|español - français - português|
UNITED STATES - SAFEGUARD MEASURES ON
7.46 In its safeguard investigation concerning imported lamb meat, the USITC defined the domestically-produced product that was "like" the imports at issue as lamb meat. The respondents in the investigation did not contest that US-produced lamb meat was "like" the imported lamb meat75, but did argue that live lambs are not "like" lamb meat. In assessing the condition of the domestic industry producing that like product, the USITC included in the industry the growers and feeders of live lambs on the one hand, and the packers and breakers of lamb meat on the other, because according to the USITC’s approach, they are all producers of lamb meat.
7.47 Australia and New Zealand claim that
because the USITC included producers of raw materials and inputs – i.e., growers and feeders of live lambs – as producers of lamb meat, the United States violated SG Article 4.1(c). In the view of
the complainants, Article 4.1(c) requires that only producers of the like product, and not producers of raw materials and inputs, can be considered to constitute the domestic industry producing a
like product. Thus, according to the complainants, the industry producing the like product should have been limited to packers and breakers of lamb meat, as live lambs are not "like" lamb meat.76
In the alternative, Australia and New Zealand argue that even if live lambs had been defined by the USITC as a "directly competitive" product to lamb meat, any such definition would not have been
legally sustainable. In this context, they cite past cases, in particular those under GATT Article III in which the question of directly competitive products has been addressed.77
7.48 The US safeguard statute, section 202(c)(6)(A)(i) of the US Trade Act of 197478 defines the term "domestic industry" in a manner virtually identical to the relevant text of Article 4.1(c) of the Safeguards Agreement, namely as
7.49 In the lamb meat investigation, the USITC explained its approach in safeguards investigations in identifying the producers as a whole of a product under investigation as follows:
7.50 In the case at issue, the USITC found that these criteria were satisfied. In particular, on the basis of these criteria, the USITC found that the domestic producers of lamb meat consisted of the growers and feeders of live lambs as well as the packers and breakers of lamb meat because:
7.51 The USITC further stated, in respect of its finding of "a coincidence of economic interests", that there was evidence of some degree of vertical integration (i.e.,
that some growers engage in both feeding and slaughtering of lambs) and evidence that "the price of lamb meat affects all four industry segments similarly (that is, when processors do well, growers
and feeders also benefit, but when processors confront lower prices, they pass the lower prices back to feeders and then growers, and all suffer to some extent)".81
(a) Australia and New Zealand
(a) Australia and New Zealand
7.52 New Zealand and Australia contend that the methodology adopted by the USITC in defining the domestic industry (i.e., continuous line of production and coincidence of economic interests) finds no basis in the text of the Safeguards Agreement. They assert that for the purposes of a safeguards investigation, the determination of what constitutes the "domestic industry" must turn on whether the producers in question produce a "product" that is "like or directly competitive with" imported lamb meat. That is, the determination of what constitutes the like or directly competitive product drives the determination of which producers constitute the industry producing that product. Hence, growers and feeders of live lambs would only fall within this definition if the live lambs produced by them were deemed a product that is "like or directly competitive" with lamb meat.82 For the complainants, the fact that the United States has traditionally used an alternative approach is irrelevant.83
7.53 Thus, the complainants argue, SG Articles 2.1 and 4.1(c) require a determination as to what industry produces a product that is "like or directly competitive" with imported lamb meat. They contend that, contrary to this, the United States has instead applied a test to determine what constitutes the abstract class of "producers as a whole". In their view, the qualifying term "as a whole" defines the scope of the producers within an industry and is not a term that defines the scope of the industry itself.84
7.54 The complainants further point to past dispute settlement cases, which they argue consistently have rejected the idea that a determination of what constitutes the relevant industry should be made on the basis of some notion of vertical integration.85 In this respect, the complainants rely largely on the reports of the panel on United States – Definition of Industry Concerning Wine and Grape Products ("US – Wine and Grapes")86 and the panel on Canada – Imposition of Countervailing Duties on Imports of Manufacturing Beef from the EEC ("Canada – Beef").87
7.55 In the alternative, the
complainants oppose the US argument that, as a factual matter, extensive integration exists between firms at different stages in the continuous line of production. In their view, most of the
integration actually found by the USITC was between growers and feeders on the one hand and packers and breakers on the other, and there is little evidence, if any, of firms which both grow live
lambs and engage in packing operations.88 Therefore, the complainants reject the US argument that the industry is so highly integrated that it is not
possible to separate respective sectors of the production process.89 (b) United States
(b) United States
7.56 The United States approaches the issue of whether the USITC's definition of the "domestic industry" is consistent with the provisions of the Safeguards Agreement from a different angle. It argues that the relevant consideration is not whether live lambs are "like or directly competitive" with lamb meat, but whether the USITC majority correctly found that growers, feeders, packers and breakers all can be considered to produce the like product, i.e., lamb meat. In the alternative, the United States contends that in any event, the USITC would have reached the same conclusions as to threat of serious injury and causation if it had limited the industry to lamb meat packers and breakers.90
7.57 The United States notes that the USITC drew on its own practice relating to anti-dumping and countervailing duties in finding that the "domestic industry" producing lamb meat included the producers of the raw product. As noted above, this methodology considers whether (1) there is a continuous line of production from the raw material (i.e., live lamb) to the processed product (lamb meat); and (2) there is a substantial coincidence of economic interest between the producers of the raw material (i.e., growers and feeders) and the processors (i.e., packers and breakers).
7.58 In support of this approach, the United States stresses that the growers and feeders together contribute approximately 88 per cent of the value of the wholesale price of lamb meat. It claims that limiting the definition of "producer" to those who contribute only limited value-added toward the final stages of production would create an artificially defined 'domestic industry', especially where extensive vertical integration exists. The United States argues that such an artificially narrow approach to defining "domestic industry" in turn would have the negative effect of denying the possibility of safeguard relief to producers of raw products even where such producers were clearly suffering from or threatened with serious injury caused by imports of processed end products.91
7.59 Furthermore, the United States maintains that any attempt to utilise a more narrow approach to the delimitation of the "domestic industry" would prove difficult since the US lamb industry "is vertically integrated in such a way that it is virtually impossible to analyse each segment of the domestic industry producing lamb meat by focusing on only one, discrete sector. … [The] inability to disaggregate the respective sectors producing the like product requires that the definition of domestic industry include all four sectors contributing to the production of the like product".92 The United States relies in this respect on the report of the panel on New Zealand – Imports of Electrical Transformers from Finland93 ("New Zealand – Transformers") which rejected that argument that the transformer industry at issue consisted of four distinguishable ranges of transformers which should have been considered separately for purposes of the injury and causation determination.
7.60 The United States also submits that it only applies the above USITC approach in investigations involving "processed agricultural products."94 Evidence of this is found in the test applied by the USITC, which provides that there needs to be a continuous line of production from the raw to the processed product. The United States concludes that this test does not "simply provide for relief to be available to input suppliers in general when they suffer injury from imports equivalent to that normally suffered by those who produce end products".95
7.61 The United States dismisses the relevance of the past GATT panel report on Canada – Beef96 because it remains unadopted. Further, the United States also distinguishes the report of the panel on US – Wine and Grapes from this case on the basis that that panel had decided that grape growers were not part of the domestic wine-producing industry because the production of wine grapes was not wholly dedicated to wine production, i.e., in a previous USITC investigation it was found that only 42-55 percent of wine grapes were used in the production of wine, and there were other major markets for wine grapes, such as table grapes and raisins. As a result, it was possible to separately identify the production of wine grapes and the production of wine.
7.62 In contrast, the United States asserts that disaggregation of the lamb industry is extremely difficult because US lambs are overwhelmingly
raised for meat rather than for wool97 and that the United States does not conduct trade in live lambs. Furthermore, unlike wine grapes, which go
through a process of treatment and fermentation prior to bottling as wine, lamb meat remains substantially the same during processing and is never transformed into a different article.98
7.63 The complainants’ claims under SG Article 4.1(c) raise the basic questions of whether the broad reading of that provision adopted by the United States is permitted, or
whether the narrow reading advocated by the complainants is required. In assessing these claims, we will consider in detail the text of the provision, taking into account past panel reports that
have addressed similar issues as well as relevant negotiating history, in particular with a view to determining whether the text can support the methodology applied by the USITC as to "continuous
line of production" and "coincidence of economic interests". (a) The definition of the "domestic industry" in SG Article 4.1(c)
(a) The definition of the "domestic industry" in SG Article 4.1(c)
7.64 SG Article 4.1(c) provides in relevant part that a "domestic industry"
7.65 We recall that in this case, the USITC found that there was a "like product", lamb meat, and did not make any finding concerning whether live lambs (or any other domestically-produced product) were "directly competitive" with the imported lamb meat. Given that the USITC99 only made a finding concerning "like product" – lamb meat – the question before us is whether the USITC's broad determination of the producers of that "like" product is consistent with the Safeguards Agreement.
