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 820
In addition, the USITC appears to have dismissed this factor in its non-attribution analysis merely on the basis that "the cost advantage enjoyed by minimills existed throughout the period of investigation." In the Panel's view, the fact that a factor existed throughout the period of investigation does not necessarily mean that it cannot play a role in causing serious injury. Moreover, changing circumstances in a market may result in a number of factors, that previously seemed harmless, playing a significant role in causing serious injury.
In failing to adequately analyse this factor, the Panel considers that the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by this factor, together with other factors, was not attributed to increased imports.
Legacy costs
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(i) supra.
Analysis by the Panel
It seems to the Panel from the USITC's Report that the USITC considered that legacy costs played a role in causing the injury that was being suffered by the domestic industry. To us, this is apparent from the following comment made by the USITC: "The funding of legacy costs is a vexing problem for the domestic industry, and evidence on the record indicates that legacy costs have prevented needed consolidation within the domestic industry from taking place".5334 We note in this regard that this statement is made in the present tense, indicating that legacy costs are currently a vexing problem and not only a problem of the past.
The Panel notes that the USITC did consider the effect on the market of legacy costs. Specifically, in pages OVERVIEW 31 – OVERVIEW 35, the USITC describes pensions and other post-employment benefits for steel company retirees. In Table OVERVIEW-9, the USITC sets out post-employment benefit data of selected steelmakers for the fiscal years 1996 – 2000.
However, even though it effectively acknowledged the role played by legacy costs in causing injury, the USITC appeared to dismiss this factor in its non-attribution analysis merely on the basis that this factor existed prior to the period of investigation. In particular, the USITC stated that "the issue of 'legacy costs' is not a new one to this industry".5335 In the Panel's view, that a factor pre-dated the period of investigation does not necessarily mean that it cannot play a role in causing serious injury during the period of investigation itself. Nor does the Panel consider that a reduction in the level of legacy costs during the period of investigation will necessarily mean that such costs could not and did not cause injury to the relevant domestic producers.
The Panel also notes that the USITC stated in its Report that "[t]he difficulties in meeting these [legacy cost] obligations were recognized before the POI, and the domestic industry was able to earn a reasonable rate of return in 1996 and 1997 despite these costs. Respondents have offered no reason why the industry's longstanding problem would cause no injury in 1996 or 1997 but then begin to depress prices and strangle revenue in 1998-2000. Legacy costs may have left certain members of the domestic industry less able to compete with low-priced imports, but are not responsible for the
5334 See para. 10.382.
5335 See para. 10.382.
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 821
low prices that have injured the industry."5336 In our view, the foregoing amounts to an acknowledgement that legacy costs compromised the competitive position of certain domestic producers. However, this effect was dismissed on the basis of the rather cursory and unsubstantiated assertion that "legacy costs are not responsible for the low prices that have injured the industry." The Panel considers that given the apparent significance of legacy costs to the situation of the domestic industry, it was incumbent upon the USITC to further examine this issue.
In failing to adequately analyse this factor, the Panel considers that the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by this factor, together with other factors, was not attributed to increased imports.
Conclusions
The Panel considers that, with respect to CCFRS, the USITC failed to comply with its non- attribution obligation contained in the second sentence of Article 4.2(b). In particular, we consider that the USITC failed to properly separate, distinguish and assess the nature and extent of the injurious effects of factors other than increased imports that were causing injury to the relevant domestic industry. This, to us, is clear from the fact that the USITC dismissed a number of factors (namely, declining domestic demand, domestic capacity increases, intra-industry competition and legacy costs) in its non-attribution analysis even though it acknowledged that those factors were causing injury to the industry.
The Panel also recalls that the USITC disregarded the effect of increases in domestic capacity, intra-industry competition and legacy costs because "they were not a cause of serious injury that was equal to or greater than the injury caused by increased imports".5337 The Panel considers that such an approach is problematic if the cumulative effects of individual factors are not analysed or assessed in cases where, individually, each of them are acknowledged to have caused some injury to the relevant domestic industry. In the case of CCFRS, by discarding factors that individually caused injury to the industry, we consider that the USITC failed to distinguish and assess the nature and extent of the effects of these other factors taken together, as distinct from those injurious effects caused by increased imports.
Therefore, the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by other factors, such as declines in demand, domestic capacity increases, intra-industry competition and legacy costs, together with other factors, was not attributed to increased imports of CCFRS.
Relevance of the product definition for CCFRS
The Panel would like to address some of the arguments made by the parties regarding the product definition for CCFRS, particularly relating to the USITC's causation analysis.
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(i) supra.
5336 USITC Report, Vol. I , p.64.
5337 See para. 10.382.
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 822
Analysis by the Panel
The Panel notes that we are not, in this section, evaluating arguments made by complainants that the USITC's grouping of the items of CCFRS is inconsistent with Article 2.1 because it violates the obligation to identify a specific imported product. Nor is the Panel dealing here with the argument that the USITC acted inconsistently with Articles 2.1 and 4.1(c) of the Agreement on Safeguards in its definition of the domestic industry that produces products that are like CCFRS. The Panel will confine its attention in this section to arguments made that the product defined as CCFRS was such that it could not be subjected to the application of the causation requirements contained in Article 4.2(b).
The Panel recalls the text of Article 4.2(b) of the Agreement on Safeguards as the starting- point for its analysis in this respect:
"The determination referred to in subparagraph (a) shall not be made unless this investigation demonstrates, on the basis of objective evidence, the existence of the causal link between increased imports of the product concerned and serious injury or threat thereof. When factors other than increased imports are causing injury to the domestic industry at the same time, such injury shall not be attributed to increased imports."
We note that, according to Article 4.2(b), the causal link must exist "between increased imports of the product concerned and serious injury or threat thereof". Serious injury is defined in Article 4.1(a) as "a significant overall impairment in the position of a domestic industry." "Domestic industry" is defined, in turn, in Article 4.1(c) as "the producers as a whole of the like or directly competitive products operating within the territory of a Member or those whose collective output of the like or directly competitive products constitutes a major proportion of the total domestic production of those products."
Reading these provisions together, it is clear that, under Article 4.2(b), a causal link must be established between, on the one hand, increased imports of the product concerned and, on the other hand, serious injury or threat thereof suffered by producers of the like or directly competitive products. In our view, the imported product and the like or directly competitive products must be defined in such a way that the causal link analysis required by Article 4.2(b) can be undertaken. More particularly, they must be defined in such a way that, for example, a coincidence or a conditions of competition analysis may be undertaken. They must also be defined in such a way that it can be established that injury suffered by producers of the like or directly competitive products caused by factors other than increased imports is not attributed to the increased imports. In our view, if the imported products or the like or directly competitive products are defined in such a way that prevents the proper application of the causation requirements contained in Article 4.2(b), the causation determination will necessarily be inconsistent with the prescriptions of Article 4.2(b).
In our view, CCFRS was defined in such a way that prevented the proper application of the causation requirements contained in Article 4.2(b). We consider that the USITC itself effectively admitted that CCFRS could not be subjected to the application of the causation requirements given the fact that, on a number of occasions, it relied upon data for the items that constituted CCFRS rather than for CCFRS as a whole without explaining why and how such specific data on such items related to the determination concerning CCFRS as a whole. In addition, the USITC itself admitted that the reliance on combined data for "the five types of certain carbon flat-rolled steel … may involve
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 823
double-counting."5338 Finally, as noted above, we do not consider that the grouping of the various products that constituted CCFRS renders it amenable to conditions of competition analysis because it becomes difficult, if not impossible, for the competent authority to identify the proper locus of competition while undertaking a conditions of competition analysis for the purposes of establishing a causal link for CCFRS.
Overall conclusion on USITC's determination of a causal link
As indicated above, the Panel found that the USITC did not provide a reasoned and adequate explanation of how the facts supported its finding that coincidence existed in this case. Nor did the USITC provide a compelling explanation that demonstrated the existence of a causal link between increased imports and serious injury suffered by domestic producers of CCFRS in the absence of coincidence. Further, the USITC's non-attribution analysis failed to separate, distinguish and assess the nature and extent of the injurious effects of declines in demand, domestic capacity increases, intra- industry competition and legacy costs so that the injury caused by these factors, together with other factors, was not attributed to increased imports. Thus, the USITC did not provide a reasoned and adequate explanation supporting a determination that there was a genuine and substantial relationship of cause and effect between increased imports and serious injury to the relevant domestic producers.
Therefore, the Panel concludes that the USITC's finding that a causal link existed between imports of CCFRS and injury caused to the relevant domestic producers is inconsistent with Articles 4.2(b), 2.1 and 3.1 of the Agreement on Safeguards.
Tin mill products
As we did in relation to our findings on increased imports for tin mill products (paragraphs 10.191-10.200 above), the Panel needs to address the issue of the divergent findings made by individual USITC Commissioners: four of the six Commissioners made findings on tin mill as a separate product5339, but the two other Commissioners (Bragg and Devaney) treated tin mill products as part of the larger CCFRS category.5340 The four who examined tin mill as a separate product made a common affirmative finding on increased imports and on serious injury, but later diverged on the question of causation, for which only Commissioner Miller made an affirmative determination.5341 Ultimately, therefore, only Commissioner Miller reached positive findings regarding tin mill as a separate product. The two Commissioners who treated tin mill as part of the CCFRS category, reached a positive conclusion on that larger category. Despite the divergent product definitions, the USITC Report concluded that three Commissioners made "an affirmative determination regarding imports of carbon and alloy tin mill products."5342
In the March Proclamation, the President did not select any of the various affirmative determinations on tin mill as the basis of the decision to impose the safeguard measure on tin mill. Rather, pursuant to domestic law, the President "decided to consider the determinations of the groups of commissioners voting in the affirmative with regard to [tin mill products and stainless steel wire] to be the determination of the [US]ITC".5343 It, therefore, is apparent that the President based his determination on the findings of all three Commissioners (Bragg, Devaney and Miller), although
5338 USITC Report, Vol. I, p. 51, note 193.
5339 USITC Report, Vol. I, pp. 71 et seq.
5340 USITC Report, Vol. I, p. 71, footnote 368 and p. 279.
5341 USITC Report, Vol. I, pp. 307-309.
5342 USITC Report, Vol. I, p. 25.
5343 Proclamation No. 7529 of 5 March 2002, Federal Register, Vol. 67, No. 45, p. 10553.
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 824
those three Commissioners did not perform their analysis on the basis of the same like product definition.
In this regard, the Panel refers to its discussion in the context of its review of the USITC's increased import determination in paragraphs 10.191-10.200 above. In sum, the Panel finds that a Member is not permitted to base its safeguard measures on an explanation that consists of alternative explanations which, given the different products upon which such explanations are based, cannot be reconciled as a matter of substance. Therefore, it is our view that the USITC Report does not contain a reasoned and adequate explanation of how the facts support the determination that increased imports of tin mill products caused serious injury to the relevant domestic industry, as required by Articles 2.1, 4.2(b) and 3.1 of the Agreement on Safeguards.
Hot-rolled bar
The Panel notes at the outset that it has focused in this section on the arguments made by the complainants that, for us, raised the most problematic aspects of the USITC's determinations on causation, that is, those aspects that more obviously entailed violations of the Agreement on Safeguards. Since we will dispose of the complainants' claims in our review below, the Panel sees no need to deal with the other arguments.
Coincidence and conditions of competition
USITC findings
The USITC's findings read as follows:
"We find that the increased imports of hot-rolled bar are an important cause, and a cause not less than any other cause, of serious injury to the domestic industry. Accordingly, we find that increased imports of hot-rolled bar are a substantial cause of serious injury to the domestic hot-rolled bar industry.
Conditions of Competition
We have taken into account a number of factors that affect the competitiveness of domestic and imported hot-rolled bar in the US market, including factors related to the product itself, the degree of substitutability between the domestic and imported articles, changes in world capacity and production, and market conditions. These factors affect prices and other considerations taken into account by purchasers in determining whether to purchase domestically-produced or imported articles.
Market participants generally agree that there are few or no substitutes for long products such as hot-rolled bar.5344 As discussed in section V.A.1. above, hot- rolled bar is used in construction, automotive equipment, and industrial applications. Hot-rolled bar encompasses a wide range of products including merchant bar, special bar quality steel bars, and light shapes.5345
5344 (original footnote) CR and PR at LONG- 78.
5345 (original footnote) See CR and PR at LONG-1.
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 825
The record indicates strong demand during the period examined, with apparent US consumption of hot-rolled bar increasing during every full-year but one of the period. Apparent consumption rose during the first three years of the period, increasing from 10.0 million tons in 1996 to 11.7 million tons in 1998. It then declined to 11.0 million tons in 1999 but increased to 11.2 million tons in 2000. Apparent consumption was lower in interim 2001, at 4.9 million tons, than in interim 2000, when it was 6.0 million tons.5346
With regard to supply of hot-rolled bar, US capacity reported in questionnaires increased slightly from 1996 to 2000, but overall industry capacity declined during the period examined. The domestic industry's capacity utilization fluctuated over the period examined. Capacity utilization for full-year periods ranged between 67.2 percent in 1996 to 74.3 percent in 1998. Foreign capacity reported in questionnaires increased from 26.7 million tons in 1996 to 29.8 million tons in 2000, and was higher in interim 2001 than in interim 2000. Foreign capacity utilization for full-year periods ranged from 74.3 percent in 1999 to 79.4 percent in 2000.5347
Price is a moderately important factor in purchasing decisions for hot-rolled bar. Price was listed as the top factor in purchasing decisions by 27.8 percent of hot- rolled purchasers in their questionnaire responses. While more purchasers listed quality than price as their top factor in purchasing decisions, they generally deemed domestically-produced hot bar and imports to be comparable with respect to the particular quality considerations most important in their purchasing decisions.5348
Analysis5349
Through price-based competition, the increased imports caused domestic hot- rolled bar producers to lose market share at the same time prices were falling. The resulting loss in revenues led to the poor operating results and plant closures discussed above.
