WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

Page 106


  1. Defining/identifying the "imported product"


    1. Specificity of the imported product


        1. The European Communities, Switzerland, Norway and Brazil recall that the Appellate Body has made clear in US – Lamb that "a safeguard measure can only be imposed on a specific 'product', namely the imported product", and that the correct definition of a "specific product" is important to ensure that a safeguard measure is only imposed "if that specific product ("such product") is having the stated effects upon the 'domestic industry that produces like or directly competitive products.' The conditions in Article 2.1, therefore, relate in several important respects to specific products".609 610


        2. According to the European Communities, Switzerland and Norway, this notion of "specific product" is distinct and narrower than the concept of "like or directly competitive" products referring to domestic versus imported products.611 This requirement was already reflected in Article XIX of the GATT 1947, entitled "emergency action on imports of particular products".612 The requirement of specificity implies that each product be identified and treated separately with respect to increase in imports and causality.613 It prevents investigating authorities from grouping together two or more imported products for the purpose of the increased imports, causation analysis and when imposing the safeguard measures, although they can establish an increase in imports and the effect of causing injury only for one of them.614


        3. According to the United States, the complainants provide no support for their allegation that "the notion 'specific product' referring to imports is distinct and more narrow than the concept 'like or directly competitive product' referring to domestic versus imported products".615 Moreover, their rationale for defining "specific imported products" first is to require authorities to consider whether such imports have increased, as a "filter", prior to conducting the like product analysis. The complainants' proposed methodology has no basis in the Agreement. Moreover, it is ironic that the complainants, who have alleged incorrectly that the USITC's like product definitions were made in order to attain a desired result, actually propose that the USITC should have conducted a results- oriented test prior to defining the domestic like product.616


        4. The European Communities also argues that the Agreement on Safeguards clearly envisages that safeguard investigations be conducted with respect to a single identified product – "a product", not into a bundle of distinct products or a bundle of selected sub-products.617 Read in the light of the object and purpose of the Agreement on Safeguards, which is to ensure that serious injury is not wrongly attributed to an imported product, it prohibits a definition of the imported product that is so


          image

          609 Appellate Body Report, US – Lamb, para. 86.

          610 Brazil's first written submission, para. 87; Switzerland's first written submission, para. 170; Norway's first written submission, para. 175.

          611 European Communities' first written submission, para. 185; Switzerland's first written submission, para. 171.

          612 European Communities' first written submission, para. 186; Switzerland's first written submission,

          paras. 171-172.

          613 European Communities' first written submission, paras. 186, 188.

          614 European Communities' first written submission, para. 186; China's written reply to Panel question No. 35 at the first substantive meeting; Norway's first written submission, paras. 176-177; Switzerland's first written submission, paras. 171-172.

          615 European Communities' first written submission, paras. 184-185.

          616 United States' first written submission, paras. 100-101.

          617 European Communities' written reply to Panel question No. 19 at the first substantive meeting; European Communities' second written submission, para. 90.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 107


          "broad" as to result in injury being wrongly imputed to a product which is not imported at increased quantities.618


        5. The United States maintains that the Agreement on Safeguards does not establish any parameters for defining "such product".619 The complainants' alleged requirement to subdivide or identify separate imported products prior to defining the domestic like product has no support in the Agreement. The complainants urge that there is support for such requirements and narrow definitions by reading interpretations into the Agreement that are not permitted by the text or purpose of the Agreement on Safeguards.620 The dictionary definition affirms that the meaning of the term "product" is quite flexible and depends on the context in which it is used. It can apply to one particular item sold, to a group of items, or to a class of items. In almost every situation in which items may be referred to as a "product", it is possible to discern both a broader "product" of which that product is a subset, and a subset of that product that itself may be referred to as a "product".621 622


        6. The United States further submits that the complainants' reliance on the Appellate Body's findings in US – Lamb in alleging that the USITC was required to define "specific imported products" is misplaced. The Appellate Body rejected imposing a safeguard measure on an imported article, lamb meat, because of the prejudicial effects that such imported article had on the domestic producers of another wholly different domestic product, live lambs, that had not been defined as a like product.623 This statement pertains to the process of defining a domestic industry consisting of producers of like or directly competitive products and does not speak to separating subject imports into categories prior to defining domestic like products as the complainants allege.624 Furthermore, in the paragraph following this finding, the Appellate Body explicitly states that "the first step … is the identification of the products which are 'like or directly competitive' with the imported product", i.e., the first step is defining the domestic like product.625 626


        7. The European Communities and China respond that it can be derived from the Appellate Body Report in US – Lamb that imports must be identified before the like domestic product.627 According to the European Communities, it can also be derived from Article 2.1 of the Agreement on Safeguards as clarified by the Appellate Body Report in US – Lamb that the bundle of domestically produced articles may not contain products that are not even like or directly competitive with each other.628 The United States seems to suggest that, had the imported product been defined as lamb meat and live lambs, then the determination of the like product and domestic industry would be different – the like and imported product would have been lamb meat and live lamb, because live lamb is an input product for lamb meat. However, the approach taken by the United States severs what the Appellate Body has called the logical continuum set forth by Article 2.1 of the Agreement on Safeguards.629 Brazil argues that the Appellate Body report in US – Lamb made it clear that an


          image

          618 European Communities' written reply to Panel question No. 138 at the first substantive meeting.

          619 United States' written reply to Panel question No. 19 at the first substantive meeting.

          620 United States' first written submission, para. 96.

          621 For example, depending on context, each of the following could be considered a "product": a disposable, retractable blue ball point pen; all disposable blue ball point pens together; all disposable ball point

          pens together; all ball point pens together; all disposable pens together; or all pens together.

          622 United States' written reply to Panel question Nos. 19 and 21 at the first substantive meeting.

          623 Appellate Body Report, US – Lamb, para. 86.

          624 Appellate Body Report, US – Lamb, para. 86.

          625 Appellate Body Report, US – Lamb, paras. 87, 92 and 94, footnote 55.

          626 United States' first written submission, para. 97.

          627 China's second written submission, para. 45.

          628 European Communities' second written submission, paras. 134-137.

          629 European Communities' second written submission, paras. 139-140.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 108


          imported product cannot be like more than a single domestic product. For each imported product there is a single domestic like product.630


        8. The United States rejects the argument that Article XIX and the provisions of the Agreement on Safeguards require that the competent authorities analyse the Article 2.1 conditions in a particular order.631 While there is admittedly a logical order to parts of the analysis, there is no specified order of analysis imposed by the Agreement on Safeguards. If the Article 2.1 conditions are met, it is irrelevant whether the competent authorities found increased imports or serious injury first. All that is required is that each of the conditions for imposing a safeguard measure is met. The United States agrees that Article XIX and the Agreement on Safeguards envision a chronological order in the evolution of an increase in imports that would meet the requirements of Article XIX: an obligation or tariff concession, then an unforeseen development, then an increase in imports, and finally, serious injury. However, this does not impose an obligation on the competent authorities to follow this chronological order in structuring their analysis. There are certainly logical first steps in the analysis. The domestic like product must be identified before determining whether the domestic industry producing the like product was seriously injured, and both the domestic like product and the imported product must be defined before determining whether imports have increased relative to domestic production. However, there is no requirement in the Agreement on Safeguards that the imported product must be defined before the domestic like product, or that unforeseen developments be established before determining whether there were increased imports. So long as the competent authorities have made all of the requisite findings, it is largely irrelevant in what order those findings are made. A failure to follow the order of findings advocated by the European Communities and Switzerland does not establish a prima facie case of a violation of the Agreement on Safeguards.632


        9. According to the complainants, it is true that the order of making findings does not always matter. However certain findings cannot properly be made in the absence of other required findings. Thus, for example, a product cannot be held to be like something that has not yet been defined. These complainants contend that the United States admits to having conducted this analysis the other way round and has, therefore, done so incorrectly.633


        10. Korea submits that under the definition adopted by the United States in this case, if the imported product is CCFRS, CCFRS is the domestic like product which is composed of slab, hot- rolled, cold-rolled, corrosion-resistant, and plate. Assuming, arguendo, that CCFRS is a separate like product, if the industry chose instead to limit its safeguard petition to imports of slab alone, the question is whether the domestic like product would continue to be CCFRS or slab. According to the United States' analysis of US – Lamb, and the United States' matching-up approach634, if the only imported product is slab then the like product would only be slabs. In contrast, Korea takes the position that if the United States is correct that slab is simply a part of a broader like product, then the fact that a petition is brought against slab alone does not change the like product. In other words, the like product for slab imports alone would still be CCFRS. Each imported product must be "like" each domestic product. Mere overlap between the imported products (i.e., slab, hot-rolled, cold-rolled, etc.) and the like products (i.e., slab, hot-rolled, cold-rolled, etc.) is not sufficient to find there is a



          image

          630 Brazil's second written submission, para. 10.

          631 United States' second written submission, paras. 36-39.

          632 Switzerland's written reply to Panel question No. 23 at the first substantive meeting.

          633 European Communities second oral statement "Scope and Standard of Review" on behalf of the complainants, para. 28.

          634 Korea's second written submission, paras. 25-33.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 109


          single like product. Korea argues that this is their fundamental disagreement with the United States' position.635


        11. The European Communities reiterates that according to Article 2.1, the determination whether "a product" fulfils certain conditions cannot be made without identifying "such product". Thus, although WTO Members are not obliged to regulate the scope of complaints or requests to start a safeguard investigation, they must ensure that their competent authorities identify the imported product concerned for the purpose of the determination. The European Communities submits that the United States has not done so. 636


        12. Brazil believes that the United States' position on whether investigations and measures apply to a single imported product or whether multiple products can be grouped into a single investigation of imports has been confused and is confusing. On the one hand, the United States appears to imply that multiple imported products can be bundled together in a single investigation so long as the domestic industry is defined as including producers that produce like products which are coextensive with the multiple imported products.637 On the other hand, the United States points out that "the USITC identifies the subject imports (or specific imported product) that corresponds or matches up to each of its like product definitions in order to conduct each of its analyses of whether increased imports of the product have caused serious injury to the domestic producers of the like product".638 Thus, the United States appears to concede that authorities cannot bundle multiple imported products and investigate whether imports of these multiple imported products have increased and caused serious injury to a domestic industry which consists of producers of multiple like products, and which have likewise been bundled together. According to Brazil, there must be discrete investigations and determinations regarding each individual imported product and the corresponding domestic like product.639 Since an imported product must correspond to a domestic product like the imported product, the like product criteria used to define domestic products must also be applied to the imported product.640


        13. According to New Zealand, the position of the United States is that it has to take the imports as they are presented to it in the petition. This, of course, ignores that in fact the USITC did group the imported products differently from the way they were set out in the petition. However, it also misconstrues the nature of the obligation under Article 2 of the Agreement on Safeguards and the need for some degree of specificity in the imported product for any coherent "like" product determination to be made.641 The term used is "product", not "products", so there is an implication at the outset of product specificity. That conclusion is reinforced both by the context and the object and purpose of Article 2 of the Agreement on Safeguards. While a safeguard measure can be imposed only where the domestic industry suffering serious injury produces a "like or directly competitive" product, equally the imported product on which the safeguard measure is imposed must be "like or directly competitive" with the domestic product. Only such products are competitive and only injury caused by an increase in competitive imports could be the subject of a safeguard measure.642 Although the United States seeks to deny that there is any requirement to demonstrate likeness between bundled products, the USITC did, in fact, engage in an attempt to show some degree of



          image

          635 Korea's written reply to Panel question No. 9 at the second substantive meeting.

          636 European Communities' second written submission, para. 99.

          637 See in this regard, the United States' arguments in para. 7.232.

          638 United States' written reply to Panel question No. 19 at the first substantive meeting .

          639 Brazil's second written submission, paras. 6-8.

          640 Brazil's second written submission, para. 9.

          641 New Zealand's second written submission, para. 3.49.

          642 New Zealand's second written submission, paras. 3.50 and 3.51.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 110


          likeness between them.643 The European Communities also argues that the Agreement on Safeguards does not allow determinations (and data collection) based on "types" or "categories" of products instead of a specific imported product. If the Agreement on Safeguards allowed, as argued by the United States, determinations based on "types of products" (between which the USITC does not see "clear dividing lines") without separating specific imported products that are not even like or directly competitive, the Agreement on Safeguards would also allow a determination on T-shirts and televisions. In conclusion, different products may be covered by one investigation procedure, as long as that investigation collects data and provides evidence relating to "such product" and the "domestic industry that produces like or directly competitive products", thereby enabling the investigating authority to make a correct determination.644


    2. Purpose of specific identification of the imported products


        1. Japan and Korea consider that imported products must not necessarily be narrower than like products.645 Rather, the boundary of the imported product should provide a reasonable basis for a meaningful like (or directly competitive, if applicable) product analysis/comparison with the domestic like product. In Brazil's view646, the grouping of products whether on the import side or the domestic side must also permit an analysis of the competitive dynamics in the market so that it can be determined whether imports are actually the cause of the industry's alleged injury. If these products are too broadly defined by bundling together products which are not like one another, the one-to-one competitive relationship between imports and domestic products cannot be established, and would result in a violation of the Agreement on Safeguards. Brazil submits that it is difficult to see how determinations of injury and causation could be appropriately made with respect to multiple imported products and multiple domestic industries producing products like those imported products.647


        2. Japan further submits that if it is ensured that the imported products and the domestic products have the proper competitive relationship, this boundary for the imported products is eventually narrowed down to meet the "like product" criteria, i.e., physical properties, end-use, consumer perception, and tariff classification. This way, the "like product" question is interrelated with the "a product" argument.648


        3. Norway argues that the product must be sufficiently defined in order to allow for an adequate evaluation of all the conditions specified in Article 2 of the Agreement on Safeguards for each and every product. The four criteria developed by the Appellate Body to establish likeness (physical characteristics, end-uses, consumer preferences, customs classification) may be taken as a point of departure for defining the imported product, but they must be applied even more narrowly than for the determination of what constitutes the "like product". Bundling of different products is not allowed, as it would undermine the criteria set out in Article 2.1. Norway also submits that the concept of "likeness" is irrelevant in this respect, as that is the second test to identify the domestic industry, not


          image

          643 New Zealand's second written submission, para. 3.55.

          644 European Communities' written reply to Panel question No. 10 at the second substantive meeting.

          645 Japan's written reply to the Panel question No. 33 at the first substantive meeting; Korea's written reply to Panel question No. 33 at the first substantive meeting.

          646 Brazil's written reply to Panel question No. 138 at the first substantive meeting. 647 Brazil's written reply to Panel question No. 137 at the first substantive meeting. 648 Japan's second written submission, para. 8.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 111


          the test to specify the imported product.649 Even if two selected distinct products are "like", it is not admissible to bundle them together for the purpose of defining a separate imported product.650


        4. Korea also submits that it is logical that the product must be sufficiently defined in order to allow for the required evaluation as to the conditions specified in Article 2.2 and elaborated in Article 4.2 of the Agreement on Safeguards. The "such" product and "like" product are obviously intrinsically linked in the Agreement on Safeguards. According to Korea, the analysis of the conditions specified in Article 2.1 and Article 4.2 must reveal the proper relationship between each product, imported and like, so that the analysis of serious injury and causation is meaningful and in compliance with the Agreement on Safeguards. If each imported and like product is distinct from the other products investigated, then an analysis which fails to take into account that distinction cannot be deemed sufficient.651 652


        5. Switzerland adds that besides not fulfilling the requirement of being "a product", bundles of different (imported) products can also not be compared to a "like or directly competitive" product, because products can only each be "like or directly competitive" with another product. If each of these different products is imported in such increased quantities that it causes or threatens to cause serious injury to the domestic industry, then an investigation is needed for each of the different products.653


    3. Grouping


        1. China asserts that the USITC is not allowed to group domestic like products and then use these groupings as imported products for the purpose of the determinations on increased imports. In doing so, the USITC uses the "bundling effects" to cover products for which the requirements of the Agreement on Safeguards are not met.654


        2. Brazil argues that an implication of grouping of products is that the imported product may include products which are neither like nor directly competitive with each other and that the effects of these imports may be measured on domestic industries producing products that are neither like nor directly competitive. In this situation, it is difficult to see how the United States could "ensure that the domestic industry is the appropriate industry in relation to the imported product" or avoid imposition of safeguard measures based on the "prejudicial effects that an imported product has on domestic producers of products that are not 'like or directly competitive products' in relation to the imported product".655


        3. The United States contends that, in the present case, the USITC's definitions of like products are coextensive with the subject imports. The like product and domestic industry definitions in this case correspond exactly to the imports subject to investigation. The USITC did not define the domestic "like products" to encompass more or different types of steel than the imported articles identified as subject to investigation. Moreover, the USITC considered the effects of only the subject imports (that corresponded to each domestic like product definition) on the domestic industry consisting of the producers of the corresponding domestic like product. The USITC's approach is


          image

          649 Norway's written reply to Panel question No. 22 at the first substantive meeting; Norway's first written submission, para. 176.

          650 Norway's written reply to Panel question No. 32 at the first substantive meeting.

          651 Appellate Body Report, US – Lamb, para. 90 (input and end products must be "like.").

          652 Korea's second written submission, para. 23.

          653 Switzerland's written reply to Panel question No. 12 at the second substantive meeting.

          654 China's second written submission, para. 40.

          655 Brazil's second written submission, para. 42.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 112


          clearly consistent with the Agreement on Safeguards and the Appellate Body's findings in US – Lamb.656 The United States also submits that the complainants fail to recognize that within almost every defined like product, and the corresponding imported product, there exist a range of goods of different sizes, grades, or stages of processing. While goods along the continuum share similar or like factors, individual items at the end of the continuum may not be as similar.657


        4. In Korea's view, the argument by the United States that the USITC merely matches up the imported product to the like product658 is nothing more than a post hoc rationalization that should be ignored by the Panel. Korea submits that the USITC record, however, contradicts even this new like product formulation: the USITC readily admits that the imported product and like product do not need to be "coextensive".659 The United States appears to advance its post hoc formulation of like product ("matching up") to permit a pretext for dismissing the relevance of US – Lamb660 (the imported product did not include lambs) and to distinguish the USITC's prior precedents in the anti-dumping and countervailing duties context.661 662


        5. Japan considers that even when two products are "coextensive", such "coextensiveness" alone does not assure the required relationship, as it still requires that the sub-component products within a grouping be like the other sub-components within that grouping in order to secure the likeness relationship between the imported and the domestic like products in their entirety.663 Bundling of unlike products, as was done by the United States for CCFRS grouping, is absurd because it results in comparing imports with unlike domestic products, and in allowing the wrong industry to benefit from imposed relief.664 Such comparisons violate Article 2.1, which requires that increased imports be found to cause serious injury, or threat thereof, "to the domestic industry that produces like or directly competitive products". According to Japan, by relying on the "coextensive" excuse, the United States has masked the true competitive dynamics between the imported and domestic products by bundling unlike products into a single "like product". Such bundling should also be rejected, particularly in light of the specific warnings of the Appellate Body in US – Lamb and US – Cotton Yarn not to benefit the wrong industry in safeguard cases.665 666


        6. Brazil further submits that the notion that competent authorities could group several domestic like products together that do not compete with each other directly and are not like each other, so long as they are coextensive with the imported products, would leave the issue of like product wide open. It would mean that the only parameter in defining the domestic product like or directly competitive with the imported product is that the domestic products must be coextensive with the imported products. Thus, the interpretation being urged upon the panel by the United States could yield absurd results such as conducting an investigation of imports of cotton shirts and televisions together, with


          image

          656 United States' first written submission, paras. 65, 98.

          657 United States' second written submission, para. 86.

          658 United States' first written submission, para. 65; United States' first oral statement, para. 16.

          659 Korea's second written submission, paras. 26-27.

          660 Appellate Body Report, US – Lamb, para. 98.

          661 Appellate Body Report, US – Lamb, paras. 105-106.

          662 Korea's second written submission, para. 32.

          663 Japan's written reply to Panel question No. 9 at the second substantive meeting.

          664 Japan does not believe it matters when the imported product is defined, as long as it is appropriately defined in accordance with the appropriate like product grouping.

          665 Appellate Body Report, US – Lamb, para. 86; Appellate Body Report, US – Cotton Yarn, para. 95.

          666 Japan's written reply to Panel question No. 10 at the second substantive meeting.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 113


          the only constraint being that the domestic industries must be defined as those producing cotton shirts and televisions.667


        7. China agrees that a single determination related to increased imports and serious injury for both cotton shirts and televisions would be absurd. While several products can be subject to one investigation, the specific imported products that will be subject to the determinations required by Article 2.1 of the Agreement on Safeguards should always be identified. The absurdity would be to have a single determination covering products that are not "specific", since this would fall short of the requirement of the Agreement on Safeguards to first identify a specific imported product. This would be the case if a single determination would cover both cotton shirts and televisions.668


        8. In Brazil's view, there must be separate investigations and determinations of increased imports, serious injury and causation of each individual imported product and the domestic industry producing the corresponding domestic like product. Competent authorities cannot collapse or bundle multiple imported products and corresponding multiple like products into the same investigations and determinations.669 Brazil also submits that the Appellate Body in both US – Lamb and US – Cotton Yarn refers to "the imported product" and not to "imported products", implying a single identifiable imported product. A single identifiable imported product must correspond with a single identifiable like product not multiple like products.670

        9. Brazil671 and New Zealand agree that the term used is "product", not the plural "products", so there is an implication at the outset of product specificity. It is difficult to see how one can read this language as permitting inclusion in a single investigation multiple imported products which correspond to multiple domestically produced like or directly competitive products. When there is bundling of different products into an imported product category, the test for determining whether that is an appropriate "product" for the purposes of the Agreement on Safeguards must be likeness. That is to say, if the products within that imported product category are not "like" each other, then there has not been a proper identification of the imported product. Only like products are competitive and since "likeness" is the basis for determining the domestic product and for determining the product on which a safeguard measure can be imposed, likeness must be the criterion for determining whether the imported good is "a product" or a set of separate products.672 Norway agrees that Article 2.1 of the Agreement on Safeguards implies that an investigation is directed towards "a" (one) product and that the grouping of products is not permitted.673 With two different products there would have to be two investigations – which can of course be carried out in parallel.674


    4. Parameters for identifying the imported product


    1. Likeness


        1. New Zealand submits that since "likeness" is the basis for determining the domestic product and for determining the product on which a safeguard measure can be imposed, then likeness must be the criterion for determining whether the imported good is a "product" or is a set of separate or different products. Thus, when there is bundling of different products into an imported product


          image

          667 Brazil's second written submission, para. 43.

          668 China's written reply to Panel question No. 12 at the first substantive meeting. 669 Brazil's written reply to Panel question No. 10 at the second substantive meeting. 670 Brazil's second written submission, para. 44.

