's finding�that is, the GOK policy to save Hynix, the ability of the GOK to control or influence Hynix's creditors, and the pressure put on those creditors by the GOK�and structured its analysis along the lines of these factual premises.261 The Panel also observed that the USDOC had based its finding on the totality of the evidence, "without attaching particular importance to one or several evidentiary factors".262 The Panel determined that it would follow the "same approach" to the evidence.263 We understand that, in so doing, the Panel implicitly accepted the reasonableness of this approach. 143. In our view, the Panel was correct in deciding to follow the agency's approach to the examination of the evidence. Despite its stated intention, however, the Panel followed a different approach, which we examine below. 144. Notwithstanding the USDOC's reliance on the totality of the evidence, the Panel maintained that "[i]n order to" follow the same approach, it was required to assess the "probative value of each evidentiary factor separately".264 Accordingly, with respect to each of the factual underpinnings of the USDOC's finding of entrustment or direction265, the Panel examined individually the pieces of evidence on which the USDOC relied to support the particular premise. 145. We see no error, in principle, in a panel's review of individual pieces of evidence under Article 1.1(a)(1)(iv), even where the investigating authority draws its conclusion from the totality of the evidence. Indeed, in our view, in many cases a panel will be able to examine the sufficiency of the evidence supporting an investigating authority's conclusion of entrustment or direction only by looking at each individual piece of evidence. 146. We find that the Panel erred, however, in the manner in which it reviewed the individual pieces of evidence. We note, first, that the Panel often appeared to examine whether each piece of evidence, viewed in isolation, demonstrated entrustment or direction. For example, the USDOC found relevant that the Financial Supervisory Commission (the "FSC") had increased the credit limits placed on banks providing loans to a single borrower so that additional funds could be provided to Hynix.266 The Panel disagreed:
We do not read the USDOC to have inferred solely from the waiver of loan limits that entrustment or direction had taken place. Nor do we consider that the USDOC's reliance on this evidence suggests that "there is government entrustment or direction every time that a loan limit waiver is provided", for this would follow only if the USDOC had based its finding of entrustment or direction exclusively on the waivers of loan limits. 147. Similarly, the USDOC also relied on documents provided during verification meetings to find that an FSC vice-chairman268 had attended a meeting of the May 2001 Creditors' Council "to urge creditor banks to execute the resolutions made by creditors".269 This evidence was relied on by the USDOC in support of its understanding that the GOK had applied pressure on Hynix's creditors to participate in the financial restructuring of the firm. The Panel dismissed the relevance of this evidence:
The USDOC did not advance the view that the attendance of the FSC official, in and of itself, "amounted to" entrustment or direction. In arriving at this conclusion, the Panel essentially faulted the USDOC for drawing a certain inference from a single piece of evidence, where, in fact, the agency did no such thing. 148. The USDOC had also pointed in its determination to internal documents from Kookmin Bank to support a link between the GOK policy to save Hynix and the actions of creditors in the bailout of the firm.271 One of those documents identified nine reasons for Kookmin Bank's participation in the December 2000 syndicated loan272, including one referring to a GOK policy relating to the role of financial institutions in the restructuring of troubled firms.273 The Panel observed that the United States did not contest that the other eight stated reasons for Kookmin Bank's participation related to commercial considerations, and then addressed the significance of the ninth reason as follows:
The Panel overlooks the fact that the USDOC did not find entrustment or direction "simply on the basis of" the one non-commercial reason supporting Kookmin Bank's decision to participate in the December 2000 syndicated loan. It is no doubt true, as the Panel states, that the fact that eight commercial reasons are provided to support the Kookmin Bank loan may affect the emphasis given by the agency to the ninth (non-commercial) reason. It is equally true, however, as the Panel failed to recognize, that that ninth reason could reasonably take on greater meaning when viewed in the light of other corroborating evidence. 149. In each of the above instances, the Panel appears to have implicitly required that entrustment or direction be established, or determined, or inferred, solely on the basis of the particular piece of evidence examined. Furthermore, these are not isolated statements, but rather, reflect a view of the Panel that is evident throughout its analysis.275 This is troubling, especially as the Panel itself initially recognized that at no point in the USDOC's determination did the agency contend that any individual piece of evidence, in isolation, would be sufficient for its finding of entrustment or direction. 150. In our view, having accepted an investigating authority's approach, a panel normally should examine the probative value of a piece of evidence in a similar manner to that followed by the investigating authority. Moreover, if, as here, an investigating authority relies on individual pieces of circumstantial evidence viewed together as support for a finding of entrustment or direction, a panel reviewing such a determination normally should consider that evidence in its totality, rather than individually, in order to assess its probative value with respect to the agency's determination.276 Indeed, requiring that each piece of circumstantial evidence, on its own, establish entrustment or direction effectively precludes an agency from finding entrustment or direction on the basis of circumstantial evidence.277 Individual pieces of circumstantial evidence, by their very nature, are not likely to establish a proposition, unless and until viewed in conjunction with other pieces of evidence. 151. Furthermore, in order to examine the evidence in the light of the investigating authority's methodology, a panel's analysis usually should seek to review the agency's decision on its own terms, in particular, by identifying the inference drawn by the agency from the evidence, and then by considering whether the evidence could sustain that inference. Where a panel examines whether a piece of evidence could directly lead to an ultimate conclusion�rather than support an intermediate inference that the agency sought to draw from that particular piece of evidence�the panel risks constructing a case different from that put forward by the investigating authority.278 In so doing, the panel ceases to review the agency's determination and embarks on its own de novo evaluation of the investigating authority's decision. As we explain below279, panels may not conduct a de novo review of agency determinations. 152. In this case, as we observed above280, the USDOC relied on the evidence to arrive at certain factual conclusions as an intermediate step in its analysis before finding entrustment or direction. These intermediate factual conclusions were: (i) the GOK pursued a policy of preventing the financial collapse of Hynix; (ii) the GOK held control or influence over Hynix's Group B and C creditors; and (iii) the GOK pressured certain of Hynix's Group B and C creditors into participating in the financial restructuring. A proper assessment by the Panel, therefore, would have considered whether the individual piece of evidence being examined could tend to support�not establish in and of itself�the particular intermediate factual conclusion that the USDOC was seeking to draw from it. By looking instead to whether such evidence directly supported a finding of entrustment or direction, the Panel determined certain pieces of evidence not to be probative when, in fact, had they been properly viewed in the framework of the USDOC's examination, their relevance would not have been overlooked. 153. The Panel ended its examination of the USDOC's evidence, in paragraphs 7.175 to 7.178 of the Panel Report, with a "global review of all the reasoning set forth by the [US]DOC". The Panel summarized its several earlier findings on the individual pieces of evidence and, on the basis of this "global review", concluded that the USDOC "could not properly have found that there was sufficient evidence to support a generalized finding of entrustment or direction with respect to private bodies spanning multiple creditors and multiple transactions over the period of investigation."281 154. We note, first, that the Panel's discussion of the totality of the evidence appears to be primarily a summation of errors that the Panel found in the course of its review of the individual pieces of evidence. Such errors undoubtedly would affect an examination of the totality of the evidence, as these pieces would constitute the evidence the Panel would consider as a whole in assessing the evidentiary support of the USDOC's finding of entrustment or direction. Nevertheless, what is absent from the Panel's "global" assessment, in our view, is a consideration of the inferences that might reasonably have been drawn by the USDOC on the basis of the totality of the evidence.282 As we have already observed283, individual pieces of circumstantial evidence are unlikely to establish entrustment or direction; the significance of individual pieces of evidence may become clear only when viewed together with other evidence. In other words, a piece of evidence that may initially appear to be of little or no probative value, when viewed in isolation, could, when placed beside another piece of evidence of the same nature, form part of an overall picture that gives rise to a reasonable inference of entrustment or direction. Although the USDOC relied on such an approach�and the Panel, not finding it unreasonable, stated its intention to emulate it�the Panel stopped short of assessing the evidence on such a global basis. 155. A few examples from the Panel Report will illustrate the Panel's error. Although the Panel found sufficient evidentiary support for the USDOC's determination that the GOK pursued a policy to ensure the financial viability of Hynix284, the Panel did not find meaningful the USDOC's reliance on Kookmin Bank's 2001 and 2002 prospectuses filed with the United States Securities and Exchange Commission, in which Kookmin Bank stated:
In its determination, the USDOC stated:
In our view, the admission by Kookmin Bank that "government policy" might lead it to extend loans that it otherwise might not offer, and the existence of a GOK policy to save Hynix, should have prompted the Panel to consider, as did the USDOC, whether the Kookmin Bank prospectuses might carry greater relevance than initially believed when the Panel examined them in isolation. 156. The Panel also found that evidence of coercion with respect to Hana Bank, a Group C creditor, would be of limited evidentiary value towards the USDOC's finding of entrustment or direction, because the evidence related to pressure placed on Hana Bank to provide funds to Hyundai Petrochemical, not to Hynix.288 The USDOC had noted in its determination:
Although we see no error in the Panel raising questions initially about the evidentiary value of coercion taking place with respect to financial transactions involving a beneficiary other than Hynix, its subsequent failure to consider the value of this evidence in conjunction with other pieces of evidence constitutes legal error. Specifically, the Panel did not take into account in this context the fact that a GOK policy existed specifically with respect to Hynix, and that Kookmin Bank (another Group C creditor) acknowledged making loans in pursuit of government policy. If the GOK was pursuing a policy to prevent the failure of Hynix, and if the GOK had previously shown a willingness to coerce private banks (Group C creditors in the Hynix context) into participating in other Hyundai Group restructurings, the Panel should have at least considered whether, in the light of these facts, it was reasonable to conclude that coercion was also likely to have taken place with respect to loans for Hynix. And, having so considered, the Panel might have had a more complete basis for evaluating whether it was reasonable to find entrustment or direction in respect of Group C creditors. The Panel's failure to approach the evidence in its totality, however, precluded such a possibility. 157. We do not raise these questions to suggest that, had the Panel conducted a proper analysis of the evidence under Article 1.1(a)(1)(iv), it would have discovered sufficient evidentiary support for the USDOC's finding of entrustment or direction. Nor do we seek to re-evaluate the evidence before the Panel; that is not our task. Rather, we are speaking about the method used by the Panel to assess the evidence. In our view, when an investigating authority relies on the totality of circumstantial evidence, this imposes upon a panel the obligation to consider, in the context of the totality of the evidence, how the interaction of certain pieces of evidence may justify certain inferences that could not have been justified by a review of the individual pieces of evidence in isolation. Having failed to undertake such an assessment, the Panel could not have arrived at a proper conclusion as to the sufficiency of the evidence underlying the USDOC's finding of entrustment or direction. 158. In sum, we are of the view that, in analyzing the USDOC's evidence under Article 1.1(a)(1)(iv), the Panel assessed the relevance of many individual pieces of evidence by examining whether each of them was sufficient to establish entrustment or direction. In so doing, the Panel failed to appreciate the circumstantial nature of the USDOC's evidence and to consider the relevance of that evidence for the particular inferences the USDOC sought to draw. This error, in turn, contributed to various findings of the Panel dismissing or discounting individual pieces of evidence relied on by the USDOC. Furthermore, in its "global" examination of the evidence, the Panel failed to consider that pieces of evidence, especially circumstantial evidence, might become more significant when viewed in their totality. For these reasons, we find that the Panel erred in failing to examine the USDOC's evidence in its totality, and requiring, instead, that individual pieces of evidence, in and of themselves, establish entrustment or direction by the GOK of Hynix's creditors.
159. In the course of making submissions before the Panel, the United States at several points attempted to rely on evidence that, although contained in the record of the CVD investigation, had not been cited in the USDOC's decision. The Panel refused to consider this evidence on the ground that submission of such evidence constituted "ex post rationalization" on the part of the United States.290 The United States acknowledges that Members may not defend the decisions of their investigating authorities on the basis of a rationale not set out in those decisions.291 The United States contends, however, that the Panel misunderstood the scope of this prohibition against "ex post rationalization".292 According to the United States, this prohibition limits only a Member's right to raise before a panel new reasons as the basis for its investigating authority's challenged decision, but not the right to rely during panel proceedings on evidence that, although contained in the record of the investigating authority, is not explicitly referred to in its decision.293 160. Korea argues that, in requiring that "all relevant information on the matters of fact and law" be included in an investigating authority's published determination, Article 22.5 of the SCM Agreement supports the Panel's decision to refuse to consider the evidence submitted by the United States.294 Korea asserts that "all relevant information" includes any evidence on which the agency relies to support its decision.295 Thus, in Korea's view, the United States' understanding that only "new reasoning"296 may properly be rejected by panels is inconsistent with the "clear" text of Article 22.5.297 161. There is no doubt that a Member may not seek to defend its agency's decision on the basis of evidence not contained in the record of the investigation. Indeed, neither participant seeks to argue otherwise.298 Moreover, Korea acknowledges that the evidence relevant to this aspect of the United States' challenge was part of the USDOC's record and was disclosed to the parties during the CVD investigation.299 162. Article 1.1(a)(1)(iv), the provision under which the Panel examined the USDOC's subsidy determination, provides that there is a financial contribution by a government or public body where:
The provision, on its face, does not speak to the evidence that a Member may (or must) adduce before a panel to demonstrate "entrustment" or "direction". The Panel itself did not explain what it understood by a prohibition on "ex post rationalization", nor on what basis such a prohibition would limit a Member's right to present evidence�as opposed to reasoning�in dispute settlement proceedings. 163. Korea suggests that Article 22.5 of the SCM Agreement provides the basis for the Panel's exclusion of the United States' evidence, particularly as it requires an investigating authority's final determination to contain "all relevant information on the matters of fact and law and reasons which have led to the imposition of final measures". We note, first, that the Panel itself did not seek to justify its treatment of the United States' evidence on the basis of Article 22.5. Moreover, Korea does not allege that the facts for which the evidence at issue here was introduced were not set out in the USDOC's final determination.300 Nor does Korea allege that those facts set out in the final determination were asserted without citation of any supporting evidence. Indeed, Korea could not so allege because the USDOC's final determination did set out those facts and did seek to support those facts by referring to record evidence, even if not the precise evidence the Panel refused to consider.301 Thus, insofar as it relates to the evidence at issue here, the USDOC's final determination provided Korea with notice of the factual bases of the finding of entrustment or direction, as well as notice of certain record evidence underlying each of those facts. 164. In these circumstances, we are of the view that Article 22.5 does not require the agency to cite or discuss every piece of supporting record evidence for each fact in the final determination.302 Therefore, even assuming arguendo that Article 22.5 of the SCM Agreement could provide the basis for a panel's exclusion of evidence, we see no reason why it would support such exclusion in this case.303 165. In the light of the above, we find no basis for the Panel's exclusion of the United States evidence in question. That evidence was on the record of the investigation and it was not put before the Panel in support of a new reasoning or rationale. We therefore find that the Panel erred, in paragraphs 7.88, 7.102, 7.116, 7.121, and 7.141 of the Panel Report, in declining to consider certain record evidence not cited by the USDOC in its published determination.