The timing of domestic producers' price declines do not correspond precisely to the timing of the import surges. The record, however, indicates that imports had a negative effect on prices and that the domestic industry used different strategies over the course of the period examined to compete with the imports. The largest increase in hot-rolled bar imports occurred in 1998, shortly following the financial crisis that led to sharply decreased steel consumption in several Asian countries. Import volumes increased by 29.5 percent from 1997 to 1998.5350 During 1998, the imports consistently undersold the domestically-produced product. Underselling margins for
5346 (original footnote) CR and PR, Table LONG-70.
5347 (original footnote) CR and PR, Table LONG-42. We have relied upon the questionnaires for foreign capacity and capacity utilization data, although such data are not complete. We acknowledge that the
domestic producers contended that the questionnaire data understated foreign capacity and overstated foreign capacity utilization.
5348 (original footnote) INV-Y-212 at 45.
5349 (original footnote) The Minimill 201 Coalition produced an economic model that attempted to
measure the relationship between imports and the domestic industry's prices and profits. We considered this model in making our determination but note its limitations. In particular, there were defects in the manner the model measured import competition, and the model did not adequately address changes in domestic competition.
5350 (original footnote) CR and PR, Table LONG-5.
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 826
the hot-rolled bar product on which the Commission collected pricing data, which hovered around 5.0 percent during the first three quarters of 1998, increased to 7.0 percent in the fourth quarter.5351
Domestic producers generally maintained their prices in 1998, but at the cost of market share. The average unit values of US shipments increased by a very slight
0.3 percent from 1997 to 1998.5352 Prices for the domestically-produced hot-rolled bar product on which the Commission collected data remained generally stable during
the first three quarters of 1998. Indeed, prices for the domestically-produced product during these three quarters exceeded prices during any other portion of the period examined. Prices for the domestically-produced product did fall slightly – by 3 percent – between the third and fourth quarters of 1998.5353
As a result of maintaining prices, the domestic industry maintained its operating margins, which declined by only three-tenths of a percentage point from 1997 to 1998. However, total operating income declined by 9.3 percent during this period.5354 The industry also lost 4.1 percentage points of market share to the imports. This was the largest drop in domestic producers' market share over the period examined.5355
In 1999 the domestic industry responded to the import competition by reducing prices in an attempt to maintain market share. Import volumes remained high, with import market share rising slightly from 20.1 percent in 1998 to 20.4 percent in 1999.5356 Moreover, inventories held by US importers had increased sharply in 1998.5357 Thus, imports continued to be a significant competitive factor in 1999 although the quantity of imports that year was below the level of 1998. Prices for the domestically-produced hot-rolled bar product on which the Commission collected data declined by 7.8 percent from the fourth quarter of 1998 to the first quarter of 1999, and fluctuated within a narrow range during the remaining three quarters of 1999. During this period, the domestic producers' prices were below those of the imports.5358 Domestic producers' average unit values showed comparable declines.5359 As a result, in 1999 domestic producers held their loss of market share to three-tenths of a percentage point.5360 Nevertheless, because declines in the domestic industry's average unit sales values exceeded declines in the average unit costs of goods sold, its operating margin fell.5361
In 2000, the domestic industry initially increased prices. Prices for the domestically-produced hot-rolled bar product for which the Commission collected data rose during the first quarter of 2000, although pricing levels remained below those of 1998. In the first half of the year, however, underselling by the imports
5351 (original footnote) INV-Y-212, Table LONG-ALT-90.
5352 (original footnote) CR and PR, Table LONG-16.
5353 (original footnote) INV-Y-212, Table LONG-ALT-90.
5354 (original footnote) CR and PR, Table LONG-27. 5355 (original footnote) CR and PR, Table LONG-70. 5356 (original footnote) CR and PR, Table LONG-70. 5357 (original footnote) CR and PR, Table LONG-C-3.
5358 (original footnote) INV-Y-212, Table LONG-ALT-90.
5359 (original footnote) CR and PR, Table LONG-27. 5360 (original footnote) CR and PR, Table LONG-70. 5361 (original footnote) CR and PR, Table LONG-27.
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 827
resumed.5362 The imports consequently gained 1.7 percentage points of market share from the conclusion of 1999 to June 2000. In response, the domestic producers again cut prices during the second half of 2000. Prices declined by 6.1 percent between the second and third quarters of 2000, and by another 2.3 percent between the third and fourth quarters.5363
These price declines mitigated, but did not eliminate, further erosion in the domestic industry's market share.5364 Indeed, the domestic industry sold less tonnage in 2000 than in 1999, although total US consumption was greater in 2000.5365 Also, price declines during the second half of the year negated the price increases during the first half of the year – average unit sales values were unchanged in 2000 from 1999.5366 The combination of lost market share, lower sales volumes, and lower prices during 2000 -- all of which were linked to the increased imports -- led to the industry's poor operating performance and closure of productive facilities. We consequently conclude that the increased imports were an important cause of the serious injury sustained by the domestic hot-rolled bar industry."5367
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.2(c) supra.
Analysis by the Panel
The Panel has examined the relevant section of the USITC Report for hot-rolled bar and notes that, in determining causal link, the USITC did not conduct a coincidence analysis. As mentioned previously, the Panel considers that if a competent authority has not examined coincidence of trends, it must, in proving causation, provide a reasoned and adequate explanation as to why such an analysis was not undertaken as well as a compelling explanation establishing the existence of a causal link. We note that for hot-rolled bar, the USITC analysed the conditions of competition in the hot-rolled bar market. Accordingly, we will now proceed to review the conditions of competition analysis undertaken by the USITC for this measure, with a view to determining whether the USITC provided such a compelling explanation.
As a starting point, we note that the essential premise for the USITC's determination of a causal link between increased imports and serious injury was the existence of price-based competition between imported and domestic products. The USITC conceded that: "The timing of domestic producers' price declines do not correspond precisely to the timing of the import surges". However, it went on to state that: "The record, however, indicates that imports had a negative effect on prices and that the domestic industry used different strategies over the course of the period examined to compete with the imports". Further, the USITC concluded that: "The combination of lost market share, lower sales volumes, and lower prices during 2000 – all of which were linked to the increased imports – led to the industry's poor operating performance and closure of productive facilities. We consequently
5362 (original footnote) INV-Y-212, Table LONG-ALT-90.
5363 (original footnote) INV-Y-212, Table LONG-ALT-90. Price declines continued through the first two quarters of 2001. Ibid.
5364 (original footnote) The domestic industry's market share was 77.0 percent in the second half of 2000, as opposed to 77.9 in the first half of the year. CR and PR, Table LONG-70.
5365 (original footnote) CR and PR, Table LONG-C-3. 5366 (original footnote) CR and PR, Table LONG-27. 5367 USITC Report, Vol. I, pp. 95-97.
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 828
conclude that the increased imports were an important cause of the serious injury sustained by the domestic hot-rolled bar industry".5368
Set out below is a graphical representation of import and domestic pricing trends during the period of investigation. This graph has been generated using USITC data. We note that at every point of the period of investigation, import prices exceeded domestic prices. This is not inconsistent with the overall observations made by the USITC regarding the relative prices for import and domestic products.5369
12,000,000 600
10,000,000 500
8,000,000 400
6,000,000 300
4,000,000 200
2,000,000 100
0 0
1996 1997 1998 1999 2000 2001
Imports Imports (semi-annual)
Domestic unit values Imports unit values
Domestic unit values (semi-annual) Imports unit values (semi-annual)
The USITC explained that domestic prices declined in an effort to mitigate the erosion of market share. Set out below is another graph, again generated using USITC data, indicating the import market share during the course of the period of investigation, which tends to support the USITC's conclusion that the domestic industry lost market share in favour of imports.5370
5368 See para. 10.424.
5369 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-5 at LONG-9; Table LONG-16 at LONG-21.
5370 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-5 at LONG-9; Table LONG-70 at LONG-67; Table LONG-C-3.
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 829
8,000,000
7,000,000
6,000,000
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
100
90
80
70
60
50
40
30
20
10
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Imports' market shares
Imports' market shares (semi-annual)
On the basis of the foregoing, overall, we find that the USITC's conditions of competition analysis was compelling in providing indications of the existence of a causal link between increased imports of hot-rolled bar and serious injury, subject, of course, to fulfilment of the non-attribution requirement.
Non-attribution
USITC findings
The USITC's findings read as follows:
"We next consider whether there is any other cause of injury to the domestic hot- rolled bar industry as substantial as the increased imports. Respondents initially contend that competition among domestic producers is at least as great a cause of injury to the domestic industry as increased imports. In particular, they assert that domestic producer Nucor is a market leader that drives down prices. They contend that, through its price leadership, Nucor has increased its market share and made its domestic competitors less profitable.5371
We observe initially that competition among domestic producers cannot provide any explanation for certain indicia of serious injury. While competition among domestic producers might explain why some individual producers gained market share during the period examined while others lost market share, it cannot explain why the domestic industry as a whole lost market share over the period examined to the imports. The imports' share of the quantity of US apparent consumption rose from
16.5 percent in 1996 to 22.5 percent in 2000, and was higher in 2000 than at any other point during the period examined.5372 As previously discussed, this loss in market share is a critical component in our causation analysis; the price declines that
5371 (original footnote) See Hot-Rolled Bar Respondents Prehearing Brief at 58-60.
5372 (original footnote) CR and PR, Table LONG-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 830
occurred during the period examined were a function of the industry's efforts to preclude or mitigate losses in market share in the face of increased import volumes.
We additionally examined data concerning Nucor to ascertain the extent to which it was a "price leader" and whether its pricing policies served to increase its market share vis a vis other domestic producers, as respondents contend. The data do not support the notion that Nucor was a primary source of pricing declines. While Nucor's average unit values were ***.5373 ***.5374
The data additionally do not establish that Nucor ***.5375 We consequently conclude that Nucor's pricing practices cannot provide any explanation for the serious injury experienced by the domestic industry. Moreover, neither Nucor's practices nor internal industry competition in general can explain why the domestic industry as a whole lost market share to the imports.
Respondents next contend that inefficient producers are a larger cause of any serious injury to the domestic industry than increased imports. They contend that domestic producers *** have much higher costs than industry averages and lost money throughout the period examined regardless of market conditions.5376
Respondents' theory fails on two accounts. First, if the difficulties of *** were due to their inefficiency relative to other domestic producers, one might expect that they would lose market share to other domestic producers that are more efficient and could therefore offer lower prices for their products. This, however, was not the case. Jointly, *** accounted for a higher proportion of the quantity of US producers' commercial sales in 2000 – at *** percent -- than they did in 1996, when they jointly accounted for *** percent of such sales.5377 Consequently, the so-called "inefficiency" of *** was not causing them to lose market to their domestic competitors. Second, if *** were aberrational performers, as respondents contend, one would expect their performance trends to differ from the other domestic producers. This was also not the case. Declines in operating performance were pervasive among hot-rolled bar producers. While *** were the only domestic producers to experience operating losses in 1997, four additional firms experienced operating losses in 1998, and four more producers beyond that experienced operating losses in 2000.5378 Thus, at most *** consistent operating losses served to make overall domestic industry operating performance consistently worse than it would have been had these two firms not been in the domestic industry. These firms' performance, however, cannot explain the overall declines in operating performance among domestic hot-rolled bar producers, the increasing incidence of operating losses, or the industry's overall loss of market share to the imports. Because neither structural problems nor the poor performance of *** can explain the domestic
5373 (original footnote) Nucor's average unit values were ***. Questionnaire Data, INV-Y-212.
5374 (original footnote) See Producer's Questionnaires.
5375 (original footnote) Nucor's share of the quantity of domestic hot-rolled bar producers' commercial sales was *** in 1996, *** in 1997, *** in 1998, *** in 1999, and *** in 2000. Questionnaire Data, INV-Y-212.
5376 (original footnote) See Hot-Rolled Bar Respondents Prehearing Brief at 80-81.
5377 (original footnote) Questionnaire Data, INV-Y-212.
5378 (original footnote) Questionnaire Data, INV-Y-212. Moreover, as previously stated, three producers that did not respond to the questionnaires declared bankruptcy and shut down production operations altogether in interim 2001.
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 831
industry's serious injury, we conclude that the alleged inefficiency of these two firms cannot be a more important cause of injury than increased imports.
We have also examined the role of changes in demand in explaining the serious injury of the domestic industry. We observe that US apparent consumption, measured by quantity, increased by 11.7 percent from 1996 to 2000. The increase was not evenly distributed throughout the period examined, and apparent consumption peaked in 1998. We observe, however, that during this period apparent consumption declined only from 1998 to 1999, when the domestic industry maintained profitable operating performance. From 1999 to 2000, however, apparent consumption rose – yet the domestic industry became unprofitable. That domestic performance reached injurious levels in 2000, a time of rising apparent consumption, indicates to us that changes in demand cannot be a cause of the serious injury evident at that time.5379
Finally, we have examined changes in input costs as a possible source of serious injury to the domestic industry. We note that costs declined during the period and observe that declines in input costs, in and of themselves, cannot be an alternative "cause" of injury. At most, a decline in input costs may indicate that a factor other than imports may be responsible for price declines.