          671 Brazil's written reply to Panel question No. 19 at the first substantive meeting.

          672 New Zealand's written reply to Panel question No. 19 at the first substantive meeting.

          673 Norway's first written submission, para. 179.

          674 Norway's written reply to Panel question No. 10 at the second substantive meeting.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 114


          category, the test for determining whether that is an appropriate "product" for the purposes of the Agreement on Safeguards must be likeness.675 Norway agrees that it will be useful to apply the four criteria for "likeness" as a point of departure, but argues that the criteria be applied even more narrowly than for the determination of what constitutes the "like product".676


        2. In contrast, the European Communities argues that the question of the "likeness" of imported products amongst themselves is not relevant. Article 2.1 of the Agreement on Safeguards employs the term "like product" only for defining the domestic industry, a distinct step which takes place after establishing that there are increased imports. However, it can safely be said that bundling together imported products which are not even "like" one another certainly does not satisfy the requirement of increased imports of specific products.677 The European Communities adds that the Appellate Body clarified in US – Cotton Yarn678 that there must be a matching between "such product" and the domestic products so as to establish that they are like or directly competitive. Such matching exercise cannot, however, be undertaken, if imported (and domestic) products may be "bundled" in a way that the components of the imported product bundle are not even like or directly competitive with all the components of the domestic product bundle.679 China further argues that it is not permissible to justify a safeguard measure on one specific imported product based on a finding of "increased imports" of a different specific product, even if "like or directly competitive".680


    2. Tariff lines


        1. The European Communities and Switzerland consider that the primary basis for identifying the specific imported product should be tariff codes. According to them, tariff classification is a conventional and generally accepted way of classifying and identifying products. Article XIX of the GATT 1994 establishes a clear link between imported products and the corresponding tariff concessions, which are made for single specific products.681


        2. In the United States' view, each tariff line generally does not correspond to a distinct or specific product. The process of defining the domestic like product and the corresponding specific imported product is based on the facts in each particular case and customs treatment is just one out of a number of criteria.682 Moreover, allegations that tariff classifications should define specific products begs the question of the appropriate level of tariff classification (i.e., four-digit level, six-digit level, eight-digit level, or ten-digit level), not all of which are harmonized among countries.683 684


        3. In response, the European Communities notes that the Harmonized Commodity Description and Coding System (HS) provides an international product nomenclature at the six-digit level. The HS is commonly used by WTO members to negotiate schedules and concessions and also referred to


          image

          675 New Zealand's second written submission, para. 3.52.

          676 Norway's second written submission, para. 55.

          677 European Communities' written reply to Panel question No. 21 at the first substantive meeting.

          678 Appellate Body Report, US – Cotton Yarn, paras. 97 and 98.

          679 European Communities' second written submission, para. 129-131.

          680 China's first written submission, para. 158.

          681 European Communities' written reply to Panel question Nos. 19-20 at the first substantive meeting.

          682 United States' first written submission, paras. 86-89.

          683 Moreover, a number of prior safeguard disputes have involved single like products covering multiple tariff classifications. US – Line Pipe; Korea – Dairy, Notification pursuant to Article 12.1(c),

          G/SG/N/10/KOR/, dated 27 January 1997; Argentina – Footwear (EC), Notification pursuant to Article 12.1(c) and Article 9, G/SG/N/10/ARG/1/Suppl.3, G/SG/N/11/ARG/1/Suppl.3, dated 17 May 1999.

          684 United States' written reply to Panel question No. 20 at the first substantive meeting.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 115


          in the analogous provisions of the Agreement on Textiles and Clothing. It should therefore be the starting point for the definition of "a product".685


        4. The United States responds that, like many of the complainants, it takes the view that while consideration of customs treatment/tariff classification may be a relevant factor in an analysis of whether there are clear dividing lines between products, depending on the facts of a particular case, it is still just one of a number of criteria and not alone dispositive.686 The Appellate Body in Japan – Alcoholic Beverages II, albeit interpreting a different agreement with a different object and purpose, reached the same conclusion and considered that tariff classifications of products could be relevant as one of a series of factors in determining what are "like products", but not as the primary or decisive factor.687 Moreover, the United State contends, it is clear that identification by tariff lines has not been the decisive factor in other safeguard actions, which have involved single like products covering multiple tariff classifications.688 689 In fact, in its recent safeguard action on steel, the European Communities also included numerous tariff classifications in each of its single imported products that correspond to its various like or directly competitive products.690 According to the United States, the European Communities' reference to tariff concessions as the basis for using tariff lines to identify products also fails to recognize that tariff concessions may include a wide range of products. The Appellate Body in Japan – Alcoholic Beverages II, interpreting a different agreement, warned that while precise tariff bindings "can provide significant guidance as to the identification of 'like products' … these determinations need to be made on a case-by case basis. … [since] tariff bindings that include a wide range of products are not a reliable criterion for determining or confirming product 'likeness.'"691 692 The United States also argues that the European Communities' rationale for defining imported products by tariff lines first is intended to require authorities to consider whether such imports have increased, as a "filter", prior to conducting the like product analysis. This methodology is neither required by the Agreement on Safeguards nor apparently followed by the European Communities in its own safeguard actions.693 The European Communities' proposal would appear to place the cart before the horse in that it focuses on what imports are increasing regardless of whether there is serious injury to an industry and before the composition of the relevant domestic industry has been defined. According to the United States, the European Communities would look at increased


          image

          685 European Communities' second written submission, paras. 123-124.

          686 Brazil's written reply to Panel question No. 20 at the first substantive meeting; Japan's written reply to Panel questions Nos. 20 and 31 at the first substantive meeting; Korea's written reply to Panel questions Nos.

          20 and 31 at the first substantive meeting; New Zealand's reply to Panel question No. 20 at the first substantive meeting. United States' first written submission, paras. 86-89.

          687 Appellate Body Report, Japan – Alcoholic Beverages II, pp. 21-22; Panel Report, Japan – Alcoholic Beverages I, para. 5.6.

          688 US – Line Pipe; Korea – Dairy, Notification pursuant to Article 12.1(c), G/SG/N/10/KOR/, dated

          27 January 1997; Argentina – Footwear (EC), Notification pursuant to Article 12.1(c) and Article 9, G/SG/N/10/ARG/1/Suppl.3, G/SG/N/11/ARG/1/Suppl.3, dated 17 May 1999.

          689 Allegations that tariff classifications should define specific products begs the question of the appropriate level of tariff classification (i.e., four-digit level, six-digit level, eight-digit level, or ten-digit level), not all of which are harmonized among countries.

          690 European Communities – Imposing Definitive Safeguard Measures Against Imports of Certain Steel

          Products, Commission Regulation No. 1694/2002/EC of 27 September 2002, paras. 10-15 (Exhibit US-84). (For example, the European Communities defined hot-rolled coils as a single "product concerned", or imported product, consisting of numerous (11) tariff classifications and found that it corresponded to a single like or directly competitive product. The European Communities made similar findings regarding other products in this action.).

          691 Appellate Body Report, Japan – Alcoholic Beverages II, p. 22.

          692 United States' second written submission, paras. 54-56.

          693 See European Communities – Imposing Definitive Safeguard Measures Against Imports of Certain Steel Products, Commission Regulation No. 1694/2002/EC of 27 September 2002 (Exhibit US-85).

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 116


          imports by tariff lines, without regard to whether tariff lines should be grouped together, as most parties agree is appropriate. The approach advanced by the European Communities would result in a scrutiny of imports to determine which imports are showing increases in volume only then to be followed by an investigation to determine the domestic like product corresponding to an import tariff line or lines and then if there was an industry somewhere that was being injured as a result of such increased imports. Therefore, looking at increases in imports by tariff lines, as proposed by the European Communities, before defining the domestic like product prevents any consideration of increases in imports relative to domestic production, since the industry would not yet been have defined, which the Agreement on Safeguards requires.694


        5. The European Communities reiterates that tariff classifications up to the six-digit level form an important starting point in determining what a product is.695 Internationally agreed customs classifications are relevant for determining the "like domestic product" and reflect the "physical properties" of products. Against that background, it is all the more incomprehensible why the United States continues to defend its untenable position that the term "product" in Article 2.1 of the Agreement on Safeguards is so flexible that it permits a competent authority to bundle an array of different individual products. According to the European Communities, although the term "product" may differ between different legal provisions, where safeguard measures that suspend tariff concessions are at stake, as in this case, tariff concessions form an important context. Although the United States agrees that tariff classifications are relevant, the USITC did not consider them in this determination. According to the European Communities, however, the key point is that the term "product" does not depend on the industry, but rather, as the Appellate Body has clarified, the definition of the domestic industry follows the definition of the imported product and the like or directly competitive domestic products.696 697


        6. Japan, Korea, Norway and Brazil argue that tariff lines are helpful, but not determinative, given that there is sometimes an overlap, particularly in terms of customer perceptions and end-uses, regarding products in different tariff lines.698 Other considerations, such as end-use, consumer perceptions, and physical properties, may be (and in this case are) more important than distinctions in tariff classifications.699 New Zealand submits that the practice of the USITC itself confirms that in reality, tariff classification is an important aid to product differentiation.700 Norway argues that the six-digit level that is internationally harmonised should be seen as the outer perimeter of the internationally agreed definition of specific product. While you will normally not have a situation where a distinct or specific product is divided into several tariff lines, you may have several products under the same tariff line.701


          image

          694 United States' second written submission, para. 57.

          695 European Communities' written reply to Panel question No. 19 at the first substantive meeting; European Communities' second written submission, paras. 119-127.

          696 Appellate Body Report, US – Lamb, para. 87.

          697 European Communities' written reply to Panel question No. 9 at the second substantive meeting.

          698 Japan's written reply to Panel question Nos. 20 and 31 at the first substantive meeting; Korea's written reply to Panel question No. 20 at the first substantive meeting.

          699 Japan's written reply to Panel question No. 31 at the first substantive meeting; Brazil's written reply to Panel question No. 19 at the first substantive meeting; Norway's written reply to Panel question No. 22 at the

          first substantive meeting.

          700 New Zealand's written reply to Panel question No. 20 at the first substantive meeting.

          701 Norway's second written submission, paras. 60-61.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 117


    3. Consensus on parameters?


      1. The United States asserts that the complainants have suggested a variety of definitions for "such imports" (tariff lines, product exclusion requests, predetermined definitions that are not universally accepted) with no clearly suggested criteria, other than like product criteria, to actually consider the evidence collected and the facts of the case in making such a finding. The complainants' own arguments, therefore, show that no such consensus on steel product definitions exists. The like product criteria (i.e., physical properties, uses, consumer tastes, and tariff classifications) suggested by the complainants do not include a market parameter or competition criteria.702


      2. The United States responds that the complainants' varied and inconsistent arguments regarding the appropriate definitions of like product demonstrate that no such universal definitions of steel products exist.703 Moreover, the complainants, submits the United States, are urging the Panel to identify a requirement for analysis in all cases, but no-one contends that such alleged universal definitions exist and should control in all cases. The United States also rejects the proposition that the complainants do not have to agree on what definition would be appropriate, but rather it is enough to show that "what the US did was too broad".704 This begs the question of how the complainants know that the USITC's definitions were too broad if complainants cannot reach a consensus on any alternative definition.705


      3. The European Communities responds that the United States exaggerates differences in the arguments of the complainants where there are none.706 The European Communities and Switzerland also argue that the allegation of the absence of universal product definitions is not true. An internationally agreed definition of products is provided under the Harmonized Commodity Description and Coding System, (the "HS"), developed by the World Customs Organization (WCO) and governed by "The International Convention on the Harmonized Commodity Description and Coding System".707 It provides an international product nomenclature at the six-digit level. All contracting parties, including the United States, undertake that "[their] tariffs and statistical nomenclatures shall be in conformity with the HS".708 The HS is commonly used by WTO Members to negotiate their schedules of concessions, and safeguards measures are a safety valve against unforeseen effects of such tariff concessions.709 Norway adds that the product definitions provided for by the HS, therefore, provide an internationally agreed basis for determining what is a specific



        image

        702 United States' first written submission, para. 103; United States' written reply to Panel question No.

        141 at the first substantive meeting.

        703 The complainants' proposals for appropriate like product definitions ranges from product definitions used in trade remedy cases under other statutes, to tariff classifications (612 classifications in all), to product

        descriptions contained in requests for product exclusions.

        704 Japan's first oral statement (like product), para. 23. Japan stated in relevant part:


        "First of all, it should be noted that none of the complainants intends to suggest that there is only one possible definition of the "like product" for the products at issue in this case. . . . The Panel need not decide which of the breakdowns presented in the complainants' submissions is most appropriate; it merely needs to find that what the US did was too broad, which it clearly was."


        705 United States' second written submission, paras. 52-53.

        706 European Communities' second written submission, para. 92.

        707 The International Convention on the Harmonized Commodity Description and Coding System,

        Brussels, 14 June 1983.

        708 Ibid., Article 3.1(a).

        709 European Communities' second written submission, para. 120; Switzerland's second written submission, para. 40; Norway's second written submission, para. 59.

        WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

        Page 118


        imported product. The six-digit level that is internationally harmonized should therefore be seen as the outer perimeter of the internationally agreed definition of specific product.710


      4. China submits also that it is not arguing that there should be universally accepted definitions of steel products, although such a classification may exist. China rather challenges the methodology used by the competent authority to define the product categories that are the subject of the investigation.711


  2. Methodology used by the USITC in determining the "imported product"


      1. The European Communities, China, Switzerland and Norway submit that the safeguard measures adopted by the United States are already fundamentally flawed because of the arbitrary and unjustified manner in which the USITC considered itself entitled to group products together for the purposes of the investigation and the determination712, contrary to the Agreement on Safeguards.


        1. Identification of the imported product


      2. The European Communities, China713 and Norway point out that the USITC completely omitted the first and fundamental step of identifying the "specific imported product". Instead of making its own determination of the specific imported product, the USITC blindly accepted the arbitrary description of "the imported product or products included within the investigation that [was] set forth in the request or petition".714 The USITC explicitly rejected identifying specific imported products it investigated when it stated in its general methodology that "[t]he Commission … is not required to consider, in the first instance, whether and how to subdivide the imported article or articles".715 Korea and Norway also argue that the USITC in the instant investigation failed to carry out an analysis of the imported product, and used as its "starting point" the four general product categories proposed by the President.716


        1. Grouping


      3. The European Communities and Switzerland recall that the President's request grouped "a wide array of steel products into four general categories: (1) CCFRS products, (2) certain carbon and alloy long products, (3) certain carbon and alloy pipe and tube products, and (4) certain stainless steel and alloy tool steel products". The USITC then established "33 product categories" for the collection of data.717 Again, a number of products were grouped together, and each specific imported product was not considered separately for the purpose of showing increase in imports and causality.718


      4. Brazil submits that the starting point was a broadly defined domestic steel industry making products ranging from carbon steel slab and hot-rolled flat products, to long products, stainless products, and tubular products. In effect, the entire steel making sector of the United States' economy


        image

        710 Norway's second written submission, para. 60.

        711 China's second written submission, para. 38.

        712 European Communities' first written submission, para. 206; China's first written submission, para. 154; Norway's first written submission, para. 202; Switzerland's first written submission, para. 192.

        713 China's second written submission, para. 33.

        714 USITC Report, Vol. I, p. 31. (Exhibit CC-6).

        715 European Communities' written reply to Panel question No. 145 at the first substantive meeting; Norway's first written submission, para. 204.

        716 Korea's first written submission, para. 22; Norway's first written submission, para. 204.

        717 USITC Report, Vol. I, p. 32.

        718 Switzerland's first written submission, paras. 195-196.

        WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

        Page 119


        was perceived as being in crisis and the sector convinced the President that imports were to blame. Thus, the starting point was not individual products and domestic facilities producing those products but, rather, the entire steel sector making virtually the entire range of steel products. While the President provided the USITC with an exhaustive list of products based on harmonized tariff categories and various product descriptions, what the USITC really received was a request to figure out how imports were injuring the steel sector and what to do about it. Thus, the starting point was not individual products, but an entire sector of the United States' economy and how to protect that entire sector from import competition.719 720


      5. Similarly, the European Communities argues that the starting point was not an individual product but a broadly defined domestic steel industry making products described by 612 tariff headings. The USTR request of 22 June 2001 was based on the perception that the entire steel sector making virtually the entire range of steel products was in crisis.721 The request responded to immense pressure by the Senate to impose steel safeguards to protect the domestic steel industry by referring to "national security" considerations.722 The European Communities submits that, given the assumption at the start – that the entire steel sector was being injured by unspecified imported products – the approach of the USITC, by its own admission, was to protect as broad an array of productive facilities as possible.723 According to the European Communities, this backward approach of first identifying a domestic industry that is allegedly being injured and then bundling a broad array of "certain imported steel products" is contrary to the Agreement on Safeguards. The United States agreed under the Agreement on Safeguards to a safeguard determination where the definition of the domestic industry depends on the proper identification of such imported products and like or directly competitive products (and not the other way round). This was also clarified by the Appellate Body in US – Lamb.724 The European Communities also clarifies that it is conceivable that a request (or petition from the industry) covers a broad array of products or even a whole industry. However, such request does not relieve a competent authority from discharging its responsibility to identify specific imported products on the basis of which it can make its determinations.725


      6. The United States stresses that in defining the domestic like product, the USITC started with the imported article (or articles) that had already been identified in the request or petition for an investigation ("subject imports") and examined the evidence in order to determine the domestic product(s) that are like the subject imported product(s).726 This request or petition identified the universe of imports subject to investigation. Second, the USITC considered whether there were domestic products that are like the subject imports. Third, the USITC applied its long-established factors to the domestic products corresponding to the subject imports to consider whether there were


        image

        719 Brazil submits that in the Bush Administration's request to the USITC to conduct a safeguards investigation, although the investigation covered "certain" steel products, the clear intent was to investigate "the effect of imports on the US steel industry", and not merely some component of that industry. Letter from Ambassador Robert Zoellick to USITC Chairman Stephen Koplan, 22 June 2001. (Common Exhibit CC-1 to Brazil's first written submission).

        720 Brazil's written reply to Panel question No. 17 at the second substantive meeting.

        721 USTR request to the USITC to initiate a safeguard investigation under Section 201 of the Trade Act of 1974, Exhibit CC-1, paras. 1-3.

        722 Press Release by Senator Rockefeller of 22 May 2001, available at http://www.senate.gov/~rockefeller/2001/pr052201.html (visited 5 January 2003), Exhibit CC-110. Resolution of the Committee on Finance of the United States Senate, transmitted to the USITC by letter of 26 July 2001, paras. 2, 3 and 7, Exhibit CC-111.

        723 USITC Report, Vol. I, p. 30.

        724 Appellate Body Report, US – Lamb, para. 87.

        725 European Communities' written reply to Panel question No. 17 at the second substantive meeting.

        726 United States' first written submission, para. 95.

        WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

        Page 120


        clear dividing lines among domestic products in order to define like products.727 The USITC traditionally takes into account such factors as the physical properties of the product, its customs treatment, its manufacturing process (i.e, where and how it is made), its uses, and the marketing channels through which the product is sold, and any other relevant factors.728


      7. The United States explains more particularly that the USITC begins with the universe of imports subject to investigation, as identified in the request or petition. After determining what domestic products are like or directly competitive with the subject imports, the USITC considers whether the domestic products corresponding to the subject imports consist of a single domestic like product or whether there are clear dividing lines among the products so as to constitute multiple domestic like products. The USITC applies its like product criteria, including production or manufacturing processes, in its analysis of whether there are such clear dividing lines as to constitute multiple like products, or that there are no clear lines and that a single like product definition is appropriate.729


      8. China responds that the USITC did more than defining or identifying the domestic product or products like or directly competitive with the imported article or articles in the petition or request. It grouped these products together and used these groupings as categories of imported products. Therefore, according to China, for these categories, none of the steps required by the Agreement on Safeguards were followed.730 In China's view, the investigating authority was not allowed to turn from the products identified as being subject to the investigation (which might eventually have been considered as "specific" enough) into broader product categories without any further justification, and in particular when these products were not "like". In this context, it is clear that the bundle under investigation cannot be considered as a "specific product". China is of the opinion that, in doing so, the USITC is, to a large extent, reshaping the range of imported products identified as subject to the investigation. Therefore, the investigation was no longer conducted on specific products, but on a group of products that are unrelated since they are not even "like".731 The USITC conducted a "likeness" analysis between these domestic products and grouped them into bundles. Therefore, the USITC did not conduct an increased imports analysis on some of those products, while in the end, they are subject to safeguard measures. This methodology is contrary to the Agreement on Safeguards.732


      9. The European Communities, China, Switzerland and Norway argue that the product "categories" created by the United States' authorities for the purposes of the safeguard investigation and imposition of measures are not justified by definitions that can allow them to be considered specific products.733 Where the USITC did consider arguments to the effect that its products groupings were incorrect734, it rejected these arguments with considerations relating not to the specificity of the products but, rather, to production processes, vertical integration and coincidence of economic interests. However, the very fact that the USITC itself needed to rely on 33 different product categories for the purpose of data collection, demonstrates the artificial nature of the broader product categories created for the purpose of the determination.735 The European Communities also


        image

        727 United States' first written submission, paras. 83-93.

        728 United States' written reply to Panel question No. 19 at the first substantive meeting, paras. 46-50.

        729 United States' second oral statement, paras. 18 and 28-31.

        730 China's second written submission, para. 50.

        731 China's second written submission, paras. 55-56.

        732 China's second written submission, para. 57.

        733 China's first written submission, para. 162.

        734 USITC Report, Vol. I, p. 37, where the Commission considered whether to "analyse certain carbon flat-rolled steel separately or as a whole".

        735 Switzerland's first written submission, paras. 197-198.

        WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

        Page 121


        argues that the USITC confirmed that the imported product must be identified for the purpose of the determination because it collected data on the basis of 33 product groups.736 In this way, a lack of increased imports for one specific product could be, and indeed was, masked by combining it with other products.737


      10. Switzerland contends that the Agreement on Safeguards and the Appellate Body require that a safeguard measure be imposed on a specific product ("such product"). Therefore, each specific product has to be determined at the outset of an investigation. Because the USITC failed to do so, it did not fulfil this first requirement of the Agreement on Safeguards.738 The European Communities argues that the arbitrary and artificial product definitions used by the USITC in the present case vitiates the whole basis of the USITC Report.739 Switzerland further considers that, having failed to establish each relevant specific product that is being imported, the United States cannot determine the domestic industry that produces the like or directly competitive product, thereby violating Article 2.1 of the Agreement on Safeguards.740


      11. The United States points out that if there are multiple domestic like products corresponding to the subject imports, the USITC identifies the subject imports (or specific imported product) that corresponds or matches up to each of its like product definitions in order to conduct each of its analyses of increased imports, serious injury and causation.741 The USITC defined 27 separate like products that corresponded to all the subject imports. Ten of these definitions corresponded to subject imports on which remedies were imposed and are subject to review by this Panel.742


      12. The European Communities responds that the United States offers two contradictory responses to what constitutes the "imported products" or the "subject imports". First, the United States frankly admits that the USITC has not identified the specific imported product concerned and denies such legal obligation. The United States subsequently attempts to sell the USITC's domestic industry definition as identification of the specific imported product.743 The European Communities also notes that the term "matching" is inaccurate in this context. What the USITC has done to identify the imported products is rather the "cloning" of the domestic industry definition and to re-label it as "imported product". The approach of the United States puts the cart before the horse, contrary to the logical continuum of Article 2.1 of the Agreement on Safeguards.744 According to the European Communities, the United States itself admitted that in "cloning" the domestic like product bundles and calling them imported product, it had not established that all components of the domestic product bundle are like all components of the imported product bundle.745 Specifically, the European Communities asserts that the United States admitted that, for example, imported slab and domestic corrosion-resistant (i.e.) coated steel are both components of a single like product, CCFRS.746 The


        image

        736 European Communities' second written submission, para. 100.

        737 China's first written submission, paras. 164, 169; European Communities' first written submission, para. 216; Norway's first written submission, paras. 205-210.