7.66 We turn first to the ordinary meaning of the relevant portion of the text, i.e., SG Article 4.1(c)'s industry definition: "producers as a whole of the like or
directly competitive products … or those whose collective output of those products constitutes a major proportion of the total domestic production of those products" (emphasis added).
(i) "Producers … of the like … products"
(i) "Producers … of the like … products"
7.67 We consider that the basic elements of SG Article 4.1(c)'s industry definition are contained in the phrase "producers … of the like or directly competitive products". To us, the ordinary meaning of this phrase is straightforward: the producers of an article are those who make that article. That is, the determination of the relevant domestic industry is derivative from the identification of the relevant "like" or "directly competitive" products. We find no basis in the text of this phrase for considering that a producer that does not itself make the product at issue, but instead makes a raw material or input that is used to produce that product, can nevertheless be considered a producer of the product.
7.68 The second part of the definition in SG Article 4.1(c), specifically the reference to the producers "whose … output" includes "those products", explicitly confirms our reading of the basic industry definition. In particular, this part of the definition underscores that the relevant industry consists of producers that themselves have "output" of the "like" or "directly competitive" products.
7.69 We find further support for our reading of the phrase "producers … of the like … products" in numerous dictionary definitions: a "producer" is variously defined as "a person or a thing which produces something",100 or "one that produces, especially one that grows agricultural products or manufactures articles".101 To "produce" means to "bring a thing into existence, bring about, effect or cause an action or result",102 or "to give being, form or shape to, make, or manufacture".103 A "product" is a "thing produced by an action, operation or natural process"104 or "something produced, or the amount, quantity or total produced".105 The term "output" means "what is produced by an industry or process" or "the action or process of supplying an output, production".106
7.70 The important common element of these dictionary meanings is that there is a clear link and close connection between the one who undertakes an action to bring an article into existence and the article resulting from this action. This supports our view that a given enterprise can be considered as a producer of only those goods that it actually makes. By this logic, a producer that makes primary or intermediate goods used in the production of further processed goods must be considered a producer of the primary or intermediate good, rather than of the processed good that it does not itself ever produce.
this ordinary meaning to the facts of this case – if not to state the obvious – points to the conclusion that growers and feeders are producers of live lambs, whereas packers and breakers of lamb
carcasses are producers of lamb meat. This is so because the good produced by growers and feeders, i.e., live lambs, is not itself the like product at issue, i.e., lamb meat. The lamb growing and
feeding operations give rise to a product which is different from the product that results from the subsequent processing operations where lambs are slaughtered and carcasses are cut into lamb meat
for final consumption. (ii) "Producers as a whole"
(ii) "Producers as a whole"
7.72 We recall that in defending the USITC's decision to include growers and feeders in the lamb meat industry, the United States relies on the phrase "producers as a whole" from the industry definition in SG Article 4.1(c).107 In particular, the United States contends that the growers and feeders form part of the producers "as a whole" of lamb meat. We further recall that the complainants disagree with this construction of the phrase "as a whole", arguing that in fact this phrase has to do with the representativeness of the data collected from producers in the industry, and not with which producers should be included in that industry.
7.73 We thus next consider whether the phrase "producers as a whole" can be seen as context relevant to the interpretation of the basic industry definition, which would permit an industry to be defined so as to include input producers, as was done by the USITC in this case. We note in this regard that the phrase "producers as a whole" is grammatically linked to, and juxtaposed with, the phrase "or those whose collective output … constitutes a major proportion of … total … production". This context implies that the phrase "as a whole" like the phrase "major proportion" relates to the representativeness of the data pertaining to the condition of the industry. That is, pursuant to SG Article 4.1(c), for purposes of determining injury or threat, the domestic industry to be investigated consists in the first instance of all producers of the relevant product in their entirety, or – at a minimum – of those producers accounting for a major proportion of the total production of the product. We recall in this regard that in response to a question from the Panel, the United States seems to acknowledge that the phrase "as a whole" – at least also – relates to the representativeness of the data concerning the industry,108 not only to the scope of the industry as it claims under its main line of argumentation.
7.74 We conclude, on the basis of the foregoing analysis, that the phrase "producers as a whole" is not related to the process of manufacturing or transforming raw materials and inputs into a final product, and thus provides no contextual support for including producers of raw materials or inputs as part of the industry producing a like product. In our view, this phrase provides a quantitative benchmark for the proportion of producers – within an industry properly defined on the basis of the like output product it makes – which a safeguards investigation has to cover. We note that – if the phrase "as a whole" could be used to widen the scope of an industry to include producers of any upstream products – competent national authorities could "tailor" domestic industries of different scope as they saw fit simply by choosing between two alternatives under SG Article 4.1(c).
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 ("SCM") and on Anti-dumping ("AD"). In particular, the three Agreements' definitions of the industry producing a like product are essentially identical.109 We also note that, while the SCM and AD Agreements refer exclusively to "like products", the SG Agreement also refers to "directly competitive products", but in the absence of a USITC finding on "directly competitive products" in this investigation, this issue is not before us. Thus the distinction between "like" and "directly competitive" products is not relevant to the complainants' claims under SG Article 4.1(c). For these reasons, we consider that particularly in the present safeguard dispute, past panel reports concerning industry definition in the context of the SCM and AD Agreements are relevant to our interpretation and application of the industry definition under the Safeguards Agreement. We discuss the past dispute settlement practice interpreting these provisions in detail below.
7.76 In our view, this reading of the industry definition is consistent with the object and purpose of the Safeguards Agreement. In particular, this reading is consistent with the Agreement's objectives of, on the one hand, creating a mechanism for effective, temporary protection from imports to an industry that is experiencing serious injury or threat thereof from imports in the wake of trade liberalization, and on the other hand, encouraging "structural adjustment", and "clarify[ing] and reinforc[ing] the disciplines of … Article XIX of GATT", in view of "the need to enhance rather than limit competition in international markets".110
7.77 If WTO law were not to offer a "safety valve"
for situations in which, following trade liberalization, imports increase so as to cause serious injury or threat thereof to a domestic industry, Members could be deterred from entering into
additional tariff concessions and from engaging in further trade liberalisation. It is for this reason that the safeguard mechanism in Article XIX has always been an integral part of the GATT.
However, we note that SG Article XIX of GATT 1994 as well as SG Article 11.1 both refer to safeguard measures as "emergency" measures, and the Appellate Body has characterized them as
"extraordinary" remedies111 A conceptual approach to defining the relevant domestic industry which would leave it to the discretion of competent
national authorities how far upstream and/or downstream the production chain of a given "like" end product to look in defining the scope of the domestic industry could easily defeat the Safeguards
Agreement's purpose of reinforcing disciplines in the field of safeguards and enhancing rather than limiting competition. These considerations based on the object and purpose of the Safeguards
Agreement thus further support a reading of the industry definition in SG Article 4.1(c) as not permitting input producers to be included as part of the industry producing the "like" end-product.112
(b) Past panel reports
(b) Past panel reports
7.78 As we have stated above, given that the industry definitions in the SCM and AD Agreements are virtually identical to that in the Safeguards Agreement in so far as "like products" are at issue (as in this case) we consider that past panel cases concerning the industry definition in disputes on antidumping, subsidies and countervailing measures are particularly relevant to our examination of the complainants' claim against the industry definition used by the USITC in this case.113 These include in particular the reports of the panels on Canada – Beef114 and US – Wine and Grapes115, but also the New Zealand – Transformers116 report. We note in this regard that the parties as well have extensively referred in their arguments concerning "domestic industry" to interpretations developed in these past panel reports.
7.79 We find quite pertinent to the question before us the adopted report of the panel on United States – Wine and Grapes under the Tokyo Round Subsidies Code, to which the parties also refer. In that case, the panel found inconsistent with the Code's industry definition a US law which mandated specifically that in countervailing duty cases involving imported wine and grape products, the domestic producers of the principal raw agricultural product (i.e., grapes) were to be included as part of the industry producing wine and grape products if they alleged injury or threat thereof caused by imports of those products.
7.80 The parties agreed that wine and grapes are not like products. The panel held that the producers of the like products could be interpreted to comprise only producers of wine.117 It also considered whether, in the light of the "close relationship" between grape and wine production, the wine-grape growers could be regarded as part of the industry producing wine. In this regard, the panel took into account that the parties agreed that in the United States, wineries did not usually grow their own grapes, but rather bought them from grape growers. Given this, the panel found that "irrespective of ownership, a separate identification of production of wine-grapes from wine … was possible and that therefore in fact two separate industries existed in the United States…"118 The Wine and Grapes panel concluded that
7.81 In reaching this conclusion, the panel took the view that "once such a separate identification was possible (e.g., because of the structure of production), economic interdependence between industries producing raw material or components and industries producing the final product" was not relevant for a like product determination.120 As discussed above, we too find no basis in the text of the Safeguards Agreement that would permit this consideration of economic interdependence or coincidence of economic interest to be taken into account in defining the domestic industry.
