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 1-7

RESPONSES OF KOREA TO QUESTIONS POSED BY THE PANEL

SECOND MEETING OF THE PANEL

(28 July 2000)

CONTENTS

1. CURRENCY CONVERSION 
    
2. UNPAID SALES 
    
3. MULTIPLE AVERAGING 

NOTE: In this submission, including the exhibits, Korea has placed information which POSCO has previously designated as business proprietary information in brackets ("{ }"). This information has been omitted and the brackets left in the text."{ }"

I. CURRENCY CONVERSION

1. As an initial matter, Korea notes that the Panel has asked several questions concerning the state of the record regarding the local sales issue. Korea respectfully submits that it is not necessary for the Panel to examine the record in the manner suggested by the Panel's questions. As mentioned at the Second Meeting, "for a panel to review a determination by reference to considerations not actually reflected in a public statement of reasons accompanying such determination would also be inconsistent with the requirements of an orderly and efficient conduct of the dispute settlement process...."¹

2. The DOC's final determinations provided three reasons for the decision to ignore the dollar prices provided by POSCO: (1) POSCO was paid in won; (2) POSCO recorded the local sales in its ledgers in won; and (3) there were differences between the exchange rates used by POSCO and those used by the United States.² No other factors were relied upon in the final determinations. Thus, the question before the Panel is whether a determination based on those factors - and those factors alone - was consistent with the requirements of the Anti-Dumping Agreement.

3. Significantly, the DOC did not rely on any failure by POSCO to provide relevant information in a timely manner. (If the DOC had relied on such a claim, it would have been required by the last sentence of Article 2.4 and the provisions of Article 6.1 to indicate to POSCO what additional information was required.) Moreover, the DOC also did not rely on the fact that the invoices for the local sales showed prices in Korean won. (To the contrary, the final determinations specifically stated that all local-sales invoices showed the prices in U.S. dollars, while only some of the local-sales invoices "also" showed prices in Korean won.) Thus, the effort by the United States to inject those issues into the Panel's consideration is improper, because those issues were not relied upon by the DOC when it made its determination.

4. In addition, as mentioned at the Second Meeting, Korea respectfully submits that it is possible for the Panel to resolve the local sales issue without deciding whether the local sales were denominated in dollars or won. The question for the Panel is whether or not the currency conversion made by the DOC was "required."³ The DOC ignored the dollar prices provided by POSCO, and instead converted the converted won amounts into dollars. That conversion was not required. There is no reason to believe that the dollar prices provided by POSCO were unreliable or that the DOC had to ignore them for some reason, such as to eliminate or reduce some type of distortion. The conversion was, therefore, inconsistent with Article 2.4.1 of the Anti-Dumping Agreement. Significantly, nothing in the DOC's final determinations (or even in the U.S. arguments before the Panel) explains why that conversion was required. That should be the end-point of the Panel's review.

5. The answers provided to the individual questions below are without prejudice to Korea's position that the Panel should limit its review of U.S. treatment of POSCO's local sales to the reasons provided in the DOC's final determinations.

Q.1. For the United States. In the Plate case, did you verify any "dollar-denominated" sales local sales? If so, did you verify the amount actually paid in respect of those sales?

6. The DOC verified three dollar-denominated local sales in the Plate case. This is evident from Exhibits 6, 23, and 24 of the SSPC Sales Verification Report.⁴ Because the Plate customers paid on a rolling basis, the DOC did not verify the amount actually paid in respect of particular local sales. The DOC did, however, verify other information sufficient to demonstrate that the local sales are denominated in dollars, including the fact that the order sheets are denominated solely in dollars and the fact that the won amount shown on the invoice was calculated from the dollar price.⁵ The DOC also verified POSCO's accounting system, which includes an account for exchange gains and losses.⁶

Q.2. For Korea. It is the understanding of the Panel that the initial questionnaire responses in both the Plate and Sheet cases reported the "local dollar-denominated sales" at issue at the invoiced price in won, rather than the amount of won actually paid. In the Plate case, the Panel is unaware that POSCO at any point informed the USDOC that the amount paid in won differed from the amount invoiced in won, nor that any evidence in the record so indicates. If Korea considers that the Panel's understanding is incorrect, please provide relevant evidence from the record in support of Korea's view.

