UNITED STATES � ANTI-DUMPING MEASURES ON
STAINLESS STEEL PLATE IN COILS AND STAINLESS
STEEL SHEET AND STRIP
FROM KOREA

Report of the Panel

(Continued)

ANNEX 3-1

FIRST SUBMISSION OF THE EUROPEAN COMMUNITIES

(5 June 2000)

CONTENTS

1. INTRODUCTION 
    
2. THE "MULTIPLE AVERAGING" METHODOLOGY

1. ARTICLE 2.4.2 
    
2. ARTICLE 2.4.1 
    
3. ARTICLE 2.4 

3. EXPORTS SALES TO A CUSTOMER WHICH LATER WENT BANKRUPT 
    
4. CLAIMS UNDER ARTICLE X.3(A) OF THE GATT  

I. INTRODUCTION

1. The European Communities (hereafter "the EC") makes this third party submission because of its systemic interest in the correct interpretation of the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994 ("the Anti-dumping Agreement").

2. Many of the issues in dispute involve questions of fact on which the EC is not in a position to comment. Accordingly, this submission will address only a number of issues of legal interpretation which are of particular interest to the EC.

3. Section II discusses some of the claims raised by Korea in connection with the so-called "multiple averaging" methodology applied by the United States in calculating the dumping margins. Section III addresses Korea's claims under Article 2.4 with respect to the export sales made to a customer that later went bankrupt. Finally, Section IV deals with Korea's claims under Article X.3(a) of the GATT.

II. THE "MULTIPLE AVERAGING" METHODOLOGY

A. ARTICLE 2.4.2

4. The US authorities divided the period of investigation into two sub-periods and compared the weighted average normal value for each sub-period to the weighted average export price for the same sub-period. The resulting dumping margins were then combined into an overall dumping margin for the entire investigation period. For that purpose "negative" margins were treated as zero. Korea alleges that the use by the United States of multiple averages instead of a single average for the whole investigation period violates Article 2.4.2.¹

5. Korea's claim rests on a misunderstanding of the scope of the requirements imposed by Article 2.4.2. The first option presented by the first sentence of Article 2.4.2 does not require a comparison of the weighted average normal value with the weighted average of all export sales of the product under investigation. Instead, Article 2.4.2 requires a comparison "with a weighted average of prices of all comparable export transactions".

6. Therefore, when all export transactions are not "comparable", Article 2.4.2 allows the investigating authorities to calculate separate averages for each set of "comparable" transactions.

7. The two textual arguments derived by Korea from the wording of Article 2.4.2² are clearly without merit:

• it is true that Article 2.4.2 refers to "a weighted average". At the same time, however, Article 2.4.2 contemplates that it may be necessary to establish the "existence of margins of dumping" (in plural) in those cases where not all transactions are comparable. Furthermore, it is significant that Article 2.4.2 alludes to "a weighted average", rather than to "the weighted average";

• the argument that Article 2.4.2 requires that all export transactions must be taken into account in the average is fallacious because, as recalled above, Article 2.4.2 refers to "all comparable export transactions". Korea's argument reads the word "comparable" out of Article 2.4.2.

8. It is obvious that the timing of the sales may affect price comparability. Indeed, it is precisely for that reason that Article 2.4 requires that the comparison of normal value with the export price shall be made "in respect of sales at as nearly as possible the same time". Thus, the timing of the sales is one of the factors which, in the EC's view, may warrant the use of multiple averages when comparing the normal value to the export price. (Other factors affecting comparability which may require the use of multiple averages are the existence of different product types or models, or the presence of different levels of trade).

9. The EC's experience is that while, under normal circumstances, all sales made within the period of investigation (normally one year in the EC practice) will be "comparable", there may be special circumstances (e.g. high inflation rates in the exporting country) which render necessary to compare quarterly or even monthly averages.

10. The EC is not a in position to express any views on the factual question of whether, in the case at hand, the US investigative authorities were justified in concluding that the sales made in the first sub-period were not "comparable" to those made in the second sub-period. The EC would note, nevertheless, that Korea does not appear to contend that they were "comparable". Rather, Korea's claim is that Article 2.4.2 does not allow the use multiple averages under any circumstances, a claim which for the reasons explained above is clearly unfounded.

11. Although Korea's submission mentions at several points the fact that the US authorities "zeroed" the negative dumping margins when calculating the overall margin for the entire investigation period³, Korea does not appear to contest that practice, but only the use of multiple averages.⁴

12. In the event that Korea were challenging also the "zeroing" of negative dumping margins, the EC submits that such practice is not addressed by Article 2.4.2. Where the investigative authorities use "multiple averages", it becomes necessary to combine the resulting dumping margins into a single margin for the product under investigation. It is at that point that the issue of zeroing arises. That stage of the calculation process, however, is not subject to Article 2.4.2.