7.82 The United States distinguishes the present case from the Wine and Grapes case, inter alia, on the basis of certain factual arguments, including that grapes were not wholly dedicated to wine production. In that case, the USITC had determined that only 42-55 per cent of wine grapes were used in the production of wine and that there were other major markets for wine grapes, such as table grapes and raisins. In contrast, the United States points out that the USITC found that lambs are overwhelmingly raised for meat rather than for wool and that the ratio of net sales/revenue for slaughter and feeder lambs in comparison to net sales/revenues obtained by US lamb growers from any other item including wool increased from 84.6 per cent in 1997 to 88.9 per cent in interim-1998.121
7.83 We recall the Wine and Grapes panel's finding, with which we agree, that the factor of economic interdependence between producers of raw, intermediate and final products is not relevant for the industry definition. Even assuming arguendo, that nevertheless criteria such as "continuous line of production" and "inputs wholly dedicated to the production of a single end-product" were at all relevant, we note that the USITC report contains no information as to the percentage of live lamb production dedicated to the production of lamb meat other than for the years covered by the safeguard investigation. It thus is unclear to what extent such predominant dedication to meat as opposed to wool production was a temporary result of the removal of the wool subsidies.122
7.84 Moreover, as in Wine and Grapes, where alternative uses for grapes were found to exist, the USITC report makes clear that there are alternative uses for live lambs, including growing mature sheep for mutton meat as well as for wool production or growing ewes for breeding purposes.123 The extent to which these alternatives are actually used may depend on amounts of imports, but also on market conditions, consumer preferences and the possibility to generate equivalent profits with these alternative uses. Thus, even assuming arguendo in the alternative that the degree of an input's dedication to a final product were relevant for the industry definition, we find no factual evidence that the situation of lamb growers and feeders in respect of the availability of alternative uses for live lambs in the longer run is fundamentally different from that of the grape growers as described in the Wine and Grapes report.
7.85 The United States also submits that unlike grapes, which undergo a process of treatment and fermentation prior to bottling as wine, lamb meat remains substantially the same during processing and is never transformed into a different article. Here again, however, in our view no such factual distinction can be drawn. In the case of both lamb and wine, we note that the agricultural input product (i.e., grapes and live lambs, respectively) is transformed into a different end-product (i.e., wine or meat, respectively).
7.86 In the light of the foregoing, we consider that the reasoning of the Wines and Grapes panel is both directly relevant to, and fully consistent with, our conclusion that the domestic industry in the lamb case should be limited to packers and breakers. Analogous to Wine and Grapes, live lambs and lamb meat not being like products to one another, producers of live lambs cannot be included as producers of lamb meat.
7.87 We also find that the reasoning of the panel in the Canada – Beef124 case is highly relevant to, and strongly supports, our reading of SG Article 4.1(c). In Canada – Beef, the EC challenged a Canadian countervailing duty investigation in which the producers and feeders of live cattle were treated as part of the domestic industry producing manufacturing beef. The factual and legal issues arising in Canada – Beef are strikingly similar to those of the present dispute.
7.88 The parties were in agreement that the "like" product was manufacturing beef, but differed on whether the domestic industry producing manufacturing beef included the producers and feeders of live cattle. Likewise, in the lamb case, the parties agree that the "like" product is lamb meat, but they disagree as to whether the industry producing lamb meat includes the growers and feeders of live lamb.
7.89 The Canada – Beef panel agreed with the parties that the like product was manufacturing beef, and that live cattle produced by ranchers and feedlots constituted a product different from the like product. The panel also observed that the relevant provision of the Tokyo Round Subsidies Code (Article 6:5)125 did not define the term "producers", but that "in common usage, one is normally considered the 'producer' of only those goods one actually makes and sells; one who produces a raw material is not normally regarded as a 'producer' of the end-product.126
7.90 Before the Canada – Beef panel, Canada argued that a narrow definition of the domestic industry was not appropriate where there was (i) a continuous sequential process of production involving the use of only one raw material input which, by undergoing relatively little processing prior to becoming an end-product, accounted for a substantial proportion of the value of the end-product; (ii) an input which was functionally dedicated to the manufacture of only one end-product and which had no economically viable alternative uses; and (iii) a situation of economic interdependence in which end-product producers were able to "pass-back" to input producers a decrease in the price of the end-product resulting from competition from subsidized imports.127 We note that these criteria are very similar to those under the two-pronged test applied by the USITC in the lamb investigation.
7.91 The Canada - Beef panel articulated concerns with respect to Canada's criteria for deciding in which cases to include input producers as producers of a processed product. Concerning the criteria used by Canada, that panel understood that these were meant to identify situations in which all or most of the adverse economic impact from subsidized imports would be concentrated on the raw material supplier.128 It found, however, that this interpretation by Canada would
7.92 The argumentation of the parties in the lamb dispute is largely similar to that before the Canada – Beef panel. The complainants argue that the USITC's above-mentioned two-prong test could lead to competent national authorities devising open-ended industry definitions, without objective limitations in practice. In contrast, the United States (as Canada did) argues that it applies its test only in the case of processed agricultural products where the inputs are wholly dedicated to the production of the processed product, and thus would not open the door to large scale tailor-making of industry definitions.
7.93 We are not however persuaded by US argumentation in the present dispute, and we too are concerned by the possibility of "open-endedness" in defining domestic industries that was highlighted by the Canada – Beef panel. We see no basis in the text of the Safeguards Agreement, nor has the United States put forward any principles of general applicability, to effectively limit the inclusion of input producers as producers of an end-product to cases involving processed agricultural products, or to any subgroup of cases.
7.94 The Canada – Beef panel also rejected the argument that industry definitions based on the like end product could cause outcomes to vary according to the degree of vertical integration which happened to exist at a particular time or in a particular country because, the panel found, the definition of "domestic industry" involves two criteria, neither of which depends on vertical integration as such:
7.95 We agree that the factors of vertical integration or common ownership are not in themselves determinative or even particularly relevant for the scope of the domestic industry. Rather, the issue is (i) whether the products at various stages of production are different forms of a single like product or have become different products; and (ii) whether it is possible to separately identify the production process for the like product at issue, or whether instead common ownership results in such complete integration of production processes that separate identification and analysis of different production stages is impossible.132
7.96 In the present dispute, the parties agree and the USITC found that the production process from live lamb to lamb meat has resulted in separate products, not products that are different forms of a single like product. Likewise, assuming arguendo that vertical integration and common ownership were at all relevant for the defining the scope of an industry, there is little vertical integration of growing and feeding operations with packing and breaking operations, and in any case it is clearly possible to separately identify the different physical stages of the production process. Moreover, according to the information contained in the USITC report, there is relatively little vertical integration in the sense of common ownership between growers, feeders, packers and breakers of lamb.133 Furthermore, to the extent that there is an overlap in activities between companies, this overlap occurs predominantly between growing and feeding operations, or between packing and breaking operations,134 but it does not occur between growers and feeders on the one hand and packers and breakers on the other.135 In other words, we find no evidence to support the US assertion that the US lamb industry "is vertically integrated in such a way that it is virtually impossible to analyze each segment … " and that "[t]he inability to disaggregate the respective sectors requires that the definition of domestic industry include all four sectors contributing to the production of the like product."136 We thus conclude that in this case it is possible to separately identify the physical production processes involved in producing live lambs on the one hand and lamb meat on the other, and, in addition, separate data clearly are available for the four industry segments, as evidenced by the fact that the USITC collected such data.137
7.97 We recall that the United
States argues that the reasoning of Canada – Beef and US – Wine and Grapes are irrelevant to the lamb case at hand because these panels applied provisions138
of the Tokyo Round SCM and Anti-dumping Codes narrowing the scope of the domestic industry which the Safeguards Agreement does not provide for. In our view, this difference does not make these past
panel reports inapposite to the present case because the provisions referred to by the United States do not address the question of the definition of the domestic industry. Rather they deal
primarily with the data collection in an investigation so as to ensure that the data reflect as closely as possible the operations pertaining to the like product, where separate identification of
those operations in a producer's records is difficult or impossible. The parallel article of the WTO SCM Agreement139 makes it even clearer that this
provision does not detract from the fact that the industry definition must be based on the "like" product and on the ability to separately identify the production processes. Thus, we find the
Canada – Beef and Wine and Grapes cases to be factually similar to the case before us, and the legal reasoning of those panels to be both relevant and persuasive140.