For hot-rolled bar, unit cost of goods sold (COGS) declined from $399 in 1996 to
$362 in 1999, and then increased to $380 in 2000; unit raw material costs declined throughout the period examined.5380 As previously stated, demand for hot-rolled bar was higher in 1999 than in 1996 and was higher in 2000 than in 1999. In times of increasing demand, producers normally need not cut their prices to reflect fully declines in cost of goods sold. Yet from 1996 to 1999, the domestic industry's declines in average unit sales values outpaced the decline in unit COGS. From 1999 to 2000, when unit COGS increased, unit average sales values remained the same. If the domestic industry could have increased its average unit sales values in 2000 to reflect increasing COGS – a reasonable expectation during a year of increasing demand – the industry could have maintained positive operating margins of at least the levels of 1999. As explained above, however, the industry could not sustain whatever price increases it initiated in 2000 because of that year's import surge. Because we cannot attribute the domestic industry's declines in operating performance in 2000 to increases in COGS, we conclude that changes in input costs cannot be as important a cause of serious injury as increased imports.
We consequently conclude that alternative causes cannot individually or collectively explain the serious injury to the domestic industry, particularly the declining market share over the course of the period examined, and the deteriorating operating performance leading to negative operating margins for the domestic industry in 2000.
5379 (original footnote) CR and PR, Table LONG-C-3. We observe that, during interim 2001, when apparent consumption fell significantly, the domestic industry experienced further declines in operating performance. The interim 2001 data merely indicate that declines in apparent consumption can lead to further deterioration to an industry that was already seriously injured.
5380 (original footnote) CR and PR, Table LONG-27.
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 832
Accordingly, we find that increased imports are a substantial cause of serious injury to the domestic hot-rolled bar industry that is not less than any other cause."5381
Factors considered by the USITC
Competition among domestic producers
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(iii) supra.
Analysis by the Panel
The Panel agrees with the United States insofar as it stated that the USITC dismissed this factor as a possible cause of injury to the industry. In particular, the USITC stated that: "We observe initially that competition among domestic producers cannot provide any explanation for certain indicia of serious injury". In addition, it stated that: "We consequently conclude that Nucor's pricing practices cannot provide any explanation for the serious injury experienced by the domestic industry. Moreover, neither Nucor's practices nor internal industry competition in general can explain why the domestic industry as a whole lost market share to the imports".5382
We note that the complainants arguments with respect to this factor are premised on the assumption that the USITC acknowledged that competition among domestic producers was a cause of injury. However, as noted above, this assumption is not valid. Further, in our view, the complainants have not put forward an alternative plausible explanation that, in fact, competition among domestic producers was a cause of serious injury.
Inefficient producers
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(iii) supra.
Analysis by the Panel
The Panel notes at the outset that the USITC stated that "the alleged inefficiency of these two firms cannot be a more important cause of injury than increased imports." It could be concluded from this statement, taken in isolation, that the USITC considered that inefficient producers were a cause of injury, albeit not a cause that was more important than increased imports. However, in the Panel's view, in light of the remainder of the USITC's analysis, it would seem that the USITC made this statement merely in keeping with domestic law requirements. On the contrary, the substance of the USITC's analysis indicates that the USITC dismissed this factor as a possible cause of injury to the industry. In particular, the USITC stated that: "Respondents next contend that inefficient producers are a larger cause of any serious injury to the domestic industry than increased imports…Respondents' theory fails on two accounts…" In addition, the USITC stated that: "These firms' performance, however, cannot explain the overall declines in operating performance among domestic hot-rolled bar producers, the increasing incidence of operating losses, or the industry's overall loss of market share to the imports. Because neither structural problems nor the poor performance of *** can explain the
5381 USITC Report, Vol. I, pp. 97-99.
5382 See para. 10.431.
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 833
domestic industry's serious injury, we conclude that the alleged inefficiency of these two firms cannot be a more important cause of injury than increased imports."
We note that the complainants' arguments with respect to this factor are premised on the assumption that the USITC acknowledged that inefficient producers were a cause of injury. However, as noted above, this assumption is not valid. Further, in our view, the complainants have not put forward an alternative plausible explanation that, in fact, inefficient producers were a cause of serious injury.
Changes in input costs
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(iii) supra.
Analysis by the Panel
The Panel notes that although the USITC did not expressly state that increases in COGS played a role in the decline in the domestic operating margin, it did explicitly state that COGS increased for hot-rolled bar from 1999 to 2000. In particular, the USITC stated that: "For hot-rolled bar, unit cost of goods sold (COGS) declined from $399 in 1996 to $362 in 1999, and then increased to $380 in 2000". In addition, the USITC stated that: "If the domestic industry could have increased its average unit sales values in 2000 to reflect increasing COGS – a reasonable expectation during a year of increasing demand – the industry could have maintained positive operating margins of at least the levels of 1999. As explained above, however, the industry could not sustain whatever price increases it initiated in 2000 because of that year's import surge. Because we cannot attribute the domestic industry's declines in operating performance in 2000 to increases in COGS, we conclude that changes in input costs cannot be as important a cause of serious injury as increased imports".
In the Panel's view, the USITC's dismissal of the effect of increases in COGS in its non- attribution analysis was not adequately reasoned. In particular, the USITC merely stated that the only reason why the domestic industry did not increase prices to recoup growing COGS was the import surge that occurred in the year 2000. This, in the Panel's view, did not amount to a reasoned and adequate explanation. Nevertheless, the Panel does consider that the USITC was probably correct in concluding that changes in input costs were not a cause of serious injury. If, indeed, COGS was playing a significant role in situation of the domestic industry, one would have expected operating margins to increase while COGS was decreasing, in particular, from 1996 until 1999 inclusive. However, as can be seen from the graph below, which has been generated using USITC data, the trends in operating margin appear to be independent of trends in COGS. While it is true that there appears to be coincidence between, on the one hand, increases in COGS from 1999 until 2000 and, on the other hand, declines in the operating margin during that period, the Panel considers that coincidence during one brief window in the period of investigation cannot detract from a lack of coincidence during the rest of the period of investigation.5383
5383 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-27 at LONG-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 834
500
450
400
350
300
250
200
150
100
50
0
250,000
200,000
150,000
100,000
50,000
0
-50,000
-100,000
1996 1997 1998 1999 2000 2001
Years Cost of goods
Cost of goods (semi-annual)
Operating margin
Operating margin (semi-annual)
Declines in demand
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(iii) supra.
Analysis by the Panel
The Panel notes that the USITC did not give any indication in its Report that it considered that demand played any role in causing serious injury to the industry. Rather, the USITC explained adequately that US apparent consumption of hot-rolled bar increased by 11.7% from 1996 to 2000, and that it increased on a year-to-year basis for every available comparison except that for 1998 to 1999. The USITC added that apparent US consumption increased from 1999 to 2000, the year that domestic industry performance reached injurious levels. Consequently, it concluded that changes in demand could not explain the industry's condition in 2000.5384 In the Panel's view the USITC examined the nature and effects of declines in demand when assessing whether increased imports of hot-rolled were causing serious injury to the relevant domestic producers. Accordingly, the Panel rejects the complainants' claims in relation to this factor.
Conclusions
In the Panel's view, the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by increases in COGS, together with other factors, was not attributed to increased imports contrary to the requirements of Article 4.2(b) of the Agreement on Safeguards. Having said this, the Panel notes that the facts do appear to support the USITC's conclusion regarding increases in COGS.
5384 United States' first written submission, para. 578.
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 835
Overall conclusion on USITC's determination of a causal link
We conclude that with respect to hot-rolled bar, although the USITC did not conduct any coincidence analysis, its conditions of competition's analysis provided a compelling explanation that indicated the existence of a causal link, subject to fulfilment of the non-attribution requirement. In this regard, we found that the USITC's non-attribution analysis failed to separate, distinguish and assess the nature and extent of the injurious effects of increases in COGS so that the injury caused by these factors, together with other factors, was not attributed to increased imports. Thus, the USITC did not provide a reasoned and adequate explanation supporting a determination that there was a genuine and substantial relationship of cause and effect between increased imports and serious injury to the relevant domestic producers.
Therefore, the Panel concludes that the USITC's finding that a causal link existed between imports of hot-rolled bar and injury caused to the relevant domestic producers is inconsistent with Articles 4.2(b), 2.1 and 3.1 of the Agreement on Safeguards.
Cold-finished bar
The Panel notes at the outset that it has focused in this section on the arguments made by the complainants that, for us, raised the most problematic aspects of the USITC's determinations on causation, that is, those aspects that more obviously entailed violations of the Agreement on Safeguards. Since we will dispose of the complainants' claims in our review below, the Panel sees no need to deal with the other arguments.
Coincidence and conditions of competition
USITC findings
The USITC's findings read as follows:
"We find that the increased imports of cold-finished bar are an important cause, and a cause not less than any other cause, of serious injury to the domestic industry. Accordingly, we find that increased imports of cold-finished bar are a substantial cause of serious injury to the domestic cold-finished bar industry.
Conditions of Competition
We have taken into account a number of factors that affect the competitiveness of domestic and imported cold-finished bar in the US market, including factors related to the product itself, the degree of substitutability between the domestic and imported articles, changes in world capacity and production, and market conditions. These factors affect prices and other considerations taken into account by purchasers in determining whether to purchase domestically-produced or imported articles.
Market participants generally agree that there are few or no substitutes for long products such as cold-finished bar.5385 As discussed in section V.A.2. above, the principal use of cold-finished bar is in automotive applications.
5385 (original footnote) CR and PR at LONG-78.
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 836
The record indicates strong demand during most of the period examined with US apparent consumption of cold-finished bar increasing during every full-year but one. Apparent consumption rose from 1.41 million tons in 1996 to 1.60 million tons in 1997 and then to 1.67 million tons in 1998. Apparent consumption then declined to 1.61 million tons in 1999 but increased to 1.64 million tons in 2000. Apparent consumption was lower in interim 2001, at 700,202 tons, than in interim 2000, when it was 905,184 tons.5386
With regard to supply of cold-finished bar, US capacity increased from 1996 to 2000 despite declines since 1998. Domestic industry capacity utilization fluctuated during the period examined. Notwithstanding that the capacity utilization data reported in the questionnaires appear to be understated, it is clear that there was additional productive capacity available to the domestic industry throughout the period examined. Foreign capacity reported in questionnaires increased from 1.6 million tons in 1996 to 2.0 million tons in 2000, and was higher in interim 2001 than in interim 2000. Foreign capacity utilization for full-year periods ranged from 75.2 percent in 1999 to 84.3 percent in 2000.5387
The record indicates that price is an important factor in purchasing decisions for cold-finished bar. Purchasers listed price second most-frequently, after quality, as the top factor in purchasing decisions, and listed price most frequently as the number two factor. Most purchasers evaluated the imports and domestically-produced cold- finished bar as comparable with respect to product consistency and product quality.5388
Analysis5389
Aggressive pricing by the imports during the latter portion of the period examined caused the domestic industry to lose market share and revenues. This resulted in the poor operating performance and serious injury discussed above.
Average unit values of the imports trended downward from 1996 to 1998, and the decline accelerated in 1999. Import average unit values declined by 1.3 percent from 1996 to 1997 and by 0.1 percent from 1997 to 1998. They then fell by
7.7 percent from 1998 to 1999.5390 Additional evidence that import prices declined dramatically in 1999 is provided by data for one-inch round C12L14, the cold- finished bar product for which the Commission obtained significant pricing data
concerning imports.5391 Between the fourth quarter of 1998 and the first quarter of
5386 (original footnote) CR and PR, Table LONG-71.
5387 (original footnote) CR and PR, Table LONG-45. We have relied upon the questionnaires for
foreign capacity and capacity utilization data, although such data are not complete. We acknowledge that the domestic producers contended that the questionnaire data understated foreign capacity and overstated foreign capacity utilization.
5388(original footnote) INV-Y-212 at 46.
5389 (original footnote) The Minimill 201 Coalition produced an economic model that attempted to measure the relationship between imports and the domestic industry's prices and profits. In particular, there
were defects in the manner the model measured import competition, and the model did not adequately address changes in domestic competition.
5390 (original footnote) CR and PR, Table LONG-6.
5391 (original footnote) The Commission collected pricing data concerning two cold-finished bar products. For one of the products, however, the reported data covered very small import volumes: less than 500
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 837
1999, import prices for this product declined by *** percent. They fell an additional
*** percent between the first and second quarters of 1999, the largest quarterly decline to that point in the period examined. Although prices rose during the next two quarters, the fourth quarter 1999 price remained 8.2 percent below the fourth quarter 1998 price.5392
Prices for domestically-produced C12L14 declined by 3.9 percent between the fourth quarter of 1998 and the first quarter of 1999 but fluctuated in a narrow range during the remainder of 1999. As a result, underselling margins were higher in the last three quarters of 1999 than in earlier periods. Between the first quarter of 1996 and the first quarter of 1999, the margin of underselling or overselling by the imports was no greater than 1.8 percent in any quarter. The underselling margin increased to 8.1 percent in the second quarter of 1999, however, and remained above
5.8 percent for the remaining quarters of that year.5393
The market did not react immediately to the price reductions by the imports. Indeed, neither the absolute volume of the imports nor their market share increased in 1999.5394 The lack of immediate reaction by the market may reflect extensive contract sales: over 40 percent of cold-finished bar purchasers made over 90 percent of their purchases on a contract basis, with contracts commonly six months to over one year in length.5395 However, the aggressive pricing by the imports continued in 2000. Compared to 1999, average unit values for all imports declined by 5.1 percent.5396 Prices for imported C12L14 declined during all but one quarter in 2000, and the price for the fourth quarter of 2000 was 14.0 percent below the price for the fourth quarter of 1999.5397
Domestic prices also declined in 2000. Average unit values for US shipments of all cold-finished bar products were lower in 2000 than in 1999.5398 Prices for domestically-produced C12L14 were 4.2 percent lower in the fourth quarter of 2000 than in the fourth quarter of 1999. Nevertheless, underselling by the
tons of imports in each quarter, and less than 100 tons of imports for each of the last six quarters for which data were collected. INV-Y-212, Table LONG-91. By contrast, reported import volume for one-inch round CL12L14 was at least 1,166 tons in every quarter during the entire period examined, and at least 2,636 tons for every quarter during 1999 and interim 2000. INV-Y-212, Table LONG-ALT92. Consequently, in our analysis of pricing we have focused on the latter, more complete data set. We also observe that, in an analysis of whether there is overselling or underselling, pricing data for a specific product can provide more probative information than average unit value data, where comparisons between values for imports and domestically- produced products can reflect variations in product mix. This is particularly true for a product such as cold- finished bar which covers a broad range of product types and values.
