tutorial video tutorial video Discover the CiteMap in 3 minutes
Source(s) of the information:
Source(s) of the information:

Lawyers, other representatives, expert(s), tribunal’s secretary

Recourse to Article 21.5 of the DSU by the European Communities - Final Report of the Panel

I. INTRODUCTION

1.1.
On 17 March 2004, the European Communities requested consultations with the United States pursuant to Articles 4 and 21.5 of the Understanding on Rules and Procedures Governing the Settlement of Disputes (DSU) and Article 30 of the Agreement on Subsidies and Countervailing Measures (SCM Agreement) with respect to the United States' new privatization methodology, its application to the determinations set out in the request and the findings in the sunset reviews of: Certain Corrosion-Resistant Carbon Steel Flat Products from France (C‑427‑810)(Case No. 9); Cut-to-Length Carbon Steel Plate from United Kingdom (C–412–815) (Case No. 8); and, Cut-to-Length Carbon Steel Plate from Spain (C–469–804) (Case No. 11)1
1.2.
The request was circulated in document WT/DS212/14 of 19 March 2004. Consultations were held on 24 May 2004 and, although they allowed a better understanding of respective positions, they failed to settle the dispute.
1.3.
On 16 September 2004, the European Communities requested the Dispute Settlement Body (DSB) to establish a panel pursuant to Articles 6 and 21.5 of the DSU, Article 30 of the SCM Agreement, and Article XXIII of the General Agreement on Tariffs and Trade 1994 (GATT 1994). The request did not include the United States' new privatization methodology.2
1.4.
At its meeting on 27 September 2004, the DSB decided, in accordance with Article 21.5 of the DSU, to refer to the original panel the matter raised by the European Communities in document WT/DS212/15.
1.5.
At that meeting, the parties to the dispute agreed that the Panel should have standard terms of reference. The Panel's terms of reference are, therefore, the following:

"To examine, in the light of the relevant provisions of the covered agreements cited by the European Communities in document WT/DS212/15, the matter referred to the DSB by the European Communities in that document, and to make such findings as will assist the DSB in making the recommendations or in giving the rulings provided for in those agreements".3

1.6.
The Panel was composed of the original members:

Chairman: Mr Gilles Gauthier

Members: Ms Marie-Gabrielle Ineichen-Fleisch

Mr Michael Mulgrew

1.7.
Brazil, China and Korea reserved their rights to participate in the Panel's proceedings as third parties.
1.8.
The Panel met with the parties on 1-2 March 2005. It met with the third parties on 2 March 2005.

II. BACKGROUND

2.1.
This dispute concerns the parties' disagreement as to the existence or consistency with the SCM Agreement and the GATT 1994 of measures taken by the United States to comply with the DSB recommendations and rulings in the dispute US – Countervailing Measures on Certain EC Products.

A. MEASURES SUBJECT TO THE ORIGINAL PROCEEDINGS

2.2.
The original panel and Appellate Body proceedings concerned the WTO-consistency of 12 countervailing duty determinations by the United States, which included original investigations, administrative reviews, and sunset reviews.4 The imposition of countervailing duties in these 12 determinations was based on the application by the USDOC of two different change-in-ownership methodologies (the so‑called "gamma" and "same person" methodologies). The current Article 21.5 proceedings only concern three of the four sunset reviews in the original panel5: Certain Corrosion-Resistant Carbon Steel Flat Products from France (C‑427‑810)(Case No. 9); Cut-to-Length Carbon Steel Plate from United Kingdom (C–412–815) (Case No. 8); and, Cut-to-Length Carbon Steel Plate from Spain (C–469–804) (Case No. 11).6
2.3.
The European Communities also brought a claim in the original proceedings concerning Section 771(5)(F) of the US Tariff Act of 1930.

B. PANEL AND APPELLATE BODY FINDINGS IN THE ORIGINAL PROCEEDINGS

2.4.
The original panel found that the 12 countervailing duty determinations were inconsistent with Articles 10, 14, 19.4, 21.1 and 21.3 of the SCM Agreement. It also found that Section 1677(5)(F) was inconsistent with the United States' obligations under the SCM Agreement and Article XVI:4 of the WTO Agreement.7
2.5.
The Appellate Body upheld the Panel's findings that the determinations of United States' Department of Commerce (USDOC) in the 12 countervailing duty cases were inconsistent with the SCM Agreement because the USDOC had failed to ascertain the continued existence of a benefit following the privatization of recipients of prior non-recurring financial contributions. However, it reversed the Panel's finding that an investigating authority must always conclude that a benefit no longer exists for a firm that has been privatized at arm's length and for fair market value (FMV). The Appellate Body also reversed the Panel's finding that the relevant United States internal legislation was, as such, inconsistent with the United States' obligations under the SCM Agreement and Article XVI:4 of the WTO Agreement.8

C. CURRENT PROCEEDINGS

2.6.
On 8 January 2003, the DSB adopted the Appellate Body Report and the Panel Report, as modified by the Appellate Body. The resulting DSB rulings included the recommendation that the United States brings its administrative practice (the "same person" methodology) and the twelve individual determinations found to be inconsistent with the SCM Agreement and the GATT 1994 into conformity with its WTO obligations.
2.7.
On 27 January 2003, the United States informed the DSB that it intended to implement the recommendations and rulings of the DSB in a manner consistent with its WTO obligations. The parties agreed under Article 21.3(b) of the DSU that the United States had until 8 November 2003 to implement the recommendations and rulings of the DSB.
2.8.
In June 2003, the United States finalized a new privatization methodology, published in the US Federal Register.9 The new methodology was then applied to the 12 individual determinations found to be inconsistent with the SCM Agreement and the GATT 1994. On 17 November 2003, the USDOC published in the US Federal Register a Notice of Implementation for these 12 countervailing duty determinations.10 According to this Notice, only in four of these re-determinations, which concerned sunset reviews, did the USDOC maintain the existing countervailing duties. In the other re-determinations, the measures were either reduced or revoked. At the DSB meeting of 7 November 2003, the United States stated that it had fully complied with the DSB recommendations and rulings on this issue. The European Communities did not agree.
2.9.
On 17 March 2004, the European Communities requested consultations under Article 21.5 of the DSU. In that request, the European Communities claimed that both the new privatization methodology and its application to three of the above-mentioned sunset review re-determinations11 were inconsistent with Article VI:3 of GATT 1994 and Articles 10, 14, 19.4, 21.1 and 21.3 of the SCM Agreement.
2.10.
After consultations had failed, the European Communities submitted a request for establishment of a panel under Article 21.5 of the DSU on 16 September 2004 this time challenging only the USDOC's re-determination of the sunset reviews on Certain Corrosion-Resistant Carbon Steel Flat Products from France; Cut-to-Length Carbon Steel Plate from United Kingdom; and, Cut-to-Length Carbon Steel Plate from Spain. This Panel was established pursuant to that request on 27 September 2004.

III. PARTIES' REQUESTS FOR FINDINGS AND RECOMMENDATIONS

A. THE EUROPEAN COMMUNITIES

3.1.
The European Communities requests the Panel to find that:

· The United States has failed to implement the recommendations and rulings of the DSB and acted inconsistently with its obligations under Articles 10, 14, 19.4, 21.1, and 21.3 of the SCM Agreement and Article VI:3 of the GATT 1994 because it did not determine properly whether the privatized French company Usinor in Certain Corrosion-Resistant Carbon Steel Flat Products from France continued to receive any benefit from financial contributions previously bestowed on state-owned producers;

· The United States has acted inconsistently with its obligations under Article 21.3 of the SCM Agreement because it failed to consider evidence on the record in Cut-to-Length Carbon Steel Plate from the United Kingdom and Cut-to-Length Carbon Steel Plate from Spain demonstrating that the benefits from programmes found to confer countervailable subsidies no longer existed;

· The United States has failed to implement the recommendations and rulings adopted by the DSB and acted inconsistently with its obligations under Articles 10, 14, 19.4, 21.1, and 21.3 of the SCM Agreement and Article VI:3 of the GATT 1994 by refusing to examine whether, and for failing to find that, the privatizations of BS plc in Cut-to-Length Carbon Steel Plate from the United Kingdom and Aceralia in Cut-to-Length Carbon Steel Plate from Spain were at arm’s length and for FMV; and, that

· The United States has failed to meet its obligations under Article 21.3 of the SCM Agreement because it did not examine in Certain Corrosion-Resistant Carbon Steel Flat Products from France, Cut-to-Length Carbon Steel Plate from the United Kingdom and Cut-to-Length Carbon Steel Plate from Spain whether expiry of the duty would be likely to lead to continuation or recurrence of material injury.12

B. THE UNITED STATES

3.2.
The United States contends that it has fully implemented the recommendations and rulings of the DSB in this dispute. With respect to Usinor's privatization – the only substantive matter that, in the view of the United States, is properly before this Panel – the USDOC’s conclusion that the employee sales did not satisfy the arm’s-length/FMV standard comports with the recommendations and rulings of the DSB, as does USDOC’s finding that a corresponding amount of the remaining benefit from pre-privatization subsidies was not extinguished by the privatization. Thus, the United States believes that the European Communities' claims against implementation of the DSB recommendations and rulings are not meritorious. The United States therefore requests this Panel to find that the United States properly implemented those recommendations and rulings.13

IV. ARGUMENTS OF THE PARTIES

A. FIRST WRITTEN SUBMISSION OF THE EUROPEAN COMMUNITIES

1. Introduction

4.1.
In the view of the European Communities, the measures taken with respect to the three sunset reviews at issue in this case failed to bring the United States into compliance with the DSB recommendations and rulings.14 The mandate of the present Panel is to determine the consistency of the measures taken to comply with the provision or provisions cited in the Panel request.15

2. Certain corrosion-resistant carbon steel flat products from France

4.2.
In its France Section 129 determination, the USDOC found that "with the exception of the employee offering, which constituted 5.16 per cent of the sale, the privatization of Usinor was at arm's length and for fair market value"16. It nevertheless concluded that the "revocation of the order on [corrosion-resistant carbon steel flat products from France] would be likely to lead to continuation or recurrence of a countervailable subsidy".17 There had been no basis for the USDOC's conclusion. The SCM Agreement obliges a Member to determine whether a privatized producer received a benefit from the financial contribution previously bestowed on state-owned producers. This cannot be deduced by considering a small part of a sale of shares in isolation.
4.3.
It is clear under WTO case law18, that in determining the likelihood of continuation or recurrence of a subsidization, a Member must determine whether the privatized producers received any benefit from the financial contributions previously granted to state-owned producers and that the producers were presumed not to have received any benefit if the privatization had been at arm's length, for FMV, and under conditions that did not seriously distort the market in which the privatization occurred. As discussed below, the privatized producers did not receive any such benefit following Usinor's privatization.

(a) The privatization of Usinor as a whole was at arm's length and for FMV

4.4.
In its Section 129 determination, the USDOC considered whether the sale of Usinor had been at arm's length and for FMV from the perspective of four separate categories of purchasers. Such an evaluation was not conclusive regarding whether the privatized firm as a whole received a benefit as required under WTO case law. The investigating authority must consider whether the entire firm had been sold for FMV, not just whether the sales to an individual category of purchasers were at FMV. The record demonstrates that the privatization of Usinor as a whole had been for FMV, because the average price paid for Usinor's shares exceeded the average price per share recognized by the Privatization Commission as necessary to recoup the value of the company. Excluding shares retained by or sold to the Government of France (GOF), directly or indirectly, the total price paid for [XXX] shares in Usinor's privatization was FF [XXX], for an average price of FF [XXX] per share.19 This price exceeds the average minimum price recommended by the Privatization Commission for Usinor, FF 84.43, based on the total number of shares to be offered, 186,544,395, at the minimum price of FF 15,750 billion.20 The average price per share was also within the range recognised by the USDOC as constituting the market clearing price (i.e., between FF 86 to 89 per share).21 Further, if the GOF's sale of 7.7 per cent of Usinor shares in 1997 were included in the calculation, the average price per share would increase to FF [XXX].22
4.5.
Thus, when considered in the context of the privatization as a whole, there had been no doubt that Usinor shares were sold at FMV. Such a finding was consistent with the USDOC's own findings in two parallel proceedings in which the USDOC found that the privatization of Usinor had taken place at arm's length and for FMV.23 As there had been no evidence that the GOF distorted the broader market conditions in any way, and the USDOC had not so found24, it follows that the privatized producers did not receive any benefit from the previous financial contributions to the state-owned producers.

(b) The shares sold through the employee/retiree offering were at arm's length and for FMV

4.6.
The European Communities believed that the above analysis was sufficient to demonstrate that the USDOC wrongly decided to continue the measures. Even considering, on an alternative basis, the employee/retiree offering independently, the evidence demonstrated that the shares sold to this category of purchasers were at arm's length and for FMV. WTO case law does not provide guidance on what type of transactions are considered to be at "arm's length". As an illustration of the meaning of an "arm's-length" transaction, therefore, the European Communities referred to the Black's Law Dictionary which defines "arm's length" as "[o]f or relating to dealings between two parties who are not related or not on close terms and who are presumed to have roughly equal bargaining power; not involving a confidential relationship".25 For the following reasons, the employee/retiree offering did not involve related parties, and, therefore, constituted an arm's-length transaction.
4.7.
At the time of Usinor's privatization, the company had been wholly owned, directly or indirectly, by the GOF.26 The buyers in the employee/retiree offering included employees of Usinor - not the GOF.27 Although the GOF participated in shareholder meetings and oversaw the financial management of the company, it did not "intervene in the day-to-day management of the Company, and the Company... conducted its day-to-day operations in a manner similar to that of other major international non-state controlled steel producers".28 Thus, the employee purchasers, including former employees, were at least two steps removed from having any relationship with the GOF. Even if some type of relationship existed between the seller and the buyers in this case, it would not have affected the price at which the GOF sold Usinor shares which had been based on independent analyses of the valuation of the company.29 In conducting its privatization analysis in the Section 129 determination, the USDOC concluded that "the employees of Usinor were related to Usinor" and, therefore, "sales of shares to Usinor employees... did not constitute an arm's-length transaction".30 The USDOC did not even consider the relationship between the former employees of Usinor (or its affiliates) and the seller in its analysis. Thus, there was no basis for the USDOC's conclusion that the employee offering had not been at arm's length, and, in fact, evidence on the record indicates otherwise.
4.8.
Further, the sale of Usinor shares in the employee/retiree offering had also been at FMV. As discussed in the USDOC's Section 129 determination, the eligible purchasers had two options: (1) they could purchase the shares at the French public offering price of FF 86 per share; or (2) they could pay a discounted price of FF 68.80 per share, with an extended payment period, if they agreed to hold the shares for a minimum of two years.31 The discount offered to employees and former employees of Usinor reflects the risk assumed by the buyers for the resale restrictions on the shares purchased. That is, because the shares purchased by employees/former employees were less liquid, they were less valuable on the market since it was unclear whether, after two years, the price for the shares would be greater or lower. Thus, the price offered to employees/former employees had been at FMV, accounting for the risk assumed by the purchasers for the non-liquidity of the shares.32

(c) The United States is not in compliance with its WTO obligations with respect to the USDOC determination in theFrench case

4.9.
As indicated above, the SCM Agreement requires investigating authorities to examine the conditions of a privatization and to determine whether the privatized producers received any benefit from the prior subsidization to the state-owned producers before deciding to continue to countervail pre-privatization, non-recurring subsidies in sunset reviews. Where a privatization is at arm's length and for FMV, there is a presumption that the benefit provided to the state-owned producer has been extinguished. Here, that presumption was warranted and had not been rebutted. Accordingly, by failing to revoke the countervailing duties imposed on corrosion-resistant carbon steel flat products from France, the United States had acted inconsistently with its obligations under Article VI:3 of the GATT 1994 and Articles 10, 14, 19.4, 21.1, and 21.3 of the SCM Agreement, and had failed to implement the recommendations and rulings of the DSB.