(iii) The New Zealand – Transformers case
(iii) The New Zealand – Transformers case
7.98 We note that the United States argues that there is support for its broad definition of the US lamb meat industry in the statement of the panel in New Zealand – Transformers141 that the domestic industry in that case should not be restricted to two kinds of transformers located along the spectrum of transformer types, because "each segment of the industry's operation made a contribution to the overall viability and profitability of a producer of transformers" and that to "allow the possibility to grant relief through anti-dumping duties to individual lines of production of a particular industry or company … would clearly be at variance with the concept of industry in Article VI."142
7.99 In our view, however, the factual and legal issues before the Transformers panel were rather different from those arising in the present dispute. First, Transformers involved a claim under Article VI:1 of GATT, not under the Tokyo Round Anti-dumping Code. Article VI:1 does not contain any reference to the concept of "like product", and thus its language is quite different from that in the Safeguards Agreement (as well as that in the parallel provisions of the WTO SCM and AD Agreements).143 The arguments of New Zealand in Transformers pertained, if anything, more to the question of the relevant domestic product to be analyzed (i.e. an issue akin to identifying the "like product") rather than to the second-step question of how broadly to define the producers of that product once identified. Second, the factual situation in Transformers was very different from that in the lamb case. In Transformers, New Zealand's domestic transformer industry essentially consisted of a single company that produced the full range of power transformers144, and the product differentiation at issue was as between different kinds of finished transformers produced by that company, i.e., differentiated kinds of the same product (transformers) at the same stage of production. By contrast, in the lamb case, there are many companies involved, most of which operate at only a single step in the production chain, and the product differentiation at issue is as between different products at different stages of production.
7.100 Moreover, to the extent that Transformers is at all relevant to the issue before us, it supports rather than undercuts our
reading of SG Article 4.1(c). In particular, it appears to us that one of the primary concerns of the Transformers panel was the possibly artificial picture of the relevant
company's/industry's condition that could result from looking at only one small slice of that company's/industry's product range, where there were no clear dividing lines either between the
products themselves or between the production processes used to produce them. In our view, this is fully consistent with our view, confirmed by the Canada – Beef panel, that separability of
production processes is a key factor in identifying the domestic producers of a like product. (iv) Criteria of continuous line of production and substantial coincidence of economic interests
(iv) Criteria of continuous line of production and substantial coincidence of economic interests
7.101 We also share the concerns of the Canada – Beef panel about the "open-endedness" of an industry definition if it is based on criteria such as (i) continuous line of production and (ii) substantial coincidence of economic interests. It is true for most processed products that there is a continuous line of production from raw materials or inputs to the final product and thus economic interdependence between operators at different stages of production. But we do not see how raw materials or inputs which are agricultural differ in this respect from industrial raw materials or inputs.
7.102 Concerning the coincidence of economic interests, moreover, whether there is a single input transformed or incorporated into a final product, whether an input is wholly dedicated to the production of a final product, or whether there are viable alternative uses at equivalent profit for that input cannot in itself be determinative of the degree of economic interdependence among industry segments. In the case of final products composed of a larger number of inputs, producers of those inputs may just as easily be highly economically dependent on the producers of the final product. But depending on the allocation of market power in the manufacturing and processing chain of a particular end-product, the opposite may also be true and producers of the final product may be dependent on producers of raw materials or intermediate inputs rather than vice versa.
7.103 Furthermore, the interests of producers in different industry segments may coincide, regardless of whether they are involved in a continuous line of production, whether there is a single or more inputs into a final product, and whether an input is wholly dedicated to a single final product. Interests may happen to coincide even if producers are engaged in entirely unrelated economic activities. Likewise, there is no certainty that economic interests of producers necessarily coincide even if there is a continuous line of production from an input which is wholly dedicated to one final product which is composed of only that input. Thus, we see nothing in the USITC's approach that limits its open-endedness.
7.104 We note that the USITC traditionally
applies the two parts of its test (continuous line of production and substantial coincidence of economic interests) on a cumulative basis and not as alternative justifications for widening the
industry definition. However, we are not persuaded that these are objective principles capable of general application that would in fact restrict the ability to include input producers to only a
narrow set of clearly defined cases. Moreover, we cannot see a textual or logical basis in the Safeguards Agreement for applying different tests in the fields of agricultural as opposed to
manufactured products. (v) Value added at different stages of the production chain
(v) Value added at different stages of the production chain
7.105 In the specific factual constellation of this investigation, we nonetheless consider the US argument significant that the inputs (live lambs) constitute a high percentage of the value added (e.g., 88 per cent of the wholesale value) and that the final product (i.e., lamb meat) derives essentially from a single input (i.e., live lambs). The US position seems to be that in such a situation, defining the industry as finishers only would mean, inter alia, that remedial measures that addressed only the effects of imports on one aspect of a continuous line of production would be inadequate to "prevent or remedy serious injury and to facilitate adjustment" since any adjustment by that industry segment would not insulate the other (higher value-added) segments from the effects of increased imports.145 This argument seems to depend on the ability of the processors to pass back any injury from increased imports to the input producers. Indeed, the USITC found that "…when processors confront lower prices, they pass the lower prices back to feeders and then growers, and all suffer to some extent."146
7.106 In our view, however, the pass-back argument in favour of broadly defining a domestic industry to include input producers does not necessarily hold true. As noted above, a high degree of economic interdependence between upstream producers and downstream processors is a commonplace in most manufacturing and processing chains. In such situations the US argument would only hold if the finishing segment in the production chain is able to "pass back" to input producers any serious injury caused or threatened by imports. In the absence of such "pass-back" effects, there is reason to assume that serious injury caused by increased imports, if any, will be felt (at least inter alia) in the finishing segment. In such a case, if a safeguard measure were applied in respect of imports of the finished product by definition this should also benefit the input producers.147
7.107 In this regard, the US "pass-back" argument could be seen to some extent as internally inconsistent. On the one hand, the argument is that the fortunes of the packers/breakers and growers/feeders rise and fall together. This is a major part of the USITC's justification for a broad industry definition, even though it does not explain why profits/losses of growers/feeders declined prior to those of packers/breakers.148 On other hand, assuming that it is true that the fortunes of all industry sectors move in tandem, a safeguard measure to assist packers/breakers would be likely also to assist growers/feeders. Thus a narrow industry definition would not necessarily preclude the benefits of a safeguard measure on the finished products from "trickling upstream" to the input producers. In other words, the "pass-back" argument suggesting that growers/feeders must be included in the industry definition only holds true if the fortunes of packers/breakers and growers/feeders do not move in the same direction (the opposite of what the USITC found and what the United States argues before us).
7.108 Furthermore, the extent to which earlier stages of input production as opposed to
processing of the final product contribute to the product's total value may change over time and may depend on the allocation of market power in the manufacturing, processing and distribution
chain, rather than on any inherent characteristics of the products involved. The availability of viable alternative uses for inputs, their ability to generate equivalent revenue and the degree to
which the inputs contribute to the value of the final product are parameters which determine, depending on market conditions, the extent of economic interdependence between input producers and
processors. We believe, however, that these parameters are not easily quantifiable or susceptible of objective assessment and cannot serve as principles of general applicability for purposes of
defining a domestic industry in a safeguard investigation. Thus, even if we were to accept arguendo that a criterion of value-added at different stages of the production chain were relevant
to the definition of a domestic industry in a safeguards investigation, we do not see how a cut-off percentage for such a test could be defined, nor at what level. (vi) Concluding remarks on past panel reports
(vi) Concluding remarks on past panel reports
7.109 In the light of the foregoing, we conclude that the reasoning of the panels in New Zealand – Transformers, US - Wine and Grapes and Canada – Beef
support the interpretation that the domestic industry should be defined as the producers as a whole of the like end-product, i.e., lamb meat in this case. We also concur with the reasoning of those
panels that separability of operations and data between different stages of production, rather than vertical integration, common ownership, continuous lines of production, economic interdependence
or substantial coincidence in economic interests are relevant for determining the scope of the industry in consistency with SG Article 4.1(c). (c) Negotiating history
(c) Negotiating history
7.110 In accordance with Article 32 of the Vienna Convention on the Law of Treaties, we refer to records of the Uruguay Round negotiations as supplementary means of interpretation in order to confirm the meaning of the text of Article 4.1(c) resulting from application of Article 31 of the Vienna Convention. Before doing so, we recall that the Canada – Beef panel's conclusion that
7.111 We thus turn to the question of whether our interpretation of SG Article 4.1(c) is confirmed by the records of the multilateral round of trade negotiations concerning contingent trade remedies following the issuance of the above-mentioned panel reports.
7.112 The Uruguay Round negotiating history reveals that the above-mentioned panel reports formed part of the basis of the discussions during the negotiations. There seems to have been a general understanding among negotiators – as suggested by the Canada – Beef panel – that broadening the industry definition standard would have required an amendment of the treaty law or at least the adoption of an agreed interpretation by negotiators.150 Given that the Canada – Beef and US – Wine and Grapes reports concerned countervailing measures, the industry definition was primarily discussed in the Negotiating Group for Subsidies and Countervailing Measures, but this question was addressed in the negotiations on anti-dumping and safeguards as well.
