EUROPEAN COMMUNITIES - ANTI-DUMPING DUTIES ON
MALLEABLE CAST IRON TUBE OR PIPE FITTINGS
FROM BRAZIL

Report of the Panel

(Continued)

4. Issue 4: constructed normal value amounts used for profit and SG&A

(a) Arguments of the parties

7.121 Brazil alleges that the European Communities erred in the calculation of constructed normal value under Article 2.2 and 2.2.2. In particular, Brazil alleges that in constructing normal value, the European Communities used data associated with sales of certain product types that "[did] not permit a proper comparison" due to "low volume" under Article 2.2.¹³⁵ Brazil argued that, as the Agreement does not define the term "in the ordinary course of trade", the chapeau of Article 2.2.2 is open to interpretation.¹³⁶ Brazil therefore submits that the European Communities violated Article 2.2.2, as Article 2.2.2, when read together with Article 2.2, requires that the amounts for SG&A and for profits should be based on the data of representative and profitable domestic sales. Where an investigating authority excludes data under Article 2.2, it follows "as a matter of construction" that the same data should be excluded under Article 2.2.2.¹³⁷ Brazil also invokes Article 2.4, alleging that, in using this same data in relation to profit margins, and not making an adjustment for the use of data relating to sales which do not permit a proper comparison, the European Communities breached the requirement to make a fair comparison between normal value and export price.

7.122 The European Communities admits that it used data relating to "low volume" sales in establishing the profit margins under the chapeau of Article 2.2.2. The European Communities argues that this approach is envisaged by the chapeau of Article 2.2.2, which requires (and permits) solely the exclusion of sales not made "in the ordinary course of trade" in establishing the amounts for profit and SG&A in constructing normal value. The European Communities contends that Tupy did not request an adjustment on this ground in the investigation, that the Article 2.4 claim is therefore inadmissible in these Panel proceedings and that no such adjustment would be warranted in any event.

(b) Arguments of third parties

7.123 Without taking a position on Brazil's allegations under Articles 2.2 and 2.2.2, the United States disagrees with Brazil that an improper calculation of constructed normal value can constitute a breach of Article 2.4, or that a putative breach of Article 2.4 can be used to bolster a claim under Article 2.2 and 2.2.2. For the United States, Brazil's arguments with respect to the calculation of constructed normal value, relate to the identification of normal value under Article 2.2 and 2.2.2, and not to its subsequent comparison with export price, under Article 2.4.

(c) Evaluation by the Panel

7.124 Our understanding of the factual situation that gave rise to this claim by Brazil is the following.

7.125 In the underlying investigation, the European Communities defined the product under consideration as: "threaded malleable cast-iron tube or pipe fittings � which are joined by a screwing joining system, falling within CN code ex 7307 19 10".¹³⁸ The European Communities also found that "malleable fittings produced by the Community industry and sold on the Community market as well as malleable fittings produced in the countries concerned and exported to the Community were like products, since there were no differences in the basic physical and technical characteristics and uses of the existing different types of malleable fittings".¹³⁹

7.126 The European Communities determined that products with Tupy's internal product codes 12, 18, 68 and 69 were all within the "product concerned"/"like product" definition. Only types within two of these product codes - 12 and 18 -- were exported to the European Communities.¹⁴⁰

7.127 The European Communities determined normal value for 1375 product types exported by Tupy. The European Communities constructed the normal value for 809 of these types.

7.128 In order to determine whether or not, based on a low volume of domestic sales, it should calculate a constructed normal value, the European Communities applied the 5% "test" referred to in footnote 2 to Article 2.2 of the Anti-Dumping Agreement -- i.e. sales in the domestic market being of such a low volume as to "not permit a proper comparison" -- first at the level of the total sales for each exporting producer¹⁴¹; and subsequently within "directly comparable"¹⁴² "types" of the exported product for the exporting producer.¹⁴³

7.129 The European Communities then considered whether sales were in the "ordinary course of trade", first for the product as a whole and then, for each product type sold by Tupy in Brazil.¹⁴⁴

7.130 For the product types where there was an insufficient level of sales (i.e. such a low volume as "not to permit a proper comparison") and/or sales were not within the ordinary course of trade, the European Communities constructed the normal value ^{145 146}

7.131 The European Communities used SG&A and profit data from domestic sales made in the ordinary course of trade (regardless of whether these sales were considered to be of sufficient volume to permit a proper comparison within the meaning of Article 2.2).¹⁴⁷

7.132 We understand that the issue before us is whether data associated with sales deemed "not to permit a proper comparison" within the meaning of Article 2.2 because of the low relative volume of domestic sales may nevertheless be used in determining profits in constructing normal value under the chapeau of Article 2.2.2. Our examination begins with the text of the relevant provisions.

7.133 Article 2.2 provides for the construction of normal value under certain identified circumstances. It states:

7.134 Article 2.2.2 governs the calculation of SG&A and profits for the purpose of constructing normal value under Article 2.2. Article 2.2.2 provides, in pertinent part:

2.2.2 For the purpose of paragraph 2, the amounts for administrative, selling and general costs and for profits shall be based on actual data pertaining to production and sales in the ordinary course of trade of the like product by the exporter or producer under investigation.