        738 Switzerland's second written submission, para. 49.

        739 European Communities' first written submission, paras. 218, 259.

        740 Switzerland's first written submission, para. 199.

        741 United States' written reply to Panel question No. 21 at the first substantive meeting, para. 40.

        742 United States' first written submission, para. 114.

        743 European Communities' second written submission, para. 92. 744 European Communities' second written submission, para.106. 745 United States' written reply to Japan's question No. 1, para. 19. 746 United States' written reply to Japan's question No. 1, para 19.

        WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

        Page 122


        United States further conceded that "individual items at the respective ends of the continuum may be less similar to each other than those in the middle of the continuum".747 748


  3. Measure-specific argumentation


  1. CCFRS


    1. Grouping


        1. Brazil submits that the United States investigated whether imports of CCFRS had increased and were causing serious injury to the domestic industry producing CCFRS. According to Brazil, the United States' determinations of increased imports, the existence of serious injury and the existence of a genuine and substantial causal link between the serious injury and increased imports were based on a single CCFRS imported product and a single domestic industry producing CCFRS.749 However, CCFRS constitutes multiple imported products produced by industries in the United States producing separate and distinguishable multiple like products.750 The European Communities and China argue that by artificially aggregating slabs, plate, hot-rolled, cold-rolled and coated products as CCFRS, the USITC attempted to disguise a sharp dive in imports of plate throughout the entire investigation period.751


        2. Japan argues that, in the case of CCFRS products, the United States has essentially decided that slab is like corrosion resistant steel and has thus created an inappropriate, overly broad grouping.752 According to Japan, the United States obviously failed to meet the like product requirement when it conjoined, among others, slab with corrosion resistant steel. Japan argues that these products are simply not like one another; nor are they like the plate, hot-rolled, and cold rolled steel with which they were conjoined. Customer perceptions, for instance, show a clear dividing line between hot, cold, and coated, rather than between tariff lines within each of these three products. This is demonstrated by the industry literature that reports the broken down prices of these products.753 Indeed, Japan argues, the USITC itself collected data based on these delineations.754 755


        3. Japan adds that the overarching point of the complainants concerning the USITC's CCFRS grouping, is that the grouping does not represent a set of products that can be logically compared with one another. The impact of imports on the domestic CCFRS industry cannot be assessed because the various parts that make up those imports are not like one another and do not compete with one another. The effect of grouping them together is to do what the Appellate Body in US – Lamb said


          image

          747 United States' written reply to Japan's question No. 1, para. 19. 748 European Communities' second written submission, para. 132. 749 Brazil's second written submission, para. 11.

          750 Brazil's second written submission, para. 12.

          751 China's first written submission, para. 169; European Communities' first written submission, para. 216.

          752 Japan's written reply to Panel question No. 33 at the first substantive meeting.

          753 PURCHASING MAGAZINE (10 Oct. 2002) at 9; METAL BULLETIN (7 Oct. 2002) (METAL BULLETIN'S

          image

          appraisal of prices for US steel); and American Iron and Steel Institute, Shipments of Steel Mill Products (Carbon) (Aug. 2002); Nucor Pricing Sheets (21 Oct. 1999) (Exhibit JPN-1).

          754 USITC Report; see also questionnaires issued by USITC in this case, available at http://info.usitc.gov (excerpted sample provided in Exhibit JPN-2).

          755 Japan's written reply to Panel question No. 20 at the first substantive meeting.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 123


          should not be done – that is, to impose a safeguard measure that results in protecting an industry that does not make like or directly competitive products.756


        4. Korea submits that imports of CCFRS, for instance, are not a particular product but rather several products. They share only a basic physical similarity due to commonalties in the production process – they are flat-rolled, made of steel and often produced in integrated facilities. By treating all CCFRS imports as a single like product, the USITC determined that all CCFRS domestic products are similar in terms of physical properties, uses, production processes, and marketing channels for the reason that each "type" (i.e., like product) of domestic CCFRS overlapped with the same "type" of imported product. Such circular reasoning results in broadening the like product simply to reflect the scope of imports identified in the petition, whether or not they are "like", in a manner not consistent with the Agreement on Safeguards.757


        5. Brazil also considers that when the United States grouped, among others, slab with corrosion resistant steel, it completely dismissed any sense of a dividing line based on physical properties, end- uses, consumer perceptions or tariff classification. Indeed, the United States appears to have ignored its own criteria. Based on customer perceptions, for instance, a clear dividing line exists at least between hot, cold, and coated flat products, rather than between tariff lines within each of these three products. For example, the industry literature reports prices broken down by these products.758 Indeed, Brazil argues, the USITC itself collected data based on these delineations.759 While the tariff schedules tend to group these products together (although with multiple tariff lines), the industry categorization for commercial purposes would seem to be a more convincing criterion.760 The impact of imports on the domestic CCFRS industry cannot be assessed because the various products that make up those imports are not like one another and do not compete directly with one another. For example, slab is imported exclusively by United States' steel mills producing the downstream finished CCFRS rolled products. Presumably, these imports benefit these mills and have an entirely different effect on the domestic industry than, for example, imports of hot-rolled flat products.761


        6. The European Communities states that the USITC Report does not provide data on the artificially aggregated product CCFRS. There are no tables on increased imports, domestic industry capacity and production and domestic industry financial performance for this product group, but only tables for individual products (and one covering all seven products in the larger grouping CCFRS).762 New Zealand adds that several disparate imported products were grouped together and termed "the imported product" CCFRS, even though the USITC had itself determined there to be five separate products for certain data collation and analysis purposes.763


          image

          756 Japan's written reply to Panel question No. 137 at the first substantive meeting.

          757 Korea's first written submission, paras. 23-24.

          758 Purchasing Magazine transaction prices (Exhibit CC-65); or US Census data on steel imports, grouped by product category (an example of this data is located at http://www.census.gov/foreign-trade/Press- Release/2002pr/09/steel/steel1cp.pdf.)(Brazil notes that the US census data groupings are the same used by the American Iron and Steel Institute to track imports and domestic shipments).

          759 Certain Steel Products, Inv. No. TA-201-73, USITC Pub. 3479 (December 2001) (hereinafter "ITC

          Report"), Purchasers' Questionnaire, p. 3 (available at http://info.usitc.gov/OINV/INVEST/OINVINVEST. NSF/0a915ada53e192cd8525661a0073de7d/f26c82f49f2bd14685256a84004ee7d1/$FILE/Purchaser- carbon+flat.PDF.).

          760 Brazil's written reply to Panel question Nos.19, 20 at the first substantive meeting.

          761 Brazil's written reply to Panel question No. 137 at the first substantive meeting.

          762 European Communities' first written submission, para. 215.

          763 New Zealand's written reply to Panel question No. 19 at the first substantive meeting.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 124


        7. The United States contends that contrary to the European Communities' allegations764, the USITC clearly considered data for the domestic industry defined as the producers of CCFRS. The USITC repeatedly referred to tables, such as FLAT-ALT-7, in its opinion in the USITC Report. Many of these tables were not included in the published USITC Report, but were released later, although it is apparent from the references in the USITC Report that they were considered.765


    2. Market and price for CCFRS


        1. New Zealand submits that the USITC identified the subject imports as those that corresponded to domestic products that were, themselves, identified by reference to the industries that produced them.766 Instead of determining the domestic industry by reference to the producers of the product that is "like" the imported product, the USITC determined the imported product by reference to its perception of products that are produced by the United States domestic industry.767 The array of different products included within CCFRS serves to underline that market parameters were not used to distinguish between different products. As a consequence, CCFRS does not represent "one sole authentic market".768


        2. Brazil argues that there is no such thing as a market for CCFRS. In contrast, there are clearly distinct markets for slab, plate, hot-rolled, cold-rolled and corrosion resistant steel. These are categories commonly used within the industry.769 The industry itself does not recognize the existence of a category of CCFRS. While there may occasionally be discussion of "sheet" products or "flat products", these never include slab. Furthermore, the industry distinguishes between plate, hot-rolled, cold rolled and corrosion resistant steel for pricing purposes, with prices of all of these products derived from a base price for each of the individual finished flat products (e.g., a separate price list with a distinct base price and extras applicable for each of these four products).770 Finally, the industry trade organizations generally use distinct hot-rolled, cold-rolled, corrosion resistant and plate categories when reporting on production, shipments, and imports, which indicates that the industry itself recognizes a clear division between these products.771 Brazil believes that the fact that the industry itself recognizes separate and distinct categories of a product constitutes a prima facie case that CCFRS does not define either a product or a like product.772

        3. Japan and Korea point out that there is no such thing as a price for CCFRS.773 Rather, there are distinct prices for each of the distinct CCFRS products subject to investigation. In fact, the USITC recognized this because the USITC did not ask for pricing data for CCFRS but rather



          image

          764 European Communities first written submission, para 215.

          765 United States' first written submission, para. 141.

          766 United States' written reply to Panel questions for the Parties, 12 November 2002, para 267.

          767 New Zealand's second written submission, para. 3.31.

          768 New Zealand's written reply to Panel question No. 141 at the first substantive meeting.

          769 Brazil's written reply to Panel question No. 141 at the first substantive meeting.

          770 The Nucor price list attached as an Annex to Brazil's written replies to Panel questions at the first substantive meeting.

          771 Shipments of Steel Mill Products – Carbon, American Iron and Steel Institute Statistical Release 10C, located at Exhibit JPN-1 of Japan's written replies to Panel questions at the first substantive meeting; US

          Census data on steel imports, grouped by product category (an example of this data is located at http://www.census.gov/foreign-trade/Press-Release/2002pr/09/steel/steel1cp.pdf.)(tracking virtually identical categories as the AISI statistical release).

          772 Brazil's second written submission, para. 12.

          773 Japan's written reply to Panel question No. 138 at the first substantive meeting.

          WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

          Page 125


          requested prices for hot-rolled, cold-rolled, etc. The United States has not ever argued that hot-rolled "competes" with corrosion-resistant steel.774


        4. The United States affirms that the USITC applied its traditional factors in determining that there was no clear dividing line between types of CCFRS and defining it as a single like product, corresponding to a single imported product. The USITC found that CCFRS at various stages of processing shared certain basic physical properties, were interrelated, had common end-uses, were generally distributed through the same marketing channels, and were essentially made by the same production processes (at least at the initial stages).775 An important fact for the USITC in defining this like product was that CCFRS at one stage of processing generally is feedstock for the next stage of processing.776 777 In determining that there was no clear dividing line between slab and other CCFRS, the USITC recognized that slab is dedicated for use in producing the next stage of steel, hot-rolled steel.778 Moreover, slab shares common metallurgical properties with CCFRS.779


        5. According to Japan, it is irrelevant that the production processes are the same "at least at the initial stages", as the United States argues. As the comparison between live lamb and lamb meat reveals, later processes can create distinguishable products in terms of physical characteristics, end- use, and/or customer perceptions.780


    3. Different remedy on slab


      1. In the European Communities' view the imposition of a different remedy on slabs confirms that there is no such "product" as CCFRS, including slabs. The possibility to apply distinctly the two remedies provided in Proclamation No. 7529 (one for slabs, one for the rest of CCFRS) rests on the fact that the competent authorities can identify and distinguish "slabs" and the other steel products grouped as CCFRS.781 Norway admits that it may always be possible to reduce a product to another sub-product (e.g. red vs. blue pens), but asserts that the imposition of a different remedy is a clear indication that the "product" (CCFRS), was not sufficiently defined.782


      2. According to Japan, if products are like one another, then the competitive dynamics between them should, by definition, be the same, thereby requiring the same remedy to address a single level of injury. The fact that the President felt compelled to impose a tariff rate quota for slab and a high tariff for the other products within the CCFRS grouping demonstrates that there are different competitive dynamics at play between slab and the finished flat products.783 Korea also argues that the "product" satisfying the conditions of Articles 2.1 and 2.2 is the same "product" against which a measure can be imposed. The decision to impose a different remedy for slab is a recognition that the conditions of competition, including supply and demand, for slab are different from the conditions of competition for the other products treated by the USITC as the same like product. It appears that this


        image

        774 Korea's written reply to Panel question No. 141 at the first substantive meeting.

        775 United States' first written submission, paras. 116-142; USITC Report, pp. 36-45.

        776 United States' first written submission, paras. 119-121, 127, and 140; USITC Report, pp. 37-42.

        777 United States written reply to Panel question No. 26 at the first substantive meeting. 778 United States' first written submission, paras. 119-121; USITC Report, pp. 37-40. 779 United States written reply to Panel question No. 29 at the first substantive meeting. 780 Japan's written reply to Panel question No. 13 at the second substantive meeting.

        781 European Communities' written reply to Panel question No. 22 at the first substantive meeting.

        782 Norway's written reply to Panel question No. 22 at the first substantive meeting.

        783 Japan's written reply to Panel question No. 22 at the first substantive meeting.

        WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

        Page 126


        is a de facto recognition by the United States that slabs are in fact a different like product from the other CCFRS because they are responsive to very different demand conditions.784


      3. Likewise, Norway argues that there cannot be various measures directed at the same product. The application of a different remedy is yet another indication that slabs (which may indeed be one or many more distinct products) are distinct from the other products within a group or bundle. The fact that the United States can easily identify "slabs" as something different from the other products in the bundle, and thus impose a different remedy, testifies to this.785


      4. Brazil also questions why it is necessary to have a different remedy for a sub-category of the products within the like product category if they all compete with each other. Presumably, if slab is competing directly with plate, hot-rolled, cold rolled and corrosion resistant CCFRS products, the remedy necessary to eliminate the injury from imports of slab is no different than that necessary to eliminate the injury from imports of these downstream products. The need for a different remedy for slab demonstrates that the products within the grouping are not like one another, do not directly compete with one another and should not be grouped together. There are different competitive dynamics at play between slab and the finished flat products, meaning that they cannot be part of the same like product.786


      5. New Zealand submits that the USITC treated slab as indivisible from CCFRS for every determination except remedy. Thus, increased imports of CCFRS were found to have caused serious injury to the domestic industry producing the competitor like product, CCFRS. Logically, the same remedy would then be applied to the whole imported product.787


      6. With regard to the slab remedy, the United States asserts that the Agreement on Safeguards permits the application of a safeguard measure at different levels to different items covered by that measure, and even the complete non-application of the measure to some items.788


  2. Tin mill products


      1. Norway considers that there are six broad categories of tin mill products, with specified sub- categories and very specific end-uses. In this regard, Norway makes reference to the explanation given of tin mill products in the USITC Report.789 Norway notes that thickness and surfaces also vary greatly depending on the end-use of the product.790 Norway adds that distinctions between the different products can easily be made and that examples of the different products comprised within this group of products may be found in the later exclusions provided by the USTR, where ten different tin mill products were excluded from the safeguard measures on 22 August 2002.791 792


        image

        784 Korea's written reply to Panel question No. 22 at the first substantive meeting. 785 Norway's written reply to Panel question No. 22 at the first substantive meeting. 786 Brazil's written reply to Panel question No. 22 at the first substantive meeting.

        787 New Zealand's reply to the Panel question No. 22 at the first substantive meeting.

        788 United States written reply to Panel questions No.22-29 at the first substantive meeting.

        789 USITC Report, Vol. II, at p. FLAT-4.

        790 Norway's first written submission, footnote 216; Norway's reply to the Panel question No. 34 at the first substantive meeting.

        791 USTR: "List of additional products to be excluded from the Section 201 safeguards measures, as established in Presidential Proclamation 7529 of March 5, 2002", dated 22 August 2002. (Available from the

        USTR-website, www.ustr.gov) (Exhibit CC-92). This list shows that 10 different tin mill products, with specific product specifications, are excluded.

        792 Norway's written reply to Panel question No. 34 at the first substantive meeting.

        WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

        Page 127


      2. Brazil believes that there was a basis for the USITC to conclude that tin mill products constitute a single imported product and that the domestic industry producing the like product is the industry producing all tin mill products.793


      3. The United States contends that tin mill products consist of a wide variety of flat-rolled carbon or alloy steel, plated or coated with tin or with chromium oxides or with chromium and chromium oxides.794 Certain complainants would accept defining tin mill products as a single like product and others, such as Norway, appear to suggest that tin mill should have been defined far more narrowly as many like products corresponding to certain requests for product exclusions.795 796 Tin mill is made of cold-rolled steel that has been coated with tin or chromium or chromium oxides. The like product, "tin mill", consists of a continuum of tin or chromium coated products that, similar to most "like product" definitions, includes a range of varying sizes, coatings, grades, etc. Four Commissioners found that the continuum of tin mill shared similar physical properties or characteristics, uses, manufacturing processes, and marketing channels.797 798


  3. FFTJ


    1. The European Communities further argues that the lack of precise definitions of the imported products concerned vitiates all the USITC's findings on increased imports, injury and causation, as illustrated by the example of "carbon and alloy fittings". Even the USITC itself recognized that this "category contains a mix of products", or diverse "pipe connection products".799 The artificial inflation of this product group fundamentally distorted the increased imports and causation analysis and resulted in the imposition of safeguard measures on a broad "mix of products" although the USITC only established a price undercutting effect for one very specific product that accounts for around 1% of all imports: "Carbon steel butt-weld pipe fitting, 6 inch nominal diameter, 90 degree elbow, long radius, standard weight, meeting ASTM A-234, grade WPB or equivalent specification".800 The European Communities further argues that the products group as "carbon and ally fittings" are not even like or directly competitive (the relevant arguments are summarised in section E.7(d)). The European Communities notes that the United States did not respond to the specific claim that the products bundled as "FFTJ" were not even like each other. Thus, all determinations based on such imported product should be found incompatible with Article 2.1 of the Agreement on Safeguards.


      1. DEFINITION OF THE DOMESTIC INDUSTRY PRODUCING PRODUCTS THAT ARE LIKE OR DIRECTLY COMPETITIVE WITH THE IMPORTED PRODUCT


        1. Introduction


    2. The European Communities, Japan, Norway and the other complainants claim that the United States acted inconsistently with its obligations under Articles 2.1 and 4.1(c) of the Agreement on Safeguards in the definition of the "domestic industry that produces like or directly competitive products". Japan adds that the United States further violated Articles X:3(a) and XIX:1 of GATT


image

793 Brazil's written reply to Panel question No. 34 at the first substantive meeting.

794 USITC Report, p. FLAT-4.

795 Norway's first written submission, para. 223.

796 United States' first written submission, para. 104.

797 United States' first written submission, paras. 143-144; USITC Report, pp. 48-49.

798 United States' written reply to the Panel question No. 27 at the first substantive meeting.

799 USITC Report, Vol. I, pp. 175 and 179.

800 USITC Report, Vol. II, p. TUBULAR-59, Table TUBULAR-61.

WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

Page 128


1994.801 In its defence, the United States contends that the complainants have not established any basis for the Panel to conclude that any of the USITC's determinations of like product are inconsistent with Articles 2.1 and 4.1 of the Agreement on Safeguards or Articles X:3(a) and XIX:1 of GATT 1994.802


    1. The United States contends that the USITC's approach regarding the definition of the like product is consistent with the Agreement on Safeguards.803 The USITC defined 27 separate like products that correspond to subject imports. Ten of these definitions correspond to subject imports on which remedies were imposed and are subject to review by this panel.804 The USITC considered the record evidence using long established factors and looked for clear dividing lines among the various types of domestic steel corresponding to the imported steel subject to this investigation. The methodology employed by the USITC was unbiased and objective. The USITC's definitions of like products were adequate, reasoned, and reasonable explanations were provided, consistent with United States' obligations under the Agreement on Safeguards.805 The USITC's like product determinations were fully justified given the specific facts with respect to each of the products in question, and the USITC's adequate and reasonable explanations of its analysis with respect to each like product determination.806


        1. Defining/identifying the "like product"


          1. Guidance in prescribing a definition


    2. The European Communities, Japan, Korea, China, Switzerland and Norway consider that the notion of "like product" in the Agreement on Safeguards should be guided by the Appellate Body's teaching that "when construing the prerequisites for taking [safeguard] actions, their extraordinary nature must be taken into account". The purpose of a safeguard determination is not, as suggested by the USITC, to provide the widest possible blanket of protection to an affected industry. The Appellate Body clarified that safeguard measures may only be resorted to "in an extraordinary emergency situation". This purpose mandates that the domestic industry is defined accurately enough to ensure a meaningful analysis of all the conditions imposed by the Agreement on Safeguards.807


    3. According to the European Communities, Japan, Switzerland, New Zealand and Brazil, one consequence is that it is not sufficient that one domestic product in a bundle is like or directly competitive with an imported product, irrespective of how specifically or broadly this has been defined. The competent authorities must establish that all components of the bundle of the domestic products are like or directly competitive with all specific imported products or all components of the bundle of imported products. Japan emphasises that in the absence of any competition between the


      image

      801 European Communities' first written submission, para. 222; Japan's first written submission, paras. 78-82; and Norway's first written submission, para. 213. For detailed discussions on the violation of GATT 1994 Article X:3(a), see, infra section VII.O.2.

      802 United States' first written submission, para. 63.

      803 United States' first written submission, para. 96.

      804 United States' first written submission, paras. 64 and 114.

      805 United States' first written submission, para. 115.

      806 United States' second written submission, para. 30.

      807 European Communities' first written submission, paras. 197-198; Japan's first written submission, paras. 95-96; Japan's second written submission, para. 27; Korea's first written submission, para. 27; China's first written submission, paras. 148-149; Switzerland's first written submission, paras. 183-184; Norway's first written submission, para. 188.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 129


      imported and domestic product, there is no basis for finding a like product.808 New Zealand adds that this means that there will be likeness between each of the products within each product bundle, and insists that a competent authority cannot group together several unlike imported products and then bundle the same unlike domestic products and claim that it has met the "like product" test of Article 2 of the Agreement on Safeguards.809


    4. Brazil argues that in the instant case, the USITC's analysis of CCFRS products only concerned itself with an analysis of "like" products and did not concern itself with "directly competitive" products. In other contexts and under other agreements, the Appellate Body has found that if an industry is defined by the "like" products it produces, the definition must be more narrowly drawn than if that industry were defined by the broader "directly competitive" products.810


    5. The United States argues that neither the rationale of safeguards as an exception to other obligations nor the statements in US – Lamb require a narrowly construed like product definition, as complainants contend.811 First, to the extent the Agreement on Safeguards is an exception, that aspect of it has already been comprehended in the text of the Agreement on Safeguards. Members are neither directed or authorized to vary the balance of rights and obligations reflected in the Agreement by appending an unstated rule of construction on the negotiated language. Moreover, the United States submits that these arguments ignore the facts of this investigation which are very different from those in US – Lamb. The like product and domestic industry definitions in this case correspond exactly to the imports subject to investigation. Thus, the effects of imports on domestic producers of goods that are not defined as like products is not at issue. The complainants' arguments apparently are more about the range of products within the investigation than the USITC's like product approach.812


      1. Relevance of definition of like product under other WTO Agreements and existing GATT/WTO jurisprudence


    6. Norway argues that, in light of the extraordinary nature of the measures, and the fact that the concept "like" is coupled with "or directly competitive" in Article 2.1 of the Agreement on Safeguards, "likeness" under the Agreement on Safeguards must be construed at least as narrowly as under Article III:2 of the GATT 1994.813 814 New Zealand also points out that "like product" in the Agreement on Safeguards stands in juxtaposition to "directly competitive product" and must, thus, be interpreted narrowly limited, so as not to encroach on situations more appropriately dealt with as "directly competitive".815


    7. Korea, Japan and Brazil argue that, as the panel in US – Lamb found, there is no reason to construe the words "like product" in the Agreement on Safeguards any differently from their


      image

      808 European Communities' first written submission, para. 200; Switzerland's first written submission, para. 186; New Zealand's first written submission, para. 4.48; Brazil's first written submission, paras. 88-89; Japan's first written submission, para. 78-79.