3. Cut-to-length Carbon Steel (CTL) Plate from the UK and from Spain

4.10.
The USDOC's "likelihood" determinations with respect to the UK and Spain were inconsistent with the SCM Agreement and the GATT 1994 because the USDOC refused to consider: (a) evidence on the record demonstrating that in both cases subsidies would not continue or recur; and (b) whether, under the new privatization methodology, BS plc had been privatized in a manner that extinguished the benefit conferred by non-recurring pre-privatization subsidies.

(a) The United States breached its WTO obligations by finding a likelihood of continuation or recurrence on the basis of insufficient evidence

4.11.
In both the UK andSpanish cases, the USDOC failed to meet its obligations under Article 21.3 of the SCM Agreement. Had the USDOC performed its "likelihood" determination as required under the SCM Agreement and considered all of the evidence on the record, it would have determined that the subsidy programmes on which it relied had expired. In the UK case, Corus and the Government of the UK (GUK) submitted evidence showing that Glynwed33 no longer produced the product under review.34.
4.12.
This point was reiterated in the European Communities' submission, which explained that "[w]ith regard to any alleged non-recurring subsidies, Glynwed sold its facilities for producing 'universal wide flats' to the United States' company Niagara LaSalle, well before the sunset review. This was an arms-length sale between two private companies, and would therefore have removed the benefit of any prior subsidies to the buyer".35 Moreover, Corus and the GUK also submitted evidence to the USDOC in the Section 129 proceeding demonstrating that all the subsidy programmes not specific to BS plc, on which the USDOC may have relied in calculating the original CVD rate for Glynwed, had been abolished or were no longer available to the UK steel industry. Specifically, they reviewed the four UK programmes36 that were not specific to British Steel, and which therefore could have been included in the USDOC's calculation of Glynwed's subsidization rate in 1993.
4.13.
These explanations as to the abolition of all the subsidy programmes not specific to British Steel, on which the original finding of subsidization for Glynwed may have been based, remain entirely unrebutted, and they eliminate any support for the assertion that the subsidization of Glynwed by either the GUK or the European Communities would continue or recur if the CVD order on CTL Plate were revoked.37 For these reasons, the USDOC acted inconsistently with the United States' obligations under Article 21.3 of the SCM Agreement by refusing to consider evidence on the record indicating that the non-privatized producer (Glynwed) no longer manufactured the product under review and that the subsidy programmes found to be countervailable in the original investigation no longer existed or were no longer available to steel producers.
4.14.
The same was true for the Spanish case. The Government of Spain (GOS) and the European Communities both submitted documents during the sunset review stating that: "most of the schemes countervailed no longer exist or have ceased to provide any meaningful benefits to the current exporters of the product concerned[,] especially in view of the fact that in the meantime most of the subsidy would have been amortized".38 The European Communities then reviewed the schemes39 that had originally been countervailed, and noted that they no longer existed or would shortly cease to exist.
4.15.
As in the UK case, the abolition of the programmes that had been found to confer subsidies on Aceralia entirely eliminated any support for the assertion that the subsidization of the privatized Spanish producer by either the GOS or the European Communities would continue or recur if the CVD order on CTL Plate were revoked. For these reasons, the USDOC acted inconsistently with the United States' obligations under Article 21.3 of the SCM Agreement by refusing to consider evidence on the record that demonstrated that the subsidy programmes found to be countervailable in the original investigation no longer existed or were no longer available to steel producers.
4.16.
The USDOC justified its failure to consider this evidence on the ground that in a Section 129 proceeding, it need only deal with the specific issue in which the WTO found a lack of compliance.40 But this restrictive interpretation of the USDOC's role in bringing its practice into conformity with the WTO's findings was based on a serious misunderstanding of a Member's obligations following a determination that one of its measures was inconsistent with the SCM Agreement. It is not sufficient for a Member merely to eliminate one erroneous aspect of the measure at issue, if to do so exposes additional failures to comply with the obligations imposed by the SCM Agreement. Nor may the Member refuse to attempt to bring those additional failures into compliance on the ground that they were not considered in previous stages of the WTO dispute settlement proceedings. To the contrary, as discussed above, in an Article 21.5 proceeding, the Panel's mandate is to "examine whether the new measure... [is] in conformity with, adhering to the same principles of[,] or compatible with" the WTO Agreement cited by the complaining Member in its Panel Request.41 Accordingly, a Member's duty remains, throughout such proceedings, to ensure that the challenged measures were brought into compliance with the Member's obligations under the WTO.

(b) The United States also breached its WTO obligations by failing to examine the privatizations of BS plc and Aceralia

4.17.
The USDOC also failed to act consistently with the SCM Agreement in these two Section 129 determinations, because it refused to examine whether the privatizations of BS plc and Aceralia were at arm's length and for FMV. In light of the arguments presented above, the USDOC could not have arrived at a conclusion of likelihood or recurrence of subsidization without having examined the nature of the UK and Spanish privatizations. Because it found independent bases for its likelihood determinations, the USDOC did not even consider whether the privatizations of BS plc or of Aceralia were at arm's length and for FMV pursuant to the methodology set forth in its Modification Notice. The USDOC's reasons for its failure were insufficient to satisfy its obligations under the SCM Agreement.
4.18.
The undisputed evidence before the USDOC in fact demonstrated that the privatizations of BS plc and Aceralia were at arm's length and for FMV, and that, therefore, the pre-privatization subsidies granted to those companies had been extinguished. Indeed, in the Section 129 determination in the UK case, Corus and the GUK submitted comments explaining that in a prior proceeding, the USDOC had "concluded that the privatization of British Steel was an arm's length, fair market value transaction".42 Based on this conclusion, the Appellate Body noted that "[t]he USDOC analyzed the sales conditions of the privatizations in two of the underlying sunset reviews [citing, inter alia, the CTL Plate from UK sunset review] and three of the original investigations..., concluding that those five privatizations took place at arm's length and for fair market value".43 Similarly, in the Spanish case, the evidence on the record showed that the privatization of Aceralia occurred at arm's length and for FMV. The GOS stated in the Section 129 proceedings that "[n]one of the purchasers [in the privatization] was related to the seller (the Spanish State) in any manner" and, therefore, "[t]he sale was at arm's length".44 As to FMV, the record indicated that three categories of purchasers of Aceralia's shares participated in the privatization process: (1) 35 per cent of Aceralia's shares were sold to the Arbed Group of Luxembourg; (2) 12.2 per cent of its shares were sold to the Spanish industrial groups Corporación J.M. Aristrain and Corporación Gestamp S.L.; and (3) 52.76 per cent of Aceralia's shares were sold to the general public by public offering.45 The share price paid by the first two categories was € 17.75, while the share price of the public offering was ESP 2080 (€ 12.50).46 When the number of shares that were purchased by each of the three categories was multiplied by their respective share prices, the resulting total value at which the company was privatized was approximately € 1,871,488,430. This amount was at the high end of the range of valuations obtained from independent experts at the commencement of the privatization process, at the request of the GOS – which ranged from € 1296.08 million to € 1895.69 million.47
4.19.
By refusing to examine the privatizations of BS plc and Aceralia in its Section 129 determinations, despite the undisputed evidence that they had occurred at arm's length and for FMV, the USDOC's determinations were not in compliance with the conclusion of the Appellate Body upholding the Panel's finding that the United States acted inconsistently with its obligations under Articles 10, 14, 19.4, 21.1, and 21.3 of the SCM Agreement by "imposing and maintaining countervailing duties without determining whether a 'benefit' continues to exist... ".48

4. Failure to consider injury

4.20.
Finally, the European Communities noted that the United States had maintained the above three measures without examining whether expiry of the duty would be likely to lead to continuation or recurrence of material injury as required by Article 21.3 of the SCM Agreement. This was a further violation of the SCM Agreement and a further reason why the United States had failed to bring all three of the measures into conformity with its obligations under the SCM Agreement.

B. FIRST WRITTEN SUBMISSION OF THE UNITED STATES

1. Claims not properly before the Panel

4.21.
The European Communities had improperly advanced claims regarding both injury and subsidization which are aspects of the determinations other than the "measure taken to comply" and therefore the Panel should reject them.

(a) Injury

4.22.
The European Communities' injury argument consists of one paragraph that simply asserts the United States erred by not examining whether the expiry of the duty would be likely to lead to continuation or recurrence of injury in accordance with Article 21.3 of the SCM Agreement. The ITC did conduct this examination as part of the original sunset review, but the European Communities did not, in the original proceedings, challenge this examination, nor, consequently, did the DSB issue rulings relating to this determination. Therefore, there was no basis for the European Communities' claim, which had not been properly brought in the context of this Article 21.5 proceeding. The recommendations and rulings made by the DSB in the underlying proceeding do not pertain to the ITC determination. Thus, there can be no "measure taken to comply" with respect to the ITC determination. As the Appellate Body stated in EC – Bed Linen (Article 21.5 – India): "we do not see why that part of a redetermination that merely incorporates elements of the original determination... would constitute an inseparable element of the measure taken to comply with the DSB rulings in the original dispute".49 Further, in that dispute, "India [sought] to challenge an aspect of the original measure which has not changed, and which the European Communities did not have to change, in order to comply with the DSB recommendations and rulings to make that measure consistent".50 Likewise, in this dispute, the ITC determination had not changed, nor did the United States have to change it in order to comply with the DSB recommendations and rulings. Therefore, the European Communities' claim regarding injury was not within the terms of reference of this Article 21.5 proceeding.
4.23.
The United States notes that the Panel and the Appellate Body made their findings in EC – Bed Linen in response to arguments advanced by the European Communities. For example, the European Communities argued that: "India did not challenge these aspects of the original determination in the original dispute, that the European Communities therefore had no implementation obligation with respect to those aspects of its original determination, and that the European Communities therefore did not modify them in the redetermination".51 Before the Appellate Body, the European Communities argued that "it was under no obligation to correct... its findings... because the original panel had not ruled that these findings were inconsistent with Article 3.5".52 The European Communities' logic as expressed in EC – Bed Linen and adopted by the DSB was equally applicable in this dispute.

(b) Subsidization

4.24.
The European Communities had argued that USDOC's findings in the revised sunset reviews about Glynwed in the UK case and the recurring subsidy programmes in the Spanish case were inconsistent with the SCM Agreement. The findings that subsidies continued to benefit Glynwed and that there were recurring subsidy programs in the Spanish case were made in the original sunset reviews. The European Communities did not challenge these aspects of the USDOC determinations during the proceedings before the original panel, and neither the original panel nor the Appellate Body made any findings concerning these programs, which did not involve use of the privatization methodology. As the Appellate Body in EC – Bed Linen noted – and argued there by the European Communities – a "measure taken to comply" does not include unchanged findings from the original determination. Because the findings concerning Glynwed and the recurring subsidy program in Spain were unchanged, and because the recommendations and rulings did not require them to be changed, the European Communities cannot use the Article 21.5 process to challenge them now.
4.25.
The European Communities had argued that the mandate of an Article 21.5 panel is to review the revised measure for its consistency with the agreement in question and that the Panel is not limited to reviewing only the matters on which the DSB adopted recommendations and rulings. In support of this proposition, the European Communities refers to the Appellate Body report in the Article 21.5 proceedings in Canada – Aircraft and the Panel report in the Article 21.5 proceedings in EC – Bananas (Ecuador). However, the European Communities misunderstands the relevance of Canada – Aircraft and fails to take into account the Appellate Body's further clarifications in EC – Bed Linen. In Canada – Aircraft, for example, the Panel correctly evaluated whether the measure taken to comply was consistent with the Member's WTO obligations; however, as noted above, the measure taken to comply does not include unchanged aspects of a measure not covered by the recommendations and rulings of the DSB. Similarly, in EC – Bananas (Ecuador), the Panel also evaluated whether the measure taken to comply had been consistent. In this dispute, USDOC's findings regarding Glynwed and the recurring subsidy programs in Spain were unchanged and did not have to be changed in response to DSB rulings and recommendations. Therefore, they were not "measures taken to comply" and were not subject to Article 21.5.
4.26.
The United States also notes that the European Communities' panel request states its claim under Article 21.5 as follows: "The United States failed to properly determine whether, in these cases, there was continuation or recurrence of subsidization and injury, because it did not examine the nature of the privatizations in question and their impact on the continuation of the alleged subsidization".53 Thus, the European Communities explicitly recognized that the privatization analysis had been the sole basis for its Article 21.5 challenge to the new sunset determination. The privatizations and their impact on the continuation of subsidization were precisely the matters that USDOC examined in the revised sunset determinations in response to the DSB recommendations and rulings. The European Communities' claims in the present proceeding regarding Glynwed and the recurring subsidy programs in Spain do not involve USDOC's treatment of the British and Spanish privatizations, or their impact and thus were not within the scope of the European Communities' Article 21.5 panel request or this proceeding.