7.135 The chapeau of Article 2.2.2 makes clear that data from sales not in the ordinary course of trade are to be excluded from the calculation of constructed normal value.¹⁴⁸

7.136 However, this does not fully resolve the question before us. We must consider whether the "ordinary course of trade" test is the sole exclusionary test that an investigating authority is permitted to apply to profits in constructing normal value under the chapeau of Article 2.2.2, and, in particular, whether or not data from so-called "unrepresentative" sales - i.e. sales deemed "not to permit a proper comparison" because of low volume within the meaning of Article 2.2 -- must or may also be excluded under Article 2.2.2 (although they may be sales in the ordinary course of trade).

7.137 It is clear that the text of the chapeau of Article 2.2.2 refers to the use of "actual data pertaining to production and sales in the ordinary course of trade of the like product". It does not refer to any additional exceptions or qualifications. The "ordinary course of trade" test is the only test that the text of the provision explicitly identified for application by a Member to exclude data in establishing SG&A and profits under the chapeau methodology.

7.138 We understand that the explicit exclusion pertaining to sales not in the ordinary course of trade in this first sentence of the chapeau to mean that where there is no other such explicit exclusion with respect to sales "not permitting a proper comparison" due to "low volume" elsewhere in the same provision, no such exclusion should be implied. From this, we discern that a Member is not permitted to exclude actual data -- on a basis other than not being made in the ordinary course of trade -- from the calculation under Article 2.2.2. In contrast to the Article 2.2 chapeau, there is no explicit exclusion, in the Article 2.2.2 chapeau, of data relating to sales the volume of which was so low as not to permit a proper comparison. On the other hand, the condition of "sales in the ordinary course of trade", mentioned in Article 2.2, is also explicitly included once again in the chapeau of Article 2.2.2. It is not our task to read into the text of the treaty words that are not there. The ordinary meaning of this phrase includes the SG&A actually incurred and the profits actually realized in the category of production and sales explicitly specified in the Agreement.

7.139 In light of these considerations, we find that Brazil has not established that the European Communities breached its obligations under Article 2.2.2 by including data relating to "low volume" sales in the construction of normal value.

7.140 Brazil also alleged that the European Communities breached the requirement to make a fair comparison between normal value and export price by using data from "low volume" sales and not making an adjustment for the use of such data under Article 2.4. However, we are of the view that Article 2.4 does not provide a legal basis for Brazil's allegation. Brazil's arguments with respect to the calculation of constructed normal value in this case relate to the identification of normal value under Article 2.2 and 2.2.2, rather than to the requirement subsequently to ensure a fair comparison with export price under Article 2.4. For this reason, we decline to consider Brazil's allegation under Article 2.4 in this context.

5. Issue 5: constructed normal value -- product codes

(a) Arguments of the parties

7.141 Brazil asserts that Article 2.2.2, read together with Article 2.6, requires that where an identical product exists, sales data relating to the SG&A and profits of that identical product must be used exclusively in constructing normal value. Only in the absence of sales of such an identical product may data relating to sales of a product with closely resembling characteristics be used. Brazil therefore submits that the European Communities acted inconsistently with Article 2.2.2 in constructing the normal value for certain product types (with codes "12" and "18") by including data relating to domestic sales of "non-identical" product types (with codes "68" and "69").¹⁴⁹

7.142 The European Communities contends that it correctly included the data in question in constructing normal value under Article 2.2.2. As the data were associated with sales of the like product, there was no reason to exclude them.

(b) Evaluation by the Panel

7.143 The Panel refers again to our understanding of the basic factual situation giving rise to Brazil's claim.¹⁵⁰ The European Communities determined normal value for 1375 product types exported by Tupy. The European Communities constructed the normal value for 809 of these types, which included (526) product types within code 12 (for which no identical types were sold domestically or were sold but in insufficient quantities and/or not in the ordinary course of trade); and (283) product types within code 18.¹⁵¹

7.144 In constructing the normal value of certain product types with Tupy internal product code "12" and "18" under the chapeau of Article 2.2.2 (i.e. using actual data for SG&A and profit from sales in the ordinary course of trade of the like product), the European Communities included data relating to sales of types in internal product codes 12, 68 and 69.

7.145 Brazil objects to the European Communities' use of data from sales of types with internal product codes "68" and "69", submitting that the only domestically-sold product types to which the EC should refer when identifying amounts for SG&A and profits under Article 2.2.2 for exported types with codes 12 and 18 are the product type with code 12 for which domestic sales are both representative and profitable. According to Brazil, the product type with code "12" is the identical product type to product type "18". Brazil asserts that types in codes "68" and "69" have differences which make them "non-comparable" with product types in code "18" and that the data from these sales should have been excluded in the calculations under Article 2.2.2. Brazil submits that Article 2.2.2, read together with Article 2.6, makes clear that where an identical product exists, data relating to its SG&A costs and profits shall be used. Only in the absence of such a product may data relating to a non-identical but similar or closely resembling product be used.

7.146 As always, the starting point for our examination of Brazil's claim is the text of the provision invoked by Brazil. We recall Article 2.2 of the Anti-Dumping Agreement , cited in full supra.¹⁵² Article 2.2.2 governs the calculation of SG&A and profits for the purpose of constructing normal value under Article 2.2. Article 2.2.2 provides, in pertinent part:

2.2.2 For the purpose of paragraph 2, the amounts for administrative, selling and general costs and for profits shall be based on actual data pertaining to production and sales in the ordinary course of trade of the like product by the exporter or producer under investigation.