7. Korea agrees that POSCO initially reported the converted won amounts shown on the invoices for the local sales. In its supplemental responses, however, POSCO corrected that mistake and reported the actual dollar prices. POSCO explained that the local sales were denominated in dollars, and argued that the United States should use the dollar price.⁷

8. Korea is unaware of any evidence in the record of the Plate case that indicates that POSCO specifically informed the DOC that the amount paid in won differed from the converted won amounts shown on the invoice. However, as discussed in response to the previous question, Korea believes that there is considerable evidence showing that the local sales were denominated in dollars. From that evidence, it follows that the amount of won actually paid would not be the same as the converted won amount shown on the invoice - unless it so happened that the exchange rate on the date of invoice was exactly the same as the exchange rate on the date of payment.

9. Furthermore, the United States should not be allowed to defend its anti-dumping measures on the ground (not mentioned in the final determinations) that POSCO should have provided additional proof that the local sales are denominated in dollars, when the Anti-Dumping Agreement squarely places the burden on the DOC to ask for all necessary information.⁸ POSCO informed the DOC that the local sales were denominated in dollars and provided proof in support of that fact and the DOC did not ask POSCO for any additional proof during the proceedings at issue. In these circumstances, it would be unfair to allow the United States to contend now that POSCO should have submitted more information when it is too late for POSCO to do so.

Q.3. For Korea. Please identify where, in the course of your argumentation in the Plate and Sheet investigations regarding whether "dollar denominated" local sales were in fact dollar or won sales, you drew the attention of the USDOC to the fact that the order sheets indicated "D" for dollar sales.

10. At the verifications in both the Plate and Sheet investigations, POSCO drew the attention of the DOC to the fact that the order sheets indicated "D" for dollar sales. This is evident from the Exhibits to the Sales Verification Reports that concern the local sales. Each Exhibit contains not only the order sheet itself, but also the "translation key" which explained in considerable detail how to read the order sheet. Indeed, the Sales Verification Reports in both investigations expressly noted that the DOC's examination of the local sales began with POSCO's order sheet.⁹

11. POSCO then argued in its case briefs in SSPC that the DOC should accept the dollar prices of local sales because, inter alia, POSCO "negotiates prices and invoices its customers in US dollars." POSCO cited Exhibit 6 of the SSPC Sales Verification Report as "demonstrating that the sales price was negotiated in US dollars and invoiced in US dollars."¹⁰ Likewise, POSCO argued in its SSSS Case Brief that, for local sales, "it negotiates prices and invoices its customers in .US dollars."¹¹ The references to the negotiations in dollars are references to the verified fact that POSCO's order sheets indicated that the price was set in dollars. (The order sheets are the only evidence of the sales negotiations in the verification reports. Consequently, the references to sales negotiations are necessarily references to the order sheets.)

II. UNPAID SALES

Q.1. For Korea. Korea characterizes the United States' "constructed export price defense" as a post hoc argument. The United States has however indicated places in the record that it considers to indicate that, in respect of sales through POSAM, the United States deducted an allocated portion of the US bad debt expenses as part of the construction of the export price (US answer to question 4 from the Panel regarding unpaid sales). The portions of the record cited by the United States seem to suggest that the adjustments in respect of POSAM sales were conducted under a legal provision applicable only to adjustments relating to a constructed export price and using a different methodology than is used for circumstance of sale adjustments. If you disagree, please explain, citing to the relevant portions of the record.

12. The final determination in Plate gives the following explanation for the DOC's decision to make adjustments to POSCO's prices for the unpaid sales (omitting the intertwined discussion of the decision to include POSCO's atypical sales in the dumping analysis):

It is the Department's practice to . . . treat the bad debt as a direct selling expense when the expense is incurred on sales of subject merchandise. See Color Televisions at 4412. As stated above, at verification, the Department found that POSAM reversed the sales in its books at year-end by issuing negative invoices to the customer for the unpaid merchandise in question. Thus, although POSAM does not maintain separate bad debt accounts, these sales have been effectively classified as a type of bad debt. . . . However, these sales led to a bad debt expense which is directly related to sales of the subject merchandise. See, AOC International v. US, 712 F. Supp. 931 (CIT 1989). For calculation, see Analysis Memorandum.¹²

Likewise, the final determination in Sheet gives essentially the same explanation, although the discussion is somewhat longer (as it addresses various related issues that do not affect the fundamental rationale for the decision). Indeed, Sheet expressly follows Plate in this respect.¹³

13. It is apparent, therefore, that the DOC gave only one reason for its decision to adjust POSCO's prices to account for the unpaid sales: The non-payment gave rise to a "bad debt expense" which is a "direct selling expense" for which adjustments are made under US practice. No other explanation was provided in the final determinations. Specifically, no mention is made in the final determinations that the adjustment was necessary to construct export price for sales through POSAM (while simultaneously making effectively the same adjustment for sales made directly by POSCO). Thus, the US argument to that effect before this Panel is post hoc and should be ignored.