166. We consider now whether, as the United States alleges304, the Panel erred in relying on evidence that was not on the record before the USDOC in the underlying CVD investigation. We first set out the uncontested facts relating to this issue and then examine the Panel's approach. 167. As noted above, one of the factual conclusions established by the USDOC in support of its finding of entrustment or direction was that the GOK had the ability to control or influence Hynix's creditors.305 One of the mechanisms examined by the USDOC, through which the GOK allegedly exercised control or influence over Hynix's creditors, was the CRPA. 168. According to the USDOC, under the CRPA, those creditors holding at least 75 per cent of a firm's outstanding debt could set the restructuring terms for all of that firm's creditors.306 Article 29 of the CRPA established a mechanism whereby those creditors that did not agree to the restructuring terms could be bought out by the remaining creditors.307 The price at which the dissenting creditors' debt would be bought was to be determined by consultations between the dissenting creditors and the remaining creditors.308 Article 29(5) of the CRPA provided that, if consultations did not result in an agreed price:
169. The restructuring terms established by the October 2001 Creditors' Council provided Hynix's creditors with three options, the third of which provided that:
170. Thus, under the third option ("Option 3"), creditors that declined to provide new loans to Hynix and declined to convert their debt to equity in Hynix could exercise their appraisal rights and be bought out by the other creditors on the October 2001 Creditors' Council. The GOK and Hynix pointed out to the USDOC during the investigation that, in fact, three311 creditors had exercised Option 3 and had "severed their ties from Hynix and the restructuring process in October 2001".312 The USDOC responded in its determination that, notwithstanding those creditors' lack of participation, the GOK's alleged ability to control or influence Hynix's creditors remained by virtue of the fact that "the terms on which these creditor banks terminated their relationship with Hynix were dictated by the banks that mattered in this case, namely the [Group A and B] creditors."313 171. The Panel found that the USDOC had not explained how the restructuring terms were "set"314 by Group A and B creditors, notwithstanding that Article 29(5) of the CRPA provided the "possibility"315 for dissenting creditors to avoid these restructuring terms and be bought out by other creditors at a price determined through mediation. The Panel asserted that "there was also evidence on the [US]DOC's record indicating that the mediation provisions had actually been invoked by three creditors in respect of the October 2001 restructuring."316 Therefore, the Panel determined that the evidence on record was sufficient to "bring into question" the USDOC's conclusion that the terms of the October 2001 restructuring were established by the Group A and B creditors.317 172. The United States challenges the Panel's finding that certain creditors of Hynix "actually ... invoked"318 their right to mediation under the CRPA. According to the United States, the evidence on the record before the USDOC contained no information relating to the exercise of mediation rights by the three creditors.319 The United States submits that the only basis for the Panel's finding that certain creditors exercised their mediation rights was evidence submitted by Korea to the Panel but never provided to the USDOC.320 By arriving at a conclusion on the basis of non-record evidence, the United States argues, the Panel acted inconsistently with its obligations under Article 11 of the DSU.321 173. Korea argues that the United States misstates the Panel's finding, which Korea understands to be that the USDOC had not undertaken a sufficient inquiry with respect to the participation of certain creditors in the October 2001 restructuring.322 This finding, according to Korea, was based not only on "actual incidents of mediation", as discernible from the 2001 Hynix Audit Report, but also on Article 29(5) of the CRPA and the opinion of one of the USDOC's experts.323 With respect to the 2001 Hynix Audit Report, Korea notes that the excerpt from this document made clear reference to mediations under the CRPA, and then provided that, "[b]ased on this clause", three creditors "raised objection[s]".324 On this basis, Korea submits, the USDOC should have understood that the raising of objections, in connection with the future payout by Hynix, necessarily implied recourse to mediation by those creditors. 174. Although the United States characterizes the error of the Panel as its reliance on non-record evidence, we note that the Panel explicitly agreed with the United States that its "review of the [US]DOC's determination should be confined to facts actually recorded on the [US]DOC's record of investigation."325 The Panel insisted that its finding was based exclusively on "evidence on the [US]DOC's record"326, namely the 2001 Hynix Audit Report. The issue raised by the United States' appeal, therefore, is not whether the evidence was contained in the record, but rather, whether the evidence contained in the record should have "indicate[d]" to the USDOC "that three of the four creditors exercising appraisal rights under option 3 actually exercised their right to seek mediation in respect of the October 2001 restructuring."327 175. The United States brings its challenge under Article 11 of the DSU, which requires that a panel "make an objective assessment of the matter before it". The Appellate Body has stated previously that, when assessing an investigating authority's determination, a panel may not fault the agency for failing to take into account facts that it could not reasonably have known.328 A panel must therefore limit its examination to the facts that the agency should have discerned from the evidence on record. Where a panel reads evidence with the "benefit of hindsight", it fails to consider how the evidence should have fairly been understood at the time of the investigation, and thereby fails to make an "objective assessment" in accordance with Article 11 of the DSU.329 176. We turn now to the evidence relied upon by the Panel. The Panel pointed to a single paragraph in Hynix's 2001 Audit Report, which was on the record of the investigation, as the basis for finding that the USDOC should have been aware that three creditors had resorted to mediation after exercising their appraisal rights under Option 3.330 That paragraph states:
177. This excerpt does not indicate explicitly that three creditors participated in a mediation. Nor does it state that a mediation in fact occurred, or that it was the mediation (rather than, for example, an agreement with the October 2001 Creditors' Council) that resulted in the payout of over 80 billion won. All that one can glean from this paragraph is that three creditors raised objections to the terms of reimbursement and that 80 billion won was "recognized". During the oral hearing, Korea insisted that this paragraph made clear that mediation had taken place, and explained that, under the CRPA, the "rais[ing]" of "objections" necessarily resulted in recourse to mediation.332 In our view, however, it is not evident from a reading of Article 29(5) of the CRPA333, the legal basis for the exercise of appraisal rights, that the raising of "objections" automatically results in mediation proceedings. Furthermore, at no point in its responses to questions during the oral hearing was Korea able to identify where in the record such an explanation could be found. Thus, we do not read the record evidence as supporting the Panel's conclusion that the USDOC should have understood the "rais[ing]" of "objections" to include the recourse to mediation by three Hynix creditors. 178. Moreover, we note that, during the investigation, the USDOC asked Hynix about creditors exercising rights under Option 3�that is, those creditors that "did not exchange their convertible bonds for the new convertible bonds with an obligation of conversion".334 The USDOC also asked Hynix, in a supplemental questionnaire, more specifically about how appraisal rights were exercised under Option 3.335 In neither response did Hynix indicate that mediation had even been requested, much less that it had taken place.336 Therefore, faced with no mention of mediation in response to questions, and given the wording of Article 29(5) of the CRPA, the USDOC, in our view, should not have been expected to read Hynix's 2001 Audit Report as indicating the recourse to mediation by three Hynix creditors. 179. In our view, therefore, the Panel erroneously concluded that the USDOC should have made a factual inference from evidence on the record that would not reasonably have suggested such an inference. We therefore find that the Panel failed to "make an objective assessment of the matter before it", as required by Article 11 of the DSU, by finding, in paragraph 7.85 of the Panel Report, that "the mediation provisions [of the CRPA] had actually been invoked by three creditors in respect of the October 2001 restructuring", in the absence of supporting evidence on the record of the underlying investigation.