2. The European Communities bears the burden of proving its claims

4.27.
It has been well established that the complaining party in a WTO dispute bears the burden of coming forward with argument and evidence that establish a prima facie case of a violation. If the balance of evidence and argument is inconclusive with respect to a particular claim, the European Communities, as the complaining party, must be found to have failed to establish that claim. The European Communities had not met its burden, for the reasons described as follows. Article 11 of the DSU sets forth the standard of review for this Panel. With respect to disputes involving a determination made by a domestic authority based upon an administrative record, the Appellate Body, in US – Cotton Yarn, summarized the role of a panel under Article 11.54 Thus, it was clear that the Panel's task was not to determine whether the privatization had been at arm's-length or for FMV, but rather whether USDOC properly established the facts and evaluated them in an unbiased and objective way. Put differently, the Panel's task was to determine whether a reasonable, unbiased person, looking at the same evidentiary record as the USDOC, could have – not would have – reached the same conclusions.

3. USDOC's application of its new privatization methodology to the revised French sunset review is consistent with the SCM Agreement and the DSB recommendations and rulings

4.28.
As noted above, both parties accepted the general principle that non-recurring subsidies may be allocated over time. The DSB found, however, that where the subsidy recipient is privatized in an arm's-length sale for FMV, the benefit stream is presumed to be extinguished. The terms "arm's length" and "fair market value" were not used in the text of the SCM Agreement and were not to be interpreted as though they were treaty text. As demonstrated below, USDOC's revised determination in the corrosion-resistant steel from France case was entirely consistent with the DSB recommendations and rulings, and was supported by the evidence on the record before USDOC. This Panel should reject the European Communities' claims in their entirety.

(a) USDOC's analysis of the Usinor privatization is entirely consistent with the SCM Agreement and the DSB recommendations and rulings

4.29.
If the investigating authority finds that a privatization was at arm's length and for FMV, then in making its subsidy determination, it must find, absent evidence to the contrary, that the pre-privatization subsidy benefits have been extinguished. USDOC's new privatization methodology implements these principles verbatim. Because the arm's-length/FMV test was not found in the text of the SCM Agreement, that Agreement does not provide a methodology for analyzing whether a transaction is at arm's-length or for FMV. It is therefore within a Member's discretion to develop a reasonable methodology for addressing these issues. Moreover, because of the fact-intensive nature of the inquiry, the appropriate analytical approach may vary from case to case. Nevertheless, the European Communities argued that USDOC was obligated to evaluate whether the average price for all of the French privatization transactions was at FMV. In the European Communities' view, therefore, it was impermissible for USDOC, in examining the French privatization transactions, to take into account the evidence that these transactions were accomplished by means of four distinct groupings of purchasers. There was simply no basis for this view. To the contrary, USDOC analyzed the issue of whether Usinor was sold at arm's length and for FMV in a manner that had been entirely consistent with those concepts as used by the Panel and the Appellate Body.
4.30.
If a privatization at arm's length and for FMV extinguishes the subsidy benefit, it follows logically that the benefit from a non-recurring financial contribution continues if a portion of the company was not sold at arm's length and for FMV. This had been precisely the approach USDOC took in the Usinor case. The Usinor privatization had been divided into four separate and distinct parts, each targeting a different group of investors and resulting in a different price. One of those groups was comprised of company employees. As discussed further below, USDOC found that the relationship between the company and its employees was not at arm's length. Moreover, USDOC found that the privatization processes were different for each of the four groups and that the 5.16 per cent of Usinor's shares sold to its employees were sold for less than FMV. Thus, USDOC determined that the 5.16 per cent of the allocated pre-privatization benefit was not extinguished. The remaining 94.84 per cent of the allocated benefit was found to be extinguished by the sale of the remaining portion of the company in arm's-length transactions for FMV.
4.31.
The analytical approach described above was entirely consistent with the Panel's finding that pre-privatization subsidies were extinguished where the privatized producer, an amalgam of the company and its new owners, receives "nothing for free". To the extent that a portion of the company was sold for less than FMV, the privatized producer received something "for free". USDOC's analysis of the Usinor privatization established that the privatized Usinor did, in fact, receive something "for free" to the extent that some of the purchasers – Usinor employees – did not pay FMV for the so-called "preferential" shares. It was therefore entirely reasonable for USDOC to determine that a corresponding portion of the pre-privatization benefit continues.
4.32.
The European Communities had failed to provide any basis for its claim that USDOC was obligated to use an "average price" analysis. The European Communities' argument, in fact, ignores the fundamental importance of determining that the privatization was at arm's length and for FMV. There was simply no basis to conclude that USDOC had been required to disregard, for purposes of the privatization analysis, the non-arm's-length/non-FMV sale of shares to the employees of Usinor. As demonstrated below, USDOC properly considered those transactions and its conclusions were supported by the evidence.

(b) USDOC properly found that the sale of shares to Usinor's employees was not at arm's length

4.33.
The evidence on record amply demonstrates that sales to Usinor's employees were not at arm's length. To the contrary, the sales were openly acknowledged to be preferential. Moreover, the evidence that the employees did not pay FMV for the portion of the company they acquired was overwhelming. The European Communities' arguments to the contrary were premised on assumptions about what constitutes an arm's-length sale that have no basis in the SCM Agreement, or in the recommendations and rulings of the DSB. The SCM Agreement does not refer to "arm's length", and neither the Panel nor the Appellate Body provided any elaboration on this term. The "arm's-length" definition55 applied by USDOC is, however, entirely consistent with the ordinary understanding of that term.
4.34.
Consistent with this definition, to determine whether sales of Usinor's shares were at arm's-length, USDOC first considered the existence of relationships that would indicate the sales were not at arm's length. USDOC found such a relationship with respect to Usinor employees. USDOC's approach recognizes thatthe respective interests of employers and employees may not be distinguishable and that members of the "corporate family" often treat each other more favourably than they do others outside the family – a point borne out by the admittedly preferential nature of the employees stock purchases in this case. USDOC then proceeded to examine the terms of the transaction between Usinor and its employees to determine whether those terms would have existed had the transaction been between unrelated parties. As USDOC explained these purchasers [employees] had two options: (1) they could purchase shares at the French public offering price of FF 86 per share, or (2) they could pay a discounted price of FF 68.80, with an extended payment period, if they agreed to hold the shares for two years. Additionally, they were eligible to receive bonus shares if they held the shares for specified periods. Thus, the employee offering was clearly distinguishable from the public offerings and was openly characterized as "preferential" in Usinor's Prospectus.
4.35.
Because there had been a preferential option available to employees that had not been available to unrelated parties, USDOC concluded that the terms of the transaction were different as a result of the relatedness of the parties and, hence, that the transactions were not at arm's length. The European Communities maintained that USDOC's arm's-length analysis had been in error because USDOC confused the company (Usinor) and its owner (the GOF), finding only that Usinor's employees were related to Usinor, not to the GOF. The facts, however, indicate that the GOF offered preferential share prices to the employees of a company that it entirely owned, demonstrating that it did not deal with those employees at arm's length for purposes of the privatization. Moreover, the DSB rationale for finding that a benefit may be extinguished by an arm's length, FMV privatization hinges on the assumption that there was little distinction between the company and its new owners. Ironically, the European Communities now appears to argue that, for purposes of determining whether a privatization transaction is at arm's length, there is a significant distinction between a company and its owners.

(c) USDOC properly found that the sale of shares to Usinor's employees was not for FMV

4.36.
USDOC's finding that the sale of shares to Usinor's employees had not been for FMV was reasonable and overwhelmingly supported by the evidence on the record. Once again, the SCM Agreement does not contain this term. In its determination, USDOC considered what the "full amount that the company or its assets (including the value of any subsidy benefits) were actually worth under the prevailing market conditions was paid, and paid through monetary or equivalent compensation". One means of performing that analysis might be to find market benchmarks in the form of actual sales of comparable companies. In those circumstances, USDOC found none and based its examination of FMV on various aspects of the transaction process. Specifically, USDOC examined the "process factors" set forth in the Modification Notice:objective analysis, artificial barriers to entry, acceptance of the highest bid, and committed investment. USDOC's approach to the FMV analysis is supported by Article 14 of the SCM Agreement, which defines the benefit from government equity infusions in terms of whether the government action is consistent with "the usual investment practice (including for the provision of risk capital) of private investors in the territory of that Member [the Member that is bestowing the benefit]".
4.37.
Because the Usinor privatization was not accomplished through a bidding process, USDOC did not consider whether the GOF accepted the highest bid. Rather, USDOC considered the sales process and whether the GOF received a price that maximized its return. Further, regarding the committed investment factor, i.e. whether buyers of Usinor's shares were required to undertake certain post-sale investments, or observe certain post-sale restrictions, USDOC found that there was no evidence to indicate that any of the committed investments distorted the amount that the buyers were willing to pay.
4.38.
USDOC found that the GOF had conducted an objective analysis that it used in structuring the Usinor privatization. That analysis indicated a minimum value of FF 15,750 billion for Usinor. Minimum value and FMV are, however, distinctly different concepts. FMV is based on the results of a market process, not merely on a pre-sale appraisal. USDOC therefore never considered the minimum value to be the FMV of the company. Regarding artificial barriers to entry for the employee sales, USDOC found that the employee shares could be purchased only by current and past employees of Usinor. Hence, by the very terms defining the employee pool, numerous potential purchasers were excluded from purchasing these shares. Unlike the terms defining the employee pool, the international offering and the French offering did not involve meaningful limitations on the competition for shares.
4.39.
As to purchase price, USDOC's general approach was to look to see whether the GOF charged a market-clearing price for its shares of Usinor. A market-clearing price is one that equates the supply of shares to the demand for shares. If the GOF set the offering price for Usinor's shares too low, the offering would have been oversubscribed and many people seeking shares would have been unable to obtain them. Conversely, if the GOF set the offering price too high, the offering would have been undersubscribed. Given the oversubscription to the French offering at the FF 86 price, the fact that shares were moved from the international offering to the French offering, and the number of shares sold at each of the two prices, USDOC found that the market-clearing price for Usinor's shares was between FF 86 and 89. Usinor's employees, on the other hand, had the option of purchasing shares for FF 68 per share – well below the market-clearing price. Based on these uncontested facts, USDOC concluded that the sale of shares to Usinor employees had not been for FMV, both because of the limitations in the purchaser pool and the failure of Usinor to set the share price for employees at anything close to a market-clearing level.
4.40.
The European Communities maintained that USDOC's FMV analysis had been in error because "[t]he discount offered to employees and former employees of Usinor reflects the risk assumed by the buyers for the resale restrictions on the shares purchased". Resale restrictions per se provide no explanation for the substantial discount afforded Usinor's employees. To the contrary, resale restrictions in the Usinor privatization were not limited to company employees. Purchasers in the French offering as well as the stable shareholders were subject to similar restrictions. Nevertheless, the share prices for the latter two groups were well above the preferential price for the employees. Moreover, if the European Communities' argument were correct, one would expect the European Communities to be able to provide evidence that employees were equally (or virtually equally) likely to participate in the French public offering as they were to participate in the employee offering. The European Communities had offered no such evidence. Nor had the European Communities offered any demonstration of how the discount offered to employees was supposed to reflect the risk assumed by the buyers for the resale restrictions. In sum, USDOC reasonably concluded that the sale of shares to Usinor's employees had not been at arm's length or for FMV. These conclusions should therefore be affirmed by the Panel.

4. In the revised UK and Spain sunset reviews, USDOC assumed that all allocable pre-privatization subsidies had been extinguished by the privatizations in question and thereby implemented the DSB recommendations

4.41.
In the revised sunset review determinations in the UK andSpanish cases, USDOC assumed for purposes of its analysis that the pertinent privatizations had extinguished all allocable pre-privatization subsidies. USDOC nevertheless concluded that, in both of those cases, there remained an affirmative likelihood of continuation or recurrence of a countervailable subsidy. Those determinations were based on evidence wholly unrelated to pre-privatization subsidies. For example, in the UK case, one company (Glynwed) was never owned by the UK government. Glynwed therefore continued to benefit from all subsidies that had been bestowed upon it. In the Spanish case, there were also recurring (nonallocable) subsidies to the privatized company (Aceralia) that continued after privatization. Despite the fact that USDOC did not rely on pre-privatization subsidies to these companies in determining that they would continue to benefit from subsidies if the countervailing duty orders were revoked, the European Communities challenged the revised determinations on the grounds that USDOC failed to examine whether the UK and Spanish privatizations were at arm's length and for FMV. The European Communities' argument elevated form over substance. USDOC assumed that the privatizations in the UK and Spain were at arm's-length and for FMV. Because USDOC did not rely on pre-privatization subsidies, its revised determination simply cannot be inconsistent with the DSB recommendations.

C. SECOND WRITTEN SUBMISSION OF THE EUROPEAN COMMUNITIES

1. The Panel's terms of reference for this Article 21.5 proceeding

4.42.
The United States asserted that the European Communities had improperly advanced claims regarding both injury and subsidization, because these claims concern part of the United States' original sunset determinations in 2000, and that, having failed to raise those issues before the original panel and Appellate Body, the European Communities was now foreclosed from doing so in this Article 21.5 proceeding.56For the following reasons, the United States' assertion was incorrect:

(a) The Section 129 determinations concerning the UK and Spain were new measures based on new reasoning

4.43.
The Section 129 determinations at issue in this proceeding were measures taken by the United States to comply with the recommendations and rulings of the DSB. The United States admitted this in its First written submission: The United States attempted to dissociate certain aspects of these determinations – the failure to examine the evidence in respect of the UK andSpanish cases and the failure to re-examine injury – from the "measures taken to comply". The legal claim that the European Communities made in the original WTO proceedings was that the United States, in maintaining countervailing duties without properly analysing the likelihood of continuation or recurrence of subsidization, violated, inter alia, Article 21.3 of the SCM Agreement. This was confirmed by the Panel who found that the measures challenged in the original proceeding were based on insufficient determinations by the USDOC. The United States could continue to maintain the challenged measures only if it were to find alternative, sufficient grounds on which they could be based. Therefore, the USDOC responded by issuing what the United States itself acknowledged were "new determinations"57, based on a revised analysis of the UK and Spanish programmes, comprising "measures taken to comply with the recommendations and rulings" of the DSB.
4.44.
The United States asserted that an aspect of a measure which was before the original panel and Appellate Body and which had not changed was not properly before an Article 21.5 panel, relying for its support on the Appellate Body's decision in EC – Bed Linen.58 However, in EC – Bed Linen, India raised in the Article 21.5 proceeding the same claim as it had in the original proceedings regarding a component of the implementing measure that remained unchanged from, and was part of, the original measure, and which had been "not found to be inconsistent with WTO obligations".59 Whereas, in the present case, the European Communities had raised new claims concerning components of the new measures which were not part of the original measure. Indeed, the situation before the present Panel was more akin to the situation which faced the 21.5 panel in Canada – Aircraft (Article 21.5 – Brazil). Contrary to the assertions in the US First written submission, the Section 129 determinations as to the UK and Spain were "new" sunset review determinations based on different reasoning. In both of the original sunset review determinations, the USDOC concluded that there was a likelihood of continuation or recurrence because none of the producers/exporters concerned participated in the review. Because of their non-cooperation, the USDOC made the adverse assumption that there was a likelihood of continuation or recurrence of subsidization, without examining any of the facts put on the record by the Governments of Spain and the United Kingdom and the European Commission respectively, and without revisiting its analysis of any of the programmes at issue,60 rather relying on findings it had made in the original investigations of the existence of certain subsidy programmes.61 Thus, the reasoning of the Section 129 determinations, based on the existence of certain subsidies purporting to justify a finding of likelihood of continuation of recurrence, was different from the USDOC's reasoning in the original sunset reviews, where the finding of likelihood of continuation or recurrence was based on assumptions drawn from the respondents' failure to cooperate. Consequently, the Section 129 determinations were new measures based on new reasoning.