14. In support of its post hoc argument, and in response to the Panel's Question Number 4 at the First Meeting, the United States has argued (or, at least, implied) that the codes at the end of the DOC's Final Analysis Memoranda show that the adjustments were made in the calculation of the "constructed export price." That argument is misleading in several respects:

• First, what matters to this Panel's review is the reasons given for the DOC's actions in the DOC's final determinations. The question for the Panel is whether the decisions actually made by the investigating authority conform with WTO rules, and not whether the defending party's lawyers can now assemble a new rationale from the facts on the record that may or may not conform with WTO rules. Since the final determinations show that concerns about the construction of the CEP did not form any part of the basis of the DOC's decision to make adjustments for the unpaid sales to both CEP and non-CEP (i.e., EP) sales,¹⁴ the fact that the United States can now point to other information about calculating CEP not mentioned in the final determinations is irrelevant.

• Second, the codes do not accurately reflect the DOC's actual methodology. For example, while the codes show that the DOC reduced POSCO's export price only for CEP sales (and not for EP sales), they do not show that the DOC increased POSCO's normal value for EP sales.¹⁵

• Third, the US argument confuses the terminology used under US law with the requirements of the Anti-Dumping Agreement. In fact, an analysis of the relevant provisions demonstrates that US law uses the terms "constructed export price" and "normal value" in a different manner than the Anti-Dumping Agreement (and Article VI:1 of GATT 1994).

- Under the Anti-Dumping Agreement (and Article VI:1 of GATT 1994), the terms export price (or constructed export price) and normal value refer to the unadjusted prices in the markets under consideration. The "allowances" made under the third sentence of Article 2.4 (for "differences affecting price comparability") and under the fourth sentence of Article 2.4 ("in the cases referred to in paragraph 3 of Article 2") are made as part of the comparison of export price and normal value, and not as part of the determination of export price and normal value.¹⁶

- Under US law, by contrast, all adjustments and allowances are made in the calculation of the export price (or constructed export price) and normal value.¹⁷ Once the export price (or constructed export price) and normal value have been calculated in this manner, no further adjustments are made in the comparison itself.

This fundamental difference in the structure of the US law and the Anti-Dumping Agreement means that the calculation of CEP under US law (based on the price net of certain statutory adjustments) is not the equivalent of the "construction" of an export price under Article 2.3 of the Anti-Dumping Agreement (based on the price before the "allowances" made in the comparison).

• Fourth, the DOC's actual practice demonstrates that the adjustments made to calculate CEP under US law are not necessarily appropriate allowances to construct an export price within the meaning of the Anti-Dumping Agreement. The DOC does not, in practice, limit the adjustments made to calculate CEP to items that relate to the affiliated US importer's operations in the United States. Instead, the DOC deducts all "direct" selling expenses from the CEP, even if those selling expenses relate solely to the foreign producers' operations.¹⁸ Expenses that relate solely to a foreign producers' operations plainly do not fall within any of the allowances permitted to construct an export price within the meaning of the Anti-Dumping Agreement. It follows, then, that the mere fact that expenses are deducted from CEP under US law does not, by itself, make them adjustments to "construct an export price" within the meaning of the Anti-Dumping Agreement.

15. In the end, however, it does not matter whether the United States believed that it was calculating a "constructed export price" under US law when it deducted the cost of non-payment from the US sales prices. The issue before this Panel is whether the United States acted in accordance with the requirements of the Anti-Dumping Agreement. As explained in Korea's previous submissions, an analysis of both the text of the fourth sentence of Article 2.4 and the objective and purpose of Article 2.3 demonstrate that the US adjustment for the cost of non-payment is not permitted under Article 2.3

16. Finally, Korea disagrees with the suggestion that the DOC used "a different methodology" to calculate the adjustment to CEP sales than it used to calculate the adjustment to EP sales. POSCO had one customer go bankrupt. That customer owed POSCO one amount for Plate sales and one amount for Sheet sales. In each case, DOC calculated one amount that was the cost of the unpaid sales. The DOC then divided that amount by POSCO's total exports (measured in metric tons). Thus, the DOC used one methodology to calculate one adjustment amount (for both direct EP and indirect CEP sales). It then simply subtracted that one amount from export price for the indirect CEP sales and added that one amount to normal value for the direct EP sales.