      809 New Zealand's second written submission, paras. 3.28, 3.21.

      810 Appellate Body Report, US – Cotton Yarn, para. 91; Appellate Body Report, Japan – Alcoholic Beverages II, at 19; and US – Lamb, para. 88, footnote 50 (implicitly recognizing this principle).

      811 European Communities' first writtensubmission, paras. 197-199; Korea's first written submission, paras. 27-28; Japan's first written submission, paras. 88 and 95.

      812 United States' first written submission, para. 99.

      813 Appellate Body statements in respect of the two concepts in Article III:2 of the GATT in Appellate Body Report, Japan – Alcoholic Beverages II, pp. 18-20

      814 Norway's first written submission, para. 188.

      815 New Zealand's first written submission, paras. 4.38, 4.41.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 130


      definition in the Agreements disciplining anti-dumping or countervailing measures.816 817 Korea and New Zealand also believe that the AD and SCM Agreements are useful to the interpretation of "like product" in the Agreement on Safeguards because "the three Agreements' definitions of the industry producing a like product are essentially identical".818 819


    8. The United States points out that the complainants' proposals for appropriate like product definitions range from product definitions used in trade remedy cases under other statutes, to tariff classifications (612 classifications in all), to product descriptions contained in requests for product exclusions. Far from universal agreement, some complainants even propose different definitions for the same item for different purposes, based on the issue contested and their desired result.820


    9. The United States points out that this dispute presents the Panel with the first occasion to examine fully the interpretation and application of the term "like products" in the context of the Agreement on Safeguards. The Appellate Body has found that the term "like products" "must be interpreted in light of the context, and of the object and purpose, of the provision at issue, and of the object and purpose of the covered agreement in which the provision appears".821 The term "like or directly competitive products", or more specifically, the term "like products" is not defined in the Agreement on Safeguards or in the GATT 1994, nor has it been at issue in dispute settlement proceedings involving the Agreement on Safeguards. Where the term "like products" has been addressed in other GATT or WTO dispute settlement proceedings, it has been in the context of provisions of GATT 1994, or other covered agreements with distinct and different purposes from those of the Agreement on Safeguards. As the Appellate Body has cautioned, the interpretation of the term "like products" for one context cannot be automatically transposed to other provisions or agreements where the phrase "like products" is used.822 For instance, it is clear that the interpretation of the term "like products" in the context of provisions of other covered agreements (e.g., Article III of GATT 1994), whose purpose is to avoid protectionism and protect an equal and competitive relationship between products, will necessarily not be identical to, and perhaps not particularly relevant for, the Agreement on Safeguards, which has the opposite purpose, i.e., permitting the temporary protection of a domestic industry under certain circumstances. The United States argues that the Panel should recognize the clear distinction between these purposes and reject, in accordance with the Appellate Body's findings, the complainants' proposals to automatically transpose interpretations made in another context to the Agreement on Safeguards.823


    10. The United States also submits that, with regard to the context of the Agreement on Safeguards, it has not been established in other GATT 1947 or WTO dispute settlement proceedings what factors are appropriate to be considered in determining whether a domestic product is like an imported product. While "general criteria, or groupings of potentially shared characteristics, provide a framework for analysing the 'likeness' of particular products … it is well to bear in mind [that such criteria are] simply tools to assist in the task of sorting and examining the relevant evidence".824


      image

      816 Panel Report, US – Lamb, para. 7.75 ("Another element of relevant context for interpreting the 'domestic industry' definition of SG Article 4.1(c) are the parallel provisions of the WTO agreements on Subsidies and Countervailing Measures and Anti-Dumping.").

      817 Brazil's first written submission, paras. 88-89; Korea's second written submission, para. 35; Japan's first written submission, paras. 89-94.

      818 Panel Report, US – Lamb, para. 7.75 (emphasis in original).

      819 Korea's first written submission, para. 30; New Zealand's first written submission, paras. 4.42-4.43.

      820 United States' first written submission, para. 103.

      821 Appellate Body Report, EC – Asbestos, para. 88.

      822 Appellate Body Report, EC – Asbestos, footnote 60 at p. 34.

      823 United States' first written submission, paras. 63-82.

      824 Appellate Body Report, EC – Asbestos, para. 102.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 131


      Moreover, it is clear that the domestic like product analysis under the Agreement on Safeguards should involve "'an unavoidable element of individual, discretionary judgement' … [and] be made on a case-by-case basis".825 As the Appellate Body has stated, "the adoption of a particular framework to aid in the examination of evidence does not dissolve the duty or the need to examine, in each case, all of the pertinent evidence".826 827


    11. According to the European Communities, there are criteria to determine "like or directly competitive products" in the Agreement on Safeguards. The Appellate Body in interpreting the analogous provision in US – Cotton Yarn distinguished between the like and directly competitive concepts for purpose of safeguard measures and confirmed that the like product relationship is narrower than "directly competitive".828 As to the criteria for determining likeness, the European Communities submits that the Appellate Body in Japan – Alcoholic Beverages II and EC – Asbestos confirmed that no matter where in the WTO Agreement the term "like" is used, "each" of the four border tax adjustment criteria "should be examined" as a framework829 and that all of the evidence should be considered. The different purposes of Article III:2 of the GATT 1994 and the Agreement on Safeguards are not the point. What is relevant is that the purpose of the like product concept is the same in both provisions, that is to analyse whether an imported product competes with a domestically produced one. The different purposes of the underlying obligations may be a reason for stretching the accordion, but the accordion and its keyboard remains the same. According to the European Communities, the United States itself agrees that the following criteria are relevant to determine likeness in a safeguard investigation: "physical properties of the product, its customs treatment, its manufacturing process, its uses". The only divergence between the United States and the European Communities on this point relates to the relevance of the "continuous line of production between an input and an end-product".830 Japan adds that the central teaching of US – Line Pipe is that the Agreement exists to prevent Members from abusing their right to protect domestic industries.831


    12. Korea also disagrees that the protection of domestic industries is the object and purpose of the Agreement on Safeguards. Such reasoning also would beg the fundamental question: "Which domestic industries can be protected?" The Agreement on Safeguards makes clear that only those domestic industries producing the like product can be the basis for a safeguard measure.832 833 The chapeau of the Agreement on Safeguards includes as its object and purpose "the need to enhance rather than limit competition". The entire context of Article XIX of the GATT 1994 and the Agreement on Safeguards is to provide only a very limited exception to increased competition from imports brought about by WTO concessions.834


      1. Physical properties, end-uses, consumer perception and tariff classification


    13. The European Communities, China, Switzerland, Norway, New Zealand and Brazil submit that an examination of likeness must involve, at a minimum, consideration of the four border tax


      image

      825 Appellate Body Report, EC – Asbestos, para. 101; see also Appellate Body Report, Japan – Alcoholic Beverages II, at pp. 20-21.

      826 Appellate Body Report, EC – Asbestos, para. 102.

      827 United States' first written submission, paras. 65, 83, 88.

      828 Appellate Body Report, US – Cotton Yarn, paras. 97 and 98.

      829 Appellate Body Report, Japan – Alcoholic Beverages II, p. 21, Appellate Body Report, EC – Asbestos, para. 109.

      830 European Communities' second written submission, paras. 244-249.

      831 Japan's second written submission, para. 27.

      832 Appellate Body Report, US – Lamb, para. 94; Appellate Body Report, US – Cotton Yarn, para. 95.

      833 Korea's second written submission, para. 37.

      834 Korea's second written submission, para. 41.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 132


      adjustment criteria: (i) the physical properties of the products; (ii) the extent to which the products are capable of serving the same or similar end-uses; (iii) the extent to which consumers perceive and treat the products as alternative means of performing similar functions in order to satisfy a particular need; and (iv) the international classification of the products for tariff purposes.835 New Zealand adds that the narrow scope of the term "like product" in the Agreement on Safeguards dictates that there will need to be a high degree of similarity across these characteristics.836


    14. Japan asserts that the Appellate Body has held in the context of GATT Article III (e.g., in Japan – Alcoholic Beverages II) that the most relevant factors are the products' physical properties, end-uses, consumer preferences, and tariff classifications. The point of these factors is clear: they help determine whether the products compete with each other. Only if the products compete with each other are they properly grouped together. Otherwise, the analysis to be performed in the injury investigation, as well as in choosing an appropriate remedy, is meaningless.837


    15. The European Communities argues that the purpose of the "like or directly competitive products" test set out in Article 2.1 of the Agreement on Safeguards is to define the domestic industry sufficiently precisely so as to ensure that only domestic producers suffering serious injury are given temporary breathing room to facilitate adjustment.838 Brazil, Japan and Switzerland submit that a safeguard measure is not permitted unless serious injury is caused by competition with imports. The four factor analysis developed in the context of Article III, which is aimed at discerning the extent of competition, therefore, is equally applicable in the safeguard context.839 Norway argues that the purpose of defining the parameters of the "like product or directly competitive product" is similar under Article 2.1 of the Agreement on Safeguards and under Article III:2 of the GATT 1994: the like or directly competitive product test is a tool to analyse the competitive relationship between imported and domestically produced products.840 841


    16. China agrees that the interpretation of "like products" given in the context of one specific agreement cannot be automatically transposed in the context of other agreements.842 However, China asserts that this can sometimes be the case and, in any case, such interpretations provide useful guidance.843 China, therefore, reaffirms that the factors identified by the Working Party on Border Tax Adjustments, and further used by the Appellate Body in Japan – Alcoholic Beverages II, are relevant in this case. 844 The suggested criteria of the Working Party on Border Tax Adjustments were meant to be used for all the different cases occurring under the GATT. It is clear that these factors can be transposed for assessing "likeness" in the context of the Agreement on Safeguards.845


    17. The United States submits that all parties agree on the reasonableness of the following criteria: physical properties/characteristics, uses, and customs treatment/tariff classification. Many


      image

      835 European Communities' first written submission, para. 202; China's first written submission, para. 150; Switzerland's first written submission, para. 188; Norway's first written submission, para. 188; New Zealand's first written submission, para. 4.45; Brazil's first written submission, para. 94.

      836 New Zealand's first written submission, para. 4.44.

      837 Japan's first written submission, paras. 98-101.

      838 European Communities' written reply to Panel question No. 51 at the first substantive meeting.

      839 Japan's, Switzerland's and Brazil's written reply to Panel question No. 51 at the first substantive

      meeting.


      840 Norway's written reply to Panel question No. 51 at the first substantive meeting.

      841 European Communities' second written submission, paras. 244-249.

      842 United States' first written submission, para. 64. 843 China's second written submission, paras. 61-62. 844 China's second written submission, para. 67.

      845 China's second written submission, para. 69.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 133


      parties agree that consideration of manufacturing processes may be appropriate. While several of the complainants maintain that consumer tastes also is an appropriate criteria, no party has objected to the USITC's consideration of marketing channels. The USITC's traditional like product criteria are consistent with the Agreement on Safeguards, there is no directly related treatment of the term in panel or Appellate Body reports to provide guidance on the issue of the appropriate criteria for the like product analysis. Moreover, resorting to the "ordinary" or "plain" meaning of the term "like" provided by the dictionary "leave[s] many interpretative questions open".846 The Appellate Body in EC – Asbestos, interpreting Article III of the GATT, noted that the dictionary definition of "like" does not resolve the following three issues of interpretation: (i) which characteristics or qualities are important; (ii) the degree or extent to which products must share qualities or characteristics; and (iii) from whose perspective "likeness" should be judged.847 848


    18. According to the United States, the Appellate Body, when addressing the term "like" pursuant to Article III of the GATT, recognized in Japan – Alcoholic Beverages II, and most recently affirmed in EC – Asbestos, that the purpose and context of the covered agreement is important in interpreting the term "like products"849 and that such interpretation for one context cannot be automatically transposed to other provisions or agreements where the phrase "like products" is used.850 In accordance with the Appellate Body's findings, this Panel should recognize the clear distinction between agreements with different purposes and reject the complainants' proposals to automatically transpose criteria established for another context, such as Article III, to the Agreement on Safeguards.851


    19. The United States further contends that the purpose of the Agreement on Safeguards is to permit the temporary protection of a domestic industry under certain circumstances.852 The United States reiterates that the focus of the analysis in a safeguards investigation is on the condition of the domestic industry and not, as it is under Article III, on whether imports are being treated in a manner different from domestic products in the home market, i.e., whether they are afforded national treatment.853 While protecting the competitive relationship between imports and domestic products and avoiding protection to domestic production is the purpose of Article III854, affording temporary


      image

      846 Appellate Body Report, EC – Asbestos, para. 92.

      847 Appellate Body Report, EC – Asbestos, para. 92.

      848 United States' second written submission, paras. 58-61.

      849 Appellate Body Report, EC – Asbestos, para. 88.

      850 Appellate Body Report, EC – Asbestos, footnote 60, at p. 34, referring to Appellate Body Report,

      Japan – Alcoholic Beverages II, at p. 20. Appellate Body Report, EC – Asbestos, para. 93.

      851 United States' second written submission, para. 62.

      852 Appellate Body Report, US – Line Pipe, para. 82; Panel Report, US – Lamb, para. 7.76 (the Agreement's objectives of "creating a mechanism for effective, temporary protection from imports to an industry that is experiencing serious injury. . . .").

      853 The term "like products" has primarily been addressed in dispute settlement proceedings regarding allegations that national treatment has not been afforded regarding 1) internal taxes pursuant to Article III:2 of

      GATT 1994, and 2) laws and regulations pursuant to Article III:4 of GATT 1994. The Appellate Body has indicated in EC – Asbestos, para. 97, regarding the purpose of Article III of GATT 1994 that:


      The broad and fundamental purpose of Article III is to avoid protectionism in the application of internal tax and regulatory measures. More specifically, the purpose of Article III "is to ensure that internal measures 'not be applied to imported and domestic products so as to afford protection to domestic production' " Article III protects expectations not of any particular trade volume but rather of the equal competitive relationship between imported and domestic products. (emphasis added)


      854 Specifically, the Appellate Body in EC – Asbestos, para. 99, stated:

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 134


      protection if necessary to the domestic industry is the purpose of the Agreement on Safeguards. The traditional like product criteria considered by the USITC focus on objective rather than subjective factors. Moreover, since the focus of the analysis in a safeguard investigation is on the condition of the domestic industry rather than the consumer or the relationship of imported and domestic products in the market, "likeness" should be viewed from the perspective of the domestic product rather than a consumer, consistent with the purpose of the Agreement on Safeguards.855 Indeed, the Appellate Body cautioned in EC – Asbestos that it may be important to consider "from whose perspective 'likeness' should be judged. For instance, ultimate consumers may have a view about the 'likeness' of two products that is very different from that of the inventors or producers of those products".856


    20. The United States submits that, in any event, the USITC's like product factors in a safeguard investigation include the three criteria on which all parties agree (physical properties, uses, and customs treatment), as well as focus on such other objective factors as the product's marketing channels and manufacturing process.857 858 These are not mandatory criteria and do not limit the USITC from considering other factors, as appropriate, in making its findings.859 No single factor is dispositive and the weight given to each individual factor (and other relevant factors) will depend upon the facts in the particular case.860 The USITC traditionally has looked for clear dividing lines among possible products and has disregarded minor variations.861


    21. The European Communities maintains its allegation that the United States failed to systematically look at each of the classical criteria and all the relevant evidence. According to the European Communities and Switzerland, the USITC has nowhere considered tariff classifications.862 The European Communities and Switzerland argue that the United States cannot defend itself by claiming that there are too many classifications at the ten-digit level and that there is no universal agreement on what constitutes specific steel products in general.863 Chapters 72 and 73 of the HS harmonize tariff classifications for steel products at the six digit level.864 Japan submits that, in fact, the United States is not discerning the proper dividing line between products, but rather between producers.865


      image


      ... a determination of "likeness" under Article III:4 is, fundamentally, a determination about the nature and extent of a competitive relationship between and among products.


      855 Appellate Body Report, US – Line Pipe, para. 82 (purpose of Agreement on Safeguards is to permit a WTO Member to "resort[] to an effective remedy in an extraordinary emergency situation that ... makes it necessary to protect a domestic industry temporarily".

      856 United States' second written submission, paras. 62-64.

      857 Panel Report, Japan – Alcoholic Beverages I, para. 5.7 (Panel was of the view that the likeness of products must be examined taking into account objective criteria (such as composition and manufacturing processes of products)).

      858 United States' first written submission, paras. 83-93.

      859 Appellate Body Report, EC – Asbestos, para. 102 (general criteria "are neither a treaty-mandated nor a closed list of criteria that will determine the legal characterization of products").

      860 United States' second written submission, para. 65.

      861 United States' first written submission, para. 84.

      862 Switzerland's second written submission, paras. 42-43; European Communities' second written submission, para. 263.

      863 Switzerland's second written submission, para. 58; United States' first written submission, paras. 65

      and 103.


      864 European Communities' second written submission, para. 263.

      865 Japan's second written submission, para. 20.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 135


    22. The United States responded that while consideration of customs treatment may be a relevant factor in such an analysis depending on the facts of a particular case, it still is one of a number of criteria.866 The USITC found that consideration of customs treatment for purposes of the definition of the like product was not a "useful factor" given the large number of classification categories (612) applicable to this investigation. The fact is, in this case the numerous tariff classifications did not provide clear distinctions between products. For instance, each of the 33 data collection categories individually have from 2 to 65 tariff classifications.867


      1. Production processes


    23. All of the complainants recall that, in US – Lamb, the Appellate Body held that the degree of integration of production processes with respect to products being considered was found to be irrelevant for the determination of the "domestic industry". "The focus", the Appellate Body said, "must, therefore, be on the identification of the [domestic] products and their 'like or directly competitive' relationship [with the imported products], and not on the processes by which those products are produced".868 869 China adds that it is clear from this statement that production processes are not an element of the like or directly competitive relationship between products.870


    24. The United States contends that, in spite of complainants' mischaracterizations, the dispute settlement proceedings in US – Lamb provides little guidance on the issue of defining the like product. There was no disagreement in that dispute regarding the definition of like product.871 Rather, the issue in US – Lamb involved the definition of the domestic industry after the USITC had already defined the like product. The findings in US – Lamb spoke to which producers could be considered members of the domestic industry producing a single domestic like product and not to defining the like product, as complainants have alleged.872 The United States further argues that, contrary to the complainants' mischaracterizations, the Appellate Body has recognized that it may be appropriate to consider the production process for a product in defining like products, particularly when the question arises as to whether two articles are separate products.873


    25. New Zealand responds that it is the United States that mischaracterizes the issue in US – Lamb. What had been defined in US Lamb was the imported product, and the issue was whether the domestic industry included the producers of a product that was not "like" the imported product. Thus, US Lamb is of direct relevance to this case.874 Japan takes the view that the United States draws a distinction without a difference. As specified in Article 4.1(c) of the Agreement on Safeguards, the scope of the like (or, if applicable, directly competitive) product defines the scope of the relevant domestic industry. Ultimately, the inquiry is about defining the "like" product and domestic industry in ways that reflect meaningful and substantial competitive interactions. Taking the United States' theory to its logical extreme, a competent authority would be authorized to combine any number of


      image


      para. 86.

      866 United States' written reply to Panel question No. 20 at the first substantive meeting.

      867 United States' first written submission, para. 86.

      868 Appellate Body Report, US – Lamb, para. 94. See also, Appellate Body Report, US – Cotton Yarn,


      869 European Communities' first written submission, paras. 191-193; Japan's first written submission,

      paras. 103-104; Korea's first written submission, para. 32; China's first written submission, para. 140; Switzerland's first written submission, para. 179; Norway's first written submission, para. 197; Brazil's first written submission, para. 96; New Zealand's first written submission, para. 4.35.

      870 China's second written submission, para. 71.

      871 Appellate Body Report, US – Lamb, para. 88.

      872 Appellate Body Report, US – Lamb, paras. 84, 90 and 95.

      873 United States' first written submission, paras. 70, 91.

      874 New Zealand's second written submission, para. 3.41.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 136


      products, regardless of the extent of their likeness. Indeed, if it were true, Japan sees no reason why the United States would not have simply conducted an investigation and imposed a measure on imports of all "steel". Japan argues that the fact that there must be some control on the scope of like product definitions under the Agreement on Safeguards is evident in US – Lamb. The Appellate Body clarified the overarching importance of ensuring that the nexus between imports and their domestic counterparts – whether like or directly competitive – is close enough to ensure that a measure is not imposed to protect industries that do not make like or directly competitive products. The point is that the like product and, in turn, the industry cannot be so broadly defined as to provide protection to producers of products with which the imports do not compete.875


    26. Korea also objects to the United States' proposition and insists that the Appellate Body in US

      – Lamb discussed the issue of like products, in conjunction with the issues of imported products and domestic industry, for the first time in the context of the Agreement on Safeguards. The Appellate Body clarified that the nexus between imports and their domestic counterparts should be close enough to ensure that a measure is not imposed to protect industries that do not make like or directly competitive products.876 Obviously, the definitions of like product and domestic industry are interrelated and interdependent concepts – Article 2.1 of the Agreement on Safeguards refers to the domestic industry that produces like or directly competitive products while Article 4.1(c), which was at issue in US – Lamb, defines the domestic industry as "producers … of the like or directly competitive products …" (emphasis added). The Appellate Body's decision in US – Lamb turned on the interpretation of the "like product" to determine the scope of the domestic industry.877 The Appellate Body further stated, "…only when those products have been identified is it possible then to identify the "producers" of those products".878 Whether the like product was improperly defined by reference to the industry's production process (in this case) or the domestic industry producing the like product was improperly defined by reference to the production process (in US – Lamb), the result is the same: "in our view, it would be a clear departure from the text of Article 2.1 if a safeguard measure could be imposed because of the prejudicial effects that an imported product has on domestic producers of products that are not 'like or directly competitive' products in relation to the imported product".879 This same improper "departure" would occur under the United States' definition of the like product in this case.880


    27. China also disagrees and stresses that the statement from the US – Lamb case confirms that the definition of the domestic industry, through an assessment of the "like or directly competitive" relationship, first requires the identification of the specific imported product. In no way can the imported product be re-defined after the domestic industry producing like or directly competitive products has been identified. China submits that the United States is misusing this statement and trying to create confusion.881


    28. In the view of the European Communities, Japan, China, Switzerland, Norway and Brazil, the Appellate Body in US – Lamb also clarified which criteria are not capable of establishing likeness between domestic and imported products. These are, first, the existence of a "continuous line of production from the raw to the processed product" and second, a "substantial coincidence of economic



      image

      875 Japan's second written submission, paras. 10-13.

      876 Appellate Body Report, US – Lamb, para. 86. 877 Appellate Body Report, US – Lamb, para. 87. 878 Appellate Body Report, US – Lamb, para. 87. 879 Appellate Body Report, US – Lamb, para. 86.