(b) The factual findings purporting to justify the reasoning in theUK andSpanish cases are clearly different

4.45.
Even assuming, arguendo, that the two forms of reasoning (i.e., the likelihood of continuation or recurrence arising from the respondents' non-cooperation, and that arising from the alleged existence of certain subsidy programmes) could be considered unchanged, the underlying bases -- that is, the factual findings purporting to justify the reasoning -- were clearly different in the two cases. In the original sunset review determinations in both the UK and theSpanish cases, although the determination of likelihood was made because of the non-cooperation of the producers/exporters, the information the USDOC reported to the United States International Trade Commission regarding the "nature of the subsidies" allegedly provided, and which were thus examined for a causal link to injury, concerned a different company or a different set of programmes from those relied upon by the USDOC in the Section 129 determinations. In its Section 129 determination in the UK case, for example, the USDOC focused exclusively on programmes that benefited Glynwed, because it assumed that all countervailable benefits that had previously been conferred upon BS plc had been extinguished through the company's privatization.62In contrast, in the sunset review determination considered in the original WTO proceedings, the USDOC identified exclusively programmes that benefited BS plc.63 There was no reference in the USDOC's analysis of the nature of the subsidies to Glynwed at all.64 Thus, assuming, arguendo, that the original sunset determination was not based on an assumption triggered by the producer/exporters' non-cooperation, but on a finding purporting to be positive evidence, the exclusive basis upon which the USDOC reached its finding in the Section 129 determination was distinct from that upon which it reached its conclusion in the sunset review at issue in the original WTO proceedings.
4.46.
Likewise, in the Spanish case, the determination at issue in the original WTO proceedings was based primarily on thirteen subsidy programmes that allegedly conferred non-recurring and recurring benefits.65 In contrast, the determination at issue in this Article 21.5 proceeding was based exclusively on programmes that allegedly provided recurring subsidies, and the USDOC had only specified one such programme.66 Thus, the bases upon which the USDOC focused its determinations regarding the UK andSpanish cases in its original sunset review and its Section 129 determination were distinct and, therefore, the measures at issue were "different".
4.47.
Because the measures at issue in this Article 21.5 proceeding were "new" and "different" from those considered in the original WTO proceedings and because they were taken to comply with the recommendations and rulings of the DSB, the function and scope of Article 21.5 proceedings discussed by the Appellate Body in Canada – Aircraft (Article 21.5 – Brazil) were applicable to this dispute. Accordingly, the new claims presented by the European Communities concerning the "new" and "different" measures at issue in this Article 21.5 proceeding were properly before this Panel.
4.48.
Likewise, the European Communities' injury argument was properly before this Panel. In the Section 129 determinations regarding the UK and Spain, the USDOC assumed for the first time that the privatizations of BS plc and Aceralia were at arm's length and for FMV. This had the effect of significantly lowering the level of the countervailable subsidies that were found by the USDOC to be likely to continue or recur in the event that the countervailing duty orders were revoked. In the UK case, the USDOC measured the countervailable subsidies provided to Glynwed at 0.73 per cent.67 As to Spain, the USDOC had previously measured the allegedly recurring subsidies provided to Aceralia at 1.73 per cent.68 These figures were far lower than the 12.00 per cent for "all others"69 and 36.86 per cent70 figures, respectively, which the USDOC had issued in the original sunset determinations. However, because such changes to the subsidy rates resulted from the USDOC's revised subsidization findings, the European Communities logically could not have argued at an earlier stage of WTO proceedings that the USDOC's redetermination of the subsidy rates required a reconsideration of the injury determination. Had the USDOC properly analysed the UK and Spanish programmes at issue in its Section 129 determinations, the USDOC would have found that no subsidies would be likely to continue or recur and that, therefore, the continuation of countervailing duties was not warranted. In an analogous situation, the Appellate Body in EC – Bed Linen (Article 21.5 – India) determined that it was appropriate for the complaining party to raise the issue of injury in an Article 21.5 proceeding where the investigating authority revised its findings regarding the amount of imports found to be dumped. Likewise here, the reductions in subsidy rates which the USDOC found in its revised Section 129 determinations would likely have an impact on the United States' finding of injury. Therefore, the United States was required to reconsider its injury determination.

(c) The claims raised by the European Communities concerning the UK and Spain are within the scope of its panel request

4.49.
The European Communities noted that in its request for the establishment of a panel in this Article 21.5 proceeding, it had challenged the propriety of the United States' determinations in the UK and Spanish cases that there was a "likelihood of continuation or recurrence of subsidization and injury". That is, that the United States had failed to meet its obligations, inter alia, under Article 21.3 of the SCM Agreement. This claim was based on a number of arguments, including that the USDOC had failed to examine the nature and also the "impact [of the privatizations] on the continuation of the alleged subsidization".71 The European Communities challengedthe USDOC's finding of a likelihood of continuation or recurrence in light of the USDOC's assumptions in its Section 129 determinations that the privatizations of BS plc and Aceralia were at arm's length and for FMV. Thus, the United States' contention that the European Communities' claims in this proceeding regarding the UK and Spain were "not within the scope of the European Communities' Article 21.5 panel request"72 was incorrect.

2. Certain corrosion-resistant carbon steel flat products from France

(a) The privatization of Usinor was at arm's length and for FMV

4.50.
The United States has asserted that the USDOC's revised determination in the French case was consistent with the DSB recommendations and rulings and was supported by the evidence before the USDOC.73 For the following reasons, the United States' assertions are incorrect.

(i) The SCM Agreement requires that a privatization be measured as to the company as a whole

4.51.
Before determining that pre-privatization, non-recurring subsidies are likely to continue, an investigating authority in a sunset review would be obliged under the SCM Agreement to examine the conditions of the privatization to determine whether the privatized producer continued to receive a benefit from prior subsidization to the state-owned producer. As the Appellate Body stated in this case:

"the obligation under Article 21.2 of the SCM Agreement... requires an investigating authority in an administrative review, upon receiving information of a privatization resulting in a change in ownership, to determine whether a "benefit" continues to exist. In our view, the SCM Agreement, by virtue of Articles 10, 19.4, and 21.1, also imposes an obligation to conduct such a determination on an investigating authority conducting a sunset review. As we observed earlier, the interplay of GATT Article VI:3 and Articles 10, 19.4 and 21.1 of the SCM Agreement prescribes an obligation applicable to original investigations as well as to reviews covered under Article 21 of the SCM Agreement to limit countervailing duties to the amount and duration of the subsidy found to exist by the investigating authority."74

4.52.
A privatization that occurs at arm's length and for FMV creates the presumption that any "benefit" from prior subsidization to the state-owned producer was extinguished and, therefore, that continuation of the countervailing duties was not warranted. As the Appellate Body explained:

"The effect of such a privatization [at arm's length and for FMV] is to shift to the investigating authority the burden of identifying evidence which establishes that the benefit from the previous financial contribution does indeed continue beyond privatization. In the absence of such proof, the fact of the arm's length, fair market value privatization is sufficient to compel a conclusion that the "benefit" no longer exists for the privatized firm, and, therefore, that countervailing duties should not be levied."75

4.53.
Although this Panel and the Appellate Body did not specify a methodology for determining under what circumstances a privatization has occurred at arm's length and for FMV, these factors were nevertheless discussed in the context of the privatization of the firm as a whole. The Panel and Appellate Body did not discuss whether each component of a privatization was at arm's length and for FMV, contrary to the USDOC's incorrect method in its Section 129 determination and contrary to the United States' suggestion in its First written submission.76
4.54.
For example, the Panel stated:

"[w]e believe that privatization calls for a (re)determination of the existence of a benefit to the privatized producer, and that fair market value payment by the privatized producer (and its owners) extinguishes the benefit resulting from the prior financial contribution (subsidization) bestowed upon the state-owned producer, because no advantage or benefit accrued to that privatized producer over and above what market conditions dictate pursuant to Article 14 of the SCM Agreement."77

The Panel also stated that:

"if upon privatization, fair market value is paid for all productive assets, goodwill, etc. employed by the state-owned producer, the Panel fails to see how the subsidies bestowed to the state-owned producer could subsequently be considered to still confer a "benefit" on the privatized producer (in the sense of the company together with its owners) who has paid fair market value for all the shares and assets, reflecting, we must assume, the value of past subsidization."78

4.55.
Likewise, quoting from its Report in US – Lead and Bismuth II, the Appellate Body stated that "the Panel made factual findings that UES and BS plc/BSES paid FMV for the productive assets, goodwill, etc. they acquired from BSC and subsequently used in the production of leaded bars imported into the United States in 1994, 1995 and 1996".79
4.56.
Moreover, considering whether the "privatized producer (and its owners)"80 as a whole paid FMV, rather than whether each component of the sale of the company was at FMV, was the only correct interpretation of the obligations established by the SCM Agreement. The contrary interpretation would allow countervailing duties to continue to be applied notwithstanding the fact that the value received by the producer as a whole was greater than or equal to the FMV of the company. The USDOC, however, reached this conclusion without taking into account the fact that a significant proportion of the purchasers of Usinor's shares (i.e., the stable shareholders) actually paid prices that were greater than the prices identified by the USDOC as market-clearing prices, and that most purchasers paid prices that were within the market-clearing range. Had the USDOC evaluated the privatization as a whole by, for example, evaluating the average price paid per share overall, the unavoidable conclusion would have been that the privatization of Usinor was for FMV.

(ii) The USDOC previously found and accepted before the Appellate Body that the privatization of Usinor was at arm's length and for FMV

4.57.
The finding by the USDOC in the French case that the privatization of Usinor was not at arm's length and for FMV was particularly troubling in light of the fact that the USDOC in two other cases reached precisely the opposite conclusion. As stated by the Appellate Body, "[t]he USDOC analyzed the sales conditions of the privatization of Usinor, in the remand determinations in Allegheny Ludlum (Case No. 1) and GTS (Case No. 2), concluding that those five privatizations took place at arm's length and for FMV."81 Further, the United States did not contest before the Panel and the Appellate Body that "the sales [i.e., all the privatizations] were at arm's length and for fair market value".82 The United States has, in this proceeding, failed to offer any explanation whatsoever for these contradictory outcomes.

Arm's length

4.58.
In considering whether a privatization has occurred at arm's length, an investigating authority must focus on the degree of post-privatization independence from the government of the purchasers of the company's shares. In the current case, it was clear that after the privatization of Usinor, there was no relationship whatsoever between the seller, the GOF, and the buyers, the employees/retirees of Usinor. After the privatization, the GOF held only an insignificant interest in Usinor, and, therefore, the relationship between GOF and the employees/retirees of Usinor was unquestionably independent (the European Communities maintains that the relationship was, in any event, independent, even before the privatization).
4.59.
Even assuming arguendo that the concept of "arm's length" refers to the non-existence of some type of special relationship between the government and the purchasers at the time of the sale, it was clear that the USDOC's analysis was inadequate. The USDOC appears to have merely assumed that the employees/retirees of Usinor and the seller of Usinor's shares (the GOF) were related because "the employee offering was clearly distinguishable from the public offerings and was openly characterized as 'preferential'."83 However, the mere fact that Usinor shares were offered at allegedly preferential rates to employees/retirees of Usinor per se does not mean that the seller and the buyers of the shares were related. Purchasers in the French public offering, for example, were also offered preferential treatment (i.e. a certain number of shares for free).84 The USDOC found that the purchasers of the French public offering "were not related to the seller" and that, therefore, those sales "constituted arm's-length transactions".85
4.60.
The United States, in its first written submission, alleged that because the GOF offered preferential share prices to employees of a company that it owned, it did not deal with those employees at arm's length.86 However, just because Usinor was owned by the GOF at the time of the sale of the shares does not lead to the conclusion that Usinor's employees were employees of the Government. In fact, according to the International Offering Prospectus, despite its ownership interest, the GOF was not involved in the day-to-day management of Usinor. Therefore, the employee purchasers were at least two steps removed from having any direct relationship with the GOF, the seller of Usinor shares.87 Further, the employee offering included former, as well as current, employees of Usinor. There was no evidence whatsoever that any relationship existed between the former employees of Usinor and the GOF that could support a conclusion that the sales to this category of purchasers were not at "arm's length".
4.61.
In conducting its privatization analysis in its Section 129 determination, the USDOC failed to evaluate how or why the employees of Usinor were allegedly related to the seller. It just assumed that they were. Additionally, the USDOC did not even consider the former employees. In light of the above, it follows that there was no basis for the USDOC's conclusion that the employee/retiree offering was not at arm's length, and, in fact, evidence on the record suggests otherwise.

Fair market value (FMV)

4.62.
Further, the sale of Usinor shares in the employee/retiree offering was at FMV. That this was so was evident from the fact that the discount offered to employees and retirees of Usinor reflects the risk assumed by the buyers for the resale restrictions on the shares purchased. The United States' argument that resale restrictions per se "provide no explanation for the substantial discount afforded Usinor's employees"88 is unpersuasive. The restrictions placed upon the French public offering and the stable shareholder offering, to the extent they were applicable, were not as restrictive as those required of the employees/retirees of Usinor. The resale restrictions imposed on purchasers in the French public offering were applicable only where purchasers acquired free shares and were less severe than for the purchasers in the employee/retiree offering who chose to purchase the shares at a discount rate (i.e., required holding period of 18 months v. 24 months). Similarly, the resale restrictions imposed on the stable shareholders were not as severe as they were in the employee/former employee offering they were only prohibited from selling all of the shares for the first 3 months and were then able to sell a limited number of shares for the next 15 months. Thereafter, they were able to sell their shares to any third party.89 In light of these facts, there was no merit to the United States' argument that resale restrictions did not provide a basis for offering a 20 per cent discount to Usinor's employees/retirees, or that the sales of shares to the employees/retirees were not FMV.