7.113 There were a number of specific negotiating proposals to redress the findings of the panels on Canada – Beef and US - Wine and Grapes, including from Canada, the United States and Australia. These proposals were intended to broaden the industry definition to encompass producers of inputs, at least in the case of processed agricultural products.151 However, a number of countries such as the EEC and other developed and developing countries submitted negotiating proposals in opposition to such amendment or agreed interpretation. These proposals favoured maintaining a narrow industry definition based upon like (or directly competitive) products for purposes of applying contingent trade remedies.152 While these proposals were made in the framework of the negotiations on countervailing and anti-dumping measures, the issue was briefly considered in the negotiations on safeguards as well.153
7.114 We thus conclude that the Uruguay Round proposals for and objections against changing the 'domestic industry' definition demonstrate that the issue was extensively
discussed in the Uruguay Round negotiations, especially in the context of subsidies, but also in respect of anti-dumping and safeguards. These negotiating documents also demonstrate that the
discussion was heavily influenced by the panel reports on Canada – Beef and US – Wine and Grapes. However, in the end the relevant Uruguay Round negotiating groups did not agree to
any broadening of the industry definitions in the texts of the Anti-dumping, SCM and Safeguards Agreements, and the relevant provisions remained unchanged from the predecessor provisions in the
Tokyo Round Codes. (d) "Directly competitive products"
(d) "Directly competitive products"
7.115 We recall, and wish to emphasize, that our analysis of the industry definition adopted by the USITC, and of the methodology applied by the USITC in arriving at that definition, have to do only with that part of SG Article 4.1(c) that pertains to the "like product" and the domestic industry producing it. That is, our analysis does not address the issue of "directly competitive" products and the industry producing them. Because the USITC explicitly did not make any determination concerning "directly competitive" products,154 this issue is not before us and we do not speculate as to whether live lambs conceivably could be considered "directly competitive" with imported lamb meat.155 Nor does the United States argue before us that they could.
7.116 Given that the USITC plurality did not make a finding on whether lamb meat and live lamb may be considered as "directly competitive", if we were to address this issue, we would substitute our own analysis and judgment for that of the USITC and would thus violate the principle that panels in disputes under the Safeguards Agreement must not engage in a de novo review of the evidence before a competent national authority.
7.117 This being said, it is clear on the face of the
Safeguards Agreement that the product coverage of a safeguard investigation can potentially be broader than in an anti-dumping or countervail case, to the extent that "directly competitive"
products are involved. In our view, this apparent additional latitude that exists under the Safeguards Agreement may be related to the basic purpose of the Safeguards Agreement and GATT
Article XIX, namely to provide an effective safety valve for industries that are suffering or are threatened with serious injury caused by increased imports in the wake of trade liberalization.
7.118 In the light of our considerations above, we find that the USITC's inclusion in the lamb meat investigation of input producers (i.e., growers and feeders of live lamb) as
producers of the like product at issue (i.e. lamb meat) is inconsistent with Article 4.1(c), and thus also with Article 2.1 of the Agreement on Safeguards.
7.119 A finding that the industry definition used by the USITC is inconsistent with SG Article 4.1(c) would appear to compromise the investigation and determination overall. In this respect, we recall the statements of the Appellate Body on "judicial economy" in the dispute on United States – Shirts and Blouses.156 But we also note that in a subsequent dispute on Australia – Measures Affecting the Importation of Salmon, the Appellate Body focuses on the need for panels to address all claims and/or measures necessary to secure a positive solution to a dispute and adds that providing only a partial resolution of the matter at issue would be false judicial economy.157 It is in the spirit of the Appellate Body's statements in Australia – Salmon that we continue with an analysis of other claims in the alternative, assuming arguendo either (1) that the USITC's industry definition were consistent with the Safeguards Agreement or (2) that, as the United States argues in the alternative, the USITC would have made a finding of threat of serious injury even if the industry definition had been limited to packers and breakers.
7.120 According to SG Article 4.1(b):
7.121 SG Article 4.2(a) enumerates relevant injury factors for safeguard investigations:
7.122 The USITC's determination concerning threat of serious injury reads as follows:
7.123 Australia and New Zealand criticise this determination as equivalent to a finding that – because there was not actual serious injury at the time of the USITC's determination – there must have been necessarily a threat of serious injury.159 The complainants submit that this is not a sufficient basis for a finding of imminent threat and that in fact increased imports caused neither actual injury of a serious degree nor threat thereof.
7.124 For the complainants, a finding of declines in certain indicators by itself, with no further explanation substantiating why these declines constitute a threat of a "significant overall impairment in the position of the domestic industry", is not sufficient to demonstrate the existence of imminent serious injury.160 The complainants argue in particular that the USITC's analysis of threat of serious injury is flawed because it was not "prospective", i.e., it was rather based on past data, and should, in line with the Korea – Resins panel findings161, instead have been based on projections as to how the industry was likely to perform in the immediate future.
7.125 The United States contends that the threat finding concerning
declines in various indicators and "other difficulties" demonstrates why the USITC regarded the industry as being on the verge of a significant overall impairment of its position. The United States
also submits that it based its threat determination on the most recent data available, in particular the year 1997 and interim 1998 (January ‑ September), which reflects the most recent trends and
is clearly most relevant for whether significant overall impairment of the domestic industry is imminent. (b) Interpretation by the Panel
(b) Interpretation by the Panel
7.126 Before discussing the USITC determination on the existence of threat of serious injury resulting from the lamb investigation in this dispute, we address the question of the relevant legal standard for a competent national authority to apply in determining threat of serious injury, and the benchmark for assessing the data gathered in an investigation against that standard.
7.127 The Safeguards Agreement contains no explicit guidance on any specific methodology that a competent national authority must employ when establishing threat of serious injury. The first sentence of SG Article 4.1(b) merely states that domestic industry must face "serious injury" – defined with reference to the injury factors listed in SG Article 4.2(a) – which is clearly "imminent". The ordinary meaning of "imminent" connotes that the industry's significant overall impairment needs to be "ready to take place"162 or "be impending, soon to happen … event, especially danger or disaster".163 The imminent injury that is threatened must be "serious".
7.128 In line with this emphasis on the imminent nature of threat, the article's second sentence requires that such a determination has to be based on facts and not on allegation, conjecture, or remote possibility. "Allegation" means "an assertion, especially one made without proof".164 "Conjecture" connotes "an opinion or conclusion based on insufficient evidence or on what is thought probable, guesswork, guess".165 In turn, remote "possibility" means "contingency, likelihood, chance".166
7.129 From these elements of SG Article 4.1(b), i.e., the emphasis on clear imminence of significant overall impairment, the requirement to base a threat determination on objective facts, and the rejection of "assertions", "opinions" and "conclusions" that are not based on sufficient factual evidence, it is possible to draw at least some inferences on how to conduct a threat analysis. These elements suggest (i) that a threat determination needs to be based on an analysis which takes objective and verifiable data from the recent past (i.e. the latter part of an investigation period) as a starting-point so as to avoid basing a determination on allegation, conjecture or remote possibility; (ii) that factual information from the recent past complemented by fact-based projections concerning developments in the industry's condition, and concerning imports, in the imminent future needs to be taken into account in order to ensure an analysis of whether a significant overall impairment of the relevant industry’s position is imminent in the near future; (iii) that the analysis needs to determine whether injury of a serious degree will actually occur in the near future unless safeguard action is taken.
7.130 Contextual guidance for safeguards cases may be found in the provisions of the Agreements on Antidumping (AD) and Subsidies and Countervailing Measures (SCM) providing specific rules for the determination of a threat of material injury in anti-dumping and countervailing duty investigations.167
7.131 In particular, AD Article 3.7 and SCM Article 15.7 state that in making a determination of threat of material injury, investigations "should consider, inter alia, such factors as a significant rate of increase in imports indicating a likelihood of substantially increased importation; sufficient freely disposable capacity in the exporting countries or an imminent substantial increase therein; the prices of the imported goods, as an indication of whether the imports are likely to suppress or depress the domestic producers' prices; and inventories of the product being imported".168 These provisions go on to say that the totality of the factors must lead to the conclusion that further dumped or subsidized imports are imminent, and that unless protective action is taken, material injury will occur.
7.132 The overall object and purpose of the Safeguards Agreement, as discussed in the section on domestic industry above, is to provide a mechanism for "emergency action" where, in the wake of trade liberalization, increased imports cause or threaten to cause serious injury to the domestic industry producing like or directly competitive products. This objective to provide for a remedy only in this type of emergency situation applies a fortiori when the relevant domestic industry is threatened with significant overall impairment of an imminent nature, but does not presently suffer serious injury. We cannot see how a future-oriented analysis of whether, in the absence of any safeguard action, injury of a serious degree is soon to occur could be carried out if it were not based on the most recent data available, combined with factual information as to expected future developments concerning imports and the condition of the domestic industry.