5392 (original footnote) INV-Y-212, Table LONG-ALT92.
5393 (original footnote) INV-Y-212, Table LONG-ALT92.
5394 (original footnote) See CR and PR, Tables LONG-6, LONG-71. A substantial increase in cold- finished bar imports occurred in 1998, shortly following the financial crisis that led to sharply decreased steel consumption in several Asian countries.
5395 (original footnote) Purchaser Questionnaire Responses.
5396 (original footnote) CR and PR, Table LONG-6.
5397 (original footnote) INV-Y-212, Table LONG-ALT92.
5398 (original footnote) CR and PR, Table LONG-17.
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 838
imports persisted, with quarterly underselling margins in 2000 ranging from 3.9 percent to 15.5 percent.5399
In 2000, the continued underselling by the imports led to significant increases in both import volume and market share. As previously stated, import quantities were
33.6 percent higher in 2000 than in 1999.5400 The imports' share of US apparent
consumption, measured by quantity, increased from 14.7 percent in 1999 to 19.2 percent in 2000.5401
Because the imports succeeded in increasing their share of the US market in 2000, the domestic industry's production and shipments declined from 1999 levels notwithstanding the increase in US apparent consumption.5402 The decline in output, together with the decline in prices, led to declining revenues and poor operating performance, with an operating margin in 2000 of only 2.8 percent.5403" 5404
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.2(d) supra.
Analysis by the Panel
The Panel has considered the relevant section of the USITC Report for cold-finished bar and notes that, in determining whether a causal link existed between increased imports and serious injury, the USITC did not conduct a coincidence analysis. As mentioned previously, the Panel considers that if a competent authority has not examined coincidence of trends, it must, in proving causation, provide a reasoned and adequate explanation as to why such an analysis was not undertaken as well as a compelling explanation establishing the existence of a causal link. We note that the USITC analysed the conditions of competition. Accordingly, we will now proceed to review the conditions of competition analysis undertaken by the USITC for this measure, with a view to determining whether the USITC provided such a compelling explanation.
We note as a preliminary point that the USITC considered that "[a]ggressive pricing by the imports during the latter portion of the period examined caused the domestic industry to lose market share and revenues. This resulted in the poor operating performance and serious injury discussed above." The USITC ultimately concluded that "[t]he decline in output, together with the decline in prices, led to declining revenues and poor operating performance, with an operating margin in 2000 of only 2.8 percent."
It is apparent from the USITC's analysis that the essential premise for its conclusion that "increased imports of cold-finished bar are a substantial cause of serious injury to the domestic cold- finished bar industry" was that "aggressive" pricing by imports caused the domestic industry to lose market share and revenues. In assessing whether the USITC has provided a reasoned and adequate explanation of how the facts support such a finding, the Panel will first consider whether import pricing can be labelled as "aggressive".
5399 (original footnote) INV-Y-212, Table LONG-ALT92.
5400 (original footnote) CR and PR, Table LONG-6.
5401 (original footnote) CR and PR, Table LONG-71.
5402 (original footnote) CR and PR, Tables LONG-17, LONG-71.
5403 (original footnote) CR and PR, Table LONG-28.
5404 USITC Report, Vol. I, pp. 104-106.
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 839
As a starting point, the Panel notes that the USITC pointed to underselling during the period, presumably in justification of its assertion that import pricing had been aggressive. Specifically, it stated that: "[U]nderselling by the imports persisted, with quarterly underselling margins in 2000 ranging from 3.9 percent to 15.5 percent.5405 In 2000, the continued underselling by the imports led to significant increases in both import volume and market share. As previously stated, import quantities were 33.6 percent higher in 2000 than in 1999.5406" The Panel notes firstly that the USITC, without any explanation as to why it did so, relied upon quarterly data for individual cold-bar products5407 when annual average data was available and such annual data had been used by the USITC in relation to other products, when available.5408 Unlike the quarterly data, the annual data indicated that imports did not undersell domestic products at any point in the period of investigation. For us, the lack of explanation regarding the data relied upon by the USITC calls into question whether "underselling" actually existed and, therefore, whether import pricing was, in fact, "aggressive" at all.
Further, we note that at no point during the period of investigation did average unit values for imports undersell average unit values for domestic products. In other words, import prices exceeded domestic prices throughout the period of investigation. This is evident from the graph below, which represents in graphical form USITC data. In the Panel's view, the fact that import prices exceeded domestic prices throughout the period of investigation tends to detract from the conclusion that import pricing was "aggressive". This is not to say, however, that the absence of underselling by imports means that pricing cannot be labelled as "aggressive". On the contrary, we concede that import overselling may, in certain circumstances, drive domestic prices downwards. However, in this case, the USITC relied upon the existence of import underselling as the basis for its assertion that import pricing was "aggressive". As noted, in fact, average unit values were always higher than domestic prices.5409
5405 (original footnote) INV-Y-212, Table LONG-ALT92.
5406 (original footnote) CR and PR, Table LONG-6.
5407 Table LONG-91 and Table LONG-ALT92.
5408 Table LONG-C-4.
5409 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-6 at LONG-10; Table LONG-17 at LONG-22.
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 840
1,600,000
1,400,000
1,200,000
1,000,000
1000
900
800
700
600
800,000
600,000
400,000
200,000
500
400
300
200
100
0 0
1996 1997 1998 1999 2000 2001
Imports Imports (semi-annual)
Domestic unit values Imports unit values
Domestic unit values (semi-annual) Imports unit values (semi-annual)
Putting aside the difficulties with the data relied upon by the USITC, the Panel notes the conclusions drawn by the USITC in its conditions of competition analysis:
"Because the imports succeeded in increasing their share of the US market in 2000, the domestic industry's production and shipments declined from 1999 levels notwithstanding the increase in US apparent consumption.5410 The decline in output, together with the decline in prices, led to declining revenues and poor operating performance, with an operating margin in 2000 of only 2.8 percent.5411"5412
The facts do appear to bear out the conclusion that "the imports succeeded in increasing their share of the United States' market in 2000," as is evident from the graph below generated using USITC data.5413
5410 (original footnote) CR and PR, Tables LONG-17, LONG-71.
5411 (original footnote) CR and PR, Table LONG-28.
5412 See para. 10.447.
5413 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-6 at LONG-10; Table LONG-71 at LONG-68; Table LONG-C-4.
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 841
1,000,000
800,000
600,000
400,000
200,000
0
100
90
80
70
60
50
40
30
20
10
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Imports' market shares
Imports' market shares (semi-annual)
The facts also appear to bear out the conclusion that production and shipments declined from 1999 levels as is evident from the graphs below, also generated using USITC data.5414
Imports and Production (Tons) | |||
1,600,000 | |||
1,400,000 | |||
1,200,000 | |||
1,000,000 | |||
800,000 | |||
600,000 | |||
400,000 | |||
200,000 | |||
0 | |||
1996 | 1997 1998 1999 2000 | 2001 | |
Years | Imports Imports (semi-annual) Production Production (semi-annual) |
5414 The data represented in the two graphs below are contained in the USITC Report, in particular in Table LONG-6 at LONG-10; Table LONG-17 at LONG-22; Table LONG-C-4.
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 842
1,600,000
1,400,000
1,200,000
1,000,000
800,000
600,000
400,000
200,000
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Domestic Shipments
Domestic shipments (semi-annual)
Despite the foregoing, we find limited support for the conclusion that "[t]he decline in output, together with the decline in prices, led to declining revenues and poor operating performance, with an operating margin in 2000 of only 2.8 percent." Indeed, we note that significant declines in revenues and operating margin began well in advance of 2000, the year when, according to the USITC "the continued underselling by the imports led to significant increases in both import volume and market share." This is evident from the graph below, based on USITC data, which illustrates the trends in operating margin together with import trends.5415
1,600,000 70,000
1,400,000 60,000
1,200,000 50,000
1,000,000 40,000
800,000 30,000
600,000 20,000
400,000 10,000
200,000 0
0 -10,000
1996 1997 1998 1999 2000 2001
Years Imports
imports (semi-annual)
Operating margin
Operating margin (semi-annual)
5415 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-6 at LONG-10; Table LONG-28 at LONG-34; Table LONG-C-4.
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 843
In light of the foregoing, the Panel finds that the USITC has not provided a compelling explanation that a causal link existed between increased imports of cold-finished bar and injury suffered by the relevant domestic industry. In particular, aside from the difficulties we have identified in relation to the data upon which the USITC relied in undertaking its conditions of competition analysis, we consider that the USITC has not provided a reasoned and adequate explanation of how the facts supported its conclusion that "[t]he decline in output, together with the decline in prices, led to declining revenues and poor operating performance, with an operating margin in 2000 of only
2.8 percent".
Non-attribution
USITC findings
The USITC's findings read as follows:
"The domestic industry's experience in 2000 serves to rebut one of the principal arguments of respondents – that declines in demand were a greater cause of the substantial injury to the domestic industry than increased imports. The domestic industry acknowledges that prices for cold-finished bar have historically tracked demand conditions.5416 Indeed, the domestic industry's decline in operating performance in 1999, a year when import volume and market penetration declined, appears to a large extent attributable to the declines in demand during that year.
However, in 2000 demand increased above the level of 1999. Nevertheless, as previously discussed, prices for US-produced product did not recover with demand, but instead declined further in the face of the import surge. The per unit difference between average unit values and COGS, although slightly higher in 2000 than in 1999, was well below the levels of any of the prior years of the period examined. Similarly, the industry's operating margin, while slightly above the level of 1999, was only 2.8 percent, less than half the levels of 1997 and 1998. The number of producers experiencing operating losses increased. When demand again declined in interim 2001, the imports maintained their significant presence in the market, and the domestic industry's performance further deteriorated. The domestic industry's poor performance despite increasing demand in 2000 indicates that it is the imports, not changes in demand, that explain the serious injury the domestic industry is experiencing.
We have also considered respondents' arguments that the domestic industry's poor performance was due more to the presence of a purportedly inefficient producer with structural problems, RTI, than to increased imports.5417 RTI's structural difficulties, however, ***.5418 ***. We consequently reject the proposition that RTI's performance was somehow anomalous or served to skew overall data for the domestic industry.
We consequently conclude that alternative causes proffered by respondents cannot individually or collectively explain the serious injury to the domestic industry, particularly the declining market share over the course of the period examined, and
5416 (original footnote) See CFTC Prehearing Brief at 7.
5417 (original footnote) See Cold-Finished Bar Respondents Prehearing Brief at 18-23.
5418 (original footnote) *** Producer's Questionnaire Response.
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 844
the poor operating performance in 2000. Accordingly, we find that increased imports are a substantial cause of serious injury to the domestic cold-finished bar industry that is not less than any other cause."5419
Factors considered by the USITC
Declines in demand
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(iv) supra.
Analysis by the Panel
In the Panel's view, the USITC clearly acknowledged that decline in demand contributed to injury that was being suffered by the domestic industry. In particular, the USITC stated that: "The domestic industry acknowledges that prices for cold-finished bar have historically tracked demand conditions.5420 Indeed, the domestic industry's decline in operating performance in 1999, a year when import volume and market penetration declined, appears to a large extent attributable to the declines in demand during that year." As is apparent from this last statement, the USITC made a clear linkage between declines in demand and operating performance, the latter being an important injury factor referred to in Article 4.2(a) of the Agreement on Safeguards.
We note that the USITC considered demand changes that occurred during the period of investigation. In particular, it noted demand declines and increases during the period of investigation. In addition, in the section in which it analysed the conditions of competition, the USITC stated that:
"The record indicates strong demand during most of the period examined with US apparent consumption of cold-finished bar increasing during every full-year but one. Apparent consumption rose from 1.41 million tons in 1996 to 1.60 million tons in 1997 and then to 1.67 million tons in 1998. Apparent consumption then declined to
1.61 million tons in 1999 but increased to 1.64 million tons in 2000. Apparent consumption was lower in interim 2001, at 700,202 tons, than in interim 2000, when it was 905,184 tons.5421"5422
Having acknowledged that demand declines contributed to the state of the domestic industry, the USITC dismissed this factor in its non-attribution analysis on the basis of the assertion that: "The domestic industry's poor performance despite increasing demand in 2000 indicates that it is the imports, not changes in demand, that explain the serious injury the domestic industry is experiencing." In the Panel's view, the mere fact that demand increased during a segment of the period of investigation during which injury persisted does not detract from the conclusion reached by the USITC itself that decline in demand contributed to injury that was being suffered by the domestic industry.
We find nothing in the report to indicate whether and how the injury caused by this factor was not attributed to increased imports. In our view, the need to separate and distinguish the effects of
5419 USITC Report, Vol. I, p. 107.
5420 (original footnote) See CFTC Prehearing Brief at 7.
5421 (original footnote) CR and PR, Table LONG-71.
5422 See para. 10.446.
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 845
declines in demand was particularly important in this case given the acknowledgement by the USITC itself that "the domestic industry's decline in operating performance in 1999, a year when import volume and market penetration declined, appears to a large extent attributable to the declines in demand during that year."