      880 Korea's second written submission, paras. 13-17.

      881 China's second written submission, para. 65.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 137


      interests" between producers of both products.882 883 Japan adds that these Appellate Body statements in US Lamb could not be more relevant to a case than the present one in which an authority has conjoined, into a single like product, products that serve as feedstock for one another, but which, in fact, have independent uses in the marketplace.884


    29. The United States insists that the USITC's methodology for analysing like product does not contravene the Agreement on Safeguards, as the complainants assert, because it employs factors different than those suggested by complainants. First, the Agreement on Safeguards does not stipulate which factors are to be considered in the like product analysis under the Agreement on Safeguards. Second, there is no Appellate Body or panel ruling that provides any interpretative guidance as to the criteria to be used in determining the like product for purposes of the Agreement on Safeguards.885 The United States insists that US – Lamb did not address the USITC's methodology for determining like product. Rather, that dispute concerned the definition of the domestic industry. Indeed, the two factors at issue in US – Lamb that are cited repeatedly by complainants – "continuous line of production" and "substantial coincidence of economic interests" – are factors used by the USITC to identify who are the appropriate domestic industry members after the like product was defined. They are cited nowhere by the USITC in its like product findings with regard to CCFRS products or any of the other like products. While the complainants try to imply that these criteria were applied in the determinations at issue here886, they fundamentally misstate the analysis actually performed by the USITC.887 888


    30. According to the United States, one of the factors considered by the USITC in defining like products is the product's manufacturing process (i.e., where and how it is made). In the context of the Agreement on Safeguards where the purpose is to allow measures to protect the domestic industry, albeit temporarily and under certain circumstances, consideration of the manufacturing process for a product is an appropriate and objective factor. The Appellate Body in US – Lamb recognized that it may be appropriate to consider the production process for a product in defining like products, particularly when the question arises as to whether two articles are separate products, as noted at footnote 55.889 890 The complainants have erroneously urged application of Appellate Body findings in US – Lamb regarding the definition of a domestic industry to the like product definition and ignored the Appellate Body's explicit recognition that consideration of production processes may be a relevant factor in defining like products.891 The Appellate Body also recognized that, when faced with products at various stages of production, a relevant factor for determining the like product definition (as opposed to the domestic industry definition) was whether products at different stages of


      image

      882 Appellate Body Report, US – Lamb, paras. 77 and 95.

      883 European Communities' first written submission, para. 201; Japan's first written submission, paras. 103-104; China's first written submission, para. 204; Switzerland's first written submission, para. 229; Norway's first written submission, paras. 194-195; Brazil's first written submission, para. 96.

      884 Japan's second written submission, para. 13.

      885 United States' second written submission, para. 27.

      886 Japan's first oral statement (like product), para. 13 ("a continuous line of production between products – a characteristic heavily relied upon by the US in this case").

      887 Panel Report, Argentina – Footwear (EC), para. 8.124.

      888 United States' second written submission, paras. 26-29.

      889 Appellate Body Report, US – Lamb, para. 94, footnote 55; Panel Report, Japan – Alcoholic Beverages I, para. 5.7 (Panel thought it was important to assess "likeness", as much as possible, on the basis of

      objective criteria, including, in particular, composition and manufacturing processes of the product, in addition to consumption habits.).

      890 United States' second written submission, para. 66.

      891 Appellate Body Report, US – Lamb, para. 94, fn. 55; Panel Report, Japan – Alcoholic Beverages I, para. 5.7.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 138


      processing were different forms of a single like product or had become different products.892 The United States submits that although most complainants agree that consideration of manufacturing processes, i.e., how a product is made, is an appropriate criteria for a safeguard investigation, they disagree with the USITC's additional consideration of where the product is made in the manufacturing process.893


    31. According to New Zealand, the United States' reliance on footnote 55 in the US Lamb Appellate Body report is misguided. The interpretation advanced by the United States would have the footnote contradict the plain meaning of the body of the Report. It is clear from reading footnote 55, that it is dealing with the question of "separate products". It does not, as the United States suggests, deal with the issue of "like products".894 Similarly, Korea also argues that footnote 55 of the Appellate Body Report in US – Lamb adds nothing to its argument. The Appellate Body was simply affirming that the production process could be relevant to the question of whether the products produced from that process were separate products (e.g., the production process could confer certain physical characteristics or could make the product adapted to certain end-uses, but the production process is not a "like" product characteristic). The inquiry is still to determine the domestically- produced like product that is to be analysed in relation to the imported product. For this reason, an evaluation of the production process could illuminate, but not determine, whether the products were "like". The like product determination must still be based on the product produced – not on the nature of the domestic production process itself.895 896


    32. Likewise, the European Communities, China, and Japan consider that the United States misunderstands this footnote by equating the terms "production process" used in that footnote with what has been outlawed by the Appellate Body in US – Lamb as a criterion to establish likeness between products which are otherwise not like: "a continuous line of production between an input product and an end-product".897 The term "production process" refers to how a product is manufactured. The Appellate Body in the above-cited footnote clarified that the production process may be a useful tool to further separate two products that have the same physical characteristics, customs classifications and uses.898 The European Communities considers that in most cases, the production process does not add to the like product determination established on the basis of the physical properties, customs classifications and end-uses. Thus, the production process may be used to further separate products, but it can never serve as criterion to bundle products that are otherwise not like based on the Border Tax Adjustment criteria.899


    33. Japan adds that, even if production process is relevant, in any event, the reliance by the United States on this passage misses an important distinction made in US – Lamb – the distinction, on the one hand, between: (i) an analysis of production processes themselves to discern the extent to which those processes create separately identifiable products; and (ii) the vertical integration of those


      image

      892 Appellate Body Report, US – Lamb, paras. 90, 92 and 94; Panel Report, US – Lamb, paras. 7.95 and

      7.96.

      893 United States' second written submission, para. 67, referring for example to Brazil's, Korea's,

      Japan's and Norway's written reply to Panel questions Nos. 69 and 150 at the first substantive meeting.

      894 New Zealand's written reply to Panel question No. 9 at the second substantive meeting.

      895 Appellate Body Report, US – Lamb, para. 94.

      896 Korea's second written submission, para. 17.

      897 European Communities' second written submission, paras. 251-255, citing Appellate Body Report,

      US – Lamb, para. 90.

      898 Japan's second written submission, para. 15; China's second written submission, paras. 74-75; European Communities' second written submission, paras. 251-255.

      899 European Communities' second written submission, paras. 251-255.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 139


      processes.900 The Appellate Body specifically stated in US – Lamb that it had "reservations about the role of an examination of the degree of integration of production processes for the products at issue".901 By suggesting in footnote 55 that production processes might be relevant to determining whether two articles are separate products, the Appellate Body was not endorsing an analysis of vertical integration. After all, vertical integration does not mean production processes are somehow blurred; it merely means they are under the same corporate hat, and perhaps located at the same general location. This is particularly relevant for flat-rolled steel. Despite vertical integration of some (not all) CCFRS production, each separate product that the USITC chose to bundle is produced on different machines, housed in different buildings. Integration, therefore, determines very little about the processes themselves and the extent to which those processes create separate products.902


    34. Brazil also argues that the United States confuses the issue of the relevance of whether one product is an input for another product, with the relevance of production processes in making a determination that products are like each other. It also confuses the issue of coincidence of economic interests and commonality of facilities with the ability to separately identify the production processes for the products at issue where there is a very high degree of integration of these processes. If anything, the Appellate Body indicated, in footnote 55 to paragraph 94, that it would not find as much relevance as did the panel, of processing and vertical integration in the identification of products. As such, the panel report would seem to describe the outer limits of the relevance of processing and vertical integration in the identification of products.903 The relevant inquiry is not whether slab is an input to downstream products, but rather what happens to slab in the downstream processing – for example in hot-rolling, cold-rolling and galvanizing – and whether the changes imparted by the processing create a different product.904 The Appellate Body simply recognized, consistent with the statements of the panel report it was reviewing, that there may be situations where an inquiry into the production processes of products may be relevant in determining whether two articles are separate products. However, the language on which the United States is relying does not in any way alter the clear findings of the Appellate Body that whether a product is an input to a downstream product is irrelevant to the like product analysis and that vertical integration is also irrelevant to this analysis.905 Read in context, the concern of the Appellate Body expressed in footnote 55 is nothing more than a common sense statement. The fact is that a production process may or may not create a separate like product; whether or not a separate product is created will depend on what happens to the product during a particular stage of processing. Therefore, one cannot assume that a further processed product is either like or unlike the prior stage product, but rather it will depend on whether and to what extent the processing changes the product itself. This does not in any way, however, qualify the Appellate Body's declaration in the same case that it is irrelevant for purposes of determining the likeness of products whether there is a continuous line of production between an input product and an end product.906


    35. The European Communities argues that the United States was not entitled to bundle domestic (or imported) products on the sole basis of a "continuous line of production between an 'input product' and an 'end-product'". This has been unambiguously said by the Appellate Body in US – Lamb907 and is in fact common sense: a pound of flour is not like a cake, although at the initial stage both flour



      image

      900 Japan's second written submission, para. 15. 901 Appellate Body Report, US – Lamb, para. 94. 902 Japan's second written submission, para. 16. 903 Brazil's second written submission, para. 33. 904 Brazil's second written submission, para. 35. 905 Brazil's second written submission, para. 39.

      906 Brazil's written reply to Panel question No. 19 at the second substantive meeting.

      907 Appellate Body Report, US – Lamb, para. 90.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 140


      and a cake are made by milling grain. Similarly, a roll of raw cotton cloth is not like a T-shirt, although both share the initial stage of cleaning and weaving cotton.908 According to China, the fact that two products correspond to two different steps of a continuous production line cannot be considered as meaning that these two products share the same production process. It is therefore not a relevant criterion for assessing "likeness".909


    36. According to New Zealand, the implication of the USITC approach is that a competent authority petitioned by a highly integrated domestic industry to investigate imports of sawn pine logs, finished lumber, treated pine and pine furniture could conclude that a depressed "domestic industry" that manufactures that bundle of products was seriously injured by increased imports even though the only imports that increased were of sawn logs. Equally, a competent authority could bundle imported ice-cream, butter, yoghurt, cheese and skim milk powder into one "imported product", and conclude that there was serious injury from increased imports to the whole of that "domestic industry" even though the part of that "domestic industry" that produced ice cream was not suffering any injury at all. Thus, bundling as the USITC has done in this case undermines the limitation in the Agreement on Safeguards that only the domestic industry that produces a product that is "like" the imported product is entitled to the benefit of a safeguard measure.910


    37. The United States responds that the complainants' position is based on the erroneous assertion that the USITC "bundled together" predefined products rather than that the USITC identified clear dividing lines, using its long-standing analysis. The USITC begins with the universe of imports subject to investigation, as identified in the request or petition. After determining what domestic products are like or directly competitive with the subject imports, the USITC considers whether the domestic products corresponding to the subject imports consist of a single domestic like product or whether there are clear dividing lines among the products so as to constitute multiple domestic like products. The USITC applies its like product criteria, including production or manufacturing processes, in its analysis of whether there are such clear dividing lines as to constitute multiple like products, or that there are no clear lines and that a single like product definition is appropriate. The complainants' position that it might be permissible to use production processes to separate articles into different products, but not to determine that articles are in fact one product, is illogical. Surely the complainants do not mean to suggest that these are in fact different exercises, requiring different criteria. If it is appropriate to use production processes to look for clear dividing lines, it must be appropriate to use production processes to determine that there are no clear dividing lines and that articles constitute one product. The Appellate Body clearly differentiated between an analysis of "domestic industry" and an analysis as to whether two articles are "separate products", and found that production processes "may be relevant" in the latter inquiry. In an inquiry into whether articles are "separate products" – which the USITC performed for slabs, plate, hot-rolled steel, cold-rolled steel, and coated steel – one possible conclusion is that they are not, and are in fact a single product.911 Consideration of manufacturing or production processes, both how and where a product is made, is particularly relevant where the inquiry involves a product at different stages of processing. The interrelationship of the manufacturing processes for a product at different stages of processing may be informative in finding clear dividing lines between the stages of processing. For example, since earlier processed CCFRS, such as slab or hot-rolled steel, is the feedstock for further processed steel, such as cold-rolled steel or coated steel, all such steel, i.e., slab, hot-rolled steel, cold-rolled steel, and

      coated-steel, is produced using essentially the same production processes in the initial manufacturing


      image

      908 European Communities' second written submission, paras. 256-257.

      909 China's second written submission, para. 76.

      910 New Zealand's second written submission, paras. 3.24-3.26.

      911 United States' written reply to Panel question No. 18 at the second substantive meeting.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 141


      stages.912 All certain CCFRS is produced from slab, with the majority of such steel further processed into hot-rolled steel on hot strip or Steckel mills.913 Substantial quantities of earlier processed steel are captively consumed by the producer in the production of further processed steel.914 This tends to blur product distinctions until the processing reaches its final stages.915


    38. According to the European Communities, China and Switzerland, in this case, the USITC did not establish "likeness" on the basis of physical characteristics, end-uses, consumer-tastes and habits and tariff classification. The USITC "focused [its] analysis in this investigation primarily on the degree to which the products in question are produced in common production facilities and using similar production processes"916, although the Appellate Body in US – Lamb had ruled out this criterion for the like product determination.917 As a result of these flaws all of the USITC's definitions of the domestic industry are inconsistent with Articles 2 and 4 of the Agreement on Safeguards.918


    39. Similarly, Brazil submits that the United States defined a domestic industry based on a continuous line of production (i.e., that each product in the sequence could be an input for the next stage product) and coincidence of economic interests (i.e., the fact of vertical integration) and then defined the imported products based on the universe of products produced by this already defined industry, because they were largely produced by the same manufacturers.919


    40. Korea submits that by focusing primarily on the production process to determine the like product920, the USITC's analysis permitted an industry's self-definition to substitute for an analysis and definition of the "industry producing the like product". The Agreement on Safeguards requires that the domestic industry be defined by the like product it produces – not the reverse.921 922


      1. Competition


    41. Japan submits that the "like or directly competitive" relationship that must exist between the imported product and the product produced by the domestic industry, as required by the Agreement on Safeguards and Article XIX:1 of the GATT 1994, should be based on the existence of competition


      image

      912 The USITC's analysis provided a detailed discussion of the five stages of processing certain carbon flat-rolled steel. The manufacturing processes for carbon steel involve three distinct stages that include: (1) melting or refining raw steel; (2) casting molten steel into semifinished form, such as slab; and (3) performing various stages of finishing operations, including hot-rolling, cold-rolling, and/or coating. USITC Report, p. OVERVIEW-7.

      913 USITC Report, pp. 40-41.

      914 Virtually all US-produced slab is internally consumed by the domestic slab producers in their production of hot-rolled steel (sheet, strip, or plate), with large shares of hot-rolled and cold-rolled steel also

      internally transferred. During the year 2000, 99.4% of the quantity of domestic producers' total US shipments of slab were internally transferred, as were 66% of the quantity of domestic producers' total US shipments of hot- rolled steel, and 58.7% of the quantity of total US shipments of domestically-produced cold-rolled steel. USITC Report, pp. FLAT-1 and 3, nn. 4 and 5.

      915 United States' second written submission, paras. 68-69.

      916 USITC Report, Vol. I, p. 191. See also pp. 30.

      917 Appellate Body Report, US – Lamb, paras. 95 and 77.

      918 European Communities' first written submission, paras. 233-234; China's first written submission, para. 181; Switzerland's first written submission, paras. 205-206.

      919 Brazil's written reply to Panel question No. 23 at the first substantive meeting.

      920 USITC Report, Vol. I, pp. 38-39 (flat-rolled), 154-155 (welded other) (Exhibit CC-6); United States' written reply to Panel question No. 69 at the first substantive meeting; United States' replies to questions from

      other Parties (15 November 2002), paras. 19-20.

      921 Appellate Body Report, US – Lamb, para. 94.

      922 Korea's second written submission, para. 18.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 142


      between the imported and domestic products to justify imposition of a safeguard measure. This requirement exists regardless of whether domestic products are deemed "like" or "directly competitive" with the imported product.923 More generally, no causal relationship can be found between an increase in imports and a sales decline in domestic products if these products do not compete for similar end-uses. As an example, Japan assumes that that sales of domestically produced semi-finished slab decline during the same period, and that sales of all (other) CCFRS products, domestic or foreign, remain unchanged. With all CCFRS products plus semi-finished slab chosen to define subject imports and their "like products" for the domestic industry, some causal link may be found between the import increase in a certain part of the subject imports (i.e., semi-finished slab), and a decrease in sales in that part of the domestically produced "like products" (again, i.e., semi-finished slab). However, this finding cannot justify the imposition of safeguard measures on imports of semi-finished slab plus all CCFRS products. While safeguard measures might be justified with respect to imports of semi-finished slab products, domestic producers of all flat-rolled steel products would enjoy protection from import competition without justification.924


    42. Korea argues that competition and the analysis of competition is fundamental to a proper safeguards investigation.925 Korea submits that "competition" has been used to analyse like product under the Agreement on Textiles and Clothing ("ATC") by the Appellate Body in US – Cotton Yarn. Japan, Korea and New Zealand recall that the Appellate Body specifically relied on its prior findings in the GATT 1994 Article III context when it indicated that "like" is a subset of "directly competitive".926 927 Japan adds that the Appellate Body dismissed the United States' argument in that case that the panel erroneously relied on Korea – Alcoholic Beverages, using the same "different provision and different agreement" argument it espouses here.928 929 Given that Article 6 of the ATC essentially has the same purpose as the Agreement on Safeguards, it is clear that the domestic industry must be of narrower scope under the Agreement on Safeguards when an authority relies solely on the words "like product", as the USITC did in this case. Brazil, Japan and New Zealand stress that, more importantly, in US – Cotton Yarn, the Appellate Body also clearly established the importance of the competitive relationship between imported and domestic products in discerning whether they are like or directly competitive with one another by stating that: "[l]ike products are, necessarily, in the highest degree of competitive relationship in the marketplace".930 931 New Zealand submits that excluding any consideration of competitive relationships from a determination of likeness is plainly absurd. Competition is, thus, at the heart of a likeness determination under the Agreement on Safeguards, and the Appellate Body jurisprudence on the meaning of "like product" under GATT Article III is of direct relevance.932


    43. Japan also submits that, as regards the existence of precedents to help discern the proper treatment of "like product" under the Agreement on Safeguards, there is relevant jurisprudence in US – Lamb and US Cotton Yarn, and the jurisprudence concerning like product delineations under Article III of the GATT 1994 is also relevant. The central purpose of the "like or directly competitive" product analysis is to define appropriately the domestic industry whose performance is


      image

      923 Japan's first written submission, para. 79.

      924 Japan's first written submission, para. 81.

      925 Korea's second written submission, para. 41.

      926 Appellate Body Report, US – Cotton Yarn, paras. 95-97, footnote 68.

      927 Korea's second written submission, para. 35; New Zealand's second written submission, para. 3.35.

      928 Appellate Body Report, US – Cotton Yarn, paras. 21, 92-94.

      929 Japan's second written submission, para. 23.

      930 Appellate Body Report, US – Cotton Yarn, para. 97.

      931 Japan's second written submission, paras. 18-19; New Zealand's second written submission, para. 3.35; Brazil's second written submission, paras. 41-42.

      932 New Zealand's second written submission, para. 3.35.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 143


      allegedly hampered by competition with imported products subject to the investigation. This competitive relationship between the domestic industry's product and imports must exist regardless of whether the domestic product is deemed "like" or "directly competitive" with the imported product. Absent this tight competitive nexus – which is required by both the "like" and the "directly competitive" standards – it makes no sense to blame imports for whatever problems the domestic industry may be experiencing.933 US – Lamb underlines the critical importance of the competitive dynamic that must exist between imported and domestic products, including between products that exist along a continuum of production processes. Regardless of whether the products are produced using processes that happen to be vertically integrated, if they do not compete with each other in the market place, their combination into a single grouping renders any findings by a competent authority regarding increased imports, serious injury, or causation null and void. Nor could an authority, as a result, devise a proper remedy as required under Article 5.1. Therefore, an improper definition of the domestic industry makes it impossible to ensure that the wrong industry is not protected.934


    44. The United States responds that substitutability is not one of the traditional factors considered by the USITC in conducting it analysis of whether there are clear dividing lines between domestic products in order to define like product(s).935 Nor has substitutability been one of the criteria suggested for the like product analysis in the context of dispute settlement proceedings regarding other covered agreements.936 The complainants' references to US – Cotton Yarn as relevant to the like product definition fail to recognize the Appellate Body's statement that "there is no disagreement ... that yarn imported from Pakistan and yarn produced by the producers of the United States ... are like products. ... It is, therefore, not necessary for us to address the meaning of the term 'like products' for the purposes of this appeal".937 The issue in US – Cotton Yarn was whether imported and domestic products determined to be like could be determined not to be directly competitive.938 There clearly is a competitive relationship between the imported and domestic products. The complainants have not disputed, and neither did the parties in the underlying investigation, that the imported and domestic products generally consist of the same types of steel, are interchangeable and thus compete with each other. Moreover, within any defined like product and the corresponding specific imported product there exists a range or continuum of goods of different sizes, grades, or stages of processing. While goods along the continuum share identical or similar factors, individual items at the extremes of the continuum may not be as similar or substitutable.939 Each like product definition must be based on the facts of the particular case and as the Appellate Body has stated, "the adoption of a particular framework to aid in the examination of evidence does not dissolve the duty or the need to examine, in each case, all of the pertinent evidence".940 The United States argues that the methodology used by the USITC is unbiased and objective. Neither Article 2 nor any other provision in the Agreement on


      image

      933 Japan's second written submission, paras. 5-6.

      934 Japan's second written submission, para. 14.

      935 The USITC has considered substitutability between products to be a factor it would consider if it made its definition(s) on the basis of a directly competitive product analysis.