(b) The United States is not in compliance with its obligations under the WTO

4.63.
By failing to revoke the countervailing duties imposed on certain corrosion-resistant carbon steel flat products from France in light of the fact that the privatization (whether considered as a whole or, alternatively, from the perspective of the employee/retiree offering) occurred at arm's length and for FMV, the United States has acted inconsistently with its obligations under GATT Article VI:3 and Articles 10, 14, 19.4, 21.1 and 21.3 of the SCM Agreement and has failed to implement the recommendations and rulings of the DSB.

3. CTL plate from the UK and from Spain

4.64.
The United States, in its first written submission, argued that the European Communities' claims challenging the USDOC's findings that subsidies continued to benefit the producers of the products involved in the sunset reviews in the UK and Spanish cases were not properly before this Panel.90 As discussed above, the United States' assertions were unfounded.91 Moreover, the United States failed to challenge the merits of the European Communities' arguments regarding the USDOC's subsidy findings. Nevertheless, it was clear that the United States breached its WTO obligations by finding a likelihood of continuation or recurrence on the basis of insufficient evidence. In both theUK andSpanish cases, the USDOC failed to consider evidence on the record demonstrating that the programmes found to confer countervailable benefits no longer existed and/or no longer applied to the steel industry.92 No contradictory evidence whatsoever exists on the record; the USDOC simply refused to consider the point. The USDOC also failed to consider evidence on the record that Glynwed, the UK producer to which the privatization issue did not apply, had sold off certain of its assets and no longer produced the product under review.93 Had the USDOC performed its likelihood determinations as required under the SCM Agreement, that is, had it considered all of the evidence on the record, it would have found no likelihood of continuation or recurrence of countervailable subsidies regarding CTL plate from the UK and Spain.
4.65.
In light of the fact that the United States admitted in its first written submission that the USDOC had assumed that the privatizations of BS plc and Aceralia were at arm's length and for FMV,94 the USDOC's sole basis for finding a likelihood of continuation or recurrence of subsidization was its reliance on programmes allegedly found to confer benefits to Glynwed and apparently one programme allegedly found to confer recurring benefits in Spain. However, because the United States had failed to consider evidence on the record demonstrating that these programmes no longer existed or applied to steel producers and that Glynwed no longer produced the product under review, the United States had acted inconsistently with its obligations under Article 21.3 of the SCM Agreement.

4. Injury

4.66.
In its first written submission, the United States failed to dispute on the merits the European Communities' argument that the United States was obliged to examine whether the revocation of the countervailing duty orders in the UK andSpanish cases would be likely to lead to continuation or recurrence of material injury as required by the SCM Agreement. The European Communities, however, maintained that Article 21.3 of the SCM Agreement obliges investigating authorities in sunset reviews to determine that the expiry of the duty would be likely to lead to continuation or recurrence of injury, separately from the determination as to the likelihood of continuation or recurrence of subsidization. Further, Article 21.1 of the SCM Agreement requires that "[a] countervailing duty rate shall remain in force only as long as and to the extent necessary to counteract subsidization which is causing injury".95
4.67.
Given the USDOC's findings in these Section 129 determinations that the applicable subsidy rates would be substantially lower than those in the original sunset reviews,96 it was highly likely that the USDOC's revised determinations would affect the causal link between those subsidies and a finding of continuation or recurrence of injury to the domestic industry.97 Accordingly, by failing to reconsider its injury determination in the UK andSpanish cases, the United States violated its obligations under Articles 21.1 and 21.3 of the SCM Agreement. Alternatively, in light of the possibility of a finding of zero subsidies (i.e., if, as discussed above, the USDOC had properly considered all of the evidence on the record),98 the need for an inquiry into the likelihood of continuation or recurrence of injury would have been obviated.

D. SECOND WRITTEN SUBMISSION OF THE UNITED STATES

1. Introduction

4.68.
Since, according to the schedule established by the Panel, the rebuttal submission of the European Communities had been due to be submitted simultaneously with this submission, and the third-party submissions were not due to be submitted until after this submission, the United States effectively had nothing to rebut. Therefore, this submission would be limited to a few additional comments regarding the EC First written submission.
4.69.
At the outset, the United States wished to emphasize the Appellate Body's finding that, where the authorities examine "a benefit resulting from a prior non-recurring financial contribution bestowed on a state-owned enterprise,"99 they must take into account a full privatization at arm's length and for FMV without regard to the ordinary legal distinction between a company and its owners.100 The USDOC's revised privatization methodology was fully consistent with the Appellate Body's finding. In determining whether a full privatization has extinguished the benefit from prior, non-recurring financial contributions, USDOC no longer makes a distinction between a company and its owners. Thus, the focus of USDOC's new methodology was on whether the privatization transaction had extinguished the benefit from allocable, pre-privatization subsidies such that the "privatized producer" – meaning the company and its new owners101 – had received no benefit. Consistent with the findings of the Appellate Body, USDOC does not perform a new benefit calculation.102 Rather, if the entire privatization is at arm's length and for FMV, USDOC presumes that the benefit from allocable, pre-privatization subsidies has been extinguished.103 If the entire privatization is not arm's-length/FMV, then all of the benefit from allocable, pre-privatization subsidies remains countervailable (except to the extent that it had already been countervailed).104 Where only a portion of the company is not sold in an arm's-length/FMV transaction, it follows that only a corresponding portion of the benefit remains countervailable. The privatized producer in the latter two situations has received something "for free" in that the government did not require FMV for all, or part, of the company.105

2. In the revised French sunset review, USDOC properly found that a benefit continued to be countervailable

4.70.
In implementing the recommendations and rulings of the DSB in the revised French sunset review – which involved the Usinor privatization – USDOC found that the privatized Usinor received something "for free" in that the so-called "preferential" shares offered to Usinor's employees were sold at less than FMV in a non-arm's-length transaction. Therefore, USDOC reasonably determined that a corresponding portion of the allocated benefit continued to be countervailable after privatization. As explained in the U.S.' first written submission, the European Communities had failed to demonstrate that USDOC's finding of a non-arm's-length transaction at less than FMV had been in error. Moreover, there is nothing in the text of the SCM Agreement or the DSB recommendations and rulings that would require USDOC to follow the European Communities' suggested method for considering FMV; i.e., averaging out the prices paid for all shares and, then, comparing the average price with FMV to make a singular judgment about the privatization.106
4.71.
The European Communities attempted to buttress its unwarranted claim that all of the shares in Usinor were sold for FMV by asserting that: "Such a finding is consistent with the USDOC's own findings in two parallel proceedings concerning the privatization of Usinor in which the USDOC found that the privatization of Usinor took place at arm's length and for fair market value".107 The findings to which the European Communities referred were two remand determinations made by USDOC in domestic litigation concerning countervailing duty orders other than the order on corrosion-resistant carbon steel flat products from France.108 The European Communities also asserted that it was a "common" practice for privatizing governments to set aside shares for company employees.109 Both European Communities assertions were misplaced. The methodology applied by USDOC in the two prior remand determinations was not the methodology described in the Modification Notice and applied in the revised French sunset review. Hence, the results of those determinations have no relevance to the present proceeding. In particular, in the remand determinations cited by the European Communities, USDOC treated Usinor's employees as "new owners," distinguishable from the company.110 For this reason, USDOC, while finding that the employees paid a subsidized price for their shares, did not take that into account in analyzing whether Usinor had been sold for FMV.111
4.72.
On the other hand, in implementing the recommendations and rulings of the DSB in the revised French sunset review, USDOC properly took into account the fact that shares were offered to Usinor's employees in a non-arm's-length transaction for less than FMV; since, consistent with the findings of the Appellate Body, its new methodology required it to consider the privatized company and its new owners together as the privatized producer.112 Furthermore, it was irrelevant whether it was a common practice when privatizing enterprises for governments to set aside shares for company employees. The only relevant issue was whether the privatized producer, Usinor and its new owners, received something "for free" by not having to pay FMV in an arm's-length transaction. In short, the European Communities had not demonstrated that USDOC erred in finding that the employee sales were not at arm's length and for FMV. Consequently, there was no basis for the European Communities' claim that USDOC erred in finding that the subsidy benefit associated with those sales had not been extinguished by the Usinor privatization, and the Panel should reject that claim.

3. In the revised UK sunset review, USDOC properly found that countervailable subsidies were likely to continue or recur

4.73.
With respect to the sunset review in the UK case, in implementing the recommendations and rulings of the DSB, USDOC assumed that the UK privatization extinguished entirely the allocated subsidy benefit from pre-privatization subsidies. By making that assumption, and by revising the sunset review in accordance with that assumption, USDOC fully complied with the DSB recommendations and rulings. USDOC nevertheless found that countervailable subsidies were likely to continue or recur in the event of revocation, for reasons explained in its first written submission.113 The European Communities cited an earlier remand determination to suggest that USDOC did not perform the proper revised analysis114, but the reliance is misplaced. What USDOC found previously regarding the nature of the privatization of British Steel plc cannot in any way render more complete USDOC's assumption in the revised sunset determination that the privatization of British Steel plc extinguished the benefit from allocable, pre-privatization subsidies. Moreover, USDOC's new privatization methodology was not applied in the remand determination cited by the European Communities.115 For that reason, even if there had been an open issue in this proceeding as to whether the privatization of British Steel plc should have been treated as satisfying the arm's-length/FMV standard – and there was not – the remand determination in question would not be germane.

V. ARGUMENTS OF THE THIRD PARTIES

A. WRITTEN SUBMISSION OF BRAZIL

1. The France Section 129 determination

5.1.
In the French Section 129 sunset review, the USDOC determination that subsidies were likely to continue or recur if the order were revoked, was premised on a finding that little more than 5 per cent of the shares in the privatization at issue in the review were not sold at arm's-length and FMV.116 Brazil found the USDOC Section 129 determination to be inconsistent with the SCM Agreement and the rulings and recommendations of the DSB.

(a) The USDOC determination that the sale of shares to Usinor employees/former employees was not at arm's-length is inconsistent with the SCM Agreement and the Appellate Body's findings and recommendations

5.2.
The USDOC found that because 5.16 per cent of Usinor shares were made available at a discount to the general offering price, the employee/former employee share offering transactions were not at arm's length.117 Brazil took exception to the US argument on a number of grounds. First, USDOC never established that Usinor employees/former employees were related to the seller, the GOF, merely offering a one sentence conclusion that the employees were related to Usinor.118 There may be no definition of "arm's-length" contained in the SCM Agreement, but the USDOC had to explain its conclusions in order to allow this Panel to consider its rationale, consistent with the Panel's standard of review under Article 11 of the DSU. The United States offered only post hoc justification. Brazil further agreed with the reasons offered by the European Communities as to why Usinor employees/former employee were in fact not related to the seller, the GOF.119 The principal United States' argument in response is that drawing distinctions between the company and the owner of the company in this situation is inconsistent with the DSB assumption that there is little distinction between the company and its owners. Brazil found the argument was simply misplaced.120 The Panel and Appellate Body were addressing whether benefits may be bestowed on a company through indirect subsidies to its owners, not whether there is an affiliation between an owner and prospective purchaser. Moreover, the Panel and Appellate Body never found that the company and the owner were one and the same for all purposes and in all instances.121
5.3.
Second, even if affiliation somehow existed between the employees/former employees of Usinor and the GOF, this is not dispositive with respect to the continuation of benefit. The arm's-length test reflects a general understanding that transactions between affiliated parties may not always be on commercial terms. Thus, a closer review of the actual terms is warranted where affiliation exists to determine if the terms reflect usual commercial practice.122 If they do, then the transaction at issue is presumptively consistent with market principles and the terms should not be used to impugn the transaction as not representing FMV. Unfortunately, the United States did not follow through on its own apparent understanding. The United States wants to colour the employee share discount as an anomaly and highly unusual, not reflecting common commercial practice whereas, in reality, such offering discounts are extremely common.

(b) The USDOC determination that the Privatization of Usinor was not for FMV is inconsistent with the SCM Agreement and the Appellate Body's findings and recommendations

5.4.
The facts of the Usinor privatization are that the market value of the firm was established, and a determination made on what was necessary to recoup the value of the company. As it turns out, the total price paid for the company and the average share price met or exceeded the recommendations of the appraisal, as well as the "market clearing price" calculated by the USDOC.123 The USDOC disregarded this fact and looked at just 5 per cent of the total sale to find that non-recurring subsidy benefits were not extinguished in the transaction. Brazil found that the approach taken by the USDOC was inconsistent with the ruling of the Appellate Body in Canada – Aircraft, as reiterated by the Appellate Body in this dispute. The Appellate Body explicitly noted: "once a fair market price is paid... [the] market value is redeemed".124

(c) The USDOC failed to limit countervailing duties to the amount and duration of the subsidy in existence

5.5.
Because the USDOC failed to properly consider both the arm's-length nature of the Usinor privatization and the market value of the transaction, its Section 129 sunset review determination in this case is also inconsistent with GATT Article VI:3 and Articles 10, 19.4 and 21.1 of the SCM Agreement. Specifically, the USDOC did not establish a proper basis for determining the amount and duration of the subsidy at issue in the review and thus it was impossible to act in accordance with these particular provisions, as interpreted by the Panel and Appellate Body, to limit the countervailing duties under the respective orders by the amount and duration of the subsidy it found to exist.

2. The UK and Spain Section 129 determinations

(a) The USDOC assumptions that the UK and Spanish privatization transactions were at arm's-length and for FMV do not constitute determinations

5.6.
In its Section 129 sunset review determinations of the UK and Spanish countervailing duty orders, the USDOC found "even assuming arguendo" that privatization in the two cases did extinguish pre-privatization subsidy benefits, it was not obliged to take any action.125 Brazil recalled that the Appellate Body resolved that a "determination" must be made with respect to the effects of privatization on pre-privatization benefit streams.126 The word "determination" denotes an official ruling; a sense of certainty and finality to a particular issue. This is not what the USDOC offered in its Section 129 determinations in the UK andSpanish cases. An assumption is not sufficient to implement the rulings and recommendations of the DSB in this case. It does not matter whether other types of subsidies or other subsidized producers were involved in the sunset review. The explicit finding of the Appellate Body is that an authority must render a determination on the effect of a privatization. In this regard, the United States had not brought its measures into compliance with its obligations under the SCM Agreement and Article VI:3 of GATT 1994, as elaborated by the Panel and Appellate Body.