The significance of this decline in operating performance in 1999 that was "to a large extent attributable to declines in demand" should be viewed in context. Below is a graph that has been generated using USITC data. This graph illustrates that the industry's operating margin dropped precipitously in 1999. Prior to 1999, the operating margin was significantly higher. Following 1999, the operating margin began to increase again.5423
1,600,000 70,000
1,400,000 60,000
1,200,000 50,000
1,000,000 40,000
800,000 30,000
600,000 20,000
400,000 10,000
200,000 0
0 -10,000
1996 1997 1998 1999 2000 2001
Years Imports
imports (semi-annual)
Operating margin
Operating margin (semi-annual)
Clearly, 1999 was a significant year in terms of the industry's operating performance. Given that "the domestic industry's decline in operating performance in 1999, a year when import volume and market penetration declined, appears to a large extent attributable to the declines in demand during that year", we consider that this illustrates that declines in demand potentially played a significant role in causing injury to the domestic industry.
Conclusions
The Panel considers that, with respect to cold-finished bar, the USITC failed to comply with its non-attribution obligation contained in the second sentence of Article 4.2(b). In particular, we consider that the USITC failed to properly separate, distinguish and assess the nature and extent of the injurious effects of factors other than increased imports that were causing injury to the domestic industry. This, to us, is clear from the fact that the USITC dismissed one factor (namely, declining domestic demand) of the two that it considered in its non-attribution analysis even though it acknowledged the importance of that factor in causing injury to the industry.
5423 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-6 at LONG-10; Table LONG-28 at LONG-34; Table LONG-C-4.
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 846
Overall conclusion on USITC's determination of a causal link
The Panel finds that the USITC failed to explain why it did not conduct a coincidence analysis and did not provide a compelling explanation indicating the existence of a causal link between increased imports of cold-finished bar and serious injury to the relevant domestic producers. Further, the Panel found that the USITC's non-attribution analysis for cold-finished bar was flawed because the USITC failed to separate, distinguish and assess the nature and extent of the injurious effects of declines in demand so that the injury caused by these factors, together with other factors, was not attributed to increased imports. We found this flaw to be significant given the acknowledgement by the USITC itself that "the domestic industry's decline in operating performance in 1999, a year when import volume and market penetration declined, appears to a large extent attributable to the declines in demand during that year." Thus, the USITC did not provide a reasoned and adequate explanation supporting a determination that there was a genuine and substantial relationship of cause and effect between increased imports and serious injury to the relevant domestic producers.
Therefore, the Panel concludes that the USITC's finding that a causal link existed between imports of cold-finished bar and injury caused to the relevant domestic producers is inconsistent with Articles 4.2(b), 2.1 and 3.1 of the Agreement on Safeguards.
Rebar
The Panel notes at the outset that it has focused in this section on the arguments made by the complainants that, for us, raised the most problematic aspects of the USITC's determinations on causation, that is, those aspects that more obviously entailed violations of the Agreement on Safeguards. Since we will dispose of the complainants' claims in our review below, the Panel sees no need to deal with the other arguments.
Coincidence and conditions of competition
USITC findings
The USITC's findings read as follows:
"We find that the increased imports of rebar are an important cause, and a cause not less than any other cause, of serious injury to the domestic industry. Accordingly, we find that increased imports of rebar are a substantial cause of serious injury to the domestic rebar industry.
Conditions of Competition
We have taken into account a number of factors that affect the competitiveness of domestic and imported rebar in the US market, including factors related to the product itself, the degree of substitutability between the domestic and imported articles, changes in world capacity and production, and market conditions. These factors affect prices and other considerations taken into account by purchasers in determining whether to purchase domestically-produced or imported articles.
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 847
Market participants generally agree that there are few or no substitutes for long products such as rebar.5424 Rebar is used solely for structural reinforcement within cast concrete structures.5425
US apparent consumption of rebar increased throughout the period examined. Apparent consumption rose every year from 1996, when it was 5.5 million tons, to 2000, when it was 8.1 million tons, a net increase of 48.1 percent. Apparent consumption was also higher in interim 2001, at 4.2 million tons, than in interim 2000, when it was 4.1 million tons.5426
With regard to supply of rebar, US capacity increased throughout the period examined. Capacity utilization fluctuated; for full-year periods it ranged between
64.9 percent in 1996 to 68.5 percent in 2000. Foreign capacity reported in questionnaires increased from 24.0 million tons in 1996 to 29.6 million tons in 2000, and was higher in interim 2001 than in interim 2000. Foreign capacity utilization for full-year periods ranged from 81.7 percent in 1996 to 86.5 percent in 2000.5427
Price is a very important purchasing factor in purchasing decisions for rebar. A majority of all purchasers listed price as the number one factor in purchasing decisions for rebar, and price was named over three times more often than any other individual factor.5428 One purchaser testified at the Commission hearing that rebar was a commodity product sold on the basis of price, a proposition not disputed by any respondent.5429
Finally, rebar imports from several countries were subject to antidumping duties during portions of the period examined. In particular, Commerce imposed provisional antidumping duties on rebar from Turkey on October 10, 1996 and issued an antidumping order on these imports on April 17, 1997.5430 Commerce imposed provisional antidumping duties on rebar from Belarus, China, Indonesia, Korea, Latvia, Moldova, Poland, and Ukraine on January 30, 2001 and issued an antidumping order on imports from these eight countries on September 7, 2001.5431
5424 (original footnote) CR and PR at LONG-78.
5425 (original footnote) CR and PR at LONG- 2.
5426 (original footnote) CR and PR, Table LONG-72.
5427 (original footnote) CR and PR, Table LONG-48. We have relied upon the questionnaires for foreign capacity and capacity utilization data, although such data are not complete. We acknowledge that the domestic producers contended that the questionnaire data understated foreign capacity and overstated foreign capacity utilization.
5428 (original footnote) INV-Y-212 at 47.
5429 (original footnote) Tr. at 1316 (Koch).
5430 (original footnote) See 61 Fed. Reg. 53203 (Oct. 10, 1996), 62 Fed. Reg. 18748 (April 17, 1997).
5431 (original footnote) See 66 Fed. Reg. 8324, 8329, 8333, 8339, 8343 (Jan. 30, 2001); 66 Fed. Reg.
46777 (Sept. 7, 2001).
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 848
Analysis5432
The increased imports put price pressure on domestic producers. This price pressure prevented domestic producers from fully achieving the benefits of cost reductions during certain portions of the period examined and from fully recovering increasing costs during others. It also prevented domestic producers from fully benefitting from the large increase in domestic consumption over the period examined. As a result, operating margins declined and by 2000 the industry's operating income was negative.
Rebar imports increased significantly in 1998, shortly following the financial crisis that led to sharply decreased steel consumption in several Asian countries. As has been observed with other long products, domestic producers did not immediately change their pricing strategy in response to the initial import surge. The average unit value of the domestic industry's US shipments declined by only one dollar per ton from 1997 to 1998.5433 For the rebar product on which the Commission collected pricing data, prices for the domestically-produced product were higher during the first three quarters of 1998 than they were during the comparable quarter of 1997. Prices did begin to fall for the domestically-produced product during the fourth quarter of 1998. Throughout 1998, however, imports undersold the domestically-produced product by margins exceeding 20 percent.5434 The imports in 1998 took nearly six percentage points of market share away from the domestic industry.5435
During 1999, imports again increased by substantial margins. The quantity of imports was 49.1 percent higher in 1999 than in 1998.5436 This surge was accompanied by price declines for both the imports and the domestically-produced product. Average unit values of the imports declined by 23.6 percent from 1998 to 1999, and average unit values of US shipments of domestically produced rebar declined by 8.9 percent.5437 For the rebar product on which the Commission collected pricing data, import prices fell by 8.8 percent from the fourth quarter of 1998 to the first quarter of 1999, and the first quarter 1999 price was 11.5 percent below the first quarter 1998 price. Similarly, for the domestically-produced product, prices declined by 5.0 percent from the fourth quarter of 1998 to the first quarter of 1999, and the first quarter 1999 price was 10.6 percent below the first quarter 1998 price. There were further price declines in the second quarter of 1999 before prices stabilized during the final two quarters of the year; the second quarter 1999 price was below the second quarter 1998 price by 12.7 percent for the domestically-produced product and by 15.6 percent for the imports.5438
5432 (original footnote) The Minimill 201 Coalition produced an economic model that attempted to measure the relationship between imports and the domestic industry's prices and profits.. We considered this model in making our determination but note its limitations. In particular, there were defects in the manner the model measured import competition, and the model did not adequately address changes in domestic competition.
5433 (original footnote) CR and PR, Table LONG-18.
5434 (original footnote) INV-Y-212, Table LONG-ALT93.
5435 (original footnote) CR and PR, Table LONG-72.
5436 (original footnote) CR and PR, Table LONG-7.
5437 (original footnote) CR and PR, Tables LONG-7, LONG-18.
5438 (original footnote) INV-Y-212, Table LONG-ALT93.
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 849
We can discern no reason other than the imports for the magnitude of price and average unit value declines during 1999. The decline was not a function of demand changes, because US apparent consumption for rebar increased by 14.1 percent from 1998 to 1999.5439 Indeed, in light of these demand conditions, we would ordinarily expect prices to have stayed stable or risen, and not to have declined by such large amounts. Changes in input costs also cannot explain the magnitude of the price decline. While there was a reduction in per unit COGS from 1998 to 1999, this reduction was less than the per unit decline in average sales values.5440 In any event, in a period of sharply increasing demand, producers normally need not cut their prices to reflect fully declines in costs of goods sold.5441 Thus the price pressure imposed by the surging volume of imports prevented the domestic rebar producers from achieving the full benefits of declining input costs in a growing market.
The imports undersold domestically-produced rebar by quarterly margins between *** and *** percent during 1999.5442 During that year, the imports gained another five percentage points of market share.5443 Nevertheless, because of the strong growth in demand, the domestic industry continued to perform profitably, although operating margins were below the levels of 1998.5444
There was not a further import surge in 2000, when import quantity and market share declined somewhat from 1999 levels. Imports did maintain a significant presence in the market in 2000, however. Import quantity and market penetration in 2000 were still both significantly above 1998 levels, not to mention those of earlier years; import quantity in 2000 was considerably more than twice the 1996 level and market penetration was nearly twice the 1996 level.5445
Imports maintained their pricing pressure as well in 2000. Average unit values of imports in 2000 increased only incrementally from their depressed levels of 1999, while the average unit values for the domestically-produced product declined further from 1999 to 2000.5446 Prices for both the domestically-produced and the imported rebar product on which the Commission collected data fluctuated within a fairly narrow range, with prices for the domestic product generally being slightly below the 1999 levels. Imports continued to undersell the domestically-produced product by margins of over 20 percent.5447
As was the case in 1999, factors in the market other than imports cannot explain why rebar pricing in 2000 continued to be at depressed levels. Demand for
5439 (original footnote) CR and PR, Table LONG-72.
5440 (original footnote) CR and PR, Table LONG-29.
5441 (original footnote) Additionally, competition between domestic producers cannot be a cause for
price declines of the magnitude observed. While cost differentials do exist among domestic producers, even the domestic producer with the lowest cost structure had per-unit COGS that was considerably above the average unit sales values of the imports. See Producers' Questionnaires. Given the importance of price in rebar purchasing decisions, the commodity nature of rebar and the magnitude of underselling by the imports, it is clear that price leadership was exerted by the imports, rather than any domestic producer.
5442 (original footnote) INV-Y-212, Table LONG-ALT93.
5443 (original footnote) CR and PR, Table LONG-72.
5444 (original footnote) CR and PR, Table LONG-29.
5445 (original footnote) CR and PR, Tables LONG-7, LONG-72. 5446 (original footnote) CR and PR, Tables LONG-7, LONG-18. 5447 (original footnote) INV-Y-212, Table LONG-ALT93.
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 850
rebar continued to increase in 2000, although this increase was less than that of the preceding years.5448 Additionally, per unit COGS increased in 2000 from 1999 levels.5449 The combination of rising demand and rising costs should have led prices of domestically-produced rebar to increase in 2000. Instead, prices generally declined -- a result we conclude is attributable to the intense price-based competition from imported rebar.5450 This decline in prices led to the poor financial performance, most notably the negative operating margins discussed above.
The data for interim 2001 indicate a continuation of the trends observed during 2000. Imports continued to maintain their presence in the market. Although import average unit values in interim 2001 were above those for interim 2000, they were still far below those from 1996 to 1998. The average unit values for US shipments of domestically-produced rebar also remained depressed, notwithstanding increasing demand. Underselling by the imports persisted. Operating performance was poor and below the level of interim 2000."5451
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.2(e) supra.
Analysis by the Panel
The Panel has considered the relevant section of the USITC Report for rebar and notes that, in determining causal link, the USITC did not conduct a coincidence analysis. As mentioned previously, the Panel considers that if a competent authority has not examined coincidence of trends, it must, in proving causation, provide a reasoned and adequate explanation why such an analysis was not undertaken as well as a compelling explanation establishing the existence of a causal link. We note that the USITC analysed the conditions of competition. Accordingly, we will now proceed to review the conditions of competition analysis undertaken by the USITC for this measure, with a view to determining whether the USITC provided such a compelling explanation.
The Panel considers that the facts before the USITC did not preclude a finding that "the increased imports put price pressure on domestic producers. This price pressure prevented domestic producers from fully achieving the benefits of cost reductions during certain portions of the period examined and from fully recovering increasing costs during others." In coming to this conclusion, we first examined average unit value data for imports and domestic products. The graph below, which has been generated using USITC data, indicates that imports undersold domestic products throughout the period of investigation and quite significantly so from 1999 onwards. This is consistent with the USITC's pricing trends findings.5452
5448 (original footnote) CR and PR, Table LONG-72.
5449 (original footnote) CR and PR, Table LONG-29.