7.133 The parties refer to the reports of the panels on Korea – Resins, US – Softwood Lumber, and Mexico – Syrup169 as relevant for developing an interpretation of the standard that is required in an analysis of threat of serious injury under the Safeguards Agreement, although these reports concerned threat analyses in antidumping disputes. We find these reports relevant as well, and in our view, they stand for the general proposition that in contingent trade remedy cases an evaluation of whether threat of injury is clearly imminent requires a fact-based, future-oriented analysis.
7.134 The Korea – Resins panel found that:
The prospective analysis referred to by the Korea - Resins panel concerned the industry's current condition as well as future trends in import volumes and prices.
7.135 The panel report on US – Softwood Lumber171 affirms that such threat analysis needs to be based on objective factual evidence. It stated that "this concept had been interpreted as requiring factual evidence of a clearly foreseen and imminent change in circumstances in which subsidised imports would cause material injury. Thus a determination of threat of material injury could not be based on mere speculation as to possible future events."172 Applying this reasoning to the safeguards context, the prospective analysis of the factual evidence would need to establish that a significant overall impairment of the industry's condition would happen soon unless safeguard action were taken.173
7.136 The panel on Mexico – Syrup made a similar finding, namely that a threat determination means that "material injury would occur in the absence of an anti-dumping duty or price undertaking".174 It also makes clear that the "threat" factors enumerated in the Antidumping Agreement must be considered in addition to, and not instead of, the factors concerning the state of the domestic industry.175 Thus, at least in the context of anti-dumping and countervailing investigations, the threat analysis must take into account, in addition to the state of the industry, factors relating to the likelihood of increased imports in the immediate future at prices that are likely to suppress or depress domestic producers' prices. The Safeguards Agreement does not provide for a list of particular "threat" factors. Thus the factors for evaluating actual serious injury listed in SG Article 4.2(a) need also to be basis for an investigation of threat of serious injury. However, we believe that the above statement of the Mexico – Syrup panel provides useful guidance also for safeguards disputes, and note that it confirms our view that an examination of the existence of threat of serious injury implies a future-oriented analysis of the domestic industry’s condition which is distinct from an examination of whether actual serious injury exists.176
7.137 In the present dispute, the complainants have raised a number of interrelated questions concerning the analytical approach used by the USITC’s threat findings. In this regard, the complainants have argued (1) that the USITC failed to consider all of the factors listed in SG Article 4.2(a); (2) that the USITC failed to conduct a "prospective" analysis in reaching its conclusion that a threat of serious injury existed; and (3) that the time period focused on by the USITC in reaching this conclusion was not the correct one. In addition, the complainants have argued that the data on which the USITC relied was not sufficiently representative of the industry as a whole. Moreover, and as addressed in another section, the complainants claim that the industry definition used by the USITC is overly broad.
7.138 As we noted above, in view of our findings in respect of industry definition, we could exercise judicial economy in respect of the claims concerning the USITC’s threat finding. We further recognize that depending on our findings regarding representativeness of the data, an issue that we take up below, there might be no need to address the analytical issues that have been raised concerning the USITC’s threat finding. However, we consider it important for our task "to make such findings as will assist the DSB"177 in carrying out its dispute settlement functions that we address the threat claims as well. We do so by taking at face value, arguendo, the data and reasoning contained in the USITC’s report, and without prejudice to our above finding concerning the definition of the domestic industry in this investigation. Furthermore, while recognising the interconnectedness of the various issues raised in the context of the threat claims, we choose, again for the sake of clarity, to address these issues separately.
76 See First Written Submission of New Zealand, Annex 2-1, at section VII.G.2(a).
77 First Written Submission of New Zealand, Annex 2-1, at paragraphs 7.42-7.49, First written submission of Australia, Annex 1-1, at paragraph 113.
78 19 U.S.C. 2252(b).
79 USITC Report, Exh. US-1, at I-12.
80 Id. at I-13.
81 Id. at I-14.
82 New Zealand First Written Submission, Annex 2-1, at paragraph 7.41.
83 New Zealand Oral Statement, First Meeting of the Panel, Annex 2-5, at paragraph 30.
84 Id. at paragraphs 27-29.
85 New Zealand's Second Written Submission, Annex 2-9, at paragraph 3.4.
86 Adopted by the SCM Committee on 28 April 1992, SCM/71, BISD 39S/436.
87 Not adopted, SCM/85, dated 13 October 1987.
88 New Zealand's Responses to Questions by the Panel, Annex 2-8, Response to Question 5.
89 US First Written Submission, Annex 3-2, at paragraph 73.
90 Closing Statement of the United States at the First Meeting of the Panel, Annex 3-5, at paragraph 14. We note in this regard that footnote 61 of the USITC's determination (Exh. US-1 at. I-16) states that "…we find that all sectors show evidence of a threat of serious injury…", without elaborating. It is not clear whether the USITC meant in this statement to equate a finding that there was "evidence" of a threat of injury in respect of all sectors with a hypothetical finding that all sectors individually were threatened with serious injury. Even if this was the USITC's meaning, we do not consider that such a statement, contained in a single sentence fragment with no supporting facts or explanation, can be viewed as constituting a finding by the USITC that all industry sectors individually were threatened with serious injury.
91 The United States argues in particular that "remedial measures that addressed on the effects of imports on one aspect of a continuous line of production would be inadequate to 'prevent or remedy serious injury and to facilitate adjustment' under Article 5.1, since adjustments made by only one segment of the line of production would not insulate it from the effects of increased imports on other segments". (See US First Written Submission, Annex 3-2, at paragraph 70.)
92 Id. at paragraph 73.
93 Panel Report on New Zealand – Imports of Electrical Transformers from Finland, adopted on 18 July 1985, BISD 32S/55.
94 The United States also argues, however, that it does not in fact limit use of this test to processed agricultural products, but rather could apply it in any situation in which the two criteria were met.
95 US Response to Question 2 from the Panel, Annex 3-7, at paragraph 28.
96 US First Written Submission, Annex 3-2, at paragraph 72.
97 Id. at paragraph 75.
98 Id. at paragraph 74.
99 The USITC investigation covered only imported lamb meat, and excluded imported live sheep and live lambs. (USITC Report, Exh. US-1, at I-3, footnote 1). The USITC plurality found that the domestic product that was "like" the imported lamb meat was domestic lamb meat (Id. at I-12). Although two individual Commissioners found that domestically produced live sheep were "directly competitive" with imported lamb meat (Id. at I-8-9, footnotes 7-8), the USITC as a whole did not rely on the concept of "directly competitive" products (Id. at I-10, footnote 10). Rather, the USITC found that the domestic industry producing lamb meat encompassed both 'growers and feeders of live lambs as well as packers and breakers of lamb meat' (Id. at I-13).
100 Oxford English Dictionary, at. 2367.
101 Webster's New Encyclopaedic Dictionary, at 805.
102 Oxford English Dictionary, at 2367.
103 Webster's New Encyclopaedic Dictionary, at 805.
104 Oxford English Dictionary, at 2367.
105 Webster's New Encyclopaedic Dictionary, at 805.
106 Oxford English Dictionary, at 2040.
107 See, e.g., US First Submission, Annex 3-2, at paragraphs 63 and 126.
108 The question posed by the Panel to the United States was: "Please comment on New Zealand's argument at para. 29 of its [first] oral statement that the term 'as a whole' in Article 4.1(c) has to do with the representativeness of data used in an investigation in respect of the entire industry, and not with the scope or breadth of the domestic industry itself." The United States replied: "The term 'as a whole' is not defined by the Safeguards Agreement. While the United States supports New Zealand’s view that the purpose of the term may be to ensure that a safeguard investigation is not limited to selected individual members of an industry, it rejects the claim that 'as a whole' is a qualifying term meant to define the scope of the producers within an industry. Contrary to New Zealand's additional assertion, the United States has not used the term 'as a whole' to expand the membership of an industry beyond those who produce the 'like or directly competitive product'." US Response to Panel Question 5, Annex 3-7 (emphasis added).
109 The respective definitions in AD Article 4.1 and SCM Article 16.1 are identical to one another in pertinent part, and read as follows: "…the domestic producers as a whole of the like products or to those of them whose collective output of the products constitutes a major proportion of the total domestic production of those products."
110 See, inter alia (1) the preamble to the SG Agreement: "Recognizing the importance of structural adjustment and the need to enhance rather than limit competition in international markets"; (2) Articles 5.1 and 7.1, which provide that a Member shall apply safeguard measures only to the extent necessary and for the period necessary to prevent or remedy serious injury and to facilitate adjustment; and (3) Article 7.2 which permits the extension of a safeguard measure beyond its initial period of application if in a new investigation it is determined that the safeguard measure continues to be necessary to prevent or remedy serious injury and that there is evidence that the industry is adjusting.