(b) The USDOC failed to limit countervailing duties to the amount and duration of the subsidy in existence

5.7.
If the United States chose to stand by the USDOC "assumptions" regarding the effect of the privatizations at issue in the UK and Spanish sunset reviews as something more than mere argument, then the USDOC Section 129 determinations in these cases would be inconsistent with GATT Article VI:3 and Articles 10, 19.4 and 21.1 of the SCM Agreement. Specifically, the USDOC should have, on the basis of its assumptions, limited the countervailing duties applied under the reviewed orders, taking into account the elimination of benefits resulting from the privatizations.

B. WRITTEN SUBMISSION OF CHINA

1. European Communities' allegation that USDOC shall consider the evidence about Glynwed

5.8.
The "measures taken to comply" should be interpreted broadly enough to accommodate the measures that a challenged Member should take to comply with the DSB recommendations and rulings. The USDOC was not obliged to consider the evidence about Glynwed in a Section 129 proceeding. Such a consideration is not a "measure taken to comply". Otherwise, the certainty of an administrative determination would be ruined. A respondent may not be allowed to cure its deficiencies in the original sunset review by having a "second-chance". The European Communities' allegation seemed to suggest that once an aspect of a sunset review was found to be inconsistent with the WTO Agreement, the investigating authority is obliged to render an entirely new review. We see no merits of this argument. We agree with the European Communities that a Member should not implement a DSB recommendation in a "formalistic" manner. However, we believe that the scope of "measures taken to comply" includes both the measures the challenged Member has actually taken and the measures the challenged Member should have taken. Because China believes that investigating authorities should not take evidence presented before them after the closure of record, China did not think the consideration of evidence in relation to Glynwed was a "measure taken to comply". More specifically, China did not think it was a measure the USDOC should have taken. Therefore, China believed that the European Communities' allegation in relation to Glynwed should not be considered by this panel.

2. European Communities' allegation that USDOC shall consider the evidence of abolishment of certain subsidy programs by the Spanish government

5.9.
China may shortcut its analysis by relying on the reasoning of the Appellate Body in EC – Bed Linen (Article 21.5 – India). In that case, India raised the non-attribution issue before the original panel but, as found by the Panel, failed to pursue it. The Appellate Body in that case denied India's claim to include such an issue in a 21.5 proceeding by reasoning that that issue does not "constitute an inseparable element of a measure taken to comply".127 China believed that if an issue had no connection with the DSB recommendations and rulings, it would not be a "measure taken to comply". Again, review of all the evidence in the original sunset review was not a measure that the challenged Member should have taken; and, accordingly, it was not a measure taken to comply. The European Communities' rationale, similarly, seemed to suggest that the investigating authority should render a new sunset review if an aspect of the review is found inconsistent with the WTO Agreement. China similarly could not agree. If the European Communities intended to argue that issue, it may well initiate a new, fresh panel. An Article 21.5 proceeding was not a proper forum.

3. European Communities' allegation that USITC shall consider injury

5.10.
In accordance with United States law, the USDOC should provide the USITC with a net countervailable subsidy which it did in its original sunset reviews for both the UK and the Spanish cases.128 The net countervailable subsidy for the UK case was 0.73 per cent for Glynwed and 12 per cent for all other producers/exporters.129 In the Spanish case, it was 36.86 per cent for all producers/ exporters.130 In the proceedings before the original panel and the Appellate Body, non-recurring allocable subsidies had been found presumably extinguished by an arm's length privatization transaction for FMV.131 In the Section 129 determinations for the UK andSpanish cases, the USDOC assumed that the privatization was at arm's length and for FMV.132 In this situation, it was reasonable to conclude that the USDOC would have had to calculate a new countervailable subsidy because certain programmes had been found to have extinguished. By examination of the Section 129 determinations for the UK and Spanish cases, China found that the USDOC did not do so. The primary reason for the USDOC not to calculate this net countervailable subsidy rate was that the likelihood determination would be made on an order-wide basis.133 However, even if it was assumed that the calculation of a net countervailable subsidy rate had no impact on the USDOC's likelihood determination because it was made on an order-wide basis, the USDOC failed to consider the potential impact of this rate on the USITC's determination.
5.11.
Accepting the rationale of the Appellate Body in the original appeal for this case, China would have deemed the net rate communicated by the USDOC to USITC in the original sunset review inaccurate. In this situation, a portion of the USITC's original likelihood determination with respect to injury could be inferred as inconsistent with the recommendations and rulings of the Panel and the Appellate Body. This portion is where the USITC relied for its determination on the rate provided by the USDOC. USDOC's determination in a Section 129 proceeding not to provide the USITC with a new net countervailable subsidy rate rendered its effort to comply with the DSB recommendations and rulings incomplete. USITC's failure to "correct" the relevant portion of its likelihood determination was also a failure to bring its measures into conformity. The separation of the investigating authorities did not immunise the USITC from taking actions to bring its measures into conformity. Both the USDOC's provision of a new net countervailable subsidy rate to the USITC and the USITC's correction of its original likelihood determination or a portion thereof, were measures taken to comply and, therefore, were within the terms of reference of the compliance panel.

C. WRITTEN SUBMISSION OF KOREA

1. The United States failed to satisfy its WTO obligations by refusing to consider evidence demonstrating the invalidity of its Section 129 likelihood determinations concerning Spain and the UK

5.12.
The United States has argued that it need not consider evidence submitted in the course of the Section 129 proceedings demonstrating that the revised measures generated new inconsistencies with the WTO Agreements. Korea submits that the United States' argument is incorrect, because it is derived from an erroneous interpretation of a Member's obligations under Article 21.5 of the DSU.
5.13.
When undertaking measures intended to comply with the recommendations and rulings of the DSB, a Member may not merely correct the errors identified by the DSB but rely on other grounds to reaffirm its determination. While refusing to consider arguments that have been raised before it, it could focus on other grounds which may create new inconsistencies with the covered Agreements. In its Section 129 determinations on Spain and the UK, the USDOC relied on grounds that had not been the basis of its original sunset determinations in deciding that, despite the privatizations of BS plc and Aceralia, there was a likelihood of continuation or recurrence of subsidization.134 In the Spanish case, the USDOC based its determination on the assertion that, independent of the non-recurring subsidies whose benefits were assumed to be extinguished by the privatization of Aceralia, "there are countervailable recurring, non-allocable subsidies provided pursuant to programs that to the best of our knowledge continue to exist".135 And as to the UK case, the USDOC based its determination on the assertion that in the original investigation, it had found a countervailable subsidy rate for Glynwed Steels Ltd. ("Glynwed").136 In both of the Section 129 determinations, while relying on these alternative bases for its conclusion that there was a likelihood of continuation or recurrence of subsidization, the USDOC refused to consider evidence submitted by the European Communities and the Governments of Spain and the UK demonstrating that the alternative bases no longer existed and, therefore, could not support its determinations. The USDOC did so on the basis of an understanding of its duty to implement that is incorrect. Moreover, the United States failed to satisfy its obligations under Article 21.3 of the SCM Agreement to exercise an active role137 when the USDOC refused to consider the evidence submitted by the European Communities, Spain and the UK that the bases for its Section 129 determinations were flawed.
5.14.
In the current case, Korea was not familiar with the subsidy programs that the USDOC relied on in its Section 129 determinations, and, therefore, took no position as to the accuracy of the arguments that those programs had been abolished or no longer applied to steel production. However, Korea submitted that the United States acted inconsistently with its obligations under Article 21.3 of the SCM Agreement when the USDOC, in the Section 129 determinations, "remain[ed] passive in the face of possible shortcomings in the evidence submitted",138 and disregarded evidence on the record demonstrating that the alternative bases on which the USDOC made its likelihood determinations no longer existed.

2. The USDOC improperly determined that there is a likelihood of continuation or recurrence of subsidization in the France Section 129 determination

5.15.
Korea recognized that, in these proceedings, the European Communities and the United States disputed whether the USDOC was correct in finding that the sales of shares in the employee offering were not at arm's length and for FMV. Even assuming for the purpose of this submission that the USDOC's factual finding was correct, Korea submitted that the USDOC erred in its legal conclusion that the privatization of the company did not extinguish the benefit conferred by the non-recurring pre-privatization subsidies and that, therefore, there was a likelihood of continuation or recurrence of subsidization. One of the flaws in the USDOC's reasoning is the lack of any connection between its factual finding regarding the non-arm's length/FMV sale of a very small proportion of the company's shares and its legal conclusion that the benefit conferred by the pre-privatization subsidies was not extinguished. The France Section 129 determination contains nothing but the bare assertion that such a connection existed which is insufficient to satisfy a Member's obligation "to make a finding on subsidization, i.e., whether or not the subsidy continues to exist",139 as required by the Appellate Body.
5.16.
Furthermore, the lack of an explanation as to a connection between the factual finding and the legal conclusion is particularly troublesome because of the insignificant proportion of the company's shares that were found by the USDOC to be sold at non-arm's length/FMV conditions. Korea did not understand that the Appellate Body intended that the "rebuttable presumption" that a privatization extinguishes the benefits from pre-privatization subsidies would arise only in cases in which 100 per cent of the privatized company's shares are sold at arm's length and FMV. To the contrary, Korea submitted that in a case such as this, in which the overwhelming majority of the shares (almost 95 per cent) were concededly sold at arm's length and for FMV, the burden remained upon the Member to "identify[ ] evidence which establishes that the benefit from the previous financial contribution does indeed continue beyond privatization".140 The USDOC failed to satisfy this burden in its France Section 129 determination.
5.17.
In addition, Korea agreed with the European Communities that a proper evaluation of a privatization must be performed for the company as a whole.141 Small individual elements of a privatization cannot be evaluated in isolation to determine whether the privatization had the effect of extinguishing the benefit conferred by pre-privatization subsidies. The contrary method of analysis adopted by the USDOC in this proceeding would permit an investigating authority to conclude that a privatization failed to extinguish such benefits on the basis of non-arm's length/FMV sales of a de minimis proportion of the company's shares, regardless of the conditions surrounding the sales of the vast majority of the company's shares – a situation closely approached in the current case. Korea agreed with the European Communities that, if the privatization of the company, when viewed as a whole, occurred at FMV, then the United States had failed to satisfy its burden "of identifying evidence which establishes that the benefit from the previous financial contribution [did] indeed continue beyond privatization".142
5.18.
Finally, in a cryptic comment to this Panel, the United States asserted that the USDOC "determined that the [sic] 5.16 per cent of the allocated pre-privatization benefit" – i.e., the same per centage as the proportion of the company's shares found not to be sold at arm's length and for FMV – "was not extinguished. The remaining 94.84 per cent of the allocated benefit was found to be extinguished by the sale of the remaining portion of the company in arm's-length transactions for fair market value".143 Such a determination regarding the proportional extinguishment of the pre-privatization benefit, however, cannot be found in the USDOC's France Section 129 determination. Moreover, if a determination that the vast majority of the pre-privatization benefit had been extinguished were in fact included in the USDOC's Section 129 determination, it becomes even less clear how the USDOC could nonetheless reach the conclusion that there was a likelihood of continuation or recurrence of subsidization sufficient to warrant continuation of the CVD order.

3. The United States erred in failing to reconsider whether there was a likelihood of continuation or recurrence of injury

5.19.
Even assuming, for the sake of argument, that some countervailable subsidies or benefits remained unaffected by the privatizations of Aceralia and BS plc, the effect of the United States' concessions before this Panel is that the remaining countervailable subsidy rates reported to the USITC must be revised significantly. The UK case provided a particularly stark example – with the elimination of the 12.00 per cent rate for "all other" producers, the only remaining subsidy rate was Glynwed's 0.73 per cent – a rate so low that it was considered de minimis for investigations under Article 11.9 of the SCM Agreement.
5.20.
Korea agreed with the European Communities that in failing to reconsider the likelihood of the continuation or recurrence of injury in light of the USDOC's revised subsidy determinations for the UK and Spain, the United States had failed to satisfy its obligations under the SCM Agreement.144 As the Appellate Body stated in the parallel situation involving anti-dumping duties in EC – Bed Linen (Article 21.5 – India): "[t]he amount of dumped imports will, of course, have an impact on the assessment of the effects of the 'dumped imports' for the purposes of determining injury. It is clear, therefore, that the revised findings on dumping and injury could have a bearing on whether a causal link exists between dumping and injury".145 Thus, when a Member's investigating authority revised its conclusions regarding the factual predicates underlying its injury determination, its obligation to ensure that it maintained countervailing duties only to the extent necessary to counteract subsidization that was causing injury required that the injury determination be reconsidered in light of those changed factual conclusions. Only by doing so could the Member ensure that it was properly discharging its obligations under Articles 21.1 and 21.3 of the SCM Agreement.

VI. INTERIM REVIEW

6.1.
On 29 April 2005, the Panel issued its Interim Report to the parties. On 13 May 2005, both the European Communities and the United States submitted written requests for the Panel to review several aspects of the Interim Report. On 20 May 2005, both parties submitted written comments on the other party's request for interim review. Neither party requested an interim review meeting with the Panel.
6.2.
The Panel has carefully reviewed the arguments and proposed amendments to the text of the Interim Report and addresses them below. Where necessary, the Panel has also made certain revisions to the Report.