7.147 Thus, Article 2.2.2 mandates the use of actual SG&A and profit data relating to certain sales of the "like product". It makes clear that data from sales of the like product not in the ordinary course of trade should be excluded from the calculation of constructed normal value. This is the only explicit indication given in the provision that data from certain sales should not be taken into account in the construction of normal value. No other data relating to sales of the "like product" are explicitly earmarked for such exclusion.

7.148 Article 2.6 of the Anti-Dumping Agreement contains a definition of the term "like product". It reads:

2.6 Throughout this Agreement the term "like product" ("product similar") shall be interpreted to mean a product which is identical, i.e. alike in all respects to the product under consideration, or in the absence of such a product, another product which, although not alike in all respects, has characteristics closely resembling those of the product under consideration.

7.149 The definition of "like product" in Article 2.6 governs how an investigating authority identifies the scope of the "like product" for the purposes of the investigation and of the Agreement. Once the investigating authority has identified the scope of the "like product", the scope of that definition remains consistent.

7.150 The chapeau of Article 2.2.2 requires the use of actual data from all relevant sales of the like product. Thus, by the express terms of Article 2.2.2 chapeau, actual data from relevant transactions relating to sales of the "like product" - as a whole - may be taken into account to construct normal value. There is no provision to the effect that constructed normal value is to be based only on a limited subset of data relating to sales of certain selective product types falling within the definition of like product, but excluding data relating to sales of other such types. It is not our task to read into the text of the treaty words that are not there.

7.151 We therefore find that Brazil has not established that the European Communities, having defined the "like product" as it did, acted inconsistently with Articles 2.2 and 2.2.2 by including data from sales of the product types of internal product codes 68 and 69, which fell within the definition of "like product", for the purposes of constructing normal value in order to reach a margin of dumping for the like product as a whole.

6. Issues 6 and 10: "fair comparison" with respect to taxation

(a) Issue 6: IPI Premium Credit

(i) Arguments of the parties

7.152 Brazil submits that Tupy obtained a 20% refund of the value of its exported fittings to the EC - the IPI Premium Credit -- to compensate for indirect taxes borne on the Brazilian market by inputs used to produce the exported product. Brazil alleges that the EC violated: Article VI:4 of the GATT 1994 by not negating the effect of the IPI Premium Credit; and Article VI:1 of the GATT 1994 and Article 2.4 of the Anti-Dumping Agreement (1) by failing to fulfil the requirement of a fair comparison between the normal value and the export price by denying allowances for differences in indirect taxation affecting the price comparability; (2) by not indicating to the Brazilian exporter what additional information with regard to the IPI Premium Credit was necessary to ensure a fair comparison; and (3) by imposing an unreasonable burden of proof upon the Brazilian exporter to demonstrate the justification of the Brazilian tax law concerned.

7.153 The European Communities submits that the it denied an allowance in respect of the IPI Premium Credit as Tupy did not demonstrate that this credit "compensated" for internal taxes "borne by the like product" when destined for domestic consumption within the meaning of Article VI:4 of the GATT 1994. Additional reasons for the rejection by the EC authorities of an adjustment to normal value for the IPI Premium Credit was that it was not consistently "booked", doubtful in value and wrongly calculated by Tupy.

(ii) Evaluation by the Panel

7.154 The Panel begins our examination of Brazil's claim in respect of the IPI Premium Credit with the text of the treaty provisions cited by Brazil. Article VI:1 of the GATT 1994 provides that in the determination of dumping:

"Due allowance 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."

7.155 Article VI:4 of the GATT 1994 states:

"4. No product of the territory of any Member imported into the territory of any other Member shall be subject to anti-dumping or countervailing duty by reason of the exemption of such product from duties or taxes borne by the like product when destined for consumption in the country of origin or exportation, or by reason of the refund of such duties or taxes."

7.156 Article 2.4 of the Anti-Dumping Agreement reads, in pertinent part:

"A fair comparison shall be made between the export price and the normal value. This comparison shall be made at the same level of trade, normally at the ex factory level, and in respect of sales made at as nearly as possible the same time. Due allowance shall be made in each case, on its merits, for differences which affect price comparability, including differences in conditions and terms of sale, taxation, levels of trade, quantities, physical characteristics, and any other differences which are also demonstrated to affect price comparability. � The authorities shall indicate to the parties in question what information is necessary to ensure a fair comparison and shall not impose an unreasonable burden of proof on those parties. (footnote omitted)

7.157 Article 2.4 imposes upon the investigating authority the obligation to make due allowance, in each case, on its merits, for differences which affect price comparability. Differences in taxation are explicitly listed as a factor that must be taken into account under Article 2.4 to the extent they may affect price comparability, and for which due allowance shall be made, in each case, on its merits. The last sentence of Article 2.4 provides that the authorities shall indicate to the parties in question what information is necessary to ensure a fair comparison. The requirement to make due allowance for such differences, in each case on its merits, means that the authority must at least evaluate identified differences in taxation with a view to determining whether or not an adjustment is required to ensure a fair comparison between normal value and export price under Article 2.4 of the Anti-Dumping Agreement , and then to make an adjustment where it determines this to be necessary on the basis of this evaluation.¹⁵³ The issue of which specific "allowances" should be made in any case depends very much on the particular facts of the case. The last part of the last sentence of Article 2.4, that the authorities "shall not impose an unreasonable burden of proof" on interested parties, does not remove the burden from interested parties to substantiate their assertions concerning claimed adjustments. In a similar vein, an investigating authority in possession of the requisite information substantiating a claimed adjustment would not be justified in rejecting outright that claimed adjustment.