Q.2. For the United States. The United States argues that, "[b]ecause Korea has made no claim under Article 2.3, the United States' decision to construct export price and the methodology it employed to do so are not issues before this Panel." US rebuttal submission, para. 22. Article 2.4 provides that "[I]n the cases referred to in paragraph 3, allowances for costs incurred between importation and resale, and of profits occurring, should also be made." Thus, it could be argued that the guidelines governing the construction of an export price pursuant to Article 2.3 may be found in Article 2.4. Please comment.

17. Moreover, as explained during the Panel's Second Meeting, the fact that Article 2.4 specifically refers to Article 2.3 should be sufficient to bring Article 2.3 within the terms of reference for this proceeding. Indeed, the decision by the Appellate Body in the Argentine Footwear case indicates that the provisions of Article 2.3 could properly be considered by the Panel to the extent that they are relevant to Korea's claims.¹⁹

18. A proper analysis demonstrates, however, that Article 2.3 is not relevant to Korea's claims, because Korea has not claimed that Article 2.3 establishes a separate basis for finding that the U.S. actions violated the requirements of the Anti-Dumping Agreement. Instead, Korea has addressed Article 2.3 simply to refute a defense offered by the United States. The Panel plainly would be permitted to consider Korea's refutation of the US defence, in order to address the merits of Korea's claims under Article 2.4 - even if Article 2.3 were not within the terms of reference.²⁰

19. Finally, to the extent that it is necessary to consider the relationship between Articles 2.3 and 2.4, Korea believes that the following points are relevant:

20. First, it is important to view Article 2.3 in context. Contrary to the US claims, Article 2.3 does not purport to address the permissible methodologies for "constructing an export price." Instead, Article 2.3 establishes only that, when exports are made to an affiliated importer, the investigating authorities may base their analysis on the importer's resale price.

21. In this regard, it should be recalled that the default rule under the Anti-Dumping Agreement is that the dumping calculations should be based on the "export price of the product exported from one country to another."²¹ If this rule were applied in all cases without exception, then the dumping analysis for sales from an exporter to an affiliated importer would have to be based on the transfer price between them - because that transfer price is the "export price of the product exported from one country to another." Moreover, if there were no export price, it would not be possible to calculate dumping margins at all.

22. Article 2.3 of the Agreement establishes the permissible exceptions to this default rule. It provides that the default rule does not apply "where there is no export price or where it appears to the authorities concerned that the export price is unreliable because of association of a compensatory arrangement between the exporter and the importer or a third party." Instead, in such situations, Article 2.3 provides that "the export price may be constructed on the basis of the price at which the imported products are first resold to an independent buyer." Significantly, that is all that Article 2.3 provides. The sole purpose of Article 2.3, then, is to authorize the investigating authorities to base their analysis on the resale price - and not on the "export price of the product exported from one country to another."

23. Second, as discussed in Korea's previous submissions, the rules governing the methodology used to construct an export price are found in the fourth sentence of Article 2.4 - and not in Article 2.3.²² That sentence describes certain "allowances" that "should also be made" in "the cases referred to in paragraph 3 of Article 2" (i.e., when there is no export price or the export price is unreliable because of association of compensatory arrangement between the exporter and the importer or a third party). There are no other provisions of the Agreement that describe the permissible allowances.

24. Moreover, the structure of the Anti-Dumping Agreement dictates that any provisions concerning the permissible adjustments would have to be found in Article 2.4. Put simply, Article 2.4 is the only provision of the Agreement that addresses the mechanics of the comparisons and the rules governing appropriate allowances. Consequently, it is logical that, when the rules governing adjustments to construct an export price were drafted, they were incorporated in Article 2.4. That Articles should, therefore, be interpreted as providing exclusive guidance on the adjustments that are permissible in the dumping comparisons - including the adjustments that are permitted to construct an export price in "the cases referred to in paragraph 3 of Article 2." Adjustments that do not conform with the requirements of Article 2.4 are not permissible in any circumstances.²³

Third, the arguments about the applicability of Article 2.3 should not obscure the obvious relevance of the first sentence of Article 2.4, which requires a "fair comparison" for all transactions - whether based on "export price" or "constructed export price." If the US treatment of the unpaid sales resulted in an unfair comparison, then that methodology was inconsistent with the "fair comparison" requirements of the first sentence of Article 2.4. Thus, even if the adjustment for the cost of non-payment had been made solely under Article 2.3 (and not under the third or fourth sentences of Article 2.4), the Panel would still have to consider whether the comparison that resulted was fair.