5450 (original footnote) Moreover, although the largest individual component of COGS – raw materials
costs – declined from 1999 to 2000 on a per unit basis, this decline was still not as great as the per unit decline in average commercial sales values. CR and PR, Table LONG-29.
5451 USITC Report, Vol. I, pp. 111-114.
5452 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-7 at LONG-11; Table LONG-18 at LONG-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 851
7,000,000 350
6,000,000 300
5,000,000 250
4,000,000 200
3,000,000 150
2,000,000 100
1,000,000 50
0 0
1996 1997 1998 1999 2000 2001
Imports Imports (semi-annual)
Domestic unit values Imports unit values
Domestic unit values (semi-annual) Imports unit values (semi-annual)
We also considered the import market share during the period of investigation. As is evident from the graph below, as imports increased from 1997 onwards, the import market share also progressively increased.5453
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
0
100
90
80
70
60
50
40
30
20
10
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Imports' market shares
Imports' market shares (semi-annual)
5453 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-7 at LONG-11; Table LONG-72 at LONG-68; Table LONG-C-5.
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 852
In addition, the graph below illustrating import trends and trends in the operating margin seems to indicate that some coincidence between increases in imports from 1997 onwards and declining operating margin existed from 1998 onwards.5454
7,000,000 100,000
6,000,000 80,000
5,000,000 60,000
4,000,000 40,000
3,000,000 20,000
2,000,000 0
1,000,000 -20,000
0 -40,000
1996 1997 1998 1999 2000 2001
Years Imports
imports (semi-annual)
Operating margin
Operating margin (semi-annual)
Taken together, it is our view that the above data supports the USITC finding that increased imports exerted downward pressure on domestic prices and that this, in turn, had an impact upon the financial performance of domestic producers. In our view, the USITC provided a compelling explanation indicating the existence of a causal link between increased imports and serious injury, subject, of course, to fulfilment of the non-attribution requirement. Therefore, we reject the complainants' claims and arguments in this regard.
Non-attribution
USITC findings
The USITC's finding reads as follows:
"In our discussion above, we have already considered and rejected several alternative causes advanced by the respondents to explain the condition of the domestic rebar industry. As discussed in the section on serious injury, the domestic industry's capacity increases cannot be deemed to be an alternative cause of injury because capacity increased far less than did US apparent consumption of rebar during the period examined; indeed, capacity utilization generally increased during the period examined. We have also discussed changes in input costs and demand and found that they cannot explain the changes in pricing that occurred during the period examined; if anything, these factors indicate that prices should have been stable to increasing
5454 The data represented in the graph below are contained in the USITC Report, in particular in Table LONG-7 at LONG-11; Table LONG-29 at LONG-35; Table LONG-C-5.
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 853
during the latter portion of the period examined. Instead, because of competition from the increased imports, prices declined."5455
Factors considered by the USITC
Domestic capacity increases
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(v) supra.
Analysis by the Panel
The Panel agrees with the United States that the USITC dismissed this factor in its non- attribution analysis. In particular, during its causation analysis, the USITC stated that: "[T]he domestic industry's capacity increases cannot be deemed to be an alternative cause of injury because capacity increased far less than did US apparent consumption of rebar during the period examined; indeed, capacity utilization generally increased during the period examined." In addition, during its injury analysis, the USITC stated that: "Reported capacity also increased during each year of the period examined, rising from 7.6 million tons in 1996 to 9.7 million tons in 2000. Capacity was higher in interim 2001, when it was 4.8 million tons, than in interim 2000, when it was 4.7 million tons in 2000.5456 The increases in capacity, however, must be viewed in the context of the increases in demand for rebar during the period examined. The 26.6 percent increase in productive capacity between 1996 and 2000 was far smaller than the 48.1 percent increase in US apparent consumption over that period. Moreover, notwithstanding the overall increases in capacity, several firms that produce rebar have shuttered production facilities during the period examined.5457"5458
We consider that the USITC provided a reasoned and adequate explanation as to why domestic capacity increases were not a cause of serious injury. In the Panel's view, the complainants have not put forward a plausible alternative explanation as to why this factor was a cause of serious injury.
Changes in input costs
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(v) supra.
Analysis by the Panel
The USITC dismissed this factor as a possible cause of injury to the industry. In particular, during its analysis of the conditions of competition, the USITC stated that: "Changes in input costs also cannot explain the magnitude of the price decline. While there was a reduction in per unit COGS from 1998 to 1999, this reduction was less than the per unit decline in average sales values.5459" Further, it stated that: "Additionally, per unit COGS increased in 2000 from 1999 levels.5460 The
5455 USITC report, Vol. I, pp. 114-115.
5456 (original footnote) CR and PR, Table LONG-18.
5457(original footnote) See Minimill 201 Coalition Posthearing Brief, vol. 3 at 5-6.
5458 USITC Report, Vol. I, p. 109.
5459 (original footnote) CR and PR, Table LONG-29.
5460 (original footnote) CR and PR, Table LONG-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 854
combination of rising demand and rising costs should have led prices of domestically-produced rebar to increase in 2000. Instead, prices generally declined – a result we conclude is attributable to the intense price-based competition from imported rebar.5461"5462 Finally, in its causation analysis, the USITC stated that: "We have also discussed changes in input costs… and found that they cannot explain the changes in pricing that occurred during the period examined; if anything, these factors indicate that prices should have been stable to increasing during the latter portion of the period examined. Instead, because of competition from the increased imports, prices declined."
The question remains as to whether the USITC was correct in discounting changes in input costs as a possible cause of injury without further analysis. It is evident from Table-LONG 29 that COGS increased quite significantly between 1999 and 2000. In addition, SG&A expenses increased significantly between 1998 and 2000.5463 Notably, the domestic industry's operating margin rose until 1998 and declined quite precipitously thereafter. In our view, at the least, the USITC should have explained why these increases in costs and expenses were not a cause of injury. It was not enough, in our view, to dismiss the effects of these increases on the mere basis that they could not account for domestic price declines.
Conclusions
In the Panel's view, the USITC failed to properly separate, distinguish and assess the nature and extent of the injurious effects of increases in COGS and SG&A expenses, contrary to the requirements of Article 4.2(b) of the Agreement on Safeguards. The Panel finds that, in failing to adequately analyse increases in COGS and SG&A, the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by these factors, together with other factors, was not attributed to increased imports.
Overall conclusion on USITC's determination of a causal link
As indicated above, the Panel found that, although the USITC did not explain why it did not undertake a coincidence analysis, it nonetheless provided a compelling explanation that indicated, leaving aside the issue of compliance with the non-attribution requirement, the existence of a causal link. However, we found that the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by increases in COGS and SG&A, together with other factors, was not attributed to increased imports. Thus, the USITC did not provide a reasoned and adequate explanation supporting a determination that there was a genuine and substantial relationship of cause and effect between increased imports and serious injury to the relevant domestic producers.
Therefore, the Panel concludes that the USITC's finding that a causal link existed between imports of rebar and injury caused to the relevant domestic producers is inconsistent with Articles 4.2(b), 2.1 and 3.1 of the Agreement on Safeguards.
5461 (original footnote) Moreover, although the largest individual component of COGS – raw materials costs – declined from 1999 to 2000 on a per unit basis, this decline was still not as great as the per unit decline in average commercial sales values. CR and PR, Table LONG-29.
5462 See para. 10.471.
5463 Even if SG&A expenses cannot be regarded as "input costs", we consider that such expenses should have been taken into account by the USITC.
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 855
Welded pipe
The Panel notes at the outset that it has focused in this section on the arguments made by the complainants that, for us, raised the most problematic aspects of the USITC's determinations on causation, that is, those aspects that more obviously entailed violations of the Agreement on Safeguards. Since we will dispose of the complainants' claims in our review below, the Panel sees no need to deal with the other arguments.
We note that the complainants have not challenged the USITC's coincidence and conditions of competition analyses. Accordingly, we will proceed directly to a review of the USITC's non- attribution analysis.
Non-attribution
USITC findings
The USITC's findings read as follows:
"We find that increased imports of welded pipe are a substantial cause of the threat of serious injury: that is, we find that serious injury – a "significant overall impairment in the position" of the domestic industry – due to imports is "clearly imminent," and that increased imports of welded pipe are an important cause, and a cause not less than any other cause, of the threat of serious injury to the domestic industry.
…
We also considered other possible causes of the current condition of the domestic industry, as well as respondents' arguments that no future threat of serious injury exists. Several respondents argued that increased domestic capacity had a negative impact on prices and therefore on the condition of the domestic industry.5464 The increase in capacity (1.5 million tons) was only modestly higher than the increase in domestic consumption of welded pipe (1.2 million tons) over the period examined. Thus, the increase was not inconsistent with the overall increase in consumption during the period examined – apparent US consumption increased by 73 percent of the amount of the increase in capacity. We do not view this differential as excessive or as contributing in more than a minor way to the condition of the industry in 2000 or interim 2001.
Joint Respondents argue that the declining profitability is explained by events pertaining to a significant domestic producer that raised the company's costs but are unrelated to imports.5465 While certain company costs appear to have increased, the main reason for the decline in the company's financial performance was the substantial drop in the unit value of company sales beginning in 1999.5466 As discussed above, this decline was largely the result of the substantial increased
5464 (original footnote) See, e.g., Joint Respondents Prehearing Injury Brief on Welded Tubular Products Other Than OCTG at 45.
5465 (original footnote) Joint Respondents Prehearing Injury Brief on Welded Tubular Products Other Than OCTG at 14-15. Our discussion of this issue is framed in general terms to avoid referencing business proprietary information.
5466 (original footnote) OINV-Y-212.
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 856
imports. Moreover, excluding this company does not substantially alter the downward trend in industry profitability described earlier.
We considered whether the antidumping orders in place on some welded non-OCTG products from several countries reduce the current or likely imminent impact of imports. The orders cover only a limited number of welded pipe products and, of those, only imports from a limited number of countries. Moreover, the orders were issued between 1984 and 1989 and thus were in place before the start of the period examined.5467 They clearly did not preclude the surge in imports in 2000 and continued high level of imports in 2001, or even prevent a surge in imports from countries covered by the orders.5468 Given these increases despite the existence of the orders, these pre-existing orders do not provide a basis to conclude that imports would not continue to increase in the imminent future.5469
Several respondents argue that the industry is not threatened with serious injury because the market for large diameter line pipe has begun to surge and will continue to expand in the imminent future.5470 We agree with respondents that available information indicates that there has been a recent increase in demand for large diameter line pipe and that projections are for continued growth due to rising demand for pipeline projects. We also agree that rising demand tends to ameliorate the impact of a given volume of imports. However, large diameter line pipe is only a portion of this industry -- an estimated 20 to 30 percent of the overall welded product category.5471 Indeed, even with a recent rise in large diameter line pipe demand, overall demand for covered welded tubular products has been relatively constant on a full-year basis since 1998, as well as between interim periods. Thus, we do not consider the likely increased demand for large diameter line pipe as eliminating the threat of serious injury.
For all of the reasons we have discussed, we conclude that increased imports pose a real and imminent threat of serious injury to the welded pipe industry."5472
5467 (original footnote) CR and PR at Table OVERVIEW-1.
5468 (original footnote) For example, imports from Thailand, which are covered by the orders, increased by 69,621 tons, or 248.2 percent, between 1998 and 2000 and undersold the domestic product by double digit
margins in 2000 and the first half of 2001. Committee on Pipe and Tube Imports Posthearing Injury Brief at 19, exhibits 3, 5. Imports have also increased by significant amounts since 1998 from Korea (68,418 tons, or 19.5 percent), Taiwan (18,762 tons, or 40.1 percent), and Turkey (30,440 tons, or 317.9 percent). In the case of Korea, such imports undersold the domestic product by margins up to 8.8 percent in 2000 and the first half of 2001, and in the case of Taiwan and Turkey generally undersold the domestic product by double digit margins in 2000 and the first half of 2001 in quarters for which data were reported. Committee on Pipe and Tube Imports Posthearing Injury Brief at 15-17, 21-22.
5469 (original footnote) The pending antidumping investigation on welded non-alloy steel pipe from China is not a basis to conclude that imports will not increase. It would be speculative to attempt to determine the outcome of that investigation or its effect on any imports in the imminent future. The Commission made an affirmative determination in the preliminary phase of this investigation in July 2001, and made negative determinations in the other investigations considered at that time. See Circular Welded Non-Alloy Steel Pipe
From China, Indonesia, Malaysia, Romania, and South Africa, Invs. Nos. 731-TA-943-947 (Preliminary), USITC Pub. 3439 (July 2001).
5470 (original footnote) See, e.g., European Steel Tube Association Prehearing Injury Brief at 11-13.
5471 (original footnote) CR at TUBULAR-55; PR at TUBULAR-43.
5472 USITC Report, Vol. I, pp. 165-166.
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 857
Factors considered by the USITC
Domestic industry overcapacity
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(vi) supra.
Analysis by the Panel
In the Panel's view, the USITC clearly considered that domestic industry overcapacity played some role in causing the injury that was suffered by the domestic industry. In particular, the USITC stated that: "The increase in capacity (1.5 million tons) was only modestly higher than the increase in domestic consumption of welded pipe (1.2 million tons) over the period examined. Thus, the increase was not inconsistent with the overall increase in consumption during the period examined – apparent US consumption increased by 73 percent of the amount of the increase in capacity. We do not view this differential as excessive or as contributing in more than a minor way to the condition of the industry in 2000 or interim 2001."