7.158 Thus, while it is incumbent upon the investigating authorities to ensure a fair comparison,¹⁵⁴ so also is it incumbent upon interested parties to substantiate their assertions concerning adjustments as constructively as possible. The duty of an investigating authority to ensure a fair comparison cannot, in our view, signify that an investigating authority must accept any claimed adjustment. Rather, the investigating authority must take steps to achieve clarity as to the adjustment claimed and then determine whether and to what extent that adjustment is merited. On this basis, we examine Brazil's claim under Article 2.4.

7.159 While there is no disagreement between the parties that legitimate differences in taxation may be the subject of adjustment under Article 2.4, the parties disagree as to whether the IPI Premium Credit fulfils the necessary conditions for an allowance to have been granted in this particular case.

7.160 We therefore examine the particular circumstances surrounding this issue. We have closely scrutinized the record of the underlying investigation, including the communications between Tupy and the European Communities pertaining to the IPI Premium Credit. The European Communities indicated and requested relevant substantiated information in the questionnaire.¹⁵⁵ Tupy indicated in its questionnaire response that according to Brazilian law, "�the Brazilian government gave to Brazilian export companies a tax of 20% refund over the FOB value for indirect taxes paid."¹⁵⁶ At the provisional stage, the European Communities stated that it would "further investigate this issue in order to establish the exact amount of indirect taxes which was actually refunded on export sales made to the Community and at the same time borne by the like product concerned when consumed in Brazil". Brazil submits that Tupy presumed from this statement that it would have had an opportunity to provide more information regarding the Brazilian legislation, if needed.¹⁵⁷ Brazil submits that the European Communities did not, however, "further investigate" the issue and that any uncertainty that remained should be ascribed to the European Communities as a violation of its obligation in Article 2.4 in that the European Communities did not indicate to the Brazilian exporter what additional information with regard to the IPI Premium Credit was necessary to ensure a fair comparison, and imposed an unreasonable burden of proof upon the Brazilian exporter to demonstrate the justification of the Brazilian tax law concerned. Brazil submits that Tupy provided all the information at its disposal with regard to the IPI Premium Credit.

7.161 On the basis of the record, we do not consider Brazil's allegation that the European Communities "did not further investigate the issue" to be a valid one. Tupy's submissions even following the Provisional Regulation continued to contain assertions and information regarding this issue¹⁵⁸, and the European Communities indicated in the Definitive Disclosure that "it had further investigated the claim for an allowance for import charges and indirect taxes made by Tupy".

7.162 In its questionnaire response, Tupy submitted a legal instrument as a basis for its claim for adjustment for the IPI Premium Credit. The European Communities determined that the domestic sales prices relied upon by the EC were "net" of four taxes identified as charged on the sales invoices¹⁵⁹, thereby obviating any need for an adjustment to be made in respect of these particular taxes, and no other taxes were explicitly identified by Tupy as being compensated for by the IPI Premium Credit.

7.163 The European Communities remarked upon the complexity of the issue and questioned the basis for the specification of a 20% adjustment. In the light of the EC indication that more substantiation was required, Tupy still did not specifically identify any particular additional tax borne by the products under consideration and continued to assert that Brazilian internal law provided for such a Credit at a level of 20 per cent.

7.164 The European Communities was faced with the citation by Tupy of its legal right under Brazilian law to receive the 20% IPI Premium Credit. It was not in possession of substantiation of that level through the provision of a copy of the relevant legislation specifying that level nor explicit specific identification of any tax for which this Credit was granted in compensation. We do not consider that the obligation imposed by Article 2.4 would necessarily compel the EC investigating authority to grant the total claimed adjustment in this situation. It may be, for example, that this legal right had not been exercised in a given period by Tupy. In any event, it would be necessary to resort to Tupy's records to discern what had actually occurred. In this respect, moreover, the EC authorities examined the factual basis for Tupy's claim and their evaluation was that they considered, inter alia, that "the real value of this tax credit [is] doubtful"; it was not consistently booked and was wrongly calculated.¹⁶⁰ The record indicates that the EC made these views known to Tupy in the course of the investigation and Tupy had an opportunity to remedy these perceived deficiencies.

7.165 Furthermore, Article VI:4 of the GATT 1994 requires that where the internal tax or duty concerned is a tax or duty borne by the like product sold on the domestic market and where such duties or taxes are refunded or where the exported product is exempted from bearing such taxes or duties, no anti-dumping duty shall be imposed in respect of such duties or taxes. It is clear that the "duties or taxes" in question must be "borne by the like product when destined for consumption in the country of origin or exportation" (emphasis added). The EC investigating authority was not satisfied that the credit in question fulfilled the requirements of Article VI of the GATT 1994.¹⁶¹ Again, the record indicates that the European Communities made these views known to Tupy in the course of the investigation and Tupy had an opportunity to remedy these perceived deficiencies.

7.166 We do not consider that the conduct of the European Communities and its decision not to make any adjustment constitutes a failure to ensure a fair comparison within the meaning of Article 2.4.¹⁶² A reasonable and objective investigating authority could have made an examination of this evidence and taken this decision on the basis of the record of this investigation.

7.167 On the basis of these considerations, and keeping firmly in mind the standard of review we are bound to apply to our examination of the matter before us, we find that Brazil has not established that the European Communities violated Article 2.4 or Article VI of the GATT 1994 in not granting an adjustment in relation to the IPI Premium Credit.