      936 Border Tax Adjustments, para. 18; quoted in part in Appellate Body Report, Japan – Alcoholic Beverages II, p. 20.

      937 Appellate Body Report, US – Cotton Yarn, para. 89.

      938 The terminology in the safeguard provision of the Agreement on Textiles and Clothing ("ATC") is different, i.e., "like and/or directly competitive products" rather than the "like or directly competitive products" language in the Agreement on Safeguards. Based on this different terminology and the findings of the underlying investigation, the Appellate Body in US – Cotton Yarn rejected a finding that a product could be part

      of the like product definition but then defined out as not directly competitive and thus not included in the definition of the domestic industry. Appellate Body Report, US – Cotton Yarn, para. 105.

      939 Moreover, goods within a single tariff line consist of a range of items as demonstrated most clearly by requests by some complainants for like products to be defined more narrowly than by tariff line.

      940 Appellate Body Report, EC – Asbestos, para. 102.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 144


      Safeguards sets forth the factors or the order that the competent authority must consider in identifying the imported product that is like or directly competitive with the domestic product.941


    45. Brazil further argues that the USITC emphasized that its mandate went beyond narrow product categories and the facilities producing those products when it talked about "serious injury to the productive resources employed in the divisions or plants in which the article in question is produced".942 Thus, the USITC viewed its mandate under Section 201 to protect not only against the effects of specific imported products on the United States' industry producing those products, but also to protect more broadly the "divisions" and "plants" in which the imported product (and other products) is produced. This approach by the USITC reinforced the problems presented by the nature of the request and ultimately led to industry definitions which were facility based (i.e. taking into account vertical integration and continuous lines of production) rather than product based.943 Within the CCFRS category, tin mill products and corrosion-resistant products both use a cold-rolled substrate.944 Tin mill products are coated with tin or chromium; corrosion-resistant products are coated with zinc or zinc-aluminum alloys.945 Yet, they were treated as separate like products, with all commissioners treating corrosion-resistant products as part of the larger CCFRS product category, and four commissioners treating tin mill products as its own separate like product. If anything, it would make more sense to consider tin mill and corrosion-resistant products as a single like product given their physical characteristics, their location in the production chain, and their sometimes common treatment in the HS. However, the USITC made the odd leap to consider, effectively, slab and plate to be more comparable to corrosion-resistant steel than tin mill products. The USITC's logic led to other bizarre results and inconsistencies similar to the tin mill product example.946 947


    46. In particular, the United States stresses that the complainants' arguments that the USITC should have defined the various like products using the same factors and with the same results as it had in anti-dumping and countervailing duty investigations involving steel fails to recognize that those definitions (as they are in a safeguard investigation) are dependent on the imports subject to that particular investigation. Contrary to the complainants' allegations, the USITC had no obligation nor reason to explain why its like product definitions in the instant safeguard action based on a different type of trade remedy investigation, with a very different scope of subject imports and a different record, were not the same as the various decisions in other types of trade remedy investigations that were based on different subject imports and different underlying facts.948


      1. Relevance of like product definitions used in the anti-dumping and countervailing duty contexts


    47. Korea, Japan, New Zealand and Brazil point out that in this case, the USITC substantially departed from the traditional like product factors it had used in the safeguards and anti-dumping or countervailing duty context to define like products, stating that the concept of industry may be more


      image

      941 United States' second written submission, paras. 71-73.

      942 USITC Report, p. 30.

      943 Brazil's written reply to Panel question 17 at the second substantive meeting.

      944 USITC Report, Vol. II, at FLAT-2 (Exhibit CC-6).

      945 Ibid.

      946 For example, despite its determination that carbon and alloy steel slab are "like" finished carbon and

      alloy flat products, the USITC elected to treat stainless steel slab and finished stainless flat products as separate like products. USITC Report, Vol. I, pp. 193-194 (Exhibit CC-6). Similarly, the USITC treated semifinished long products – billets – as a separate like product, distinct from finished long products. Ibid., pp. 82-83.

      947 Brazil's first written submission, paras. 113-114.

      948 United States' first written submission, para. 106.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 145


      broadly defined than in anti-dumping and countervailing duty cases.949 In light of the stated purpose of Section 201, the USITC expanded its like product consideration, making "both the productive facilities and processes and the markets for these products … [a] … fundamental concern in defining the scope of the domestic industry".950 The result was not only inconsistent with the USITC's past practice, it was also erroneous and inconsistent with the USITC's findings on other products simultaneously under investigation in this case.951


    48. The United States stresses that the complainants' arguments that the USITC should have defined the various like products using the same factors and with the same results as it had in anti- dumping and countervailing duty investigations involving steel fails to recognize that those definitions (as they are in a safeguards investigation) are dependent on the imports subject to that particular investigation. Contrary to the complainants' allegations, the USITC had no obligation nor reason to explain why its like product definitions in the instant safeguard action based on a different type of trade remedy investigation, with a very different scope of subject imports and a different record, were not the same as the various decisions in other types of trade remedy investigations that were based on different subject imports and different underlying facts.952


        1. Comparison of imported product and domestic like product


    49. China, the European Communities, Norway and Switzerland assert that the United States has not only acted inconsistently with the first step of properly identifying the like or directly competitive product, but that it also acted inconsistently with its obligations under Articles 2.1 and 4.1(c) of the Agreement on Safeguards at the second step, i.e., the definition of the "domestic industry that produces like or directly competitive products".953


    50. In the view of the European Communities, China and Switzerland, the USITC fails to compare imported and domestic products to establish whether they are like or directly competitive. Instead, the USITC endeavoured to explain at great length why its groups of domestically produced products form one single like domestic product.954 For example, the USITC considered whether to analyse specific types of CCFRS separately or as a whole for the purpose of the domestic industry definition.955 The USITC declined to identify specific products by essentially pointing to: (i) physical properties such as a common metallurgical base of those products956; (ii) the "single common production base"957 (iii) and the common end-use of all products in the automobile and construction industry.958 959



      image

      949 USITC Report, Vol. I, p. 30 (footnotes omitted) (Exhibit CC-6).

      950 USITC Report, Vol. I, pp. 30-31 (ExhibitCC-6).

      951 Korea's first written submission, paras. 34-36; Japan's first written submission, paras. 129-148; New Zealand's first written submission, paras. 4.68-4.70; Brazil's first written submission, para. 99, 115.

      952 United States' first written submission, para. 106.

      953 European Communities' first written submission, para. 222; China's first written submission, para. 172; Switzerland's first written submission, para. 200; Norway's first written submission, para. 213.

      954 European Communities' first written submission, paras. 224-225; China's first written submission, paras. 174-175; Switzerland's first written submission, paras. 202-203.

      955 USITC Report, Vol. I, p. 37.

      956 USITC Report, Vol. I, pp. 37-38.

      957 USITC Report, Vol. I, pp. 30, 31, 37.

      958 USITC Report, Vol. I, pp. 43-44.

      959 European Communities' first written submission, para. 231; China's first written submission, para. 179.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 146


    51. The European Communities asserts that the complainants have made a prima facie case that the USITC has only justified the bundling of certain domestic producers by arguing that the products they produce are like between themselves, but failed to carry out the essential comparison between imported and domestic products.960 Japan submits that comparing the imported and domestic products is indeed the essential step in the analysis to ensure the proper scope of the domestic like product. The United States failed to do this. This failure incidentally resulted in the fact that the United States has also failed to justify the bundling of domestic products together. Its only justification for this bundling is the integration of certain production processes, which we know is irrelevant to the question of whether products are distinct or not. Hence, the domestic products are just as distinct among themselves as they are when compared "across borders".961


    52. The United States responds that the USITC found that the evidence demonstrated that domestic and imported steel consisted of mainly the same types of steel, and thus that imported steel competes with corresponding domestic steel. The European Communities does not contend that the facts do not support the finding that imported and domestic steel are generally the same types of steel. The parties, including steel industry experts representing both domestic and foreign producers and importers, did not dispute these findings in the underlying proceeding. While there is no challenge to the USITC's factual findings, the European Communities seeks to misrepresent the USITC's approach. The USITC considers whether there are domestic products that are like the subject imports. This analysis considers whether subject imports and domestic products generally share similar physical properties, uses, production processes and marketing channels. This comparison shows whether domestic and imported CCFRS are similar and whether they are interchangeable, and as such whether they compete with each other. For example, in its CCFRS analysis, the USITC found that the evidence showed that imported CCFRS consists mainly of the same range of carbon steel as the domestic CCFRS.962 The USITC found that imported and domestic CCFRS share the same basic physical attributes and are generally interchangeable, have similar uses with the same metallurgic composition, thickness, width, and amount of processing, generally were not produced by significantly different production processes, and overlap in the marketing channels for domestic and imported CCFRS. The domestic like product, CCFRS, is like and coextensive with the imported CCFRS used in the USITC's injury analysis. The USITC performed this comparison between imported and domestic products in making its like product determination, in its report.963


    53. The European Communities and Switzerland submit that in respect of five products, i.e. hot- rolled bar, cold-finished bar, rebar, welded pipe, as well as carbon and alloy fittings, the USITC did not make any finding that imported and domestically produced are alike.964 965

    54. In response to a related question of the European Communities966, the United States refers to the passages in the USITC Report containing findings that domestic and imported products were like


      image

      960 European Communities' second written submission, para. 239.

      961 Japan's written reply to Panel question No. 18 at the second substantive meeting.

      962 The complainants do not take issue with the USITC's findings regarding this comparison nor that the evidence showed that domestic and imported CCFRS consisted mainly of the same range of carbon steel. European Communities' first written submission, paras. 223-233.

      963 United States' written reply to Panel question 18 at the second substantive meeting.

      964 USITC Report, Vol. I, pp. 81-83, 147-157.

      965 European Communities' first written submission, para. 226; Switzerland's first written submission, para. 204.

      966 European Communities' question No. 4 to the United States at the first substantive meeting.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 147


      for each of the products.967 According to the United States, the complainants have not disputed, nor did parties to the underlying proceeding, that the imported product and domestic like product generally are of the same types of steel and are like.968

    55. The European Communities responds that for the "products" CCFRS969, tin mill products, stainless steel bar, stainless steel rod, and stainless steel wire970, the USITC provided some cursory assertions, e.g., that "domestic CCFRS is like the imported CCFRS", because in "terms of physical properties, imported and domestic CCFRS share the same basic physical attributes and are generally interchangeable". These assertions are certainly not a sufficient analysis, in particular because there is no comparison addressing physical characteristics, common end-uses, consumer perceptions and tariff classifications.971


    56. In particular, the European Communities and China submit that the USITC did not establish that all components of the bundle of domestic products are like or directly competitive with components of the groups of imported articles. For example, although the USITC explicitly acknowledged the existence of different types of CCFRS, it neither showed that, e.g., domestically produced slabs are like imported cold-rolled sheets, nor that domestically produced hot-rolled steel is like imported coated steel. Similarly it is not shown that imported flanges are like domestically produced fittings, although the USITC explicitly recognized the heterogeneity of FFTJ, and that this "category contains a mix of products".972 Instead, the USITC Report erroneously compares domestic products within each of the product groupings between themselves. 973


    57. Norway adds that the methodology employed by the USITC – according to which each domestically produced article in principle is "like" at least one of the many different imported products within the same bundle – would also have allowed it to consider "steel" to be one specific product, and all domestically produced steel to be the "like product", and thereafter to slap a tariff on stainless steel wire for alleged injury to slab-producers. This methodology allows for great possibility of "cross-fertilization" of increases in imports of one product to be counted against alleged injury to producers of a product that is "like" an imported product not subject to increase, thus clearly contradicting the explicit statements by the Appellate Body in US – Lamb.974 975


    58. The United States contends that the complainants do not take issue with the USITC's findings regarding the comparison nor that the evidence showed that domestic and imported CCFRS consisted mainly of the same range of carbon steel.976


    59. In response, the European Communities insists that it challenges the United States' like product determination on the basis of a vitiated comparison (or no comparison at all) between bundles


      image

      967 USITC Report, p. 79 for hot-rolled bar, cold-finished bar and rebar; pp. 147 and 158 for certain welded pipe; pp. 147 and 175 for carbon and alloy fittings; pp. 36-37 for certain carbon flat-rolled steel; p. 49 for tin mill products; p. 190 for stainless steel bar, stainless steel rod and stainless steel wire.

      968 United States' written reply to the European Communities' question No. 4 to the United States at the first substantive meeting.

      969 USITC Report, Vol. I, p. 36.

      970 USITC Report, Vol. I, pp. 49, 196, 198 and 201.

      971 European Communities' first written submission, paras. 227-228.

      972 USITC Report, Vol. I, pp. 175 and 179.

      973 European Communities' first written submission, paras. 229-230, 261; China's first written submission, paras. 177-179.

      974 Appellate Body Report, US – Lamb, para 86.

      975 Norway's second written submission, paras. 66-68.

      976 United States' first written submission, para. 118, footnote 142.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 148


      of imported and domestic slab, plate, hot-rolled sheet, cold-rolled sheet and coated sheet claimed to be like. For all products but CCFRS and tin mill, the USITC Report contains, if at all, the tautological assertion that "there are ten domestic industries producing articles like and corresponding to the similar imported articles subject to investigation within the long products category".977 Only for CCFRS products and tin mill, the USITC saw itself in a position to at least write a few more lines, however, as demonstrated by the complainants these do not sufficiently establish likeness because these only compare "types" or "ranges" of products according to a "no clear dividing line" test without any basis in the Agreement on Safeguards.978


        1. "Directly competitive" products


    60. The United States submits in response to a question from the Panel that if, despite the United States' analysis of like products, the Panel finds that its findings are consistent with a directly competitive product analysis, it cannot find that the USITC's findings are inconsistent with the Agreement on Safeguards.979


    61. The European Communities responds that the United States cannot rescue its flawed domestic industry definition by relying on a "directly competitive product" determination that was somewhat implicit in its like product determination.980 A competent authority that only makes a finding on "like product" has not given a reasoned and adequate explanation of its "like product" determination if the country concerned later argues that the products of the domestic industry are "directly competitive" even if not like. The Panel cannot conduct a de novo investigation and examine whether the competent authority could have considered the products "directly competitive". The Panel is, therefore, confined to reviewing the determinations actually made in the report.981 Japan, Korea, China, Norway, New Zealand and Brazil agree with the European Communities. Korea and Norway add that the USITC did not, and could not have made, a de facto "directly competitive" analysis because the USITC explicitly said that it was not doing so. The USITC stated as follows: "Having identified domestic producers of an article that is like the imported article, we are not required to, and do not in this case, look further for an industry producing articles that are directly competitive but not like the imported article".982 983


    62. The United States responds that, contrary to certain allegations, the United States has not shifted its position to urge this Panel to approach its findings on the basis of a directly competitive product analysis. The USITC conducted its analysis and made its findings on the basis of a like product analysis and did not, de facto, carry out a directly competitive product analysis. Moreover, the USITC applied a like product analysis in this investigation consistent with the United States'


      image

      977 USITC Report, p. 79 for hot bar, cold bar and rebar. The same applies to the determinations for welded pipes, pp. 147 and 157 and fittings, pp. 147 and 175. The European Communities argues that for SS bar, rod and wire, the United States' reference to USITC Report, p. 190 cuts and pastes the same meaningless language.

      978 European Communities' written reply to Panel question No. 146 at the first substantive meeting (for flat products); the European Communities refers to Norway's submission for tin mill products; European

      Communities' written reply to Panel question No. 9 at the second substantive meeting.

      979 United States' written reply to Panel question No. 65 at the first substantive meeting.

      980 United States' written reply to Panel question No. 65 at the first substantive meeting. See also United States' reply to Panel's question No. 67 at the first substantive meeting.

      981 European Communities' written reply to Panel question No. 21 at the second substantive meeting.

      982 USITC Report, Vol. I, p. 45, footnote 139 (flat-rolled) (Exhibit CC-6); USITC Report, Vol. I,

      p. 147, n. 893 (pipe and tube) (Exhibit CC-6).

      983 Japan's, Korea's, China's, Norway's, New Zealand's and Brazil's written replies to Panel question No. 21 at the second substantive meeting.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 149


      obligations under the Agreement on Safeguards. However, if the Panel finds that the USITC analysis

      • which the United States considers as defining a "like product" under both the Agreement on Safeguards and United States law – actually falls within the realm of "directly competitive product" for Agreement on Safeguards purposes, its characterization as a "like product" analysis would not affect its consistency with WTO rules.984


          1. Identification of domestic producers


    63. The European Communities and Norway contend that, at least for some product groups, the USITC did not provide an explanation for how it determined the producers of the products it had grouped together, contrary to Article 3.1 of the Agreement on Safeguards. For example, with respect to the CCFRS group, the USITC only noted that "***% of total production of certain carbon flat- rolled steel [were] made by producers of at least 4 of the 5 types of certain carbon flat-rolled steel" but unjustifiably confidentialized the data relating to domestic producers.985 On this basis, it is not possible to assess the accuracy of the determination of the producers with respect to the quantitative element.986


        1. Burden of proof


    64. The United States contends that the complainants have not met their burden of making a prima facie case that the United States' measure is inconsistent with the Agreement on Safeguards because of the manner in which like products were defined.987


    65. According to the United States, the complainants do not specifically challenge six of the ten like product determinations made by the USITC in the ten safeguards measures at issue. The failure of the complainants to specifically challenge the majority of the like product determinations made by the USITC by itself suggests that the USITC's methodology is not inconsistent with the requirements of the Agreement on Safeguards. Indeed, those like product factors considered by the USITC are entirely consistent with the United States' obligations under the Agreement on Safeguards. As a result, the only basis on which the Panel can consider whether complainants have met their burden of proof with respect to like product is to examine whether the specific factual findings made by the USITC with respect to each of the contested like product findings cannot support a finding of "likeness" consistent with the ordinary meaning of the Agreement on Safeguards, in its context and in light of the object and purpose of the agreement. In this regard, the United States recalls that the Panel is not to engage in a de novo review but, rather, to engage in a review that is limited to an objective assessment, pursuant to Article 11 of the DSU, of whether the domestic authority has considered all relevant facts, including an examination of each factor listed in Article 4.2(a), of whether the published report on the investigation contains adequate explanation of how the facts support the determination made, and consequently of whether the determination made is consistent with the United States' obligations under the Agreement on Safeguards.988 989


    66. The European Communities submits that the United States wrongly asserts that only four of the ten "like product determinations" have been specifically challenged and that this suggests that the



      image

      984 United States' written reply to Panel question No. 21 at the second substantive meeting.

      985 USITC Report, Vol. I., pp. 39 and 50.

      986 European Communities' first written submission, para. 256; Norway's first written submission, paras. 233-234.

      987 United States' second written submission, para. 73.

      988 Panel Report, Argentina – Footwear (EC), para. 8.124.

      989 United States' second written submission, paras. 26-29.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 150


      other six are sound.990 The methodologies used were inappropriate and therefore all the determinations are unsound. For example, the "subject imports" were selected on the basis of unknown criteria (relating apparently to whether there were "potential import problems")991, and then they were divided into bundles that waxed and waned over the course of the proceeding and according to the Commissioner concerned. In particular, the product bundles are too broad and contain gaps.992 For none of the product bundles, the USITC carried out the essential comparison between imported and domestic products (see section E.3 above).


    67. Korea asserts that if a prima facie case has been made that the United States determination with respect to the definition of the like product is in violation of the Agreement on Safeguards, it is the United States that has the obligation to demonstrate that its like product determination is in accord with the requirements of the Agreement regardless of whether it agrees with the like product formulations of complainants.993


        1. Measure-specific argumentation


      (a) CCFRS


      1. General


    68. The European Communities and China claim that the USITC's definition of the domestic industry with respect to CCFRS is inconsistent with Articles 2.1 and 4.1(c) of the Agreement on Safeguards.994 Japan and Brazil submit that because the USITC did not find each of the five CCFRS products to be "like" the imports under investigation, and did not even try to establish whether they were "directly competitive" in relation to the imports under investigation, its determination to combine all CCFRS products into a single like product and its consequent decision to define the domestic industry by such products is inconsistent with Article XIX:1 of GATT 1994 and Articles 2.1 and 4.1(c) of the Agreement on Safeguards.995


    69. The United States asserts that the USITC found that domestic CCFRS is like the corresponding imported CCFRS that is subject to this investigation and defined CCFRS as a single like product. The USITC considered the facts using long established factors and looked for clear dividing lines among the various types of domestic CCFRS corresponding to imported CCFRS subject to this investigation. The methodology employed by the USITC was unbiased and objective based on data analysed using a long-standing and transparent methodology and factors. The USITC's definition of certain CCFRS as a single like product is consistent with Articles 2.1 and 4.1 of the Agreement on Safeguards and should not be disturbed by the Panel.996


    70. New Zealand submits that the USITC divided the request category CCFRS into three categories, namely CCFRS, GOES, and tin mill products. By doing this, it effectively also divided


      image

      990 United States' second written submission, para. 28.

      991 United States' written reply to the European Communities' question No. 2 at the first substantive meeting, para. 5.

      992 European Communities' second oral statement "Scope and Standard of Review" on behalf of the complainants, para. 18.

      993 Korea's second written submission, para. 64; Norway makes a similar argument with regard to the identification of domestic producers of the "like product" in Norway's second written submission, paras. 70-71.

      994 European Communities' first written submission, para. 236; China's first written submission,

      para. 182.