A. REQUEST OF THE EUROPEAN COMMUNITIES

6.3.
The European Communities has asked the Panel to address three points. First, the European Communities questions the accuracy of the comparison of the restrictions and incentives applicable to the French public offering and the employee share offering. The European Communities then proposes to amend paragraph 7,153 of the Interim Report (paragraph 7,153 of this Report).146 The United States also proposes to amend the paragraph and states that its version follows the Panel's original draft more closely.147
6.4.
The Panel considers that the description of the various restrictions and incentives applicable to both share offerings in paragraph 7,153 of its Interim Report was not inaccurate although brief. The Panel does not object to expanding its description as requested by the parties and has therefore amended paragraph 7,153 of the Interim Report (paragraph 7,153 of this Report) accordingly.
6.5.
Second, as regards paragraphs 7,175 and 7,176 of the Interim Report (paragraphs 7,175 and 7,176 of this Report), the European Communites argues that the Panel's reasoning that the US action is further justified by the fact that it has a retrospective system of duty collection and exporters can request annual or changed circumstances reviews is erroneous and dangerous.148 The European Communities argues that an investigating authority's obligation to conduct periodic reviews is independent of the obligation to properly conduct a sunset review under the SCM Agreement. The European Communities contends that distinguishing between systems of duty collection suggests that an investigating authority in a retrospective system can take certain actions that an investigating authority in a prospective system cannot. In addition, the European Communities asserts that the United States never argued the points at issue in paragraphs 7,175 and 7,176 and thus request that the Panel delete paragraph 7,175 and a portion of paragraph 7,176.149 The United States disagrees and believes the European Communities' proposed deletion is inappropriate. The United States maintains that the Panel is not implying that a retrospective system permits a Member to derogate from its obligations relating to sunset reviews.150. The United States emphasizes that the Panel does not mean that Members with prospective systems incur greater obligations than those with retrospective systems. Instead, the United States understands the Panel to mean that the margin established in the original investigation is not necessarily the rate at which duties are collected because the exporter has the option to seek review to establish that rate.151 In addition, the United States disagrees with the European Communities' statement that paragraph 7,175 of the Interim Report (paragraph 7,175 of this Report) is unnecessary to the Panel's reasoning. The United States insists that it did argue throughout the proceedings that assessment reviews, as opposed to sunset reviews, are the means by which an individual company can obtain a modified assessment rate.152 Finally, the United States disagrees with the European Communities' proposal to delete a phrase in paragraph 7,176 and explains that the reasoning in paragraph 7,176 is related to sunset reviews and does not have to be deleted even upon the deletion of paragraph 7,175, which the United States considers as related to the original investigation.153
6.6.
Contrary to the European Communities' characterization of the Panel's reasoning, the Panel does not consider that the United States' retrospective assessment system somehow further justifies the USDOC's continuation of the measure. In paragraphs 7,175 and 7,176, the Panel is simply addressing the issue of whether the continuation of the measure is inconsistent with Article VI:3 of GATT 1994 and Articles 10, 19.4 and 21.1 of the SCM Agreement as regards the obligation to ensure that the duties collected are limited to the amount and duration of the subsidy found to exist by the investigating authority. We note that the Panel had already concluded in paragraph 7,172 of the Interim Report (paragraph 7,172 of this Report) that the USDOC's consideration of Usinor's privatization when making its likelihood-of-subsidization analysis is not inconsistent with Articles 14 and 21.3 of the SCM Agreement. If the Panel would have found that, on the contrary, the USDOC's consideration of Usinor's privatization was inconsistent with Articles 14 and 21.3 of the SCM Agreement, then the measure at issue would automatically be inconsistent with the provisions requiring the United States to limit the duties collected to the amount of subsidization; at that point, any duty would be unjustified. However, having found that the USDOC's consideration of Usinor's privatization is not inconsistent with Articles 14 and 21.3 of the SCM Agreement, the Panel went on to analyse the European Communities' claim that the United States acted inconsistently with its obligations under Article VI:3 of the GATT 1994 and Articles 10, 19.4 and 21.1 of the SCM Agreement. In doing so, the Panel does not find that a retrospective system somehow renders the USDOC's affirmative likelihood-of-subsidization re-determination not inconsistent with the relevant provisions of the GATT 1994 and the SCM Agreement. Rather, the Panel notes that the United States' retrospective system functions in a way that ensures that excess duties are not collected pursuant to a sunset review. Moreover, the Panel never states that an investigating authority in a retrospective system can take certain actions that an investigating authority in a prospective system cannot. On the contrary, an investigation authority in a prospective system can always adopt a mechanism to ensure that excess duties are not collected pursuant to a sunset review. Given the nature of a prospective system, that mechanism would probably be different from that of the United States, for example, a refund mechanism. The Panel, therefore, has not deleted the contested portions of paragraphs 7,175 and 7,176 as requested by the European Communities. However, to further clarify this Panel's ruling, we have amended the text of the relevant paragraphs accordingly.
6.7.
Finally, the European Communities identifies the erroneous numbering of an exhibit as "EC Exhibit-11" instead of "EC Exhibit-10" in footnotes 375, 377, 378, 379, and 380, and a line of typing error in footnote 390 of the Interim Report (footnotes 422, 424, 425, 426, 427, and 437 of this report).154
6.8.
The Panel has amended the text of the relevant footnotes to correct these typographical errors.

B. REQUEST OF THE UNITED STATES

6.9.
In its request for interim review, the United States comments on several portions of the Interim Report. The United States does not, however, suggest any specific amendments to it.
6.10.
Regarding paragraphs 7.16-7.17 of the Interim Report (paragraphs 7.16-7.17 of this Report), the United States submits that the Interim Report provides no explanation as to how the USDOC had "changed the basis" of its affirmative likelihood-of-subsidization finding in the Spain Section 129 determination.155 Regarding paragraph 7,293 of the Interim Report (paragraph 7,293 of this Report), the United States again submits that paragraph 7,293 and paragraphs 7.16-7.18, which are referenced therein, contain no analysis of how the basis of the affirmative finding changed in the Spain Section 129 determination.156
6.11.
The Panel notes that paragraphs 7,289 and 7,290 of the Interim Report specifically indicate that, in the Spain Section 129 determination, the USDOC based its likelihood-of-subsidization re-determination on the existence of recurring subsidies to Aceralia, whereas in the original sunset determination, the USDOC based its likelihood-of-subsidization determination on the insufficiency of evidence due to lack of cooperation from foreign producers. The Panel however notes the comment of the United States and has amended paragraph 7.17 of this Report to further clarify the changed bases of the affirmative likelihood-of-subsidisation re-determinations as set out in the UK Section 129 determination and the Spain Section 129 determination. Having done this, the Panel does not consider that paragraph 7,293, which cossreferences paragraph 7.17, needs further clarification.
6.12.
Regarding paragraph 7.69 of the Interim Report (paragraph 7.69 of this Report), the United States submits that the European Communities never identified any new evidence placed on the record during the Spain Section 129 proceedings157 and that, regarding the UK Section 129 determination, it is "inaccurate to state that the evidence regarding Glynwed’s cessation of production is 'new evidence, which could not have been presented before.'".158 The United States also argues that the Panel inaccurately characterized the UK Section 129 determination and the Spain Section 129 determination when the Panel stated that the determinations are based on different reasoning than the original sunset reviews. The United States challenges the Panel's statement that the different or relative "weight" given to the evidence in the re-determination regarding Glynwed's subsidisation is an aspect of the measure taken to comply. The United States emphasizes that the Report does not provide any reasoning as to how the Spain Section 129 determination was based on different reasoning.159 The European Communities disagrees and proposes inserting the phrase "or evidence which was not considered in the original sunset reviews was re-submitted" after "was submitted" in the sixth sentence of paragraph 7.69.160
6.13.
The Panel agrees with the United States that the European Communities has not identified any new evidence that was submitted during the Spain Section 129 proceedings. As regards the evidence on Glynwed's sale of its production facilities, the Panel considers it as new evidence since, as submitted by the European Communities161 and not contested by the United States, it was presented for the first time during the Section 129 proceedings. Whether the evidence could have been submitted before has no bearing on whether it is new evidence submitted for the first time during the Section 129 proceedings. The Panel has nevertheless amended paragraph 7.69 to remove its statement that the relevant evidence "which could not have been submitted before". As regards the United States' comments on the Panel's findings on the relevance of the weight given to the evidence during the Section 129 determinations, the Panel considers that it has sufficiently explained its reasoning in the Report.
6.14.
Regarding the analysis of the Panel's mandate in paragraph 7.71 of the Interim Report (paragraph 7.71 of this Report), where the Panel states that the United States introduced the issue of the treatment of evidence by revising the determination and by changing the legal basis of the determination, the United States insists that the USDOC did not change the legal basis of the affirmative determination.162
6.15.
The Panel considers that it has sufficiently addressed this issue in paragraph 6.11 above.
6.16.
Regarding paragraphs 7,214-7.215 of the Interim Report (paragraphs 7,214-7.215 of this Report), the United States argues that Article 21.3 of the SCM Agreement requires a sunset review determination, not a privatization determination, and the Interim Report does not explain how an analysis of privatization is equivalent to an analysis of a sunset review.163
6.17.
The Panel recalls that, as explained in paragraphs 7,199 and 7,209 below, the Appellate Body upheld this Panel's findings in the original proceedings that, before deciding to continue to countervail pre-privatization, non-recurring subsidies, the USDOC is required to examine the conditions of the privatization and to determine whether the privatised producer continues to benefit from pre-privatization, non-recurring subsidies.164 In paragraphs 7,214-7.215 of its Report, the Panel has referred to this finding from the original proceedings. Given the existence of the privatizations concerned, the USDOC is obliged to examine the conditions of these privatizations and to determine whether the privatised producer received any benefit from the financial contributions bestowed on the state-owned producers. In other words, the latter "determination" is one specific determination as required by Article 21.3 in cases of sunset reviews involving privatizations.
6.18.
Regarding paragraph 7,216 of the Interim Report (paragraph 7,216 of this Report), the United States notes that there is no textual analysis to support a statement of the Panel, which the United States interprets as requiring "that a determination must be made in a sunset review for purposes of assisting respondent interested parties in an assessment proceeding".165
6.19.
The Panel disagrees with the United States' interpretation of the Panel's reasoning in paragraph 7,216 of this Report. Requiring the USDOC to examine the conditions of the privatisation and determine whether the benefit continues is not for purposes of assisting respondent interested parties in an assessment proceeding. The reference to the situation where an assessment is requested is used simply to illustrate the legal uncertainty that would arise from an assumption..
6.20.
Regarding paragraphs 7,234-7.238 of the Interim Report (paragraphs 7,234-7.238 of this Report), the United States asserts that the Panel is "imputing the requirements necessary to make a determination of likelihood to a finding regarding the effects of privatization".166 The United States argues that neither Article 11.3 of the Anti-Dumping Agreement nor the Appellate Body addressed "the issue of whether an administering authority may make an assumption where there are sufficient other bases for making the likelihood determination in question".167 The United States insists that the assumption about privatization does not compromise or affect the overall likelihood-of-subsidization determination; therefore, the assumption about privatization did not prevent the USDOC from making its likelihood-of-subsidization determination on the basis of positive evidence. Consequently, the United States maintains "that it is inaccurate to conclude that Commerce erred in assuming, rather than determining, that privatization extinguished the benefit".168
6.21.
The Panel notes that paragraphs 7,234-7.238 simply review a number of Appellate Body and panel reports regarding the legal requirements on the treatment of evidence during sunset reviews conducted under Articles 21.3 of the SCM Agreement and Article 11.3 of the Anti-Dumping Agreement. The issue of why an assumption is not sufficient to meet the requirement of making a determination is a different issue and has been analysed in paragraphs 7,198 to 7,217 below.
6.22.
Regarding paragraph 7,245 of the Interim Report (paragraph 7,245 of this Report), the United States raises two issues. First, the United States characterizes as inaccurate the Panel’s statement that the USDOC based its affirmative likelihood-of-subsidization re-determination in the UK Section 129 determination on a finding from the original investigation that there was a subsidy benefit to Glynwed.169 The United States argues that the findings in the UK Section 129 determination were based on the evidence and findings from the original sunset review proceedings, not the original investigation. In the United States' view, the finding in the original sunset review determination that there was insufficient evidence to conclude that the benefit had been extinguished is the same as the finding in the UK Section 129 determination that a subsidy continued for Glynwed.170 Second, the United States also characterizes as inaccurate the Panel's statement that the insufficiency of information on the termination of subsidy programmes due to the non-cooperation of respondents was the basis of the original sunset review determination and that this basis is different than the basis of the UK Section 129 determination. The United States emphasizes that the affirmative finding in the original sunset review was based on the fact that there was no evidence that the benefit had been extinguished; respondents' non-cooperation is only one of the reasons for the absence of evidence since any other interested party could have put forth evidence.171 The European Communities disagrees and asserts that the USDOC did not assess the evidence provided by the European Commission and the GOUK. The European Communities insists that the only explanation for the USDOC's refusal to asses the information provided by the European Communities and the GOUK is that the relevant exporters/producers had not participated.172
6.23.
As regards the first comment, the Panel notes that the original sunset review determination focused almost exclusively on the benefit to BS plc. The text of the original sunset review determination only mentioned Glynwed in a description of findings from the original countervailing duty order and as part of the description of the domestic interested parties' arguments, never in the USDOCs analysis. While the USDOC also restated the rate for Glynwed at the end of the determination, the USDOC explained that it normally selects the rate from the original investigation.173 Moreoever, the programme descriptions provided by the USDOC only mention the subsidization of BS plc, not Glynwed.174 In addition, the reasoning of the original sunset review determination as set out in "The Department's Position" focused only on the "subsidy programmes" and the insufficiency of evidence demonstrating that some programmes had been fully amortized.175 Therefore, the Panel considers that the "basis" or the "reasoning" of the original sunset determination as set out in "The Department's Position" was not specifically that a benefit continued to exist for Glynwed.176 In the UK Section 129 determination, however, the USDOC focused solely on the benefit to Glynwed.177 The Panel therefore disagrees with the United States' assertions. As regards the Unites States' second comment, the Panel notes that the respondents' non-cooperation is one of the reasons for the insufficiency of evidence in the original sunset review. Since nevertheless the focus of the Panel's statement was the insufficiency of evidence, we agree to delete the reference to non-cooperation in paragraph 7,245 of this Report.178
6.24.
Regarding the United States' comments on paragraphs 7,248179 and 7,281180 of the Interim Report (paragraphs 7,248 and 7,281 of this Report), the Panel has addressed these concerns in paragraph 6.11 above.
6.25.
As regards the United States' comments on paragraph 7,252 of the Interim Report (paragraph 7,252 of this Report)181, the Panel refers to its comments in paragraph 6.13 above.
6.26.
Regarding paragraph 7,291 of the Interim Report (paragraph 7,291 of this Report), the United States characterizes as inaccurate the Panel's statement that "the 1987 Government Delegated Commission on Economic Affairs: Fund For Employment Promotion and Early Retirement was the programme that the USDOC found to be 'recurring' and that the USDOC used as the basis for its affirmative likelihood-of-subsidization determination in the Spain Section 129 determination". Quoting page three of the UK Sunset Review Issues and Decision Memo, the United States insists that the programme is an example of a recurring subsidy programme.182 The European Communities comments that the United States identified no other programme as recurring and therefore proposes amending paragraph 7,291 as follows: "was the only programme that the USDOC found identified to be as 'recurring'".183
6.27.
The Panel notes the United States' clarification that the recurring subsidy programmes collectively form the basis of the Spain Section 129 determination, even though the USDOC has not identified any other recurring programmes in the determination. The Panel has therefore clarified this point in paragraph 7,291of this Report.
6.28.
Regarding paragraph 7,291, footnote 492 of the Interim Report (paragraph 7,291, footnote 539 of this Report), the United States notes that page 17 of the Issues and Decision Memorandum cited does not exist.184 The European Communities states that it does exist.185 The Panel has verified that the page exists and therefore does not need to amend footnote 539 of this Report.
6.29.
Regarding paragraph 7,296 of the Interim Report (paragraph 7,296 of this Report), the United States characterizes as inaccurate the Panel's statement that the USDOC rejected evidence on the termination of subsidy programmes as insufficient due to the non-cooperation of respondents. Citing page thirteen of the Spain Sunset Review Issues and Decision Memo, the United States emphasizes that the determination states that no evidence of termination was provided and therefore the respondents' arguments were unsubstantiated.186 The European Communities disagrees and asserts that the USDOC did not assess the evidence provided by the European Commission and the Government of Spain. The European Communities insists that the non-cooperation of respondents was the only reason that the USDOC provided for not assessing the evidence at issue.187
6.30.
The Panel notes that in the original sunset review determination, the USDOC summarized the European Communities and the GOS' arguments that a number of subsidy schemes no longer existed and that the subsidization of the steel sector was prohibited since 1997 following the adoption of a series of European Commission Decisions. In the section entitled "The Department's Position", the USDOC did mention that "the Department did not receive a response from the foreign producer/exporter and, pursuant to section 351,218(d)(2)(iii) of the Sunset Regulations, this constitutes a waiver of participation".188 Immediately following this statement, the USDOC stated "[b]ecause there have been no administrative reviews of this order and no evidence has been submitted to the Department demonstrating the termination of the countervailable programs, the Department assumes that these programs continue to exist and are utilized."189 The USDOC further based its affirmative determination on the following grounds: "[absence of] evidence that the programs have been terminated, [absence of evidence] that the benefits from programs for which benefits are allocated over time will not continue beyond this sunset review, or [lack of] participation in this review of a foreign producer/exporter...".190 It remains unclear which of those grounds are the ones on which the USDOC relied in this case. In any event, the Panel will delete the reference to non-cooperation in paragraph 7,296 of this Report and will add a footnote referring to those grounds.