25. As discussed in detail in Korea's previous submissions, the US treatment of the unpaid sales was unfair - because it penalized POSCO for an event beyond its control, and created dumping margins due to the inclusion of atypical sales. Thus, the US methodology violated the "fair comparison" requirement of Article 2.4.

Q.3. For the United States. Is there anything in the record of the investigations indicating whether POSAM extended the credit in respect of the unpaid sales and thus bore the risk of non-payment, or whether, to the contrary, the risk of loss was borne by the exporter? Where in the record does the USDOC make such a determination in respect of its treatment of unpaid sales as a cost incurred by POSAM between importation and resale.

26. The risk of non-payment was borne by POSAM. The DOC verified that POSAM paid POSTEEL (which in turn paid POSCO) for the goods that were then re-sold to the ABC Company, and that POSAM was not subsequently reimbursed by POSTEEL or POSCO for the non-payment.²⁴

27. Korea believes that the question of which entity bore the risk of non-payment was not relevant to the DOC's decision to make adjustments for the non-payments.

28. Moreover, the fact that POSAM bore the risk of non-payment does not mean that the risk constituted a cost incurred between importation and resale for which an adjustment was appropriate under the fourth sentence of Article 2.4. To begin with, the risk of non-payment is not a cost; it is a probability. Moreover, the risk exists only after there has been a resale - and not between importation and resale.

29. Finally, as discussed in Korea's previous submissions, the DOC did not make an adjustment for the risk of non-payment that existed at the time of resale. Instead, it made an adjustment for the actual cost of non-payment that occurred only after the resale. Thus, the DOC's adjustment cannot be justified as an adjustment for a cost incurred between importation and resale.

Q.4. For the United States. If unpaid sales were a cost incurred by POSAM between importation and resale in respect of those sales handled by POSAM, why did the USDOC consider that those sales should be allocated over all sales of the subject merchandise in the United States, irrespective of whether or not those sales were handled by POSAM?

30. As discussed in Korea's previous submissions, the allocation methodology chosen by the DOC was plainly arbitrary. An allocation of the actual costs of non-payment over all US sales would have been appropriate only if the evidence demonstrated that all US sales had the same risk of non-payment, and that no other sales in other markets had a similar risk. In these cases, however, there was no such evidence. As discussed in Korea's previous submissions, there was no evidence indicating that the risk of non-payment was different for US sales than for sales in other markets.

31. The error in the US methodology might be seen from the following example: Suppose that an automobile insurer covers ten cars - five of which are black and five of which are navy blue. Now suppose that one of the blue cars has an accident for which the insurer must pay. When analyzing the future risk of accidents, it plainly would be absurd for the insurer to assume that blue cars have a one-in-five risk of accidents, while black cars have a zero risk of accidents. Yet that is essentially what the DOC did in this case. It took the fact that one US customer did not pay as evidence that US customers had a high risk of non-payment, and that Korean customers had no risk of non-payment.

32. As explained in Korea's previous submissions, the only reasonable methodology for analyzing the risk of non-payment would have been to consider the historical experience of non-payment in the two markets. In the absence of such an analysis, there simply is no basis for concluding that the risk of non-payment in the United States was any different from the risk of non-payment in any other markets in which POSCO extended credit. And, to the extent that the non-payment was simply the result of a risk of non-payment that was the same in all markets, there was no basis for allocating the cost of the non-payment solely to sales in one market.

Q.5. For the United States. Without prejudice to your position that the construction of the export price is not within the Panel's terms of reference, please respond to Korea's argument in paragraphs 56-59 of Korea's rebuttal submission.