111 Appellate Body Report on Argentina – Footwear, at paragraph 94.
112 Our conclusion is subject to the caveat that – in a factual situation where a so-called input product can be considered to be "like" to the final product – producers of that input product could be included in the domestic industry producing the final product.
113 We note that the reports in the latter two cases were adopted, while that in the Canada – Beef case was not. In this regard, we recall the Appellate Body's statements in Japan – Alcohol that "adopted panel reports are an important part of the GATT acquis. They are often considered by subsequent panels. They create legitimate expectations among WTO Members, and, therefore, should be taken into account where they are relevant to any dispute. However, they are not binding, except with respect to resolving the particular dispute between the parties to that dispute. (footnote omitted)." The Appellate Body further agreed with the Panel in that case that "'a panel could nevertheless find useful guidance in the reasoning of an unadopted panel report that it considered to be relevant'". See Appellate Body Report on Japan – Taxes on Alcoholic Beverages, adopted on 1 November 1996, (WT/DS8/10/11/AB/R), pp. 14-15, citing the panel report, at paragraph 6.10
114 Report of the Panel on Canada - Imposition of Countervailing Duties on Imports of Manufacturing Beef from the EEC, dated 13 October 1987, not adopted, SCM/85.
115 Report of the Panel on United States – Definition of Industry Concerning Wine and Grape Products, adopted by the Committee on Subsidies and Countervailing Measures on 28 April 1992, SCM/71, BISD 39S/436.
116 Panel Report on New Zealand – Imports of Electrical Transformers from Finland, adopted on 18 July 1985, BISD 32S/55.
117 Panel Report on United States – Wine and Grapes, op. cit., paragraph 4.2.
118 Id., at paragraph 4.3.
119 Id., at paragraph 4.6 (emphasis added).
120 Id., at paragraph 4.5.
121 US First Written Submission, Annex 3-2, at paragraph 75; USITC Report, Exh US-1, at II-4, II-26.
122 In its interim review comments, the United States argues that information on this point was contained in a 1995 USITC study on lamb meat ("Lamb Meat: Competitive Conditions Affecting the US and Foreign Lamb Industries") which was before the USITC in the safeguard investigation. We note however that this study was not part of the record before us, nor was information derived from it reproduced in the USITC's report on the safeguard investigation. Rather, only the title of the study was cited in the USITC Report. Furthermore, while in its interim review comments the United States argues that this study shows that income received by live lamb producers from shorn wool declined from 12 percent in 1990 to 5 percent in 1993, New Zealand in its comments on the US comments argues that the very same study shows that income from shorn wool plus the Wool Act payments was higher during that period, declining from 30 percent to 23 percent, and would be even higher if income from wool pelts and slipe wool were included. In any case, we recall our view that economic interdependence between producers of inputs and final products is not relevant for industry definition, and thus see no reason to further consider these statistics.
123 USITC Report, Exh. US-1, at I-30.
124 Report of the Panel on Canada – Imposition of Countervailing Duties on Imports of Manufacturing Beef from the EEC, dated 13 October 1987, not adopted, SCM/85.
125 The industry definition in Article 6.5 of the Tokyo Round Subsidies Code is identical to that in Article 16.1 of the WTO SCM Agreement
126 See, Canada – Beef, op. cit. at paragraph 5.2. In that case, virtually all of the processing operations were under separate ownership from the live cattle operations.
127 See, Canada – Beef, op. cit., at paragraph 3.12. The reasons for the Canadian Import Tribunal to define the industry in this way were that (i) "the production of manufacturing beef in Canada was a continuous sequential process commencing with the live cattle and ending with the boxed grinding beef"; (ii) "there was a high degree of functional dedication and economic dependence in this sequential process"; (iii) "no one disputed that the primary purpose of raising beef cattle was to produce beef, and that grinding beef was merely one of the product forms produced by the cattlemen". See, Canada – Beef, op. cit., at paragraph 2.2.
128 The Canada – Beef panel also recalled that although there had been proposals in the negotiations leading to the Anti-dumping Code of 1967 to allow a certain flexibility in defining the domestic industry, so as to encompass producers whose products were "competitive" or in "close competition" with the imported product, in the end the "narrow" definition of domestic industry based on the like product concept was adopted in the 1967 Code. The panel noted that that definition was imported unchanged into the Tokyo Round Code. See Canada – Beef, op. cit., at paragraph 5.11. As noted, the same definition was subsequently introduced, again unchanged, into the WTO Agreements on Anti-dumping, Subsidies and Safeguards.
129 See, Canada – Beef, original footnote 5 to paragraph 5.12: "The criterion requiring that the input in question account for 'a substantial portion of the value of the end-product' has nothing to do with the severity of the economic harm that subsidized imports may cause to any particular input supplier. In addition, while the fact that an input has 'no economically viable alternative uses' is certainly relevant, the existence of an alternative market will cushion the impact of subsidized imports only to the extent that prices in the alternative market are equal to or higher than the import-depressed price in the principal market."
130 Panel Report on Canada – Beef, op. cit., at paragraph 5.12. Emphasis added, footnotes in part omitted.
131 Id., at paragraph 5.14 (emphasis added).
133 USITC Report, Exh. US-1, at II-11ff, II-12, I-14.
134 Id. at II-29, II-33.
135 Id. at II-11-16.
136 First Written Submission of the United States, Annex 3-2, at paragraph 73.
137 "A major US packer (Transhumance) also owns both a breaker operation and Superior Farms, which is a lamb feeder". (See USITC Report, Exh. US-1, at I-14, footnote 47, II-14). Apparently, these commonly owned companies are legally separate entities and therefore, separate business operations can be identified.
138 Article 6.6 of the Tokyo Round Subsidies Code: "The effect of subsidized imports shall be assessed in relation to the domestic production of the like product where available data permit the separate identification of production in terms of such criteria as: the production process, the producers' realization, profits. When the domestic production of the like product has no separate identity in these terms the effects of subsidized imports shall be assessed by the examination of the production of the narrowest group or range of products, which includes the like product, for which the necessary information can be provided."
139 Article 15.6 of the WTO SCM Agreement: "The effect of subsidized imports shall be assessed in relation to the domestic production of the like product when available data permit the separate identification of production on the basis of such criteria as the production process, the producers' sales and profits. If such separate identification of that production is not possible, the effects of subsidized imports shall be assessed by the examination of the production of the narrowest group or range of products, which includes the like product, for which the necessary information can be provided."
140 We recall our reference to the Appellate Body's statement in Japan – Alcohol in footnote 113, above.
141 Panel Report on New Zealand – Imports of Electrical Transformers from Finland, adopted on 18 July 1985, BISD 32S/55.
142 Panel Report on New Zealand – Transformers, op. cit., at paragraph 4.6.
143 The language in the Tokyo Round Code in respect of like product and the domestic industry definition (Article 4.1) is identical to that in the WTO Anti-dumping Agreement (also Article 4.1), which as discussed above is essentially identical to the part of the language of Article 4.1(c) of the Agreement on Safeguards which is relevant to this case.
144 Panel Report on New Zealand – Transformers, op cit., at paragraph 4.6.
145 US First Written Submission, Annex 3-2, at paragraphs 69-70.
146 USITC report, Exh. US-1, at I-14 (emphasis added).
147 We note that the United States does allude to this possibility, but argues that if such benefits were to reach input producers, such safeguard actions would escape multilateral control. (See US First Written Submission, Annex 3-2, at paragraph 70.) We do not see why this would be the case, as any safeguard measure that would benefit a domestic industry, to be permitted, would have to comply with all of the relevant WTO rules, including those pertaining to defining the domestic industry. We note that the WTO rules on safeguard measures do not concern the effects of safeguard measures on any economic actors other than the domestic producers of the like or directly competitive products.
148 See Question 8 by the Panel to the United States and US Response, Annex 3-7.
149 Panel Report on Canada – Beef, at paragraph 5.13.
150 For example, Canada proposed that a "special provision" be made to clarify the term 'domestic industry', and Australia proposed to "develop an agreed and more reasonable interpretation" of the definition of "domestic industry" in the case of agricultural products.
151 Canada noted that the industry definition "under current rules" could preclude the use of countervailing duties, particularly in respect of processed agricultural products, even where subsidized imports were shown to be directly causing injury. Canada thus proposed introduction of "special provisions" to clarify the term "domestic industry" in such situations. (See MTN.GNG/NG/10/W/25, Framework for Negotiations – Communication from Canada, 28 June 1989, at Section 2(c). Canada made an identical proposal later in 1989 in the context of the anti-dumping negotiations. See MTN.GNG/NG8/W/65, dated 22 December 1989.) The US proposal explicitly referred to "at least two disputes" over what constitutes the 'domestic industry' in countervailing duty investigations involving processed agricultural products, and suggested a review of the relevant provisions focusing on the relationship between primary and processed product producers where the production of the primary product was wholly or primarily dedicated to production of the processed product. (See MTN.GNG/NG10/W/1, Communication from the United States, dated 16 March 1987, at section II.E. The United States made an identical proposal in the anti-dumping negotiations later in 1987. See GNG.MTN/NG8/W/22, dated 14.12.87.) Australia's proposal voiced concern over a panel's "unduly narrow" interpretation of the term "domestic industry", which in Australia's view would deny any remedy against injurious subsidization to producers of agricultural and other raw materials destined for transformation into a commonly traded form, and proposed the development of an agreed interpretation of the domestic industry definition, in relation to this type of product. (MTN.GNG/NG10/W/15, Communication from Australia, 30 November 1987, at paragraphs 14-15).