VII. FINDINGS

A. GENERAL CONSIDERATIONS FOR THIS DISPUTE

1. The Panel's mandate

(a) The measures taken to comply in these proceedings

(i) Arguments of the parties

(ii) Evaluation by the Panel

The USDOC's treatment of evidence in the likelihood-of-subsidization determination

The likelihood-of-injury determination

Other issues

The failure to revoke the countervailing duty orders as the measure taken to comply

The implementation status of the Section 129 determinations at issue

(b) Scope of the claims in these proceedings

(i) Arguments of the parties

(ii) Evaluation by the Panel

New claim not included in the Panel request

Other disputed new claims

New claim on evidence

New claim on likelihood-of-injury

Conclusion

2. Standard of review

7.78.
The standard of review applicable to SCM Agreement-related disputes is the general "objective assessment" standard of Article 11 of the DSU. In EC – Hormones, the Appellate Body clarified that "the applicable standard [pursuant to Article 11 of the DSU] is neither de novo review as such, nor 'total deference', but rather the 'objective assessment of the facts'".296
7.79.
In US – Cotton Yarn, the Appellate Body indicated that its Reports in Argentina – Footwear (EC), US – Lamb and US – Wheat Gluten, all concerning disputes under the Agreement on Safeguards, "spell out key elements of a panel's standard of review under Article 11 of the DSU in assessing whether the competent authorities complied with their obligations in making their determinations". The Appellate Body stated:

"This standard may be summarized as follows: panels must examine whether the competent authority has evaluated all relevant factors; they must assess whether the competent authority has examined all the pertinent facts and assessed whether an adequate explanation has been provided as to how those facts support the determination; and they must also consider whether the competent authority's explanation addresses fully the nature and complexities of the data and responds to other plausible interpretations of the data. However, panels must not conduct a de novo review of the evidence nor substitute their judgement for that of the competent authority."297

7.80.
Precisely, in the context of the analysis of Usinor's privatization, we note that the United States, with reference to the Appellate Body Report in US – Cotton Yarn, argues that the Panel’s task under Article 11 of the DSU is not to determine whether the privatization was at arm’s-length or for FMV, but rather whether the USDOC properly established the facts and evaluated them in an unbiased and objective way. "Put differently, the Panel’s task is to determine whether a reasonable, unbiased person, looking at the same evidentiary record as [the USDOC], could have – not would have – reached the same conclusions."298
7.81.
The Panel notes that several panels and the Appellate Body have addressed the scope of a panels' task in the specific context of sunset reviews. Most of these disputes concerned the anti-dumping sunset review provision under Article 11.3 of the Anti-Dumping Agreement. We note that the Appellate Body has considered that its interpretation of Article 21.3 of the SCM Agreement applies mutatis mutandis to Article 11.3 of the Anti-Dumping Agreement due to the provisions’ (almost) identical wording.299 We also note that the Declaration on Dispute Settlement Pursuant to the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994 or Part V of the Agreement on Subsidies and Countervailing Measures recognizes "the need for the consistent resolution of disputes from anti-dumping and countervailing duty measures".300 On this basis, the Panel considers that the following description of a panel’s task in considering the investigating authorities' behaviour, even if referring to anti-dumping sunset reviews, should also apply to sunset reviews under the SCM Agreement.
7.82.
In this respect, the Appellate Body in US – Oil Country Tubular Goods Sunset Reviews cited with approval the panel's description of the investigating authority's duties in an anti-dumping sunset review in US – Corrosion-Resistant Steel Sunset Review.301 According to this description, Article 11.3 of the Anti-Dumping Agreement imposes an obligation on the investigating authority to make a determination on the basis of positive evidence and with sufficient factual basis to allow it to draw reasoned and adequate conclusions concerning the likelihood of such continuation or recurrence.302 The Appellate Body concluded that a panel must evaluate the investigating authority's sunset review determination in light of that obligation and asses whether the investigating authorities properly established the facts and evaluated them in an unbiased and objective manner:

"These obligations of investigating authorities inform the task of a panel called upon to evaluate the consistency of an investigating authority's determination with Article 11.3 of the Anti-Dumping Agreement. The task of the panel is to assess whether the investigating authorities properly established the facts and evaluated them in an unbiased and objective manner.303 We agree with the Panel that '[its] task [was] not to perform a de novo review of the information and evidence on the record of the underlying sunset review, nor to substitute [its] judgment for that of the US authorities'.304 If the panel is satisfied that an investigating authority's determination on continuation or recurrence of dumping or injury rests upon a sufficient factual basis to allow it to draw reasoned and adequate conclusions, it should conclude that the determination at issue is not inconsistent with Article 11.3 of the Anti-Dumping Agreement.305"306

7.83.
We shall therefore apply the above standard of review when evaluating the affirmative likelihood-of-subsidization re-determinations at issue in these proceedings.

3. Burden of proof

7.84.
The Appellate Body summarised its prior findings on the burden of proof in US – Wool Shirts and Blouses, indicating that the burden of proof rests upon the party, whether complaining or defending, who asserts the affirmative of a particular claim or defence. If that party adduces evidence sufficient to raise a presumption that what is claimed is true, the burden then shifts to the other party, who will fail unless it adduces sufficient evidence to rebut the presumption.307
7.85.
The Appellate Body also specifically considered the application of the burden of proof in the context of an Article 21.5 proceeding. In Canada – Aircraft (Article 21.5 – Brazil), it ruled that the examination of "measures taken to comply" must be based on the relevant facts proved by the complainant to the Article 21.5 panel during the panel proceedings.308 In its sister case, Brazil – Aircraft (Article 21.5 – Canada), the Appellate Body confirmed that the fact that the measure at issue was 'taken to comply' with the 'recommendations and rulings' of the DSB does not alter the allocation of the burden of proving a defence.309
7.86.
In determinations on subsidization, we recall that in the original proceedings in US – Countervailing Measures on Certain EC Products, the Appellate Body found that, when the investigating authority determines that a privatization has taken place at arm's length and for FMV, the investigating authority has the burden to prove that the benefit still passed through to the privatized producer:

"We understand the Panel to be stating that privatization at arm's length and for fair market value privatization presumptively extinguishes any benefit received from the non-recurring financial contribution bestowed upon a state-owned firm. The effect of such a privatization is to shift to the investigating authority the burden of identifying evidence which establishes that the benefit from the previous financial contribution does indeed continue beyond privatization. In the absence of such proof, the fact of the arm's-length, fair market value privatization is sufficient to compel a conclusion that the 'benefit' no longer exists for the privatized firm, and, therefore, that countervailing duties should not be levied. This is an accurate characterization of a Member's obligations under the SCM Agreement.

... Privatization at arm's length and for fair market value may result in extinguishing the benefit. Indeed, we find that there is a rebuttable presumption that a benefit ceases to exist after such a privatization. Nevertheless, it does not necessarily do so. There is no inflexible rule requiring that investigating authorities, in future cases, automatically determine that a 'benefit' derived from pre-privatization financial contributions expires following privatization at arm's length and for fair market value. It depends on the facts of each case."310

4. Panel's approach

7.87.
Having concluded that the measures taken to comply in these proceedings are the affirmative likelihood-of-subsidization re-determinations set out in the three Section 129 determinations at issue, we will address the claims raised by the European Communities concerning each of these three re-determinations.

B. CERTAIN CORROSION-RESISTANT CARBON STEEL FLAT PRODUCTS FROM FRANCE

1. Background

7.88.
On 8 January 2003, the DSB adopted the original panel and Appellate Body Reports. The Panel found that the sunset review determination on Certain Corrosion-Resistant Carbon Steel Flat Products from France (C‑427‑810) (Case No. 9), which was based on the so-called gamma methodology, was inconsistent with the SCM Agreement, since the USDOC did not examine whether the privatization, that occurred after the original imposition of countervailing duties, was at arm's length and for FMV and had failed to determine whether the privatized producer received any benefit from the financial contributions previously bestowed on the state-owned producer. The Panel considered that by failing to determine the likelihood of continuation or recurrence of a subsidization, prior to its decision to maintain countervailing duties, the United States had violated Articles 14, 19.4, 21.1 and 21.3 of the SCM Agreement, which prohibit a Member, pursuant to a sunset review, from maintaining countervailing duties where there has not been any determination of likelihood of continuation or recurrence of subsidization and thus of a continued need for countervailing duties. It also concluded that, since the United States had maintained countervailing duties that were inconsistent with Articles 14, 19.4, 21.1 and 21.3 of the SCM Agreement, it had also violated Article 10 of the SCM Agreement which requires that countervailing duties be imposed or maintained consistently with the SCM Agreement.311 The Appellate Body upheld these Panel findings.312
7.89.
Following the adoption of the Panel and Appellate Body Reports by the DSB, the USDOC published a new privatization methodology ("Modification Notice"), which is not being challenged by the European Communities in these proceedings.313 The USDOC also reviewed its determination of the likelihood of continuation or recurrence of subsidization contained in the original sunset review, which had been found to be inconsistent with the SCM Agreement and the GATT 1994. Accordingly, the USDOC issued its Issues and Decision Memorandum for the Section 129 Determination: Corrosion-Resistant Carbon Steel Flat Products from France; Final Results of Expedited Sunset Review of Countervailing Duty Order of 23 October 2003 ("France Section 129 determination").314
7.90.
In the France Section 129 determination, the USDOC applied its new privatization methodology to Usinor's privatization. According to the France Section 129 determination, the Government of France ("GOF") incrementally privatized Usinor over a three-year period (July 1995-August 1998). Prior to the privatization, the GOF wholly owned Usinor, holding 80 per cent directly and 20 per cent through the government-controlled Crédit Lyonnais (or later, its division, Clindus).315 In effectuating the privatization, the GOF issued four types of share offerings to four different classes of purchasers: 1) French resident nationals and European Communities or European Economic Area nationals in France ("French public offering"); 2) current and qualifying former employees of Usinor throughout the world ("employee/former employee offering"); 3) stable shareholders comprising various institutional investors, both public and private ("stable shareholder offering"); and 4) the general public in the French and international financial markets ("international offering"). The GOF subjected the four types of share offerings to different restrictions.316
7.91.
According to the France Section 129 determination, by the end of 1995, private shareholders held the majority of shares with over 75 per cent. The GOF and Clindus had reduced their holdings to 9.8 per cent and 2.5 per cent, respectively. Stable shareholders held 15.8 per cent. Of those stable shareholders, three were directly or indirectly government-controlled ("public stable shareholders"). Usinor employees held only 3.55 per cent.317
7.92.
Also according to the France Section 129 determination, the USDOC stated that in January 1997, the GOF delivered approximately 1.2 per cent of total Usinor shares to French individual shareholders and employees who had complied with the share restrictions. In October 1997, the GOF sold another approximately 7.7 per cent of total Usinor shares. By year-end 1997, private shareholders held over 85 per cent of total Usinor shares. The GOF held 0.93 per cent directly. "Clindus and the public stable shareholders remained unchanged at 2.5 per cent and 10.1 per cent, respectively".318 Usinor employees held 5.16 per cent.319 In August 1998, the GOF completed the privatization by delivering the remaining 0.93 per cent to shareholders as free shares. The GOF no longer held any direct ownership in Usinor.320
7.93.
After analysing Usinor's privatization in the France Section 129 determination, the USDOC found that the privatization of Usinor was at arm’s length and for FMV with the exception of the employee/former employee offering, which constituted 5.16 per cent of the sale. The USDOC therefore reaffirmed the affirmative likelihood determinationin its original sunset review, indicating that "[t]he sale of shares to Usinor employees at prices below fair market value did not extinguish certain allocable, nonrecurring, pre-privatization subsidies that continue at an above de minimis rate beyond the original sunset review".321 The USDOC affirmed its original likelihood-of-subsidization determination, which reported the net countervailable subsidy rate likely to prevail upon revocation of the order as an ad valorem country-wide countervailing duty rate of 15.13 per cent.322