33. In Korea's view, this question is uniquely directed to the United States.

Q.6. For both parties. It is a generally accepted principle that certain costs are to be expensed while others are to be allocated over time or over product. See, e.g., ADP Agreement, Article 2.2.1.1. In respect of the treatment of unpaid sales, for example, the United States allocated the "cost" of unpaid sales incurred by POSAM over all subject merchandise. Assuming for the sake of argument that the unpaid sales were "costs . . . incurred between importation and resale," it could be argued that those costs should be allocated over a period of time longer than the POI. Please comment.

34. In Korea's view, an amortization of the actual costs of non-payment over an extended period of time would not constitute an acceptable substitute for a proper measurement of the risk of non-payment. As explained in Korea's previous submissions, a proper adjustment would have to reflect differences in the risk of non-payment that were known at the time the sale was made - and not the actual costs of non-payment that could not have been known until after the sale occurred.

35. In any event, even if an amortization methodology were acceptable, the amortization period would have to reflect the risk of the non-payment. If too short an amortization period were used, the result would be to inappropriately overstate the costs assigned to the initial years and understate the costs assigned to subsequent years.²⁵

III. MULTIPLE AVERAGING

Q.1. Korea acknowledges that "zeroing" per se is outside this Panel's terms of reference. It argues however that it does claim that the "multiple averaging" methodology used by the United States violates the AD Agreement, and that, to the extent that "zeroing" is an integral part of that methodology, it is properly within the Panel's terms of reference. Korea's request for establishment (WT/DS179/2), however, states that

"[I]n the final determinations of sales at less than fair value . . . the DOC divided the period of investigation into two sub-periods and calculated separate weighted average normal values for each sub-period. That methodology, however, is inconsistent with Article 2.4.2. . . ." (emphasis added).

It could be argued that "zeroing" is not part of the "methodology" regarding which Korea complains in its request for establishment of a panel. Please comment.

36. As mentioned at the Second Meeting, Korea believes that it is possible for the Panel to decide the multiple-averaging issue without addressing zeroing. If the Panel finds that it was improper for the United States to split the investigation periods to account for the devaluation of the won during the periods, then zeroing simply disappears from this case. In other words, because splitting the periods was itself a violation of the Anti-Dumping Agreement and the zeroing at issue necessarily occurred after the periods were split, the DOC never should have been in the position to engage in the zeroing at issue in the first place. As stated by the Panel in the Audio Tapes case, "the issue of 'zeroing' would not arise in cases where the comparison made was between an average normal value and an average export price."²⁶

37. That is not to say, however, that zeroing (as applied in the multiple-averaging methodology) is outside the Panel's terms of reference. To the contrary, Korea has brought the DOC's multiple-averaging methodology before the Panel and zeroing is an inherent part of that methodology.

38. The Panel has observed that it is arguable that zeroing is not part of the methodology described in the request for establishment of a Panel. Apparently, the argument would be that the "methodology" at issue stops after the period-splitting and the calculation of separate averages. In other words, under that view, the "methodology" at issue would not include anything after the calculation of separate averages, including the recombination of those separate averages into one overall average. Korea does not believe that would be a reasonable interpretation of the "methodology" at issue.

• First, splitting the period, calculating separate averages for the sub-periods, "zeroing out" the negative averages, and recombining the separate averages into an overall average were all part of a single methodology. It would not have been possible for the DOC to split the investigation period without later recombining the separate averages; they are simply two parts of a unitary whole. From the decision to split the period, all the rest followed. The US methodology should not be artificially divided into its component stages.

• Second, even the US arguments show the integral link between zeroing and the decision to engage in multiple-averaging (and all that entailed). The US argues that the key distinction between Preserved Mushrooms from Indonesia and the Plate and Sheet cases is that period-splitting would have had no effect in Mushrooms but it did have an effect in the cases at issue.²⁷ That distinction can mean only one thing: In Mushrooms, neither potential sub-period had a negative margin to be zeroed, so there was no reason to split the investigation period. This shows that the DOC split the periods in Plate and Sheet precisely in order to engage in zeroing.²⁸ It would be artificial indeed to exclude zeroing from Panel consideration of the methodology of which it is not only an integral part, but was in fact the motivating force.

• Third, beyond being the motivating force for the DOC's decision to split the investigation period, zeroing is also the motivating force for Korea's decision to challenge the period-splitting before this Panel. But for zeroing, splitting the period would not have had a significant effect on the dumping margin. Splitting the period still would have been inconsistent with WTO rules, to be sure, but there would have been no practical reason to challenge it. It would be an odd interpretation of the Panel request indeed if it excluded consideration of the factor motivating a claim.