180. We turn now to the United States' allegation that the Panel failed to examine the USDOC's subsidy determination consistently with the applicable standard of review. The Panel began its analysis by observing that Article 11 of the DSU sets forth the applicable standard of review. On the basis of this provision and the Appellate Body's discussion on standard of review in US � Lamb, the Panel stated:
181. The United States does not contest the Panel's articulation of the standard of review based on US � Lamb. The United States contends, instead, that the errors alleged in sub-sections C through F above, viewed collectively, amount to error in the Panel's application of the proper standard of review prescribed by Article 11 of the DSU, and as clarified by the Appellate Body in US � Cotton Yarn and US � Lamb.338 Korea contends that the Panel properly engaged in the "in-depth review"339 contemplated by the Appellate Body decision in US � Lamb, and that, as is evident from a review of the challenged Panel findings in this case, the United States' appeal is based on "partial or chopped and twisted quotations from the Panel Report".340 182. Article 11 of the DSU sets out the proper standard of review to be applied by panels when examining Members' subsidy determinations.341 That provision states, in relevant part:
183. The Appellate Body has observed that, with respect to a panel's review, in accordance with Article 11, of facts established by an investigating authority:
184. The Panel and both participants343 have recognized that the Appellate Body has in the past elaborated on the standard of review mandated by Article 11 with respect to factual and legal issues in the context of claims under the Agreement on Safeguards.344 The standard of review articulated by the Appellate Body in the context of agency determinations under that Agreement is instructive for cases under the SCM Agreement that also involve agency determinations.345 Nevertheless, we recall that an "objective assessment" under Article 11 of the DSU must be understood in the light of the obligations of the particular covered agreement at issue in order to derive the more specific contours of the appropriate standard of review.346 In this respect, we are especially mindful, in this appeal, of Articles 12, 19, and 22 of the SCM Agreement. 185. We have noted above that Article 12.2 requires that an investigating authority's determination of entrustment or direction be "based on" evidence.347 We also note that, under Article 19.1, countervailing duties may be imposed only where the investigating authority has "determin[ed]", inter alia, the "existence" of a subsidy. The existence of a subsidy is "determined", in turn, by reference to the definition of "subsidy" set out in Article 1.348 Finally, Article 22.5 requires an investigating authority's affirmative determination to include the "reasons" for the decision349 as well as "the basis on which the existence of a subsidy has been determined".350 186. In the light of the above, we are of the view that the "objective assessment" to be made by a panel reviewing an investigating authority's subsidy determination will be informed by an examination of whether the agency provided a reasoned and adequate explanation as to: (i) how the evidence on the record supported its factual findings; and (ii) how those factual findings supported the overall subsidy determination.351 Such explanation should be discernible from the published determination itself. The explanation provided by the investigating authority�with respect to its factual findings as well as its ultimate subsidy determination�should also address alternative explanations that could reasonably be drawn from the evidence, as well as the reasons why the agency chose to discount such alternatives in coming to its conclusions.352 187. A panel may not reject an agency's conclusions simply because the panel would have arrived at a different outcome if it were making the determination itself. In addition, in the absence of an allegation that the agency failed to investigate sufficiently or to collect certain information353, a panel must limit its examination to the evidence that was before the agency during the course of the investigation, and must take into account all such evidence submitted by the parties to the dispute. In other words, a panel may not conduct a de novo review of the evidence or substitute its judgement for that of the investigating authority. A failure to apply the proper standard of review constitutes legal error under Article 11 of the DSU.354 188. These general principles reflect the fact that a panel examining a subsidy determination should bear in mind its role as reviewer of agency action, rather than as initial trier of fact. Thus, a panel examining the evidentiary basis for a subsidy determination should, on the basis of the record evidence before the panel, inquire whether the evidence and explanation relied on by the investigating authority reasonably supports its conclusions. In the context of reviewing individual pieces of evidence, for example, a panel should focus on issues such as the accuracy of a piece of evidence, or whether that piece of evidence may reasonably be relied on in support of the particular inference drawn by the investigating authority. As we observed above355, however, the Panel in this case examined whether certain pieces of evidence were sufficient to establish certain conclusions that the USDOC did not seek to draw, at least solely on the basis of those pieces of evidence. Moreover, it failed to examine the evidence in its totality.356 The Panel thus failed to assess the agency's determination. Instead, the Panel's examination reflected its own view of whether entrustment or direction existed in this case; the Panel thereby engaged, improperly, in a de novo review of the evidence before the agency.357 189. Furthermore, with respect to the Panel's refusal to admit certain evidence submitted by the United States, we note that the Panel did not indicate that the evidence was not contained in the record of the underlying investigation. Nevertheless, the Panel excluded such evidence from its consideration in the absence of any legal basis to do so.358 In addition, the Panel erred in concluding that the USDOC should have been aware of a fact that was not reasonably based on evidence in the agency record, namely, that three creditors exercised mediation rights under the CRPA.359 In so doing, the Panel essentially "second-guessed" the investigating authority's analysis of the evidence and thus overstepped the bounds of its review. 190. Taken together, these errors lead us to conclude that the Panel went beyond its role as the reviewer of the investigating authority's decision and, instead, conducted its own assessment, relying on its own judgement, of much of the evidence before the USDOC. Accordingly, we find that the Panel failed to apply the proper standard of review and, therefore, failed to comply with its obligations under Article 11 of the DSU.