We note that the USITC identified and considered changes in domestic capacity during the period of investigation. In particular, the USITC stated that:
"Domestic capacity rose 22 percent during the period examined, from 6.86 million short tons in 1996 to 8.37 million short tons in 2000, with the largest one-year increase occurring in the middle of the period, between 1997 and 1998 (7.1 percent). Domestic capacity has increased by smaller amounts recently (by 4.4 percent between 1999 and 2000, and by 0.5 percent between interim 2000 and interim 2001).5473
…
Domestic welded pipe capacity increased during the period examined, and was at its highest level in 2000.5474 US capacity growth largely tracked the increase in apparent US consumption of welded pipe.5475 However, the recent decline in domestic production coupled with the increase in domestic capacity resulted in a significant decline in capacity utilization beginning in 1999 and continuing through 2000, and in interim 2001 compared to interim 2000. The capacity utilization rate for the industry fluctuated during the first three years of the period examined (66.7 percent in 1996,
71.9 percent in 1997, and 70.7 percent in 1998), and then declined sharply to 63.8
5473 (original footnote) CR and PR at TUBULAR-C-4.
5474 (original footnote) US producers' average capacity was 6.86 million short tons in 1996, and increased to 7.04 million short tons in 1997, 7.54 million short tons in 1998, 8.02 million short tons in 1999, and
8.38 million short tons in 2000. US producers' average capacity was 4.69 million short tons in interim 2001 (half year basis), virtually the same as in the same period of 2000 (4.67 million short tons). CR and PR at Table TUBULAR-C-4.
5475 (original footnote) The increase in average annual capacity of approximately 1.5 million short tons during the period examined was slightly above the 1.2 million short ton increase in domestic consumption that
occurred during that period. We note that US producers maintain capacity to export, and that exports have accounted for as much as 475,000 tons of production during the period examined. CR and PR at Table TUBULAR-C-4.
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 858
percent in 1999 and 56.2 percent in 2000. This rate was 53.2 percent in interim 2001 as compared to 53.4 percent in the same period of 2000.5476"
The Panel further notes that the USITC dismissed this factor in its non-attribution analysis on the basis that it was not regarded as contributing to injury suffered by the domestic industry in "more than a minor way." In our view, that a factor contributes to injury in no "more than a minor way" does not detract from the implicit acknowledgement that it, nevertheless, contributes to the injury. In our view, the need to separate and distinguish the effects of domestic industry over-capacity was particularly pertinent in this case given its apparent inter-relationship with a number of the injury factors referred to in Article 4.2(a), namely, domestic production and capacity utilization. This relationship was referred to by the USITC itself when it stated that: "However, the recent decline in domestic production coupled with the increase in domestic capacity resulted in a significant decline in capacity utilization beginning in 1999 and continuing through 2000, and in interim 2001 compared to interim 2000."
As a further point, the Panel notes that the apparent premise upon which the USITC dismissed domestic industry overcapacity in its non-attribution analysis was that: "The increase in capacity (1.5 million tons) was only modestly higher than the increase in domestic consumption of welded pipe (1.2 million tons) over the period examined." Even though the USITC dismissed this factor on the basis that it contributed to injury in no "more than a minor way", the Panel considers that the USITC implicitly acknowledged the importance of this factor. In particular, it noted that capacity increased by a not insubstantial amount of 22% during the period of investigation.
The Panel considers that in dismissing this factor in its non-attribution analysis, the USITC failed to meet its obligation to establish, through a reasoned and adequate explanation, that the injury caused by it, together with other factors, was properly distinguished and not attributed to increased imports. At the very least, the USITC should have specifically identified what it considered to be the "minor" contribution that domestic industry over-capacity played in causing serious injury to the industry.
Aberrational performance of one member of the industry
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(vi) supra.
Analysis by the Panel
The Panel notes that the USITC addressed the aberrational performance of one member of the domestic industry in its Report. In particular, it stated that:
"Joint Respondents argue that the declining profitability is explained by events pertaining to a significant domestic producer that raised the company's costs but are unrelated to imports.5477 While certain company costs appear to have increased, the main reason for the decline in the company's financial performance was the
5476 (original footnote) CR and PR at Table TUBULAR-C-4.
5477 (original footnote) Joint Respondents Prehearing Injury Brief on Welded Tubular Products Other Than OCTG at 14-15. Our discussion of this issue is framed in general terms to avoid referencing business proprietary information.
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 859
substantial drop in the unit value of company sales beginning in 1999.5478 As discussed above, this decline was largely the result of the substantial increased imports. Moreover, excluding this company does not substantially alter the downward trend in industry profitability described earlier."
The Panel considers that the USITC's decision to dismiss the aberrational performance of one member of the domestic industry in its non-attribution analysis was not adequately reasoned. The USITC states that the "main" reason for the decline in the company's financial performance was the drop in units sales caused "largely" by imports. In our view, words such as "main" and "largely" indicate subjective judgement on the part of the USITC, which should have been the subject of further explanation. We believe that the USITC should have identified and considered possible reasons other than the asserted "main" one for the company's decline, which apparently were identified in the Joint Respondents Prehearing Injury Brief for welded pipe.5479 In addition, we note that the USITC stated that excluding the poor performer from the analysis would not have "substantially" affected the downward trend in profitability. By implication, the exclusion had some effect, albeit not substantial. In the Panel's view, this effect should have been identified, evaluated and explained.
Conclusions
The Panel considers that, with respect to welded pipe, the USITC failed to comply with its non-attribution obligation contained in the second sentence of Article 4.2(b). In particular, we consider that the USITC failed to properly separate, distinguish and assess the nature and extent of the injurious effects of factors other than increased imports that were causing injury to the domestic industry. This, to us, is clear from the fact that the USITC dismissed a number of factors (namely, domestic industry overcapacity and aberrational performance of one member of the industry) in its non-attribution analysis even though it acknowledged that those factors were causing injury to the industry.
We thus find that the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by other factors such as domestic industry overcapacity and aberrational performance of one member of the industry, together with other factors, was not attributed to increased imports.
Overall conclusion on USITC's determination of a causal link
In the Panel's view, the USITC failed to properly separate, distinguish and assess the nature and extent of the injurious effects of domestic industry overcapacity and the aberrational performance of one member of the industry, contrary to the requirements of Article 4.2(b) of the Agreement on Safeguards. The USITC, therefore, failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by these factors, together with other factors, was not attributed to increased imports. Thus, the USITC did not provide a reasoned and adequate explanation supporting a determination that there was a genuine and substantial relationship of cause and effect between increased imports and serious injury to the relevant domestic producers.
Therefore, the Panel concludes that the USITC's finding that a causal link existed between imports of welded pipe and injury caused to the relevant domestic producers is inconsistent with Articles 4.2(b), 2.1 and 3.1 of the Agreement on Safeguards.
5478 (original footnote) INV-Y-212.
5479 See the complainants' Common Exhibit CC-78.
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 860
FFTJ
The Panel notes at the outset that it has focused in this section on the arguments made by the complainants that, for us, raised the most problematic aspects of the USITC's determinations on causation, that is, those aspects that more obviously entailed violations of the Agreement on Safeguards. Since we will dispose of the complainants' claims in our review below, the Panel sees no need to deal with the other arguments.
Coincidence and conditions of competition
USITC findings
The USITC's findings read as follows:
"We find that the increased imports of fittings, flanges, and tool joints are an important cause, and a cause not less than any other cause, of serious injury to the domestic industry. Accordingly, we find that increased imports of fittings, flanges, and tool joints are a substantial cause of serious injury to the domestic industry.
Conditions of Competition
Pipe connection products are diverse (flanges, butt-weld fittings, other fittings, including couplings and nipples, and tool joints), but in general are used to join or cap pipe. Many of the products are commodity grade, produced to standards and specifications established by standards and testing bodies such as ASTM, API, and AWWA. Fittings and flanges are often distributed with other tubular products, and purchasers stated that demand for them is driven by utilities, automotive products, and import competition in downstream markets.5480 Demand for tool joints is connected with OCTG demand, since tool joints are used in manufacturing finished drill pipe.5481 Purchasers of fittings and flanges reported that imported and domestically produced fittings and flanges produced to the same grade and specification are used in the same applications.5482 Once the standards are met, price and cost competitiveness often become the most important factor.5483
Apparent US consumption of fittings and flanges increased by 9.7 percent between 1996 and 2000, with most of this increase occurring between 1996 and 1997. Demand was less volatile thereafter, until interim 2001, when it rose by 10.4 percent over interim 2000.5484
Domestic producers' capacity increased by 7.4 percent over the period examined, somewhat less than the growth rate in consumption. Domestic capacity reached its highest level of the period examined in 1999, and declined by 5.2 percent in 2000, and by an additional 4.6 percent in interim 2001 compared to interim 2000.5485 As indicated above, Trinity Fitting Group, a domestic producer, has closed plants in
5480 (original footnote) CR at TUBULAR-55; PR at TUBULAR-43. 5481 (original footnote) CR at TUBULAR-55; PR at TUBULAR-43. 5482 (original footnote) CR at TUBULAR-62; PR at TUBULAR-50.
5483 (original footnote) CR at TUBULAR-59; PR at TUBULAR-47; Tr. at 2514 (Berger); Tr. at 2516
(Zidell); Tr. at 2524 (Keilers).
5484 (original footnote) CR and PR at Table TUBULAR-C-6.
5485 (original footnote) CR and PR at Table TUBULAR-C-6.
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 861
Kentucky, Arkansas, Mississippi, and in early 2001, Texas, effectively exiting the flange business. Domestic production fluctuated during the period examined, and was 5.3 percent lower in 2000 than in 1996; domestic production was 11.6 percent lower in interim 2001.5486
Foreign producers' reported capacity increased throughout the period examined, and was 19.5 percent higher in 2000 than in 1996. It rose in interim 2001 compared to interim 2000. Foreign producers' production, on the other hand, fluctuated, and was higher in 2000 than in 1996, and higher in interim 2001 than in interim 2000. Foreign producers became more export-oriented during the period examined. Their share of total shipments exported also fluctuated, but was higher in 2000 at 60.5 percent (60.6 percent in interim 2001) than at the beginning of the period examined (58.9 percent in 1996). The share shipped to the US market also fluctuated but was at its highest level at the end of the period examined, 19.0 percent in 2000 and 19.2 percent in interim 2001. Foreign producers' capacity utilization rate also fluctuated during the period examined, and was 58.4 percent in 2000 and 70.4 percent in interim 20015487, indicating available capacity for additional production.
Analysis of Factors
As indicated above, imports of fittings and flanges have increased in both actual terms and relative to domestic production. Imports increased in actual terms by 30.8 percent (as measured in quantity) during the course of the investigation, and by 15.3 percent between 1999 and 2000. Imports were 32.1 percent higher in interim 2001 than in the same period of 2000. Imports increased in each year of the period examined and were at their highest level of the period in 2000.5488
Imports have taken an increasingly larger share of the domestic market each year since 1997, with the largest increase occurring in 2000. The market share captured by imports also increased sharply in interim 2001 as compared to the same period of 2000. The share of the domestic market held by imports was 35.0 percent in 1996 and fell to 32.9 percent in 1997 and then rose to 35.5 percent in 1998, 37.7 percent in 1999, and 41.7 percent in 2000. The share of the market held by imports was 46.7 percent in interim 2001, well above the market share of 39.0 percent in the same period of 2000.5489 The steady increase in volume of imports, and the increase in import market share, especially since 1997, coincided with the deterioration of the condition of the domestic industry described above.
Information on prices was mixed. The AUVs of domestic shipments fluctuated from 1996 to 1998, then fell somewhat from 1998 to 2000; they were lower in interim 2001 compared to interim 2000. The AUVs of imports fluctuated but increased
5486 (original footnote) CR and PR at Table TUBULAR-C-6.
5487 (original footnote) CR and PR at Table TUBULAR-36.
5488 (original footnote) The ratio of imports to domestic production also increased significantly during
the period examined, from 50.5 percent in 1996 to 69.7 percent in 2000, and was at its highest full-year level in 2000. This was significantly above the level of 55.3 percent in 1998 and 63.0 percent in 1999. The ratio in interim 2001 (88.8 percent) was substantially above the level of the same period of 2000 (59.4 percent). CR and PR at Table TUBULAR-15.
5489 (original footnote) CR and PR at Table TUBULAR-C-6.
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 862
overall during the period. Import AUVs were generally above domestic AUVs.5490 By contrast, pricing information gathered by the Commission on a butt weld fitting product5491 showed that imports from non-NAFTA sources and Mexico (there were no reported imports from Canada) undersold the domestic product in each quarterly period for which data were provided. The data further showed that the margin of underselling was at its highest level in 2000 and January-June 2001. Non-NAFTA imports have been priced at more than 20 percent below the domestic product since the fourth quarter of 1999.5492 Domestic prices for the butt-weld product fell slightly during the period, before rising in the final quarter.5493 Import prices for this product fell significantly over the period, particularly since 1998.
Purchasers of tubular products indicated that price was a key factor in their purchasing decisions, behind only quality.5494 Moreover, nearly all purchasers indicated that imported and domestic fittings and flanges made to the same grade and specification may be used in the same applications. We find that such broad interchangeability indicates that price plays an important role in the market. In light of these facts, we find the product-specific evidence of underselling to be significant.
In sum, the steady and large increase in imports, which captured an increasing share of the US market, led to erosions in such industry indicators as production, capacity utilization, shipments, and employment indicators. Lower production and shipments meant fewer sales over which to spread fixed costs, contributing to increased unit costs. The increasing presence of imports, in at least some cases at substantial underselling margins, prevented the industry from recouping increased costs through higher prices; instead, prices fell somewhat over the period. Accordingly, we find that imports are a substantial cause of serious injury.5495
Respondents argued that none of the injury data in the Commission prehearing staff report can be correlated to import volumes. They allege that when import volumes increased by the greatest margin, domestic industry operating income increased by the greatest margin.5496 The evidence in the record does not support respondents' contentions. Imports increased by the greatest margin of the period examined in 2000
5490 (original footnote) We are cautious of placing undue weight on AUV information, as it may be influenced by issues of product mix.