• Finally, Korea considers that it is important that the terms of reference should be understood in light of the entirety of Korea's request for the establishment of this Panel. After identifying in detail the measures at issue, Korea listed ten paragraphs of reasons why the US measures are inconsistent with WTO rules. Korea expressly stated at the outset that those ten paragraphs were "without limitation." Then, Korea reinforced the point as follows: "The above summary is designed to briefly describe the legal basis of the complaint sufficient to present the problem clearly, but is not to be taken as restricting the arguments that Korea may develop before the Panel."²⁹

Q.2. For the United States. Do you agree that multiple averaging is permitted by Article 2.4.2 only to the extent it enhances the comparability of the normal value and export prices being compared?

39. In Korea's view, Article 2.4.2 does not permit multiple averaging to "enhance" the comparability of normal value and export price. Instead, Article 2.4.2 requires that a single average export price and a single average normal value be calculated for each set of comparable transactions (unless a transaction-to-transaction methodology is employed, or the exceptional circumstances authorizing the use of an average-to-transaction methodology exist). The determination whether transactions are "comparable" must be made before constructing the appropriate averages under Article 2.4.2.

40. For these purposes, the "comparability" of transactions is to be determined based on the provisions of the Agreement governing permissible comparisons. Since movements in exchange rates do not provide a basis for considering transactions non-comparable under any provisions of the Agreement, exchange-rate movements do not justify the division of the investigation period into separate averaging periods.³⁰

1 Korea - Anti-Dumping Duties on Imports of Polyacetal Resins from the United States, Report of the Panel, ADP/92, adopted on 27 Apr. 1993, para. 210; see also Mexico - Anti-Dumping Investigation of High Fructose Corn Syrup (HFCS) from the United States, Report of the Panel, WT/DS132/R, adopted 24 Feb. 2000, paras, 7.104, 7.192.

2 SSPC Final Determination, at 15456 (ROK Ex. 11); SSSS Final Determination, at 30678 (ROK Ex. 24).

3 Korea notes that, in both the SSPC and SSSS investigations, POSCO expressly argued the local sales issue in terms of whether or not the double-conversion was "required." See POSCO�s SSPC Case Brief, at 5 (ROK Ex. 7); POSCO�s SSSS Case Brief, at 4 (ROK Ex. 20).

4 See ROK Ex. 6, Ex. 84.

5 See id. Note in particular the hand-written calculations based on the exchange rate for the date of invoice, which was provided in the exchange rate chart on the last page of SSPC Sales Verification Exhibit 6 (ROK Ex. 6).

6 See SSPC Sales Verification Report, at 3-4 (ROK Ex. 6).

7 See Korea�s Responses to Panel Questions at the First Meeting, at 22-22 (Response D.3).

8 See Anti-Dumping Agreement, art. 2.4 (chapeau, last sentence); id., art. 6.1.

9 See SSPC Sales Verification Report, at 4 (ROK Ex. 6); SSSS Sales Verification Report, at 14 (ROK Ex. 19).

10 See POSCO�s SSPC Case Brief, at 4 (ROK Ex. 7) (emphasis added).

11 See POSCO�s SSSS Case Brief, at 4 (ROK Ex. 20) (emphasis added).

12 SSPC Final Determination, at 15448-49 (ROK Ex. 11).

13 See SSSS Final Determination, at 30674 (ROK Ex. 24).

14 Indeed, Korea submits that it is not logically possible for concerns about the construction of the CEP to form part of the basis of the DOC�s decision to make adjustments to POSCO�s EP sales, because by definition concerns about construction of CEP are irrelevant to EP sales. It follows that concerns about the construction of the CEP could only form part of the basis for the decision to make adjustments to CEP sales, if there were a decision to make the adjustment for CEP sales that was separate from the decision to make the adjustment for EP sales. There is, however, simply no basis in the final determinations for the view that the DOC made two separate decisions.