191. Following its examination of the evidence in relation to the factual premises of the USDOC's finding of entrustment or direction, the Panel concluded:
192. Although the Panel may have intended to convey a broader meaning of the terms "entrusts" and "directs" than what might be understood from a strict reading of the terms "delegation" and "command", we have found it useful, nevertheless, to modify the Panel's interpretation of "entrusts" and "directs" in order to clarify the meaning of these terms in accordance with the interpretation we set out above.361 We have also found multiple errors in the Panel's analysis of the USDOC's finding of entrustment or direction. In particular, we have found that the Panel erred in (i) applying Article 1.1(a)(1)(iv) so as to examine the USDOC's evidence piecemeal rather than in its totality, notwithstanding the Panel's stated intention to follow the USDOC's approach362; (ii) refusing to admit certain record evidence submitted by the United States; and (iii) faulting the USDOC for its failure to address facts that were not on the record of the investigation. On the basis of these errors, we have found that the Panel failed to apply the proper standard of review in accordance with Article 11 of the DSU. In our view, these errors, taken together with the modification we found necessary to the Panel's interpretation of the terms "entrusts" and "directs", invalidate the basis for the Panel's conclusion, quoted above, that there was not sufficient evidence to support the USDOC's finding of entrustment or direction.363 193. Because this conclusion is the sole basis for the Panel's finding of inconsistency with Article 1.1(a)(1)(iv) of the SCM Agreement, we reverse the Panel's findings, in paragraphs 7.178, 7.209, and 8.1 of the Panel Report, that the USDOC's determination of GOK entrustment or direction of Hynix's Group B and C creditors is inconsistent with Article 1.1(a)(1)(iv) of the SCM Agreement. 194. We note that neither participant requested, in its written submissions, that we complete the analysis by undertaking our own review of the USDOC's finding of entrustment or direction if we were to reverse the Panel's finding of inconsistency with Article 1.1(a)(1)(iv). In response to questions posed at the oral hearing, Korea stated that, although it had not made such a request in its written submissions, it did want the Appellate Body to complete the analysis and find the USDOC's subsidy determination inconsistent with Article 1.1(a)(1)(iv) of the SCM Agreement. Korea suggested that, because the USDOC's finding of entrustment or direction rests on its finding that a "single [subsidy] program" exists, we could complete the analysis by reviewing only the evidentiary basis of the USDOC's finding of a "single program".364 195. We might have been able to resolve this dispute solely by addressing the USDOC's finding of a "single program", as Korea suggests, only if this were one of the indispensable bases for the USDOC's ultimate finding of entrustment or direction. We are not persuaded, however, that this is the case. In support of its "single program" finding, the USDOC pointed to conclusions it had drawn in the course of its entrustment and direction analysis.365 As we read the USDOC's determination, therefore, the characterization of the alleged subsidy as a "single program" appears to follow from the USDOC's analysis of entrustment or direction, rather than serve as a premise of this analysis. As a result, even if we were to determine that the USDOC's finding of a "single program" lacked sufficient evidentiary support, as Korea contends, that alone would not undermine the USDOC's finding of entrustment or direction. Examining the USDOC's finding of a "single program" would therefore not provide us with a basis to arrive at a definitive answer as to the consistency of the USDOC's subsidy determination with Article 1.1(a)(1)(iv). 196. Moreover, in our view, the nature of the errors we have found in the Panel's decision, especially with respect to the approach taken by the Panel to the admissibility and probative value of several individual pieces of evidence, is such that completing the analysis would require us to examine anew the entire USDOC finding of entrustment or direction. We have stated above that the determination of entrustment or direction will hinge largely on the particular facts of the case.366 Thus, in completing the analysis�that is, in examining the legal question whether the USDOC could have arrived at a finding of entrustment or direction on the basis of the evidence and explanation provided�we would need to engage in a thorough examination of the evidence, particularly as the Panel improperly excluded certain evidence from its consideration. 197. Furthermore, we do not consider that the participants have addressed sufficiently, in their submissions, those issues that we might need to examine if we were to complete the analysis in this case, including, for example: (i) whether the probative value of certain pieces of evidence is affected by our modification of the Panel's interpretation of the terms "entrusts" and "directs"; (ii) the probative value of the United States evidence improperly excluded by the Panel; (iii) the relevance of certain factual disagreements that the Panel considered unnecessary to resolve in the light of its legal analysis367; and (iv) the inferences that may reasonably be drawn from an analysis of the evidence in its totality. In these circumstances, we believe it is more appropriate to limit our examination to a review of the issues of law covered in the Panel Report and the legal interpretations developed by the Panel. Therefore, we do not complete the analysis to arrive at our own conclusion on the consistency of the USDOC's subsidy determination with Article 1.1(a)(1)(iv) of the SCM Agreement. 198. This does not mean that we are hereby expressing any view as to whether the USDOC's determination of entrustment or direction�which is a necessary component of its determination of financial contribution�is necessarily supported by sufficient evidence. We conclude only that the Panel's finding of inconsistency, which resulted from its flawed approach to reviewing the evidence, is in error. Of course, the inquiry into the existence of a subsidy does not end with a determination of a financial contribution. Such an inquiry must proceed to examine whether the alleged financial contribution confers a "benefit".368 In addition, the SCM Agreement requires that a subsidy be "specific" in order for countervailing measures to be imposed.369 We turn to these issues in the following Section of our Report. 199. The United States appeals the Panel's findings relating to the USDOC's determination of benefit and, in respect of Group B and C creditors, specificity. 200. As to the determination of benefit, the Panel noted that the USDOC's benefit analysis "was predicated almost entirely upon the [US]DOC's determination that Group B and C (except Citibank) creditors were entrusted or directed by GOK to participate in the four financial contributions at issue".370 "[T]hese creditors", the Panel explained, "were rejected as market benchmarks ... because the [US]DOC found that they were acting pursuant to government entrustment or direction, rather than market principles, when participating in the Hynix restructuring".371 The Panel reasoned that, because the USDOC could not properly have found that these private creditors had been entrusted or directed by the GOK, "government entrustment or direction of these creditors could not have been a proper basis for the [US]DOC to reject them as market benchmarks".372 Therefore, the Panel found that "the [US]DOC's benefit determination is inconsistent with Article 1.1(b) of the SCM Agreement".373 201. The Panel made the following finding in respect of the USDOC's determination of specificity relating to Group B and C creditors:
202. On appeal, the United States requests the Appellate Body to reverse these findings because "the Panel's conclusions are based solely on its erroneous finding that the [US]DOC's determination of GOK entrustment or direction of certain Hynix creditors is inconsistent with Article 1.1(a)(1)(iv) of the SCM Agreement."375 203. At the oral hearing, Korea disagreed with the United States' assertion that the Panel's finding relating to benefit is premised exclusively on the Panel's finding on entrustment or direction. Korea stated that the Panel also criticized the USDOC's rejection of Citibank as an appropriate benchmark for the determination of benefit.376 204. We find no such "criticism" in the paragraph of the Panel Report referred to by Korea; in that paragraph, the Panel is merely describing the approach taken by the USDOC, including its rejection of Citibank as a benchmark. In paragraph 7.191 of the Panel Report, the Panel expressly indicates that, in the light of its finding that the USDOC's determination of benefit was inconsistent with Article 1.1(b) because it was premised on an improper finding of entrustment or direction, "it is not necessary ... to examine other issues raised by the parties regarding market benchmarks". Thus, we agree with the United States' assertion that the Panel's finding concerning the USDOC's determination of benefit is premised exclusively on the Panel's finding relating to entrustment or direction. 205. Having reversed the Panel's findings that the USDOC's determination of entrustment or direction is inconsistent with Article 1.1(a)(1)(iv), there is no basis for us to uphold the Panel's finding on benefit. Consequently, we also reverse the Panel's finding, in paragraphs 7.190, 7.209, and 8.1 of the Panel Report, that the USDOC's benefit determination is inconsistent with Article 1.1(b) of the SCM Agreement.377 206. As to specificity, Korea agreed at the oral hearing that the Panel's finding of inconsistency in respect of the USDOC's determination of specificity relating to Group B and C creditors is premised exclusively on the Panel's finding on entrustment or direction. Korea acknowledged that a reversal of the Panel's finding relating to entrustment or direction would necessarily result in the reversal of the Panel's finding of inconsistency concerning the determination of specificity in respect of Group B and C creditors. We agree. In paragraph 7.206 of the Panel Report, the Panel explains its view that "the [US]DOC's finding of GOK entrustment cannot provide a proper basis for a determination of specificity in respect of alleged subsidies provided by Group B and C creditors". The Panel provides no other basis for its finding. 207. Accordingly, we reverse the Panel's findings, in paragraphs 7.208, 7.209, and 8.1 of the Panel Report, that the USDOC's finding of specificity is inconsistent with Article 2 of the SCM Agreement insofar as it relates to alleged subsidies by Group B and C creditors. 208. The Panel did not examine these issues further. Consequently, there are neither sufficient findings by the Panel nor undisputed facts contained in the record to allow us to conduct our own analysis of Korea's claims regarding benefit and specificity.378 We recall that it is not sufficient to determine that there is a "financial contribution by a government or any public body" in order to find that there is a "subsidy" under Article 1.1 of the SCM Agreement. This provision also requires that "a benefit is thereby conferred". Article 1.2 requires, in addition, that the subsidy be "specific". Because the Panel's findings on benefit and specificity were premised exclusively on its conclusion relating to entrustment or direction, there is insufficient basis for us to examine the consistency of the USDOC's benefit and specificity determinations with the SCM Agreement. Even though we reverse the Panel's findings, we offer no view as to the consistency of the USDOC's underlying determinations of benefit and specificity. VIII. Findings and Conclusions 209. For the reasons set out in this Report, the Appellate Body:
210. Based on these findings, the Appellate Body makes no recommendation to the Dispute Settlement Body pursuant to Article 19.1 of the DSU.