13. The view that the first sentence of Article 2.4.2 prohibits "zeroing" would upset the finely balanced compromise achieved by the negotiators of the Anti-dumping Agreement. As confirmed by the Audio Cassettes report⁵, there is nothing inherently "wrong" or "unfair" about "zeroing". It would be a mistake to assume that simply because Article 2.4.2 now provides for the use of an average-to-average methodology at the first stage of the dumping margin calculation, the same methodology should be extrapolated to the entire process

B. ARTICLE 2.4.1

14. Korea contends that by dividing the investigation period into two sub-periods to take into account the devaluation of the won, the US authorities violated Article 2.4.1.⁶

15. In the EC's view, this claim is unfounded. Article 2.4.1 is concerned exclusively with the selection of exchange rates for converting normal value and export price into the same currency, prior to comparing them. The rules contained in Article 2.4.1 do not preclude the possibility that, after that conversion, the prices of some export transactions may still not be "comparable" within the meaning of Article 2.4.2, due to factors such as the depreciation of the local currency, which would warrant the use of "multiple averages".

C. ARTICLE 2.4

16. Korea alleges that the "multiple averaging" methodology is inconsistent with the fair comparison requirement contained in Article 2.4 because it resulted in a finding of dumping based solely on pre-devaluation sales, while the finding of injury was based primarily on post-devaluation sales.⁷

17. The EC considers that this claim does not concern the comparison between normal value and export price, which is the only issue governed by Article 2.4, but rather the entirely different issue of whether there is a causal link between dumping and injury. Accordingly, the EC is of the view that this claim should have been brought under Article 3.5, and not under Article 2.4.

III. EXPORTS SALES TO A CUSTOMER WHICH LATER WENT BANKRUPT

18. The Korean exporter made some export sales to a US customer that later went bankrupt (the ABC company). The US authorities treated those sales as "bad debt" and deducted their cost by way of adjustment from the export price for the remaining transactions. Korea alleges that in doing so the US authorities violated, inter alia, Article 2.4.⁸ Korea contends that the US authorities should have excluded the sales to ABC from the export price calculation.

19. The EC agrees with the United States that it was not required by Article 2.4 to exclude the sales to ABC from the export price.⁹ The EC also agrees with the United States that the "bad debt" expenses incurred by a related importer may be deducted from the resale price when calculating the constructed export price in accordance with Article 2.3.¹⁰

20. On the other hand, the EC disagrees with the US view that differences in "bad debt" expenses may give raise to an adjustment under Article 2.4.¹¹ In the EC's view, "bad debt" expenses do not "affect price comparability" within the meaning of Article 2.4 because they have no bearing on the customer's decision to make a purchase.

21. Contrary to what is argued by the United States, the risk that a customer may go bankrupt is not a "condition" of sale in the meaning of Article 2.4.¹² The word "conditions", as used in Article 2.4, is largely synonymous with the word "terms". Both words allude to the conditions agreed by the seller and the purchaser, and not to the general conditions prevailing in the market in which the sale takes place. The US extensive interpretation of the term "condition" would require to make an adjustment for virtually any difference in SGA expenses.

22. The analogy drawn by the United States to warranty expenses and credit costs¹³ is inapt. While it is true that the eventual cost of the warranty to the exporter is not known at the time where a sale is made, the provision of the warranty is part of the sales contract (or is required by law) and, therefore, influences the customer's purchasing decision. As regards credit costs, the EC practice is to make an adjustment only where the payment terms are expressly agreed in the sales contract or, at least, in the invoice.

IV. CLAIMS UNDER ARTICLE X.3(A) OF THE GATT

23. The EC considers that the requirements imposed by Article X.3(a), and in particular the requirement of "reasonableness", should be interpreted consistently with the requirements imposed upon the investigative authorities by the Anti-dumping Agreement.

24. The Anti-dumping Agreement contains very detailed substantive and procedural requirements for the imposition of Anti-dumping measures. Those rules must be assumed to reflect the Members' agreement as to what is "reasonable" in that particular field. Article X.3(a) should not be used in order to second-guess the drafters of the Anti-dumping Agreement.

25. Many, if not all the claims submitted by Korea under Article X.3(a) concern matters that are specifically addressed in the Anti-dumping Agreement, as evidenced by the fact that those claims have been submitted together with other claims based on the Anti-dumping Agreement. In addressing those claims, the Panel should consider first the relevant provisions of the Anti-dumping Agreement, in accordance with the Appellate Body's admonition in EC - Bananas that the more specific WTO agreement must be applied first.¹⁴ If the Panel finds that the measures are in conformity with the relevant provisions of the Anti-dumping Agreement, it should find that they are "reasonable" also for the purposes of Article X.3(a).

1 Korea�s submission, paras. 4.45-4.48.

2 Ibid., para. 4.46.

3 Korea�s submission, paras. 3.46 and 4.43.

4 Korea�s panel request does not refer to the issue of zeroing (cf. point (d)of the request).

5 Panel report on EC - Anti-dumping Duties on Audio Tapes in Cassettes originating in Japan, ADP/136, distributed on 28 April 1995, unadopted, at paras. 347-366.

6 Korea�s submission, paras. 4.49-4.53.

7 Ibid., paras. 4.59-4.63.

8 Ibid., para.4.10 et seq.

9 US Submission, paras.65-73.

10 Ibid., paras. 74-80.

11 Ibid., paras. 81-90.

12 Ibid. , para. 83

13 Ibid., paras. 83-86.

14 Report of the Appellate Body in EC - Regime for the Importation, Sale and Distribution of Bananas, WT/DS27/AB/R, at para.204.

To continue with ANNEX 3-2

Return to Table of Contents