15 For Plate, Page 8 of the Final Analysis Memorandum (ROK Ex. 12) shows that DOC increased DIREXPU (Direct Expenses - US) by { }, the "amount associated with unpaid bankrupt sales." Page 9 shows that DOC subtracted DIREXPU from NETPRIU (Net Price - US) for CEP sales, but not for EP sales. By contrast, Page 13 shows only one method for calculating NETPRIH (Net Price - Home Market, or normal value). No distinction is made between calculating NETPRIH in one way for CEP sales and in another way for EP sales. Specifically, no indication is given that for EP sales DIREXPU is added to NETPRIH. Yet, by the United States� own admission, in fact the DOC did add DIREXPU (including the adjustment for unpaid sales) to NETPRIH for EP sales.

Likewise, for Sheet, Page 9 of the Final Analysis Memorandum (ROK Ex. 25) shows an increase of { } to DIREXPU. Page 10 differentiates between EP and CP sales in the calculation of NETPRIU, but Page 13 fails to make the same distinction in the calculation of NETPRIH - notwithstanding the US admission that it either reduced export price or increased normal value for all sales.

16 The relevant provisions under GATT 1994 are as follows: The second sentence of Article VI:1 of GATT 1994 provides that the existence of dumping is to be determined by comparing the "price of the product exported from one country to another" to the normal value - where normal value is essentially defined to be the domestic price (or, "in the absence of such domestic price," third-country prices or constructed value). The third sentence of Article VI:1 then provides that "due allowances shall be made in each case for differences in conditions and terms of sale, for differences in taxation, and for other differences affecting price comparability." Significantly, Article VI:1 does not describe these allowances as adjustments made in calculating the export price or normal value. To the contrary, the allowances are described only after the export price and normal value have been defined. See GATT 1994, Art. VI:1.

The Anti-Dumping Agreement adopts a similar structure. Article 2.1 provides, again, that the dumping margins are to be calculated based on a comparison of the export price to the domestic price. Article 2.2 sets forth detailed rules for determining normal value. Article 2.3 provides that, when the merchandise is exported to an affiliated importer, a constructed export price based on the importer�s resale price may be used in lieu of the export price. Finally, Article 2.4 describes the manner in which the price comparison is to be made, and specifies certain "due allowances" and other adjustments that are required for the comparison. Once more, the allowances and adjustments do not affect the determination of export price or normal value under the Anti-Dumping Agreement. Instead, the allowances and adjustments are made as part of the comparison, after the export price and normal value have been determined.

17 Thus, the US statute indicates that the "export price" or "constructed export price" mean "the price at which the subject merchandise is first sold ... as adjusted under" the applicable subsections of the US statute. See Tariff Act of 1930, as amended, � 772(a) and (b), 19 USC. � 1677a(a) and (b) (ROK Ex. 1). Similarly, the US statute defines normal value (in the first instance) as the price for home-market sales - and then indicates that this "price" is to be "increased," "reduced" or "increased or decreased" for various adjustments. See Tariff Act of 1930, as amended, � 773(a)(1)(B) and (a)(6), 19 USC. � 1677b(a)(1)(B) and (a)(6) (ROK Ex. 1).

18 For example, where a foreign producer pays royalties on its sales of the merchandise, the DOC will deduct the amount of the royalties paid on exports from the constructed export price. See, e.g., Certain Fresh Cut Flowers From Ecuador, 64 Fed. Reg. 18878, 18882 (Apr. 16, 1999) (Preliminary Determination); Stainless Steel Plate from Sweden, 63 Fed. Reg. 36877 (July 18, 1998) (Preliminary Determination); Dynamic Random Access Memory Semiconductors of one Megabit or Above from Korea, 63 Fed. Reg. 11411, 11413 (Mar. 9, 1998) (Preliminary Determination); Static Random Access Memory Semiconductors from Korea, 63 Fed. Reg. 8934, 8935-36 (Feb. 23, 1998) (Final Determination). There clearly is no basis for claiming that royalties paid on production in the foreign country constitute an appropriate adjustment to "construct" an export price.

19 In Argentine Footwear , the Appellate Body observed that:

We note that the very terms of Article 4.2(c) of the Agreement on Safeguards expressly incorporate the provisions of Article 3. Thus, we find it difficult to see how a panel could examine whether a Member had complied with Article 4.2(c) without also referring to the provisions of Article 3 of the Agreement on Safeguards.... What is more, we fail to see how any panel could be expected to make an "objective assessment of the matter," as required by Article 11 of the DSU, if it could only refer in its reasoning to the specific provisions cited by the parties in their claims.

30 See Korea�s Second Submission, paras. 102-107.

To continue with ANNEX 2-1

Return to Table of Contents