(b) Issue 10: PIS/COFINS

(i) Arguments of the parties

7.168 Brazil argues that under Brazilian law, Tupy received a PIS/COFINS credit amounting to 5.37% over input of the exported final product. Brazil submits that the European Communities violated Article VI:1 of the GATT 1994 and Article 2.4 of the Anti-Dumping Agreement by failing to make a fair comparison between the normal value and the export price by denying full allowances for resulting differences in indirect taxation affecting price comparability and by applying an arbitrary, manipulative and punitive "sampling" methodology in its assessment of the PIS/COFINS tax and credit that lacks any legal foundation.

7.169 The European Communities does not dispute that an allowance for the PIS/COFINS credit is appropriate.¹⁶³ It argues that the EC authorities took the initiative in calculating the adjustment as Tupy made no claim for adjustment relating to this credit in its questionnaire response. Tupy referred to an "estimated Credit PIS/COFINS of 5.37% over input" for the first time in replying to a deficiency letter, but did not request an adjustment or rectify incorrect data it had submitted in its questionnaire Response. The European Communities argues that its calculation of the adjustment on the basis of data from the 20 most exported types sold on the domestic market was a reasonable and appropriate methodology representative of the products sold.

(ii) Evaluation by the Panel

7.170 The Panel recalls that the chapeau of Article 2.4 requires that "a fair comparison shall be made between the export price and the normal value". As we have already stated, differences in taxation are explicitly listed as a factor that must be taken into account under Article 2.4 to the extent they may affect price comparability, and for which due allowance shall be made, in each case, on its merits. The last sentence of Article 2.4 provides that the authorities shall indicate to the parties in question what information is necessary to ensure a fair comparison. The requirement to make due allowance for such differences, in each case on its merits, means that the authority must at least evaluate identified differences in taxation with a view to determining whether or not an adjustment is required to maintain price comparability and to ensure a fair comparison between normal value and export price under Article 2.4 of the Anti-Dumping Agreement , and then to make an adjustment where it determines this to be necessary on the basis of this evaluation.¹⁶⁴

7.171 Differences in taxation are explicitly listed as a factor that should be taken into account under Article 2.4 to the extent they may affect price comparability. The divergence between the parties centres on the methodology applied by the European Communities in calculating this adjustment, which, according to Brazil, is inherently unfair and resulted in this case in an adjustment that Brazil alleges is less than the full amount required. We understand Brazil to allege that the EC failed to conduct a "fair comparison" within the meaning of Article 2.4 of the Anti-Dumping Agreement by calculating an adjustment for PIS/COFINS on the basis of a methodology involving data pertaining to transactions involving the twenty "most-exported" types of pipe fittings from Brazil that were also sold domestically, representing approximately 33% of the total quantity exported.

7.172 We see the issue we must decide as whether the obligation to conduct a "fair comparison" under Article 2.4 permits, or does not preclude, the use by an investigating authority of a representative subset of data relating to certain transactions in calculating adjustments, or whether an investigating authority has an obligation to take into account a comprehensive data set in calculating adjustments.¹⁶⁵

7.173 There is no dispute between the parties that the total amount of PIS/COFINS refunded to Tupy in the IP was 2,491,000 Real. The European Communities calculated the PIS/COFINS adjustment as follows: it divided that total amount by the value of total export sales and then multiplied this percentage by the ratio between the export prices and domestic sales of the product concerned.¹⁶⁶ It thereby granted an adjustment to normal value of 0.88%. The European Communities indicated that it considered that the difference between the adjustment originally requested by Brazil for indirect taxation and 0.88% was misleading and that the entire claimed adjustment could therefore be totally rejected. "Given the complexity of the issue", the European Communities nevertheless decided to grant the adjustment.¹⁶⁷ Tupy contested the accuracy of the EC calculation of the adjustment questioned the legal basis for this adjustment during the investigation.^{168 169}

7.174 The European Communities justifies its methodology, using data pertaining to the 20 most exported types, as being "representative of the products sold"¹⁷⁰, "necessary to allocate the total amount of the refund among the different types of fittings¹⁷¹ and reasonable "given the practical constraints under which investigators act".¹⁷² In response to questioning, the EC specified that its use of the methodology was not exclusively "because of personnel and time constraints",¹⁷³ but that, "[t]he judgement made by the EC investigators was that, in the circumstances of the investigation, given the data in question and the significance of the outcome of the calculation relative [to] the size of the dumping margin, the use of a key based on the 20 most-exported types was appropriate and reasonable."¹⁷⁴ The European Communities asserts that Brazil has not shown that a more expansive methodology would have been more advantageous to Tupy. Brazil now submits to us a calculation -- on the basis of 40 types -- which it argues would have resulted in a more favourable amount for Tupy. On this basis, Brazil contests the EC assertion that there is no indication that its methodology produced less favourable results than another methodology or the use of a complete data set.¹⁷⁵ We are conscious of the constraints placed upon us by Article 17.5(ii). Tupy submitted no such alternative methodology in the course of the underlying investigation.