Signed in the original in Geneva this 9th day of June 2005 by:
_________________________ Georges Abi-Saab Presiding Member
199 Panel Report, para. 7.117. (footnote omitted) 200 Ibid., para. 7.107. 201 Ibid., para. 7.108. 202 Ibid. 203 Ibid., footnote 136 to para. 7.117. 204 Korea's other appellant's submission, para. 18. 205 Ibid., para. 29. 206 Korea does not challenge "the Panel's factual determination that there was [government] coercion with respect to KFB". (Ibid., para. 4) 207 United States' appellee's submission, para. 5. 208 Ibid. (footnote omitted) 209 United States' appellee's submission, para. 5 (quoting Panel Report, para. 7.117). (original emphasis; footnote omitted) 210 See, for example, Panel Report, para. 7.114; and Figure US-4 submitted by the United States to the Panel. 211 We explained earlier that in most cases government entrustment or direction would include a threat or inducement. (See supra, para. 116) 212 See Appellate Body Report, Canada � Dairy, para. 87, where the Appellate Body explained that "a 'subsidy' involves a transfer of economic resources from the grantor to the recipient for less than full consideration." 213 Panel Report, para. 7.177. 214 Ibid., para. 7.178. 215 United States' appellant's submission, para. 47. 216 Ibid. (referring to Panel Report, paras. 7.35 and 7.46). 217 Ibid., paras. 58-87. 218 Ibid., para. 94 (referring to, inter alia, Panel Report, paras. 7.88, 7.102, and 7.141). 219 Ibid., para. 101. 220 United States' appellant's submission, para. 119. 221 In its determination, the USDOC referred to public bodies as "government authorities". (United States' response to Question 3 posed by the Panel at the First Panel Meeting, Panel Report, p. E-37, para. 10 and footnote 9 thereto) See also Issues and Decision Memorandum, supra, footnote 6, pp. 15-17. 222 See Issues and Decision Memorandum, supra, footnote 6, p. 49. 223 Panel Report, para. 7.8 (referring to Figure US-4 submitted by the United States to the Panel). 224 United States' response to Question 3 posed by the Panel at the First Panel Meeting, Panel Report, p. E-37, para. 10 and footnote 9 thereto. 225 Some of Hynix's creditors were treated by the USDOC as Group B creditors despite the fact that the GOK held a 100 per cent ownership interest. The Panel noted that, in its view, the USDOC might have been entitled to treat these 100 per cent-owned firms as "public bodies", but having refused to so classify them, the USDOC was required to establish entrustment or direction with respect to such creditors. (Panel Report, footnote 29 to para. 7.8 and footnote 80 to para. 7.62) 226 Panel Report, para. 7.8; United States' responses to Questions 4-6 posed by the Panel at the First Panel Meeting, Panel Report, p E-38, para. 17. 227 Panel Report, para. 7.8; United States' responses to Questions 4-6 posed by the Panel at the First Panel Meeting, Panel Report, p E-38, para. 19. 228 Figure US-4 submitted by the United States to the Panel. 229 Issues and Decision Memorandum, supra, footnote 6, p. 19. 230 Ibid., p. 23. 231 Ibid., p. 19. 232 Issues and Decision Memorandum, supra, footnote 6, p. 20. We understand that the references in the Issues and Decision Memorandum to Global Depository Shares relate to company shares offered for sale globally, that is, in multiple markets around the world, by virtue of receipts that are issued by banks in several countries and that evidence the shareholder's ownership interest in that company. (See P. Moles and N. Terry, The Handbook of International Financial Terms (Oxford University Press, 1999), p. 256) 233 Dynamic random access memory semiconductors (DRAMS) and memory models containing DRAMS are herein collectively referred to as "DRAMS". 234 Issues and Decision Memorandum, supra, footnote 6, p. 20. 235 The exercise of appraisal rights involved a creditor seeking not to continue financing Hynix, which would have its Hynix debt purchased by the other Hynix creditors at a price determined either through consultation or with the assistance of an accounting firm. (See infra, paras. 169-170) 236 Issues and Decision Memorandum, supra, footnote 6, p. 20. 237 Ibid., pp. 49-50. See also United States' first written submission to the Panel, para. 37; and United States' second written submission to the Panel, para. 10. 238 Issues and Decision Memorandum, supra, footnote 6, pp. 50-59. See also United States' first written submission to the Panel, para. 37; and United States' second written submission to the Panel, para. 10. 239 Issues and Decision Memorandum, supra, footnote 6, pp. 57 and 60-61. See also United States' first written submission to the Panel, para. 37; and United States' second written submission to the Panel, para. 10. In its Issues and Decision Memorandum, the USDOC also categorized the second and third inferences under one broader finding that "evidence on the record establishes a pattern of practices on the part of the GOK to act upon that policy to entrust or direct lending decisions as part of the restructuring." (Issues and Decisions Memorandum, supra, footnote 6, p. 49) 240 Issues and Decision Memorandum, supra, footnote 6, pp. 61-62. Based on its findings in previous investigations, the USDOC determined that "the lending and credit practices of Citibank are not directed by the GOK." (Ibid., p. 17) 241 United States' appellant's submission, para. 47. 242 Panel Report, para. 7.35. 243 Ibid., para. 7.46. 244 United States' appellant's submission, footnote 56 to para. 48. 245 Ibid., paras. 48-49. 246 United States' appellant's submission, para. 49 (quoting The New Shorter Oxford English Dictionary, supra, footnote 36, Vol. 1, p. 458; and Oxford Dictionary of English, C. Soanes and A. Stevenson (eds) (Oxford University Press, 2004) 2nd edn, p. 352). 247 Korea's appellee's submission, paras. 96-97. 248 Ibid., para. 82. 249 Korea's and the United States' responses to questioning at the oral hearing. 250 In contrast, Article 15.1 of the SCM Agreement provides that a "determination of injury � shall be based on positive evidence". 251 Other provisions of the SCM Agreement relevant to a finding of entrustment or direction similarly suggest the investigating authority is to "base" its finding on evidence, but they do not clarify further the nature or quantum of evidence required to support such a finding. (See Articles 12.5, 12.8, 22.4, and 22.5 of the SCM Agreement) 252 Panel Report, paras. 7.35 and 7.46. 253 United States' appellant's submission, para. 49 (referring to Oxford Dictionary of English, supra, footnote 246, p. 352). 254 Ibid. (quoting The New Shorter Oxford English Dictionary, supra, footnote 36, p. 458). 255 Panel Report, para. 7.177. (emphasis added) 256 United States' appellant's submission, para. 47. 