      995 Japan's first written submission, para. 124; Brazil's first written submission, para. 112.

      996 United States' first written submission, para. 142.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 151


      the imported "product" into three separate product categories. It did this on the basis that there were separate domestic industries producing these products.997 In aggregating a range of different products into the category of CCFRS, even though the requisite degree of likeness between these products and the imported products could not be demonstrated, the United States incorrectly defined the "domestic industry that produces like … products". The "bundling" approach employed by the USITC in this case made it impossible to apply the crucial "like product" requirement, which is essential to the identification of the relevant domestic industry, in the way intended by the Agreement on Safeguards.998 In effect, the United States determines "like product" by reference to the "domestic industry", rather than determining the "domestic industry" by reference to its production of a "like product". It accepted as an imported "product" a range of different and distinguishable CCFRS and then defined its "domestic industry" by reference to the producers of that broad range of distinct products, notwithstanding the fact that the products within that range were not like each other. The result was that each of the products included within the domestic product category were equally not like all of the products within the imported product category.999


    71. New Zealand further argues that, while the category of domestically produced CCFRS may be the same as the category of imported CCFRS, each of the products within that category is very different. CCFRS is not a single identifiable product – it is a product category comprised of four distinct finished products, finished flat steel plate, hot-rolled steel, cold-rolled steel, and coated steel, together with the semi-finished product slab. Instead of focussing on the differences or similarities amongst these products, the USITC focussed on commonalties within the industry producing the range of CCFRS products. In effect, the USITC did exactly what the Appellate Body in US – Lamb said should not be done – it focussed on the producers rather than the product. Although the jurisprudence indicates that the concept of "like product" in the Agreement on Safeguards is to be construed narrowly, the USITC argues that the appropriate policy is to provide the widest possible blanket of protection to the relevant industry.1000 1001 New Zealand adds that in this case, imports of slab, a semi-finished input product, were apportioned part of the blame for alleged serious injury to producers of a highly finished end-product, coated steel. This result is particularly perverse since the only increase in slab imports during the period of investigation was from 1998-19991002 which coincided with a period of coated steel profits.1003 This illustrates the point that there can be no causal relationship between increased imports of one product and serious injury to an industry that produces products that are unlike the imported products, but this is entirely concealed by the bundling approach. The USITC does not demonstrate what would be the only relevant point, that is, whether increased imports of slab caused serious injury to the domestic producers of the like product – slab – because this, too, is concealed by the bundling approach.1004


    72. In the view of the United States, the USITC found that domestic CCFRS is like the corresponding imported CCFRS that is subject to this investigation and defined CCFRS as a single like product. The Agreement on Safeguards includes no definition of "like product" nor addresses what factors to consider in determining whether to define separate like products and corresponding domestic industries. The USITC's methodology and definition of CCFRS as a single like product was adequate, reasoned and reasonable explanations were provided. The USITC started this analysis with



      image

      997 New Zealand's second written submission, para. 3.30.

      998 New Zealand's second written submission, para. 3.21.

      999 New Zealand's first written submission, paras. 4.49-4.50.

      1000 USITC Report, Vol. I, p 30.

      1001 New Zealand's first written submission, paras. 4.53, 4.54 and 4.58.

      1002 New Zealand first written submission, para 4.92, Figures 12 and 13.

      1003 USITC Report, Vol. I, p 53.

      1004 New Zealand's second written submission, para. 3.23.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 152


      the range of steel broadly categorized as certain carbon and alloy flat products, all of which had been identified as imports subject to this investigation in the President's request (as well as the Senate Committee on Finance's request). After examining the evidence and conducting its analysis regarding the corresponding domestic CCFRS products, the USITC found clear dividing lines so as to define three separate like products from this category.1005 In comparing the domestic steel to the imported steel, the evidence indicated that imported CCFRS consists mainly of the same range of carbon steel as the domestic CCFRS.1006 1007


    73. Japan insists that no one in the world, whether now or 1993, thinks of CCFRS as a single product. The various CCFRS products do not comprise one authentic market. They are each distinct in their physical properties, use, customer perceptions, general tariff classifications, and even production processes. Japan submits that even the United States industry breaks down its marketing and pricing materials in the manner that Japan proposes. Plate is sold and marketed separately from hot-rolled, which is distinct from cold rolled, which is distinct yet again from corrosion resistant steel.1008


    74. The United States insists that the USITC applied its traditional factors in determining that there was no clear dividing line between types of CCFRS and defined such steel to constitute a single like product. The USITC found that CCFRS at various stages of processing shared certain basic physical properties, were interrelated, had common end-uses, were generally distributed through the same marketing channels, and were essentially made by the same production processes (at least at the initial stages). The USITC also recognized that there were some differences in physical properties and end-uses.1009 Since CCFRS in an earlier processed form is the feedstock for further processed CCFRS, all such steel is produced using the same production processes at the initial stages, with downstream steel merely employing additional stages of processing. All CCFRS is produced from slab, with the majority of such steel further processed into hot-rolled steel on hot strip or Steckel mills. Substantial quantities of earlier processed steel are internally transferred for production of further processed steel. The USITC found that this tends to blur product distinctions until the processing reaches its final stages since earlier stages of steel comprise feedstock for the next stage.1010 As part of its consideration of the manufacturing process (i.e., where and how it is made), the USITC also recognized that there is substantial commonality in production facilities and vertical integration in the industry. The USITC recognized that the interrelationship between the production processes and integration of the producers demonstrates that the market for each type of CCFRS is not isolated, but directly affected by the markets across the spectrum of all CCFRS.1011


      image

      1005 Four Commissioners found clear dividing lines so as to define three separate like products from this category and two Commissioners determined that this category was a single like product. Commissioners Okun, Hillman, Miller, and Koplan defined the following three separate like products: 1) certain carbon flat- rolled steel ("CCFRS"); 2) grain-oriented electrical steel ("GOES"); and 3) tin mill products. Commissioners Bragg and Devaney defined a single like product, carbon and alloy flat products (including slab, hot-rolled sheet and strip, cut-to-length plate, cold-rolled sheet and strip, corrosion resistant, grain oriented electrical steel, and tin mill products).

      1006 USITC Report, pp. 36-37.

      1007 United States' first written submission, paras. 116-118.

      1008 Japan's second written submission, para. 36.

      1009 United States' second written submission, paras. 77-78.

      1010 USITC Report, pp. 38-39.

      1011 United States' second written submission, paras. 79-82.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 153


      1. Like product criteria


      General


    75. China argues that, contrary to the WTO standard for likeness and its own stated methodology, the USITC did not establish the "likeness" of the five distinct downstream products (slabs, plate, hot- rolled steel, cold-rolled steel and coated steel) on the basis of their physical characteristics, end-uses, consumer-tastes and habits and tariff classification. The USITC essentially relied on commonalties in the chain of production and the end-users of all five types of flat steel – the two criteria that had already been ruled out by the Appellate Body.1012


    76. Similarly, Japan argues that the USITC's determination to conjoin the five products into a single "CCFRS" like product is inconsistent with Articles 2.1 and 4.1(c) of the Agreement on Safeguards and Article XIX:1 of GATT 1994. There are wide differences between subject imports and domestic products in terms of the factors identified by the Appellate Body for a determination on "like products": the products' physical properties, end-uses, consumer perceptions, and tariff classifications. It is evident from the USITC's own findings that semi-finished slab produced by domestic producers is not a like product of any of the imported corrosion-resistant steel, cold-rolled, hot-rolled, or plate products, because of the stark differences in terms of product properties and end- uses. Domestic corrosion-resistant steel products are not "like products" of any imported semi- finished slab, plate, hot-rolled, or cold-rolled products. Domestic cold-rolled products are not like products of any flat-rolled steel products except for cold-rolled products. The same is true for plate and hot-rolled. In a nutshell, imports and domestic products falling within the same category – semi- finished slab, plate, hot-rolled, cold-rolled, or corrosion-resistant steel products – might be "like products" of each other, but imports and domestic products that fall within different categories are definitely not "like" one another.1013


    77. Brazil argues that in expanding its like product consideration to "both the productive facilities and processes and the markets for these products"1014, the USITC determined that semi-finished slab and other major finished flat-rolled carbon and alloy steel products – plate, hot-rolled, cold-rolled, and corrosion-resistant steel sheet – constitute a single like product. Where the USITC had found in previous cases that each of these products was a distinct like product, these same products were now somehow one like product.1015 In fact, application of the USITC's own traditional like product factors

      • including most of those identified by the Appellate Body as appropriate to separate products from one another – to the information contained in its own report compels the conclusion that the various products within the CCFRS category are distinct from one another. The USITC's report admits, without thoughtful analysis, many distinctions, and the foreign producers in the case produced even more information that leads to the same conclusion.1016


    78. The United States asserts that many of the specific allegations raised by the complainants regarding the USITC's CCFRS like product definition are based on their erroneous interpretation of what factors the USITC was either "required or not permitted" to consider in making its like product definitions. The complainants can identify nothing in the Agreement on Safeguards addressing what



      image

      1012 China's first written submission, para. 181. 1013 Japan's first written submission, para. 117. 1014 USITC Report Vol. I at 30-31 (Exhibit CC-6).

      1015 Brazil's first written submission, paras. 99 and 115.

      1016 Joint Respondents' Prehearing Framework Brief (11 September 2001) (filed by the Law Firm of Willkie Farr & Gallagher) at 17-24 (Exhibit CC-50); Joint Respondents' Prehearing Brief on Slab (11 September 2001) (filed by the Law Firm of Willkie Farr & Gallagher) at 3-10 (Exhibit CC-51).

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 154


      factors may or may not be considered in determining like products. They instead assert that the USITC was bound to use the four factors suggested by the Working Party on Border Tax Adjustments. These factors, which were suggested for use in border tax adjustments, were for a different purpose, and the Appellate Body has recognized that "[n]o one approach to exercising judgement will be appropriate for all cases".1017 Thus, the USITC was not required to consider the four factors derived from the Working Party that are urged by the complainants.1018 The United States notes that the USITC's like product factors in a safeguards investigation include the three criteria on which all the parties agree (physical properties, uses, and customs treatment), and also focus on such other objective factors as the product's marketing channels and manufacturing process.1019


    79. New Zealand and Brazil note that three of the six factors relied upon by the USITC to support the grouping of CCFRS related to the issue of vertical integration and commonality of facilities linked to vertical integration. These are, in fact, the same factors as the continuous line of production and coincidence of economic interests at issue in US – Lamb. Japan concurs in this argument. The Appellate Body could not have stated more clearly that the fact that one product is an input to another product, even if there is no other use for the input product, is simply not relevant to the determination of whether products are like each other. Similarly, the Appellate Body was clear in its statement that a coincidence of economic interests between the producers of upstream and downstream products is also not relevant to the determination of whether products are like each other. That is, vertical integration and commonality of facilities are not relevant to the like product determination.1020 According to Brazil, the discussion of physical properties relates to the physical properties resulting from a stage prior to slab, steelmaking, and not to physical properties imparted as a result of production of the downstream products. Finally, end-use applications are very broadly defined, although there is an admission that the various products are not generally substitutable between the various products within CCFRS. There is, of course, no mention of the fact that slab is not sold at all to the automotive or construction industries (or, indeed, any industry other than steel) and a variety of other factors traditional to the United States' analysis (e.g., channels of distribution/marketing channels) are not addressed at all.1021


    80. The United States responds that as part of its consideration of the manufacturing process in this particular case (i.e., how and where it is made), the USITC recognized that the interrelatedness of CCFRS at different stages of processing resulted in substantial captive consumption, with a concomitant commonality of production facilities and vertical integration in the industry. This interrelationship between the production processes and integration of the producers demonstrated that distinctions in markets for each type of CCFRS were blurred, all types of CCFRS were directly affected by the markets for the whole spectrum of CCFRS, given that each type of CCFRS constituted the feedstock for the next processed stage of steel within the overall category. Considering the manufacturing processes of steel at various stages of processing, particularly the fact that they are feedstocks, is a "product-oriented" and not "producer-oriented" analysis, as alleged by the complainants.1022 The complainants' arguments fail to recognize that within almost every defined like product, and the corresponding imported product, there exist a range of goods of different sizes,



      image

      1017 Appellate Body Report, EC – Asbestos, para. 101. 1018 United States' first written submission, para. 124. 1019 United States' second written submission, para. 65.

      1020 New Zealand's second written submission, paras. 3.42-3.43; Brazil's second written submission,

      paras. 29-30; Japan's first written submission, paras. 121-122.

      1021 Brazil's second written submission, para. 16.

      1022 United States' second written submission, para. 70.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 155


      grades, or stages of processing. While goods along the continuum share similar or like factors, individual items at the end of the continuum may not be as similar.1023


    81. Korea submits that the United States' analysis was inadequate as demonstrated by the following table:1024


      Distinguishing Characteristics in Defining "Like Product"

      Flat-Rolled Products

      Physical Properties*

      Customs Treatment*

      Manufacturing Process*

      Uses*

      Marketing Channels*

      Slabs

      Semi-finished form, usually 4 in. Slab has a rectangular cross-

      section with width at least two times the thickness. Low fracture

      toughness and high porosity.

      Three distinct HTS headings: 7207.12, 7207.20 a nd 7224.90

      Molten steel cast by continuous casting. No rolling takes place.

      All further processed into hot-rolled steel or plate. No independent use.

      99.6% to end-users, 0.4% sold to distributors

      Plate

      It is made from slab, and it is

      Ten distinct HTS headings:

      Hot-rolled on Steckel mill or

      Designed for heavy industry uses,

      45.2% to e nd-users and 54.8% to

      thicker and stronger than the other

      7208.40, 7208.51, 7208. 52,

      reversing mill.

      such as bridgework, machine parts

      distributors

      flat products. It typically ranges

      7208.90, 7210.90, 7211. 13,

      (body or fram e), transmission

      between 3/16" to more than 1 foot

      7211.14, 7225.40, 7225. 50,

      towers and light poles, buildings,

      in thickness. Thick gauge and

      7226.91

      self-propelled machinery such as

      superior strength.

      cranes and bulldozers, railway cars,

      tanks, oceangoing ships and floor

      plate or formed into pipe, oilwell rigs

      and platforms.

      Hot-Rolled

      >2mm in thickness. The least

      Sixteen distinct HTS headings:

      Hot-rolled in hot-strip mill or Steckel

      For cold-rolling/galvanizing /coatin g;

      60% to end users, 40% to

      refined steel sheet product.

      7208.10, 7208.25, 7208. 26,

      mill.

      Formed and welded to make pipe;

      distributors

      Thinner and weaker than plate, but

      7208.27, 7208.36, 7208. 37,

      Cut to length for sheet; HR sheet is

      heavier and less smooth than cold-

      7208.38, 7208.39, 7208. 40,

      sold for uses where surface finish

      rolled. Lower quality tha n bo th cold-

      7208.53, 7208.54, 7211. 14,

      and light weight are not crucial,

      rolled and coated.

      7211.19, 7225.30, 7225. 40, and

      such as structural and internal parts

      7226.91

      of autos and appliances.

      Cold-Rolled

      25-90 % reduction in thickness of

      Sixteen distinct HTS headings:

      After being hot-rolled on strip mill

      Feedstock for corrosion-resistant

      71.3% to end users and 28.7% to

      hot-rolled to under 2mm ; Special

      7209.15, 7209.16, 7209. 17,

      a nd undergoing other processing

      steel, tin mill, and GOES. CR sheet

      distributors

      mechanical properties or surface

      7209.18, 7209.25, 7209. 26,

      similar to hot-rolled, it is finished on

      sold for applications where surface

      texture. Thin-gauge, smooth

      7209.27, 7209.28, 7209. 90,

      a cold-reduction mill.

      finish and light we ight are important

      surface and high strength to weight

      7211.23, 7211.29, 7211. 90,

      consideration (panels in electrical

      ratio, created by the cold-rolling

      7225.19, 7225.50, 7226. 19 and

      equipm ent and appliances, utensils,

      process.

      7226.92

      cutting tools, cutlery, seat belt

      components). Also sold for

      unexposed auto body parts such as

      automotive transmission

      components.

      Coated

      Has metallic or non-metallic

      Nineteen distinct HTS headings:

      Electro-galvanizing or hot-dip

      Used mostly in applications

      64.3% to e nd-users and 35.7% to

      coating; coated, clad or plated with

      7210.20, 7210.30, 7210. 41,

      galvanizing. There are seven

      requiring protection against the

      distributors

      metals and alloys for corrosion-

      7210.49, 7210.61, 7210. 69,

      alternative processes fo applying

      weath er and other corroding agents

      resistance and improved

      7210.70, 7210.90, 7212. 20,

      coatings.

      (Auto parts such as mufflers and

      aesthetics.

      7212.30, 7212.40, 7212. 50,

      trunk lids, construction uses such

      7212.60, 7225.91, 7225. 92,

      as roofing and siding, garbage

      7225.99, 7226.93, 7226. 94 and

      cans, storage tanks and buildin g

      7226.99

      products, gasoline tanks, chem ical

      containers, oil filters, television

      chassis, highway equipment and

      agricultural buildings and

      equipm ent.)

      * Identified by the ITC as a distinguishing factor the it takes into account when determining what constitutes the "like product." ITC Determination, Views on Injury of the Commission, Vol. I, p. 30. (Exhibit CC-6) Note: HTS classifications in bold font indicate the only 5 HTS classifications that are shared among the 55 distinct HTS classifications.

      Sources: ITC cites are all found in Exhibit CC-6.

      Slab: ITC at 39 and at FLAT-1 (physical characteristics), at FLAT-1 (customs treatment), at 40 (manufacturing processes) at 40 and OVERVIEW-13 (end uses); and at 44 (marketing channels). Also, First Submission of Japan, at 41 (physical characteristics and end uses).

      Plate: ITC at 41 and at FLAT-1 (physical characteristics), at FLAT-1 (customs treatment), at 40-41(manufacturing processes), at FLAT-I (end-uses ) and at 44 (marketing channels). Also, First Submission of Japan at 40 (physical characteristics).

      Hot-Rolled: ITC at 44 (physical characteristics), at FLAT-2 (customs treatment), at 40 (manufacturing processes) at 38 and FLAT-2 (end uses); and at 44 (marketing channels). Also First Submission of Japan at 39 (physical characteristics and end uses), at 40 (manufacturing process) and Exhibit CC-55 (Exhibit 2) (physical characteristics and end uses).

      Cold-Rolled: ITC at 41and 44 (physical characteristics), at FLAT-3 (customs treatment), at 41, OVERVIEW -13 and FLAT-2 (manufacturing process and end-uses), and at 44 (marketing channels). Also, First Submission of Japan at 39 (phsycial characteristics and end uses) and Exhibit CC-55 (Exhibit 2) (end uses).

      Coated: ITC at 42 and FLAT-3 (physical characteristics and manufacturing process); at FLAT-3 (end uses); and at 44 (marketing channels).

      Also, First Submission of Japan at 37 (physical characteristics) and at 39 (physical characteristics and end uses) and Exhibit CC-55 (Exhibit 2) (manufacturi ng proces s).


      Physical properties


    82. The European Communities and China submit that the USITC's finding that all five CCFRS products share "certain basic physical properties and are interrelated to a certain degree"1025 is


      image

      1023 United States' second written submission, para. 86.

      1024 Korea's second written submission, para. 68.

      1025 European Communities' second written submission, para. 275; USITC Report, Vol. I, p. 37.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 156


      untenable. It is only based on the commonplace of "a common metallurgical base".1026 However, the basic chemical combination of carbon and iron is not sufficient to demonstrate that these products share the same physical properties and may therefore be regarded as one specific product: indeed, the very same chemical mix in metallurgy can also be found in other steel product categories that have been classified as separate products (e.g., carbon and alloy long steel products or stainless steel).1027 In order to establish common physical properties, the USITC would have needed, in accordance with its own stated methodology and the general definition of physical properties under WTO law1028, to look at size, shape and texture of the five flat products. The USITC itself acknowledged significant difference in "thickness" (slabs are 4 inches while cold-rolled steel is reduced to size to below 2 mm) and other qualities (corrosion resistance and surface texture).1029 1030


    83. New Zealand adds that a proper comparison of physical properties cannot stop at the conclusion that CCFRS has a "common metallurgical base", a finding that the United States merely notes rather than tries to defend.1031 The fact that ice cream and cheese share certain compositional characteristics which mean they have more in common with each other than with a banana does not mean they are "like". Nor are sawn pine logs, planks, fence posts and furniture all "like" because they all started life as trees.1032


    84. The European Communities and China argue that the mere generalization that the five CCFRS share certain chemical characteristics that are also common to other steel product categories identified by the USITC was certainly not an adequate and reasoned explanation to dismiss the contradicting information in its own Report.1033


    85. Japan and Brazil insist that the products each have different physical properties. The USITC found, for instance, that cold-rolled steel differs from hot-rolled steel in terms of thickness, mechanical properties, and surface texture, while coated steel differs from cold-rolled steel in terms of its coating of zinc or other materials.1034 Further, New Zealand also stresses that each of the discrete products has very different properties in terms of their shape, thickness, degree of finishing and physical performance1035, and that these physical differences are fundamental. Clearly, slab is thicker and less refined than hot-rolled steel.1036 According to the European Communities and China, the USITC Report also explicitly concedes a lack of interchangeability between the five products where it holds that the "vertical nature of the relationship between CCFRS at different stages limits interchangeability between products".1037 In other words, these parties contend, the USITC admits



      image

      1026 USITC Report, Vol. I, pp. 37 and 38.

      1027 European Communities' first written submission, para. 237; China's first written submission, para. 184.

      1028 The USITC Report, Vol. I, p. 29 refers to "appearance, quality and texture". The Appellate Body in EC – Asbestos, para. 92 considered "composition, size, shape, texture, and possibly taste and smell" as relevant physical properties.

      1029 USITC Report, Vol. I, pp. 38-42 with detailed descriptions.

      1030 European Communities' first written submission, para. 238-239; China's first written submission, para. 185-186.

      1031 United States first written submission, para 119.

      1032 New Zealand's second written submission, para. 3.46.

      1033 European Communities' first written submission, para. 241; China's first written submission, para. 193.

      1034 Japan's first written submission, para. 117; Brazil's first written submission, para. 106.

      1035 The USITC highlights some such differences in "Appendix A" at pp. 9-10 of the USITC Report,

      Vol I.


      1036 New Zealand's first written submission, para. 4.59-4.60.

      1037 USITC Report, Vol. I, p. 44.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 157


      that part of the producers it wanted to group in a single domestic industry did not produce products interchangeable with the imported ones.1038 China submits that through the process of coating, cold- rolled steel becomes a different product that cannot meet the definition of a "like product". Indeed, the coating process is likely to alter the physical "properties" of the raw product (cold-rolled steel), through the addition of a new substance (metal or non-metallic substance). The new "properties" of the product are, in particular, illustrated by the fact that the coated product becomes corrosion- resistant.1039


    86. In the United States' view, the USITC found that imported and domestic CCFRS share the same basic physical attributes and are generally interchangeable, have similar uses with the same metallurgic composition, thickness, width, and amount of processing, generally do not employ significantly different production processes, and have an overlap in the marketing channels for domestic and imported CCFRS. The USITC found that the domestic article, CCFRS, is like the imported CCFRS. The USITC then applied its long established factors1040 in considering whether to analyse specific types of CCFRS separately or as a whole.1041 The USITC found that CCFRS at different stages of processing share certain basic physical properties and are interrelated to a certain degree.1042 Specifically, the USITC found that this steel has a common metallurgical base, with desired properties and essential characteristics embodied in the steel prior to the casting or semifinished stage.1043 The mix in metallurgy depends on the requirements of the end-use, whether the end-use is at the same or different stages of processing. Thus, the chemical content of such steel essentially is determined at the melt stage of processing with some reductions in carbon content possible through subsequent hydrogen annealing.1044


    87. The United States stresses that it is clear from the USITC's determination that it did not ignore evidence of differences in physical properties and end-uses and in fact generally acknowledged such evidence in its analysis. Rather, it is the complainants who ignore the evidence of the interrelationship of CCFRS at different stages of processing. The complainants fail to acknowledge, although they do not dispute, the fact that CCFRS at one stage of processing generally is feedstock for the next stage of processing, which tends to blur product distinctions until the processing reaches its final stages since earlier stages simply are feedstock for the next stage. This interrelationship between


      image

      1038 European Communities' first written submission, para. 240; China's first written submission, para. 187.

      1039 China's first written submission, para. 192.

      1040 The USITC traditionally has taken into account such factors as the physical properties of the product, its customs treatment, its manufacturing process (i.e., where and how it is made), its uses, and the marketing channels through which the product is sold in determining what constitutes the like product in a safeguards investigation. These are not statutory criteria and do not limit what factors the USITC may consider

      in making its determination. Appellate Body Report, EC – Asbestos, para. 102 (general criteria "are neither a treaty-mandated nor a closed list of criteria that will determine the legal characterization of products."). No single factor is dispositive and the weight given to each individual factor (and other relevant factors) will depend upon the facts in the particular case. The decision regarding the like or directly competitive article is a factual determination. The USITC traditionally has looked for clear dividing lines among possible products and has disregarded minor variations. USITC Report, p. 30.