152 See MTN.GNG/NG10/W/7, Communication from the EEC, 11 June 1987; MTN.GNG/NG10/W/30, Communication from the Nordic Countries, 27 November 1989; MTN.GNG/NG10/W/11, Communication from Korea, 22 October 1987, as well as MTN.GNG/NG10/W/36 and MTN.GNG/NG8/W/10, dated 30.09.87), also from Korea; MTN.GNG/NG10/W/14, Communication from Egypt, 30 November 1987; MTN.GNG/NG10/W/24, Communication from Brazil, 10 November 1988; MTN.GNG/NG10/W/33, Communication from India, 30 November 1989.
153 A note by the Secretariat reporting on the 7 and 10 March 1988 meeting of the negotiating group on safeguards indicates that "[m]any delegations stressed that 'domestic producers' and 'like or directly competitive products' had to be clearly defined in order to avoid the abusive use of safeguard actions. One delegation said that there should be limits on both the upstream and downstream of products to qualify as like or directly competitive products". (See MTN.GNG/NG9/5, dated 22.04.98).
154 In particular, the USITC plurality defined lamb meat as the like product, and identified the growers, feeders, packers and breakers as producers of that like product. The USITC plurality did not define any product as "directly competitive" with lamb meat, and indeed explicitly stated that it had not made such a determination in respect of live lambs. (See USITC Report, Exh. US-1, at I-11). Therefore, it is not relevant to this Panel's review of the domestic investigation and determination that two individual Commissioners stated their view that domestically produced live sheep were "directly competitive" with imported lamb meat (Id. at I-8-9, footnotes 7-8) because the USITC as a whole did not rely on the concept of "directly competitive" products (Id. at I-10, footnote 10).
155 The interpretation of the phrase "directly competitive products" in SG Article 4.1(c) has not been addressed by any panel to date. Indeed, GATT Article III is the only context in which the concept of directly competitive products has been addressed in GATT/WTO dispute settlement practice (See Panel Report on Chile – Taxes on Alcoholic Beverages, adopted on 12 January 2000, WT/DS87/110/R, paragraphs 7.14ff. Appellate Body Report on Japan – Taxes on Alcoholic Beverages, adopted on 1 November 1996, WT/DS8/10/11/AB/R, paragraph 6.28; Panel and Appellate Body reports on Korea – Taxes on Alcoholic Beverages, adopted on 17 February 1999, WT/DS75/84/R and WT/DS75/84/AB/R, paragraph 10.38; Report of the Panel on Japan – Customs Duties, Taxes and Labelling Practices on Imported Wines and Alcoholic Beverages, adopted on 10 November 1987, BISD 34S/83; Report of the Working Party on Border Tax Adjustment, adopted on 2 December 1970, BISD 18S/97, paragraph 18). But it is not clear whether or to what extent the interpretation of this concept in the context of GATT Article III would necessarily be relevant to SG Article 4.1(c).
156 In United States – Shirts and Blouses, the Appellate Body stated:
"Nothing in this provision or in previous GATT practice requires a panel to examine all legal claims made by the complaining party. Previous GATT 1947 and WTO panels have frequently addressed only those issues that such panels considered necessary for the resolution of the matter between the parties, and have declined to decide other issues. Thus, if a panel found that a measure was inconsistent with a particular provision of the GATT 1947, it generally did not go on to examine whether the measure was also inconsistent with other GATT provisions that a complaining party may have argued were violated. …". (Footnotes omitted). See Appellate Body Report on United States – Measures Affecting Imports of Woven Wool Shirts and Blouses, adopted 23 May 1997, WT/DS33/AB/R, at.18.
157 In Australia – Salmon, the Appellate Body stated:
"The principle of judicial economy has to be applied keeping in mind the aim of the dispute settlement system. This aim is to resolve the matter at issue and 'to secure a positive solution to a dispute'. To provide only a partial resolution of the matter at issue would be false judicial economy. A panel has to address those claims on which a finding is necessary in order to enable the DSB to make sufficiently precise recommendations and rulings so as to allow for prompt compliance by a Member with those recommendations and rulings 'in order to ensure effective resolution of disputes to the benefit of all Members.'" (Footnotes omitted). See the Appellate Body Report on Australia – Measures Affecting the Importation of Salmon, adopted on 6 November 1998, WT/DS18/AB/R, paragraph 223.
158 USITC Report, Exh. US-1, at I-21.
159 Australia and New Zealand state that the USITC found that there was no present serious injury, citing, in answer to question 11 from the Panel, the following statements which were made in the USITC’s remedy recommendations: "[W]e have taken into account that the US lamb industry is not currently experiencing serious injury, but rather is threatened with serious injury" (USITC Report, Exh. US-1, at I-29); and "[W]e found a threat of serious injury … as opposed to present serious injury" (USITC Report, Exh. US-1, at I-33, fn 166).
The United States contends that there was no express statement by the USITC that there was no actual serious injury.
160 For example, Australia argues that "[t]here is no analysis in the USITC Report how 'the declines' and 'other difficulties' during the period of investigation proved that serious injury was clearly imminent in February 1999…". Australia's Response to the Panel's Question 7.
161 Panel Report on Korea – Anti-dumping Duties on Imports of Polyacetal Resins from the United States (ADP/92), adopted by the Committee on Anti-dumping Practices on 27 April 1992, BISD 40S/205.
162 Webster's New Encyclopaedic Dictionary (1994), at 496.
163 Oxford English Dictionary, at 1316.
164 Oxford English Dictionary, at 54.
165 Oxford English Dictionary, at 480.
166 Oxford English Dictionary, at 2302.
167 These provisions refer, inter alia, to those factors which the USITC took into account in its causation analysis (and which the US argues are relevant to its threat finding).
168 The list in the SCM Agreement also includes the "nature of the subsidy or subsidies", and the likely trade effects thereof (presumably referring to whether the subsidies are export subsidies or import subsidies, as opposed to production or other "domestic" subsidies). The AD Agreement in a footnote provides as an example "though not an exclusive one", "convincing reason to believe that there will be, in the near future substantially increased importation of the product at dumped prices".
The Tokyo Round Anti-dumping Code under which the Korea - Resins dispute was adjudicated did not elaborate on the nature of the factors to be examined in a threat case except to cite as a possible example the convincing reason to believe that there would be substantially increased dumped imports. The Tokyo Round Subsidies Code in the context of threat cited only the nature of the subsidy and its likely trade effects. Thus, it was in the Uruguay Round that the more elaborated framework for assessing threat, which refers almost exclusively to future developments in imports, and which closely resembles the analysis set forth in Korea - Resins was explicitly introduced into the relevant provisions concerning anti-dumping and countervailing duty investigations.
169 Panel Report on Mexico – Anti-Dumping Investigation Of High Fructose Corn Syrup (HFCS) from The United States, WT/DS132/R and Corr.1, not appealed, adopted 24 February 2000.
170 Panel Report on Korea – Resins, op. cit. at paragraph 271.
171 Panel Report on United States – Measures Affecting Imports of Softwood Lumber from Canada, adopted by the Committee on Subsidies and Countervailing Measures on 27-28 October 1993, SCM/162, BISD 40S/358.
172 Panel Report on US – Softwood Lumber, op.cit., at paragraph 402.
173 In the context of safeguard measures under the Agreement on Textiles and Clothing, the panel on US – Underwear noted that a threat finding has to "demonstrate that unless action is taken, damage will most likely occur in the near future." That panel also affirmed the need for a prospective analysis. See Panel Report on United States – Restrictions on Imports of Cotton and Man-Made Fibre Underwear, WT/DS24/R, adopted on 25 February 1997, at paragraph 7.55.
174 Panel Report on Mexico – Syrup, op. cit., at paragraph 7.125.
175 Panel Report on Mexico – Syrup, op. cit., at paragraph 7.131 et seq.
176 See, also Argentina – Footwear, op. cit., at paragraph 8.284, in which the Panel found that an analysis of threat of serious injury in the safeguards context is separate from an analysis of actual serious injury: "[t]he question of threat, whether instead of or in addition to a finding of present serious injury, must be explicitly examined in an investigation and supported by the evidence in accordance with Article 4.2(a-c).
177 See Article 7.1 of the DSU.