257 We note that third participants China and the Separate Customs Territory of Taiwan, Penghu, Kinmen, and Matsu similarly understand the Panel not to have imposed additional obligations on Members merely by requiring evidence to be "probative and compelling". (China's third participant's submission, para. 25; Third participant's submission of the Separate Customs Territory of Taiwan, Penghu, Kinmen, and Matsu, para. 7) 258 United States' appellant's submission, paras. 58-73. 259 Ibid., paras. 74-85. 260 Ibid., para. 87. 261 See Panel Report, para. 7.45 and Section VII.C.1(b), sub-sections (i)-(iii). 262 Ibid., para. 7.45. 263 Ibid. 264 Ibid. 265 Supra, para. 135. 266 Issues and Decision Memorandum, supra, footnote 6, p. 51. 267 Panel Report, para. 7.101. 268 The Panel observed that Korea disputed this fact, claiming instead that the meeting had been attended by an official of the Financial Supervisory Service, which was "not a governmental organization, but a special public corporation affiliated with FSC [and] functioning as an executive arm of the FSC". The Panel found that it did not need to resolve this disagreement because of its "finding regarding the DOC's treatment of this issue". (Panel Report, footnote 166 to 7.141) Similarly, given the basis for our finding of error in the Panel's approach to reviewing individual pieces of evidence, we do not need to resolve this factual disagreement of the participants. 269 Issues and Decision Memorandum, supra, footnote 6, p. 60 (quoting GOK Verification Report, p. 19 (Exhibit US-12 submitted by the United States to the Panel)). 270 Panel Report, para. 7.141. 271 Issues and Decision Memorandum, supra, footnote 6, p. 59. 272 Supra, para. 132. 273 Panel Report, para. 7.167. The reason relating to the GOK policy cannot be quoted because Korea identified it before the Panel as business confidential information. (See ibid., para. 7.166 and footnote 188 thereto) 274 Ibid., paras. 7.167-7.168. 275 See, for example, Panel Report, paras. 7.62, 7.77-7.78, 7.129, 7.141, and 7.167-7.168. 276 We note that the European Communities makes a similar observation:
(European Communities' third participant's submission, para. 20) (original underlining) 277 We agree with the United States, and third participants the European Communities and Japan, that this approach is particularly relevant in cases of entrustment or direction under Article 1.1(a)(1)(iv), where much of the evidence that is publicly-available, and therefore readily accessible to interested parties and the investigating authority, will likely be of a circumstantial nature. (United States', the European Communities', and Japan's responses to questioning at the oral hearing) Moreover, strictly speaking, entrustment or direction is not a pure fact. It is, rather, a legal assessment based on a proven set of facts. 278 This is not to say that a panel is prohibited from examining whether the agency has given a reasoned and adequate explanation for its determination, in particular, by considering other inferences that could reasonably be drawn from�and explanations that could reasonably be given to�the evidence on record. Indeed, a panel must undertake such an inquiry. (See infra, para. 186) 279 We address infra, in sub-section G, the implications of the Panel's approach to the evidence for its application of the proper standard of review. 280 Supra, para. 135. 281 Panel Report, para. 7.177. 282 As a result of its approach to the individual pieces of evidence, (see supra, paras. 144-152), several pieces of evidence erroneously deemed irrelevant by the Panel were not part of the Panel's "global review". 283 Supra, para. 150. 284 Panel Report, para. 7.51. 285 Kookmin Bank Prospectus (10 September 2001), p. 24, Exhibit US-45 submitted by the United States to the Panel (quoted in Issues and Decision Memorandum, supra, footnote 6, p. 58). 286 Kookmin Bank Prospectus (18 June 2002), p. 22, Exhibit US-46 submitted by the United States to the Panel (referred to in Issues and Decision Memorandum, supra, footnote 6, p. 58). 287 Issues and Decision Memorandum, supra, footnote 6, p. 58. 288 Panel Report, para. 7.127. 289 Issues and Decision Memorandum, supra, footnote 6, p. 61. 290 Panel Report, paras. 7.88, 7.102, and 7.141. See also paras. 7.116 and 7.121. 291 United States' appellant's submission, paras. 90 and 93. 292 Ibid., para. 90. 293 The United States does not appeal the Panel's conclusion of ex post rationalization with respect to a new factual argument advanced by the United States before the Panel. The United States argued before the Panel that the GOK had disciplined various credit rating agencies for giving Hynix a low credit rating. This was advanced in support of the USDOC's determination that the GOK exercised control or influence over Hynix's creditors. The Panel found that the USDOC had not referred in its determination to the coercion of credit rating agencies and, therefore, argument and evidence relating to such coercion "[fell] outside the scope of [the Panel's] proceedings." (Panel Report, para. 7.135) 294 Article 22.5 of the SCM Agreement provides, in relevant part:
295 Korea's appellee's submission, paras. 183-184. 296 Ibid., para. 185 (quoting United States' appellant's submission, para. 93). 297 Ibid. 298 United States' appellant's submission, para. 102; Korea's appellee's submission, para. 196. 299 Korea's response to questioning at the oral hearing. 300 The facts for which the United States sought to submit the excluded evidence were: (i) the influence held by the GOK over Hynix's creditors by virtue of the CRPA (Panel Report, paras. 7.72-7.74); (ii) decisions taken during the Economic Ministers' meetings that facilitated the extension of further credit to Hynix (Ibid., para. 7.102); (iii) pressure applied to KFB by the GOK (Ibid., para. 7.115); (iv) pressure applied to KorAm Bank by the GOK (Ibid., para. 7.121); and (v) attendance by FSC officials at a meeting of Hynix creditors. (Ibid., para. 7.141 and footnote 161 to para. 7.138) 301 See Issues and Decision Memorandum, supra, footnote 6, pp. 53-55 (discussing the influence held by the GOK over Hynix's creditors by virtue of the CRPA); ibid., pp. 50-52 (discussing the Economic Ministers' meetings); ibid., pp. 60-61 (discussing pressure applied to KFB by the GOK); ibid., p. 60 (discussing pressure applied to KorAm Bank by the GOK); ibid., p. 60 (discussing attendance by FSC officials at a meeting of Hynix creditors). 302 On this point, Japan expressed the view that "no provision of the DSU or the SCM Agreement requires that an [investigating] authority discuss each and every individual item of evidence, which the authorit[y] assessed, or was supportive of its conclusion, in a particular determination." (Japan's third participant's submission, para. 30) 303 We recall that previous decisions of the Appellate Body have addressed, in the context of the Anti-Dumping Agreement, the evidence a panel may consider when reviewing decisions of investigating authorities. (See Appellate Body Report, Thailand � H-Beams, paras. 