7.176 We also asked the European Communities to identify the legal basis that permits or does not preclude the use by an investigating authority of data from a representative selection of transactions or the use of an "allocation key" for the purposes of calculating adjustments in an investigation (not involving "facts available" or "sampling" within the meaning of the Anti-Dumping Agreement). The European Communities drew our attention to Article 6.14.¹⁷⁶

7.177 Article 6.14 refers to the "procedures set out above", which we take as a reference to the procedures set out in Article 6 of the Anti-Dumping Agreement . As we understand that the EC is not specifically basing its methodology in this instance upon any provision of Article 6 (in particular, Articles 6.8 or 6.10) we do not believe that Article 6.14 is specifically applicable in this context.

7.178 An investigating authority must act in an unbiased, even-handed manner and must not exercise its discretion in an arbitrary manner. This obligation also applies where an investigating authority confronts practical difficulties and time constraints. We do not find, in Article 2.4, or in any other relevant provision in the Agreement, any specific rules governing the methodology to be applied by an investigating authority in calculating adjustments. In the absence of any precise textual guidance in the Agreement concerning how adjustments are to be calculated, and in the absence of any textual prohibition on the use of any particular methodology adopted by an investigating authority with a view to ensuring a fair comparison, we consider that an unbiased and objective authority could have applied this methodology applied by the European Communities and calculated this adjustment on the basis of the actual data in the record of this investigation. Moreover, Tupy had an opportunity to substantiate its claimed adjustment.

7.179 Recalling that we are bound by our standard of review, we find that Brazil has not established that the European Communities has breached its obligation to ensure a fair comparison under Article 2.4 of the Anti-Dumping Agreement or its obligations under Article VI of the GATT 1994 by the methodology it applied in calculating the PIS/COFINS adjustment.

7. Issue 7: advertising expenses (claims withdrawn)

7.180 Brazil withdraws its claims regarding advertising expenses under Issue 7.¹⁷⁷ We therefore do not examine these claims.

8. Issue 8: packing costs

(a) Arguments of the parties

7.181 Brazil asserts that Article 2.4 of the Anti-Dumping Agreement obligates the investigating authority, as opposed to the exporter, to ensure a fair comparison. Brazil alleges that, in breach of Article 2.4, the European Communities wrongly denied Tupy an adjustment relating to greater packing costs associated with domestic as compared to export sales: imposed an unreasonable burden of proof on the Brazilian exporter and did not indicate to the Brazilian exporter what information was necessary to ensure a fair comparison. According to Brazil, packing costs were allocated (essentially on the basis of labour) 25% for export sales and 75% for domestic sales, and this reasonable allocation key was provided to the EC investigating authorities, who could then have used the on-the-spot verification exercise to assess the physical conditions for packing at the company itself.

7.182 The European Communities argues that, under Article 2.4, the primary responsibility for justifying adjustments is on those claiming them. According to the European Communities, no data were available in respect of Tupy either for packing materials or for working time that distinguished between domestic and export sales. Tupy was unable to produce an allocation key for packing expenses that it could show had been historically used for such expenses. The European Communities denies that its officials ever denied invitations to view relevant evidence, and asserts that -- in the questionnaire and in the verification letter -- it indicated the information that would be necessary to ensure a fair comparison. For the European Communities, verification is essentially a documentary exercise, and no documentary evidence supporting Tupy's request for an adjustment was provided.

(b) Evaluation by the Panel

7.183 The Panel recalls our earlier examination of the obligations imposed by Article 2.4.¹⁷⁸ In this respect, we underline that we agree with the view of a previous panel that the obligation in Article 2.4 to make due allowances for differences that affect price comparability is intended to neutralize differences in transactions that an exporter could be expected to have reflected in its pricing.¹⁷⁹ We further recall that the requirement to make due allowance for such differences, in each case on its merits, means that the authority must at least evaluate identified differences with a view to determining whether or not an adjustment is required to maintain price comparability and to ensure a fair comparison between normal value and export price under Article 2.4 of the Anti-Dumping Agreement , and to then to make an adjustment where it determines this to be necessary on the basis of this evaluation.

7.184 It is not disputed that packing costs may, in principle, be a difference affecting price comparability for which an allowance must be granted in order to ensure a fair comparison of export price and normal value under Article 2.4. Rather, the parties differ in their view of the nature of the evidence that should be submitted in support of a claim for such an adjustment and whether it is the investigating authority or the exporter that bears the burden of identifying and substantiating the claimed adjustment.

7.185 Because the details relating to the exchange of information between Tupy and the EC investigating authorities in the underlying investigation are critical to an understanding of Brazil's claim, we begin by outlining the relevant developments in the underlying investigation.

7.186 The EC questionnaire¹⁸⁰ sent to Tupy contained precise instructions as to the nature of the information requested in respect of packing costs in order to provide the basis for a fair comparison between normal value and export price. In particular, the questionnaire directed Tupy to specify the packing costs for the product concerned; to list material and labour costs separately; to describe packing materials and any special or extraordinary procedures used in preparing the product concerned for shipment to the European Communities; and to report the adjustment transaction-by-transaction. Furthermore, the European Communities sent a letter to Tupy prior to the verification¹⁸¹, which stated, inter alia, "[y]ou are hereby requested to have all supporting documents available for the investigation, including all worksheets used to prepare the reply for the questionnaire�" and "[i]f you claimed in the reply to the questionnaire allowances�you should be prepared to justify these and to have all relevant substantiating evidence readily available."