      1041 USITC Report, pp. 36-45.

      1042 USITC Report, pp. 37-38.

      1043 The USITC found that all certain carbon flat-rolled steel originally is made of raw materials that include carbon and iron.

      1044 United States' second written submission, paras. 78-79.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 158


      CCFRS at different stages is "product-oriented" rather than "producer-oriented" and clearly was an important factor in the USITC's analysis and finding.1045 1046


    88. The European Communities submits that the United States nowhere rebuts the European Communities' argument concerning the different length, width and other qualities of the five different CCFRS products1047, but merely insists on the "same common metallurgical base". This analysis fails to take account of the significant physical differences between the five CCFRS products1048 and is also inconsistent with the investigation of the same criteria with respect to stainless steel products, where the USITC found that the sole metallurgical composition was not a sufficient element to define one single industry producing stainless steel products.1049 1050


      End-use


    89. Brazil submits that by making gross generalizations, the USITC found that "the primary end- use applications for commercial shipments of CCFRS are the automotive and construction industries".1051 1052 Japan and Brazil submit that, in fact, the products differ in terms of end-uses. For example, the USITC found that while hot-rolled steel and coated steel are both used in automotive applications, they are almost always used for different applications.1053 The overlap at issue is not of a type that may amount to the "likeness" of products under the Agreement on Safeguards. For example, the USITC found that hot-rolled and corrosion-resistant steel products are subject to similar demand trends in the automotive industry, but that they are not used for the same end-use. This finding indicates that the products are not substitutable with each other, but at most complementary. They therefore might not even be directly competitive, much less "like" one another.1054


    90. According to the European Communities, the USITC did not show that the five products have common end-uses. Its finding that "all types are used in the production of automobiles, albeit in different applications" and "for end-use applications in the construction industry" admits that the end- uses, i.e., application of each product, are very different. Some (e.g. coated cold rolled) is used for making certain car parts. Other products (e.g. slab) are used purely as an input good in the production of downstream steel products.


    91. The European Communities, Korea, China and New Zealand submit that, instead of showing common end-uses (e.g., body parts for automobiles), the USITC Report relies on common ultimate end-users of the various products in question (e.g., the automotive industry). This is not one of the criteria acknowledged by the Appellate Body to establish likeness. Moreover, a focus on end-users leads to some nonsensical results as any range of diverse input goods used by particular industries could be grafted into a single product category. Under the USITC's approach, also automotive leather and windscreen would also be like products to CCFRS and should have been analysed together.1055


      image

      1045 Appellate Body Report, US – Cotton Yarn, para. 86.

      1046 United States' first written submission, para. 127.

      1047 European Communities' second written submission, para. 275.

      1048 European Communities' first written submission, paras. 238 to 239.

      1049 USITC Report, Vol. I, pp. 190-205.

      1050 European Communities' second written submission, para. 275.

      1051 USITC Report, p. 43.

      1052 Japan's first written submission, paras. 119-120; Brazil's first written submission, paras. 100, 102 104, 105, 109.

      1053 Japan's first written submission, para. 117; Brazil's first written submission, para. 106.

      1054 Japan's first written submission, para. 120; Brazil's first written submission, para. 104.

      1055 European Communities' first written submission, para. 243; Korea first written submission, para. 54; China's first written submission, paras. 194-195; New Zealand's first written submission, para. 4.61.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 159


      The European Communities also argues that the USITC does not consistently apply its criterion of common end-users: had it done so, also certain tubular products that are used in the automotive industry (precision tubes to conduct forces) should have formed part of the CCFRS product category or vice versa, because they share common end-users.1056


    92. Korea further submits that the USITC record information confirms that the correct analysis – "end-use" – demonstrates that each of the five CCFRS products had different uses, and thus, is a separate like product.1057 Slab is internally consumed for downstream products, such as sheet, strip, and plate; plate is used for bridgework, machine parts (e.g., the body of the machine or its frame), transmission towers and light poles, buildings, self-propelled machinery such as cranes and bulldozers, railway cars, oceangoing ships, and floor plate or formed into pipe and oilwell rigs and platforms; hot-rolled steel is internally consumed to make cold-rolled and/or galvanized or other coated products, formed and welded to make pipe, or cut to length to produce discrete sheet, it is also used in the manufacture of structural parts of automobiles and appliances; cold-rolled steel is internally consumed to produce coated or tin products, and it is also used to make panels in electrical equipment and appliances, body parts for automobiles, where surface finish or strength-to-weight ratio is important but corrosion-resistance is not, auto transmission and seat belt components, utensils, cutting tools, cutlery. Coated steel is used for corrosion-resistant auto parts, garbage cans, storage tanks, building products, gas tanks, chemical containers, oil filters, television chassis, highway equipment, agricultural buildings, and equipment (guardrails, bridgedeck, signs).1058


    93. The United States submits that the USITC recognized that the interrelationship between the production processes and integration of the producers demonstrates that the market for each type of CCFRS is not isolated, but directly affected by the markets across the spectrum of types of CCFRS. The primary end-use applications for commercial shipments of CCFRS are the automotive and construction industries. Thus, the USITC found that all types of CCFRS are substantially affected by the collective demand of these two markets.1059 The USITC also found that the primary end-use applications for commercial shipments of CCFRS are the automotive and construction industries.1060 The USITC found that all types of CCFRS are substantially affected by the collective demand of these two markets.1061 The USITC also recognized that the vertical nature of the relationship between CCFRS at different stages may result in differences in uses between stages of CCFRS.1062 Nevertheless, the USITC found that the evidence demonstrated that in some situations, there may be some substitution for use between products from one stage to another, e.g., coated steel can be adapted for use in applications that typically use cold-rolled steel and vice versa, and hot-rolled has


      image

      1056 European Communities' first written submission, para. 243.

      1057 USITC Report, Vol. II, pp. FLAT-51-53 (Exhibit CC-6). See also Respondents' Joint Framework Brief, p. 22 (Exhibit CC-50).

      1058 Korea's first written submission, paras. 54, 57; source: Joint Respondents' Posthearing Brief, Flat-

      Rolled Steel, Inv. No. TA-201-73, Exhibit 2: "Application of Like Product Factors to Flat-Rolled Products", pp. 11-12 (Exhibit CC-55); see also USITC Report, Vol. II, pp. FLAT1-4 (Exhibit CC-6).

      1059 United States' first written submission, para. 122.

      1060 United States' first written submission, para. 122; USITC Report, pp. 43-44.

      1061 The USITC recognized that while hot-rolled steel may not be used in place of, or substituted for, a coated sheet in a car fender, all certain carbon flat-rolled steel is directly affected by the demand for

      automobiles, since all types are used in the production of automobiles, albeit in different applications. The USITC also found that similarly, but to a lesser extent, all types of such steel are used for end-use applications in the construction industries. Thus, all types of certain carbon flat-rolled steel are substantially affected by the collective demand of these two markets. USITC Report, pp. 43-44.

      1062 USITC Report, p. 44.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 160


      some limited interchangeability with cold-rolled steel.1063 1064 The United States also submitted that the complainants fail to recognize that within almost every defined like product, and the corresponding imported product, there exist a range of good of different sizes, grades, or stages of processing. While goods along the continuum share similar or like factors, individual items at the end of the continuum may not be as similar.1065


    94. Further, in the view of the United States, and contrary to the complainants' contentions, the USITC was not required to consider whether each type of CCFRS was substitutable with each other.1066 The complainants fail to recognize that the substitutability or competitive relationship was found to be relevant in the context of Article III of GATT 1994, as discussed by the Appellate Body in EC – Asbestos.1067 Protecting the competitive relationship between imports and domestic products is a purpose of Article III. However, it is not the purpose of the Agreement on Safeguards, whose purpose is to permit protection of a domestic industry under circumstances.1068 Thus, the competitive relationship or substitutability of domestic and imported products is not a necessary factor regarding the like product definition in the context of the purpose of the Agreement on Safeguards.1069 In considering uses for the types of CCFRS, the USITC recognized that similarity or interchangeability in uses were limited for CCFRS, as would be expected for feedstock or input products.1070 The complainants' attempts to construe this recognition regarding uses of feedstock as consideration of a substitutability factor by the USITC is misplaced.1071


    95. China responds by reaffirming that, in accordance with the Appellate Body ruling in US – Lamb, this fact is not relevant for the assessment of likeness. Moreover, China considers that this element rather underlines the fact that the different products included in the category CCFRS cannot share common end-uses.1072 The United States only acknowledges that there are no common end-uses for the various CCFRS products, by referring to markets and industries in general as end-uses applications. In particular, the fact that all types of CCFRS could be substantially affected by the collective demand of the automotive and construction industries does not demonstrate in any way that


      image

      1063 USITC Report, p. 44. Specifically, several US companies produce hot-rolled sheet in thicknesses (i.e., light-weight gauges) that have been more typically characteristic of and competitive with cold-rolled sheet. Although the overlap between hot-rolled steel and cold-rolled steel has traditionally been considered to begin at approximately 2 mm and thinner, improvements in hot-rolling have allowed mills to hot-roll below 2 mm. In addition, while cold-rolled steel generally is used as the feedstock for coated steel, coated hot-rolled sheet is a growing product niche. USITC Pub. 3446, p. I-8, and nn.18 and 19.

      1064 United States' second written submission, paras. 83-84.

      1065 United States' second written submission, paras. 72 and 86.

      1066 Japan's first written submission, paras. 79 and 101; Korea's first written submission, paras. 38-39 and 58-59; European Communities' first written submission, para. 240.

      1067 Specifically, the Appellate Body in EC – Asbestos, para. 99, stated:


      . . . a determination of "likeness" under Article III:4 is, fundamentally, a determination about the nature and extent of a competitive relationship between and among products. In saying this, we are mindful that there is a spectrum of degrees of "competitiveness" or "substitutability" of products in the marketplace. . . .


      1068 Appellate Body Report, US – Line Pipe, para. 82.

      1069 Moreover, in this investigation the USITC made its definition on the basis of like product analyses and not on the basis of directly competitive analyses. Thus, while the consideration of substitutability may be

      relevant to the directly competitive analysis, it is not germane to, and should not be transposed to, the like product analysis.

      1070 USITC Report, p. 44.

      1071 United States' first written submission, para. 136.

      1072 China's second written submission, para. 77.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 161


      all types of CCFRS can be used for the same purposes and applications in these industries. Once again, the market on which a product is finally sold cannot be considered as its "end-use".1073


    96. Korea and China also submit that substitutability is relevant in the assessment of "likeness" since like products are necessarily also directly competitive and, therefore, substitutable. Contrary to what the United States asserts, substitutability is a necessary factor regarding the like product definition in the context of the purpose of the Agreement on Safeguards, since, if imported and domestic products were not substitutable, they could in no way be considered as being "like".1074


    97. The United States responded that there clearly is a competitive relationship between the imported and domestic products. The complainants have not disputed, and neither did the parties in the underlying investigation, that the imported and domestic products generally consist of the same types of steel, are interchangeable, and thus compete with each other.1075


    98. The European Communities submits that the United States fails to respond to the complainants' argument that there is no similar end-use. The United States' rebuttal on this point only repeats the error of equating end-uses and end-users but does not respond to any of the arguments attacking the validity of the substantial coincidence of economic interests argument. What is more, the United States exclusively focused on the "commercial market".1076 However, the European Communities submits that the USITC itself admits that the main end-use for up-stream products is to be a feedstock for down-stream products, out of the commercial market. How could the USITC conclude that coated products sold to the automotive industry have the same end-use as slabs, of which 99.4% of the United States' production is internally transferred? Second, the automotive and construction industry identified as the main end-user for CCFRS account respectively for only 20 and 11% of the US shipments.1077


    99. Japan submits that the United States' argument that these products have common end-uses is not credible. They may be sold to the same industries, but to suggest that steel products have common applications because they are used in a specific industry is to suggest that steel, plastic, and glass should be a single like product because they are all sold to the automotive industry. End use is not the same as end-user. The fact is, no one would ever use slab to make a car; nor would hot-rolled steel be used for the same car-part as corrosion resistant steel. They are simply different products, used for different purposes. Furthermore, they each have a base price, reflected in the companies' price sheets and in the trade literature. There is no such thing as a price for "CCFRS" (as the USITC defined it). This proves not only that the industry and customers recognize the distinctions, but also that any analysis of this grouping performed by the USITC is meaningless because there is no "CCFRS" price that can be used to determine price effects in a causation analysis; rather each individual product must be analysed and then somehow combined with the other individual products.1078 Such an analysis distorts the true competitive dynamics in the marketplace.1079 The finding that most finished CCFRS products are sold into the automotive and construction markets is


      image

      1073 China's second written submission, para. 78.

      1074 China's second written submission, para. 80; Korea's second written submission, paras. 39-40, with reference to the Appellate Body Report, US – Cotton Yarn, paras. 91(d), 97.

      1075 United States' second written submission, para. 72. 1076 United States' first written submission, para. 122. 1077 USITC Report, Vol. I, p. 43, footnote 127.

      1078 USITC Report, Vol. I, pp. 61-62.

      1079 Japan's second written submission, para. 37.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 162


      analogous to its earlier finding of a "coincidence of economic interests" between producers of live lambs and lamb meat.1080


      Consumer perception


    100. According to China and the European Communities, if "consumers" are deemed to be the auto manufacturers and construction sites that use finished products, then the five different products grouped into the category "CCFRS" cannot be seen as substitutable. As the USITC acknowledged itself, "hot-rolled steel may not be substituted for a coated sheet in a car fender"1081 and neither may a slab be substituted for a coated sheet in a car-fender.1082


    101. Brazil, Japan and New Zealand argue that if the USITC had considered the factor identified as important by the Appellate Body, consumer tastes and habits1083, it would have found additional distinguishing characteristics, such as: slab purchases are restricted to producers of downstream rolled products1084; consumers regard hot-rolled steel as ideal for applications where strength is more important than appearance – for example, in the use of automobile structural parts1085; consumers view cold-rolled steel as ideal for applications where appearance and thinness take precedence over strength, and exposure to corrosive elements is not an issue1086; consumers view galvanized steel as ideal in exposed applications, where corrosion-resistance is important.1087 1088 New Zealand adds that the very fact that import trends differ amongst the products making up the category of CCFRS indicates that consumer needs and preferences differ product by product.1089


    102. The United States points out that while the Working Party in Border Tax Adjustment suggested that consumers' tastes and habits may be a criterion to be considered in finding like products for purposes of border tax adjustments, contrary to complainants' arguments, this is not a required factor in a safeguard investigation. The Panel in Japan – Alcoholic Beverages recognized that consumer habits were variable in time and discounted consumer views in considering whether vodka was "like" shochu and thus whether the like product should consist of vodka and shochu.1090 The consideration of consumer tastes and habits seems to conflict with the purpose of a safeguard investigation.1091 Since the purpose of the Agreement on Safeguards is to permit protection of a


      image

      1080 Japan's second written submission, para. 40.

      1081 USITC Report, Vol. I, p. 43.

      1082 European Communities' first written submission, para. 245; China's first written submission, para. 197.

      1083 In EC – Asbestos, the Appellate Body recognized the additional factor of "consumer tastes and

      habits", as one of four general criteria in analysing the "likeness" of two products, the others being physical characteristics, end uses, and tariff classification. Appellate Body Report, EC – Asbestos, para. 101.

      1084 USITC Report Vol. II at FLAT-1 (Exh. CC-6).

      1085 Ibid., p. FLAT-2

      1086 Ibid.

      1087 Ibid., p. FLAT-3.

      1088 Japan's first written submission, para. 118; New Zealand's first written submission, para. 4.64; Brazil's first written submission, para. 107.

      1089 New Zealand's first written submission, para. 4.65.

      1090 Panel Report, Japan – Alcoholic Beverages I, para. 5.7 ("Panel was of the view that the 'likeness' of products must be examined taking into account not only objective criteria (such as composition and manufacturing processes of products) but also the more subjective consumers' viewpoint (such as consumption

      and use by consumers) "but also recognized that "consumer habits are variable in time" and "traditional Japanese consumer habits with regard to shochu provided no reason for not considering vodka to be a 'like' product").

      1091 The Appellate Body cautioned in EC – Asbestos that it may be important to consider "from whose perspective "likeness" should be judged. For instance, ultimate consumers may have a view about the "likeness"

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 163


      domestic industry under certain circumstances1092, a focus on the subjective consumers' views of the product or market rather than producers or both is one-sided and misplaced. The USITC instead focused on such objective factors in its traditional analysis of like products such as the product's physical properties, uses, marketing channels and manufacturing process.1093


    103. New Zealand responds that consumer perceptions must be a critical component in assessing the "highest degree of competitive relationship" which the Appellate Body in US – Cotton Yarn has stressed as crucially important. It can and should be supplemented by other factors, but how purchasers behave in a market cannot be ignored.1094


      Tariff classification


    104. The European Communities, Japan, Korea, China and Brazil, submit that, while the USITC impermissably discounted the customs treatment of the various products due to "the large number of classification categories"1095 1096, this fact is no excuse for not considering them at all for the purpose of the like product determination. On the contrary, this suggests that the products concerned are not alike. Each CCFRS product is classified under a separate HS number, but the USITC expressly discounted this factor.1097 In EC – Asbestos, the Appellate Body noted that "tariff classification clearly reflects the physical properties of a product" and provided important indications for the like- product determination which must be considered.1098 The fact that so many HS classifications have been grouped into a single imported product and a single like product is just one more indication that the USITC failed in its determination. In fact, contrary to the assertion of the USITC, HSUS classifications of CCFRS at the four-digit level (depending on the width and thickness of the product) do provide a break-out of the various like products: semi-finished (including slab), hot-rolled, cold- rolled, corrosion-resistant, and plate. Moreover, only five of the 55 distinct HSUS classifications at the six-digit (international) level are shared between two CCFRS steel products. The European Communities adds that a closer look at the 55 classifications for what the USITC calls CCFRS reveals that even the HSUS classifications distinguish at the four-digit level (depending on the width and thickness of the product) between "semi-finished products" including slabs, hot-rolled, cold-rolled, corrosion resistant and plate.1099 This is a high level of delineation and reflects fundamental product distinctions recognized internationally for purposes of classifying imports.1100 1101


    105. The United States noted that in this case the numerous tariff classifications did not provide clear distinctions between products. For instance, each of the 33 data collection categories individually have from 2 to 65 tariff classifications. The United States also noted that the facts did not provide support for complainants' allegations that consideration of tariff classifications at the 4-digit level would have provided clear product distinctions. For example, at the 4-digit level, there are nine


      image

      of two products that is very different from that of the inventors or producers of those products." Appellate Body Report, EC – Asbestos, para. 92.

      1092 Appellate Body Report, US – Line Pipe, para. 82; USITC Report, p. 9.

      1093 United States' first written submission, para. 137.

      1094 New Zealand's second written submission, para. 3.39.

      1095 USITC Report, Vol. I, p. 37, footnote 71.

      1096 Brazil's first written submission, para. 102; European Communities' first written submission, para. 246.

      1097 Japan's first written submission, para. 117; Brazil's first written submission, para. 106.

      1098 Appellate Body Report, EC – Asbestos, paras. 102, 109 and 124.

      1099 Presidential Proclamation Annex I, pp. 25 ff.

      1100 "Chapter 72 Flat-Rolled HTS Descriptions at the Four and Six-Digit Level", (Exhibit CC-83).

      1101 European Communities' first written submission, paras. 246-248; Korea first written submission, paras. 51-52; China's first written submission, paras. 198-200.

      WT/DS248/R, WT/DS249/R, WT/DS251/R, WT/DS252/R, WT/DS253/R, WT/DS254/R, WT/DS258/R, WT/DS259/R

      Page 164


      separate tariff classifications covering the like product defined by the USITC as CCFRS. Of the nine 4-digit classifications, two classifications (7225 and 7226) apply to steel at four (hot-rolled steel, CTL plate, cold-rolled steel, and coated steel) of the five stages of processing defined as CCFRS; one classification (7211) applies to three stages; two classifications (7208 and 7210) apply to two stages; and four classifications (7207, 7209, 7212, and 7224) apply to one stage of CCFRS. Thus, rather than provide clear product category distinctions, tariff classifications at this level demonstrate an interrelationship between the physical properties of steel at different stages of processing which led to the USITC defining these types of steel collectively as CCFRS.1102


    106. Further, in New Zealand's view, the differences amongst the various uses to which the various CCFRS products are put are further reflected in the differing tariff classifications attributed to each of them.1103 The role of tariff classification in like product determinations has been well accepted in the jurisprudence. Tariff classifications reflect international consensus as to the degree of similarity and difference between products. Clearly, it is not open to a competent authority simply to ignore tariff classification.1104 Several of the five CCFRS products do not even share the same classification to four digits, let alone eight or ten.1105 This simply reinforces the lack of likeness between them.1106


    107. The European Communities maintains that the HS clearly distinguished the different products on the basis of the different stages of processing, as is demonstrated in the relevant paragraphs of Chapter 72 of the HS in exhibits CC-105. The United States is wrong in asserting that the distinction between products, which are part of a different stage of the production process is not reflected at the four digit level. As can be discerned from the tariff heading 72.07-72.09 even at the four-digit level there is a distinction between slabs (semi-finished) hot-rolled, cold-rolled and coated.1107 The differentiation is even clearer when considering the internationally agreed tariff classifications at the six-digit level.1108 1109


    108. According to Korea, the tariff numbers have aided the USITC's classification of "like" steel products in past anti-dumping and countervailing duty investigations for over 20 years. The USITC, both in its determination and in its Staff Report, breaks the seven CCFRS products down into "like" product categories traditionally used in the various anti-dumping and countervailing duty investigations performed by the USITC: slabs, plate (cut-to-length and clad), hot-rolled, cold-rolled, grain-oriented silicon steel, coated steel and tin mill products.1110 1111



image

1102 United States' first written submission, paras. 86 and 89.

1103 New Zealand's first written submission, para. 4.66. 1104 United States' first written submission, paras. 86-87. 1105 USITC Report Vol I, pp. 9-10.

1106 New Zealand's second written submission, para. 3.48.

1107 For example, products under heading 72.07 which are referred to as semi-finished products of iron

or non-alloy steel (slabs) are clearly distinguished from products under 72.08, which are referred to as flat-rolled products of iron or non-alloy steel, of a width of 600 mm or more, hot-rolled, not clad, plated or coated.

1108 At the six digit level products are further defined and separated from one another. For example, take products that fall under paragraph 72.11, which at the four-digit level are described as "Flat-rolled products of iron or non-alloy steel, of a width of less than 600 mm, not clad, plated or coated". At the six digit level, these products are further defined on the basis of whether they are not further worked than hot-rolled (721113,

721114, 721119) or whether they are not further worked than cold-rolled (cold-reduced) (721123, 721129, 721190).

1109 European Communities' second written submission, paras. 272-274.

1110 USITC Report, Vol. I: "Appendix A", pp. 9-10 and Vol. II, pp. FLAT-1-4 (Exhibit CC-6).

1111 Korea's first written submission, para. 53.