109-110; and Appellate Body Report, EC � Tube or Pipe Fittings, paras. 131-132) 304 United States' appellant's submission, para. 100. 305 Supra, para. 135. 306 Issues and Decision Memorandum, supra, footnote 6, p. 55. 307 Article 29(1) of the CRPA, Exhibit GOK-22(c) submitted by Korea to the Panel. 308 Article 29(4) of the CRPA, Exhibit GOK-22(c) submitted by Korea to the Panel. 309 Article 29(5) of the CRPA, Exhibit GOK-22(c) submitted by Korea to the Panel. 310 Issues and Decision Memorandum, supra, footnote 6, p. 20. 311 At the oral hearing, Korea stated that, although the 2001 Hynix Audit Report referred to three creditors that exercised their rights to mediation, the correct number was four, as the GOK and Hynix submitted to the USDOC. (Issues and Decision Memorandum, supra, footnote 6, p. 61) Korea suggested at the oral hearing that the reference to three creditors in the 2001 Hynix Audit Report reflected an error in translation. We need not resolve this discrepancy for purposes of resolving this issue and refer throughout our discussion to three creditors. 312 Issues and Decision Memorandum, supra, footnote 6, p. 61. 313 Ibid. 314 Panel Report, para. 7.84 (quoting Issues and Decision Memorandum, supra, footnote 6, p. 55). 315 Ibid. 316 Ibid., para. 7.85. 317 Ibid., para. 7.87. 318 Ibid., para. 7.85. 319 United States' appellant's submission, paras. 108-111. 320 Ibid., paras. 112-114. 321 Ibid., para. 117. 322 Korea's appellee's submission, para. 198. 323 Korea's appellee's submission, paras. 198-199 and 210. 324 Ibid., para. 203. 325 Panel Report, para. 7.84. 326 Ibid., para. 7.85. 327 Ibid., para. 7.82. 328 See Appellate Body Report, US � Cotton Yarn, para. 78. 329 Ibid. 330 Although the Panel also referred to Article 29(5) of the CRPA and the testimony of one of the USDOC's experts, the Panel did not find that either of these pieces of evidence could have informed the USDOC about the recourse to mediation�rather than the "possibility" of such recourse�by three Hynix creditors. (Panel Report, para. 7.84 and footnote 98 to para. 7.87) 331 Hynix Audit Report (2001), p. 40 (Exhibit US-125 submitted by the United States to the Panel) (quoted in Panel Report, para. 7.85). 332 Korea's response to questioning at the oral hearing. 333 Supra, para. 168. 334 In its questionnaire, the USDOC had asked:
(Hynix's response to the USDOC Questionnaire, 28 January 2003, p. 60 (Exhibit US-129 submitted by the United States to the Panel))
(Ibid.) 335 In its supplemental questionnaire to Hynix, the USDOC had asked:
(USDOC Supplemental Questionnaire to Hynix, 11 February 2003 (Exhibit US-130 submitted by the United States to the Panel) Question 54, p. 10) 336 Korea's response to questioning at the oral hearing. 337 Panel Report, para. 7.3. 338 United States' appellant's submission, para. 119; United States' response to questioning at the oral hearing. 339 Korea's appellee's submission, para. 214. 340 Ibid. 341 Appellate Body Report, US � Lead and Bismuth II, para. 51. 342 Appellate Body Report, US � Steel Safeguards, para. 299 (referring to Appellate Body Report, Argentina � Footwear (EC), para. 121). 343 Panel Report, paras. 7.2-7.3; United States' appellant's submission, footnote 181 to para. 123; Korea's appellee's submission, paras. 213-214. 344 See, for example, Appellate Body Report, US � Lamb, paras. 103 and 106. 345 In this respect, we note that disputes under the Agreement on Safeguards as well as the SCM Agreement are subject only to the standard of review in Article 11 of the DSU, whereas the Anti-Dumping Agreement contains a specific standard of review in Article 17.6, which must be applied in conjunction with Article 11 of the DSU for disputes arising under the Anti-Dumping Agreement. (Appellate Body Report, US � Hot-Rolled Steel, para. 55) 346 See, for example, Appellate Body Report, US � Lamb, para. 105; and Appellate Body Report, US � Cotton Yarn, paras. 75-78. 347 Supra, para. 138. 348 We understand the relevant definitions of the term "determine" to include "[c]onclude from reasoning or investigation, deduce" as well as "[s]ettle or decide (a dispute, controversy, etc., or a sentence, conclusion, issue, etc.) as a judge or arbiter". (Appellate Body Report, US � Corrosion-Resistant Steel Sunset Review, para. 110 (quoting Shorter Oxford English Dictionary, 5th edn, W.R. Trumble, A. Stevenson (eds) (Oxford University Press, 2002), Vol. 1, p. 659). (emphasis added)) 349 Article 22.5 of the SCM Agreement. 350 Article 22.4(iii) of the SCM Agreement (incorporated by reference into Article 22.5). (emphasis added) 351 Compare Appellate Body Report, US � Steel Safeguards, paras. 276-279; and Appellate Body Report, US � Lamb, para. 103. 352 Compare Appellate Body Report, US � Lamb, para. 106. 353 Ibid., para. 114; and Appellate Body Report, US � Wheat Gluten, paras. 55-56. 354 Appellate Body Report, US � Wheat Gluten, para. 162. 355 Supra, paras. 149-152. 356 Supra, paras. 154-157. 357 We observe that the European Communities agrees that, by failing to consider the evidence in its totality, the Panel effectively conducted a de novo review of the USDOC's finding of entrustment or direction. (European Communities' third participant's submission, para. 20) 358 Supra, para. 165. 359 Supra, para. 179. 360 Panel Report, para. 7.177. 361 Supra, para. 118. 362 Panel Report, para. 7.45. 363 We note that Korea agrees that a finding of error in this case under Article 11 of the DSU, with respect to the Panel's standard of review, would require a reversal of the Panel's finding that the USDOC's subsidy determination is inconsistent with Article 1.1(a)(1)(iv) of the SCM Agreement. (Korea's response to questioning at the oral hearing) 364 Issues and Decision Memorandum, supra, footnote 6, p. 48; see also Panel Report, paras. 7.146-7.155. 365 Ibid., pp. 48-49:
366 Supra, para. 116. 367 See, for example, Panel Report, footnote 166 to para. 7.141 and para. 7.167. 368 Article 1.1(b) of the SCM Agreement. 369 Article 1.2 of the SCM Agreement. 370 Panel Report, para. 7.190. 371 Ibid. (footnote omitted) 372 Ibid. (footnote omitted) 373 Ibid. 374 Panel Report, paras. 7.206 and 7.208. (footnote omitted) The Panel found that the USDOC could have properly found specificity in respect of Group A creditors. (Ibid., para. 7.207) 375 United States' appellant's submission, para. 131. (footnote omitted) 376 Korea's response to questioning at the oral hearing (referring to Panel Report, para. 7.180). 377 Because we have reversed the Panel's findings of inconsistency with Article 1.1(a)(1)(iv) and 1.1(b), we do not address whether a Member may be found to be acting inconsistently with a definitional provision, such as Article 1 of the SCM Agreement. (Appellate Body Report, US � FSC (Article 21.5 � EC), para. 85)
378 Appellate Body Report, US � Softwood Lumber IV, para.
113. |
|