7.187 In its questionnaire response,¹⁸² Tupy claimed an allowance in respect of the difference between the amounts of packing expenses between the Brazilian and the EC markets. In its submissions in the course of the EC investigation, Tupy asserted that its packaging costs in the domestic market were greater than in the export market and that an adjustment to the normal value was therefore needed to ensure a fair comparison.¹⁸³ In particular, Tupy asserted, at the verification visit, that packing costs were allocated (25%/75%) in respect of labour used, because packing orders for the domestic market required the use of three persons, whereas packing for export orders required only one person.¹⁸⁴ Further, Tupy asserted, the packing process for products to be sold on the domestic market involves the use of more boxes, of varying sizes. According to Tupy, these differences could and should have been verified at the warehouse during the verification visit and an adjustment should not be denied because the Commission abstained from verifying the packing process.¹⁸⁵

7.188 The European Communities rejected Tupy's claim for adjustment for packing in the Provisional Regulation,¹⁸⁶ and subsequently. The European Communities indicated that the reason for the refusal was insufficient substantiating evidence for the cost estimate provided by Tupy, including during verification, and that Tupy had not previously raised the issue of different packing materials.¹⁸⁷ Tupy submitted further objections to the EC assertion that the European Communities had not been provided with sufficient information during the verification visit substantiating the claim for a packing adjustment.¹⁸⁸ The European Communities maintained its refusal to grant an allowance, on the basis of its view that Tupy had not provided sufficient evidence.¹⁸⁹

7.189 Against this factual background, we turn to Brazil's allegation that the European Communities failed to indicate to Tupy what information was necessary in order to ensure a fair comparison within the meaning of Article 2.4. We find ample indication of the European Communities' requests for precise information and further objective substantiation in respect of the packing cost allowance claimed by Tupy in the questionnaire and in subsequent communications with Tupy and record documents.¹⁹⁰ On the basis of these indications in the record of the investigation, we do not find that the European Communities failed to indicate to Tupy the information that was necessary to ensure a fair comparison. The chief difficulty identified by the European Communities in the investigation was the lack of substantiation for the "allocation key" for packaging costs as submitted by Tupy. The European Communities indicated that it considered Tupy's allocation key was not supported by evidence¹⁹¹, that further objective substantiation was required, and that the allocation key "was completely at odds with any figure that could be derived from the volumes of sales on the domestic and export markets".¹⁹² This is an evaluation by the EC investigating authority with respect to the facts before it. The record reflects that the European Communities evaluated the claim for adjustment with a view to considering whether or not an adjustment was merited. We recall our standard of review, which precludes us from substituting our judgment for that of the investigating authority.

7.190 We understand Brazil to reproach the European Communities for not having accepted Tupy's "allocation key" and for not having attempted to verify the approach in the allocation key through physical (i.e. non-documentary) inspection during the verification visit at the premises of Tupy. We therefore examine whether the European Communities acted inconsistently with its obligations under Article 2.4 in not accepting, or in not verifying through non-documentary/physical means, the labour cost "allocation key" submitted by Tupy in support of its request for adjustment for packing costs and whether the European Communities imposed an unreasonable burden of proof upon Tupy.

7.191 We do not agree with Brazil's argument that Article 2.4 required the European Communities to base the adjustment on a visual/physical inspection of the working activities and practices in the packaging area at the company's premises. Rather, we view verification as an essentially "documentary" exercise that may be supplemented by an actual on-site visit. On-site verification is provided for, but not mandated by, the Agreement. Thus, it would seem incongruous to require the European Communities to use a methodology that would have necessitated substantiation through on-site verification.

7.192 An essentially documentary approach to verification -- which focuses upon documented support for claims for adjustment - seems to us to be entirely consistent with the nature of an anti-dumping investigation¹⁹³ and, is, indeed, critical for the purposes of dispute settlement and meaningful Panel review under the DSU and the Anti-Dumping Agreement . We recall that pursuant to the DSU and Article 17 of the Anti-Dumping Agreement , compliance by a Member with the obligations of the Anti-Dumping Agreement is subject to review by a panel and the Appellate Body.¹⁹⁴ A contemporaneous written record, including of evidence substantiating a claimed adjustment by an interested party, is essential as a basis for such multilateral review. Moreover, in this particular case, Tupy was given a clear indication of the EC intention to conduct a verification visit that was predominantly documentary in nature.¹⁹⁵ In these circumstances, in particular, in light of the specific information requested by the European Communities and the continued absence of any specific documentary evidence for adjustment for packing costs on the record of the investigation substantiating Tupy's claim for adjustment for packing costs forming part of the record of the underlying investigation - for example, audited and confirmed historical data clearly distinguishing between packing material or labour costs allocated to domestic and foreign sales -- we cannot find that the European Communities violated its Article 2.4 obligations by not having granted a packing cost adjustment or that it imposed an unreasonable burden of proof upon Tupy.

7.193 On the basis of these considerations, and recalling the standard of review we are bound to apply to our examination of the matter before us, we find that Brazil has not established that the European Communities acted inconsistently with its obligations under Article 2.4 of the Anti-Dumping Agreement by denying an adjustment with respect to packing costs, by failing to indicate to Tupy what information is necessary to ensure a fair comparison or by imposing an unreasonable burden of proof on Tupy in respect of packing costs.

To continue with 9. Issues 9 and 18: currency conversion and opportunities to see relevant information

Return to Table of Contents

135 Brazil's original allegation was that the European Communities wrongly included "low volume" profit data in constructing normal value, and that the European Communities used different data from a different set of transactions for SG&A, on the one hand, and profit, on the other. However, Brazil does not reiterate in its second submission its allegation concerning the use of different data for SG&A and profits. Rather, we understand Brazil's argumentation to have developed to allege that the European Communities violated Article 2.2.2 by including amounts for both SG & A and profits pertaining to sales of product types for which domestic sales were "not representative" within the meaning of Article 2.2 and footnote 2 thereto (see Brazil second written submission, para. 49).

136 Brazil second written submission, paragraph 45.

137 Brazil second written submission, paras. 46-47.

138 Provisional Regulation, recitals 9 - 12, confirmed in Definitive Regulation, recital 9.

139 Provisional Regulation, recital 13, confirmed in Definitive Regulation, recitals 14 - 19.

140 The following is a summary of these four different types (with Tupy internal product codes 12, 18 , 68 and 69) indicating basic characteristics and domestic and export sales to the EC:
 

Types of product concerned	Sold in domestic market (Brazil)	Exported for sale in EC market
12 (BSP threading)	yes	yes
18 (BSP threading)	no	yes
68 (NPT threading)	yes	no
69 (NPT threading)	yes	no

In the Provisional Regulation, the European Communities used the product control numbers proposed by Tupy in the questionnaire, and normal values based on domestic sales were based only on product types 12. In the Definitive Regulation, the EC altered its approach by using internal product numbers, taking into account also data for product types "68" and "69".

141 The volume of domestic sales by Tupy was 22.8 million units, compared to 22.3 million exported to the EC. See EC response to Panel question 36 following the first Panel meeting, para. 36, Annex E-3. Definitive Disclosure, BRL - 16, Annex II, page 8, point 2.11; Provisional Disclosure, BRL-11, Annex II, page 1, point 1.1; Provisional Regulation, recital 20.

142 I.e. whether for each of the product types exported, the quantity sold on the domestic market was at least 5% of the identical type exported to the EC.

143 Provisional Disclosure, Exhibit BRL - 11, Annex 4; Definitive Disclosure, BRL-16, Annex II, Annex 4.

144 See Provisional Regulation, Exhibit BRL-12, recital 23.

145 We do not understand Brazil to be challenging the fact that the EC applied certain tests at two different levels in its calculations, nor the EC's model-by-model analysis.

146 Provisional Regulation, Exhibit BRL-12, recitals 26 and 27.

147 Provisional Regulation, Exhibit BRL-12, paras. 20-27. When the sales to independent customers at prices equal to or above the cost of production represented at least 10% of the total of domestic sales volume of the product concerned by the company concerned. Where this criterion was not met, a weighted average profit margin of the other companies with sufficient sales in the ordinary course of trade in the country concerned was used. We do not understand that Brazil is alleging any inconsistency with the Anti-Dumping Agreement with respect to the EC�s application of this 10% test.

148 Article 2.2.1 addresses when sales of the like product may be treated as not "in the ordinary course of trade". It is clear that data from sales not in the ordinary course of trade should be excluded under the chapeau methodology. We find support for this view in Appellate Body Report, European Communities - Anti-Dumping Duties on Imports of Cotton-Type Bed Linen from India ("EC - Bed Linen "), WT/DS141/AB/R, adopted 12 March 2001, para. 82.

149 Brazil does not refer in its second written submission to its claim under Article 2.4 that, having included data from such sales, the European Communities violated Article 2.4 by refusing to make adjustments for differences in physical characteristics affecting price comparability. We take Brazil�s response to Panel Question 45 and Brazil's second written submission to indicate that Brazil pursues this claim under Article 2.2.2, rather than Article 2.4. In any event, we recall our view supra, para. 7.140, that Brazil�s arguments with respect to the calculation of constructed normal value here relate to the identification of normal value under Article 2.2 and 2.2.2, rather than to the requirement subsequently to ensure a fair comparison with export price under Article 2.4.

150 Supra, paras. 7.124 ff.

151 See Provisional Regulation, recitals 20-27, 35-37; EC response to Panel question 39 following the first Panel meeting, Annex E-3, para. 46 ff.

152 Supra, para. 7.133.

153 We find support for our view in Panel Report, Argentina - Definitive Anti-Dumping Measures on Imports of Ceramic Floor Tiles from Italy ("Argentina - Ceramic Tiles "), WT/DS189/R, adopted 5 November 2001, para. 6.113.

154 We recall the view of the Appellate Body that the obligation to ensure a fair comparison under Article 2.4 "lies on the investigating authorities" and not on exporters. Appellate Body Report, US-Hot-Rolled Steel, supra, note 40, para. 178.

155 The questionnaire states, inter alia, "List all internal taxes imposed on the domestic market products, which were either rebated upon exportation or not collected on the products exported to the European Community.

For each tax listed above, provide English translations of statutes and regulations authorising the collection of the tax, including documents which explain the method of calculation, assessment, and payment of the tax.
For each tax listed above, separately provide information on the tax base or taxable price, the tax rate, the amount of taxes assessed, any deductions or offsets to the tax and the formula used to calculate the tax amount.
Indicate when you are legally obligated or liable for tax payment. Report when you actually paid taxes and whether you maintain separate accounts for these taxes.
Report in the transaction-by-transaction listing below (see points 11 and 12) the amount of such taxes applicable to each domestic sale.
Explain how you calculated this amount�.�

Prepare a listing named �DMALLUR� (computer file - for details see Section H6) of all adjustments you claim for direct sales to independent customers on the domestic market on a transaction-by-transaction basis.