5491 (original footnote) Carbon steel butt-weld fitting, 6 inch nominal diameter, 90 degree elbow, long radius, standard weight, meeting ASTM A-234, grade WPB or equivalent specification.
5492 (original footnote) CR and PR at Table TUBULAR-61.
5493 (original footnote) We observe that the domestic producers' attempt to raise prices in that final quarter, even in a period of rising demand, resulted in a substantial loss of volume.
5494 (original footnote) CR and PR at Table TUBULAR-53. Domestic producers of fittings and flanges
and a distributor of fittings testified that price was an important consideration in customer purchasing decisions. Tr. at 2516 (Zidell); Tr. at 2518 (Graham); Tr. at 2520 (Ketchum); Tr. at 2523 (Bernobich).
5495 (original footnote) Domestic producers cited increased imports as the cause of injury to the domestic industry. In the questionnaire sent to fittings producers, the Commission asked recipients to identify the factors, from a list of 13, including imports, that are adversely impacting the domestic industry. Recipients were given the option of identifying more than one factor. Of those responding, 16 producers identified
imports, and one identified the general economic downturn. No other factors were identified. Persons testifying at the public hearing also cited imports. One company official asserted that declining sales volumes and profits caused by imports have forced his firm to shelve plans for capital investment, severely impairing the firm's competitiveness and efficiency. Tr. at 2517 (Zidell).
5496 (original footnote) Bebitz et al. posthearing brief at 11-17.
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 863
(15.3 percent), and the domestic industry operated at a loss that year, its worst year of the period examined. This occurred notwithstanding a 4.3 percent increase in apparent US consumption of fittings and related products that year.5497 While it is true that industry profit margins also fell sharply in 1999 when the quantity of imports increased by only a small amount (0.3 percent), the unit value of imports fell that year by 7.1 percent, domestic consumption fell by 5.5 percent, and the share of the market held by imports that year increased to 37.7 percent from 35.5 percent in 1998.5498
Respondents also contend that segments of the market are wholly or partially closed to imports due to Approved Manufacturers' Lists.5499 However, it is questionable how much, if any, impact that such lists have on limiting import competition in fittings and flanges. Domestic fittings and flanges producers who appeared at the Commission's injury hearing testified that approved manufacturer lists have been expanded to include many foreign producers of fittings and flanges, and approved lists of butt-weld pipe fittings suppliers include firms in Italy, Thailand, Japan, the United Kingdom, Austria, France, Germany, Canada, and Mexico.5500 More generally, approved manufacturer lists do not appear to have been an insurmountable hurdle to imports entering the US market, as they increased by over 30 percent from 1996 to 2000, and by another 32 percent between interim 2000 and 2001."5501
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.2(f) supra.
Analysis by the Panel
At the outset, the Panel notes that the USITC undertook a coincidence analysis for FFTJ and concluded that coincidence existed. Accordingly, we will consider whether these findings provide a reasoned and adequate explanation of how the facts support this finding.
In particular, we note that USITC found that:
"Imports have taken an increasingly larger share of the domestic market each year since 1997, with the largest increase occurring in 2000. The market share captured by imports also increased sharply in interim 2001 as compared to the same period of 2000. The share of the domestic market held by imports was 35.0 percent in 1996 and fell to 32.9 percent in 1997 and then rose to 35.5 percent in 1998, 37.7 percent in 1999, and 41.7 percent in 2000. The share of the market held by imports was 46.7 percent in interim 2001, well above the market share of 39.0 percent in the same period of 2000.5502 The steady increase in volume of imports, and the increase in import market share, especially since 1997, coincided with the deterioration of the condition of the domestic industry described above."5503
5497 (original footnote) CR and PR at Table TUBULAR-C-6.
5498 (original footnote) CR and PR at Table TUBULAR-C-6.
5499 (original footnote) Bebitz et al. Posthearing Injury Brief at 11-17. 5500 (original footnote) Tr. at 2516-17 (Zidell); at 2522-23 (Bernovich). 5501 USITC Report, Vol. I, pp. 174-177.
5502 (original footnote) CR and PR at Table TUBULAR-C-6.
5503 See para. 10.505.
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 864
The Panel recalls that, when examined, coincidence needs to be established between the movements or trends in imports and the movements or trends in injury factors. The Panel has considered coincidence between a number of the injury factors mentioned in Article 4.2(a), including those referred to by the USITC, and imports on the basis of facts that were available to the USITC in making its determination.
First, the Panel considers that coincidence does appear to exist between upward trends in imports and downward trends in production for the duration of the period of investigation.5504
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
50,000
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Production
Production (semi-annual)
With respect to the relationship between increased imports and net commercial sales, the Panel again considers that coincidence does appear to exist between upward trends in imports and downward trends in net commercial sales during the period of investigation.5505
5504 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-14 at TUBULAR-17; Table TUBULAR-C-6.
5505 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-20 at TUBULAR-24; Table TUBULAR-C-6.
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 865
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
50,000
1996 1997 1998 1999 2000 2001
Imports
Net commercial sales Imports (semi-annual)
Net commercial sales (semi-annual)
Similarly, the Panel considers that coincidence does appear to exist between upward trends in imports and overall downward trends in employment during the period of investigation.5506
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
50,000
3,500
3,000
2,500
2,000
1,500
1,000
500
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Employment
Employment (semi-annual)
The Panel also considers that coincidence does appear to exist between upward trends in imports and downward trends in operating margin during the period of investigation. More particularly, the level of operating margin dropped quite precipitously from 1997 onwards as import
5506 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-14 at TUBULAR-17; Table TUBULAR-C-6.
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 866
levels started to rise. In this regard, the Panel does not consider that rising operating margin during interim 2001, which was accompanied by rising imports, detracts from our conclusion.5507
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
50,000 -5,000
1996 1997 1998 1999 2000 2001
Years Imports
imports (semi-annual)
Operating margin
Operating margin (semi-annual)
The Panel does not discern coincidence between productivity trends and imports trends. While it is true that coincidence can be discerned at the very end of the period of investigation when productivity declined quite significantly and imports increased, the Panel considers that the trends during the rest of the period of investigation do not demonstrate any coincidence. In particular, between 1997 and 1999 when imports started to rise, productivity levels remained more or less unchanged. From 1999 until 2000, as import levels increased further, productivity also increased.5508
5507 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-20 at TUBULAR-24; Table TUBULAR-C-6.
5508 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-14 at TUBULAR-17; Table TUBULAR-C-6.
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 867
250,000 29
225,000
28
200,000
175,000 27
150,000
125,000 26
100,000
25
75,000
50,000 24
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Productivity
Productivity (semi-annual)
The Panel considers that coincidence does appear to exist between upward trends in imports and downward trends in capacity utilization during the period of investigation. More particularly, as import levels started to rise from 1997 onwards, capacity utilization also declined.5509
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
50,000
100
90
80
70
60
50
40
30
20
10
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Capacity utilization
Capacity utilization (semi-annual)
Conclusions
On the basis of the foregoing, the Panel concludes that, overall, clear coincidence exists between the upward trend in imports and the downward trend in the injury factors, except for productivity. Accordingly, we consider that the USITC was justified in concluding that: "The steady increase in volume of imports, and the increase in import market share, especially since 1997,
5509 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-14 at TUBULAR-17; Table TUBULAR-C-6.
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 868
coincided with the deterioration of the condition of the domestic industry." We note, however, that the USITC did not provide a reasoned and adequate explanation of how the facts support the finding that that coincidence existed. Indeed, apart from the quoted sentence from the USITC Report (at paragraph 10.508), we cannot find anything further in the USITC Report that demonstrates that movements in imports coincided with movements in injury factors. Given that the USITC failed to provide a reasoned and adequate explanation that demonstrated the existence of coincidence between movements in imports and movements in injury factors, it was for the USITC to provide a compelling explanation as to why a causal link was considered, nevertheless, to exist. We proceed now to the USITC's analysis of the conditions of competition for FFTJ.
The Panel considers that the following observations made by the USITC are seminal to its conditions of competition analysis. First, the USITC stated that "[i]mports have taken an increasingly larger share of the domestic market each year since 1997". We agree with this observation on the basis of the graph below, which has been generated using USITC data.5510
150,000
125,000
100,000
75,000
50,000
100
90
80
70
60
50
40
30
20
10
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Imports' market shares
Imports' market shares (semi-annual)
Secondly, the USITC observed that "Import AUVs were generally above domestic AUVs. By contrast, pricing information gathered by the Commission on a butt weld fitting product showed that imports from non-NAFTA sources and Mexico (there were no reported imports from Canada) undersold the domestic product in each quarterly period for which data were provided." We also agree with the USITC's observation that import average unit values were generally above domestic average unit values on the basis of the graph below, which, again, has been generated using USITC data.5511
5510 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-45 at TUBULAR-38; Table TUBULAR-C-6.
5511 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-14 at TUBULAR-17; Table TUBULAR-C-6.
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 869
250,000 2500
225,000
200,000
2000
175,000
1500
150,000
125,000
1000
100,000
75,000
500
50,000 0
1996 1997 1998 1999 2000 2001
Imports Imports (semi-annual)
Domestic unit values Imports unit values
Domestic unit values (semi-annual) Imports unit values (semi-annual)
We note that the USITC questioned the relevance of average unit values in the context of its conditions of competition analysis, stating that: "We are cautious of placing undue weight on AUV information, as it may be influenced by issues of product mix." While we consider that the USITC should have provided a more detailed explanation of why it so readily dismissed such data, we do not necessarily consider that the USITC was wrong to rely upon data for the "butt weld fitting product " to the exclusion of the average unit value data. Indeed, at page TUBULAR-54 of Volume II of the USITC Report, it is stated that this product, otherwise referred to as "Product 22", is a "high volume" fitting product. This suggests to us that Product 22 was a reasonably representative basis for the USITC's conditions of competition analysis. We also agree with the USITC's observation that "pricing information gathered by the Commission on a butt weld fitting product showed that imports
… undersold the domestic product in each quarterly period for which data were provided." This much is evident from Table TUBULAR-61.5512
Finally, we note that the USITC found "the product-specific evidence of underselling to be significant" and concluded that "[t]he increasing presence of imports, in at least some cases at substantial underselling margins, prevented the industry from recouping increased costs through higher prices; instead, prices fell somewhat over the period. Accordingly, we find that imports are a substantial cause of serious injury." As noted above, there was evidence of import underselling for Product 22, which we consider to be sufficiently representative to have formed the basis of the USITC's conditions of competition analysis. We find that the existence of underselling together with the increasing level (and market share) of imports, as evidenced above in relation to our review of the
5512 USITC Report, Vol. II, 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 870
USITC's coincidence analysis, coincided with the decline in the situation of the domestic industry and tends to support the USITC finding above.5513
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
50,000 -5,000
1996 1997 1998 1999 2000 2001
Years Imports
imports (semi-annual)
Operating margin
Operating margin (semi-annual)
Therefore, on the basis of the foregoing, the Panel concludes that the USITC provided a compelling explanation that indicated, subject to the fulfilment of the non-attribution requirement, that a causal link existed between increased imports and serious injury.
Non-attribution
USITC findings
The USITC's findings read as follows:
"Respondents also alleged that causes other than imports were responsible for any injury experienced by the domestic industry. First, respondents assert that the industry's performance is related to factors such as the business cycle in the oil and gas industry.5514 A certain portion of domestic production is used for oil- and gas- related purposes and thus would be affected by market dynamics in that sector. However, to the extent that the industry's performance is related to the business cycle in the oil and gas industry, this should mean that the industry's financial performance should have been strong in 2000 and into 2001 because demand for OCTG and other oil and gas related products was very strong during that period. In fact, consumption of fittings and flanges was 4.3 percent higher in 2000 than in 1999, and was 10.4 percent higher in interim 2001 than in interim 2000. However, the financial performance of the fittings industry was at its lowest level in 2000, and the profit
5513 The data represented in the graph below are contained in the USITC Report, in particular in Table TUBULAR-8 at TUBULAR-10; Table TUBULAR-20 at TUBULAR-24; Table TUBULAR-C-6.
5514 (original footnote) Joint Respondents Prehearing Injury Brief on Product 22, Carbon Steel Flanges, Fittings, and Tool Joints, at 49.
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 871
level in interim 2001, while positive, remained well below the level of earlier years in the period examined on an annualized basis.5515
Respondents also claim that the domestic industry's capacity expansion and intra- industry price competition led to injury.5516 The industry did add capacity over the period examined, but at a rate less than the increase in apparent consumption.5517 Thus, the increase in capacity would not be expected to place substantial pressure on domestic prices. Nor have respondents identified what has changed over the period examined such that competition among domestic producers alone would turn a solidly profitable industry into one experiencing operating losses.
Respondents allege that the decreasing profitability of the domestic has resulted from industry facilities that are inefficient or outdated, and that domestic producers are unable to obtain sufficient forgings used in domestic production.5518 These allegations are not supported by record information.
Respondents also claim that the industry suffered from a shortage of qualified workers.5519 While a few producers noted worker shortages at certain times, the claim of a worker shortage is inconsistent with the fact that the domestic industry reduced its production workers by 6 percent from 1998 to 1999, another 8.7 percent from 1999 to 2000, and by 4.5 percent between interim 2000 and interim 2001. These reductions coincided with reduced industry production, shipments, and market share, as imports increased.
Finally, respondents claim that purchaser consolidation explains any negative price effects experienced by the industry.5520 In support, respondents cite one domestic producer who indicated that consolidation had negatively impacted price levels, but also had the benefit of reducing shipping costs. In general, purchaser consolidation would be expected to place some pressure on domestic prices. However, any consolidation would not explain the reduction in domestic production, shipments, employment, and other non-price indicators that occurred during the period examined.
In summary, we find that the increase in imports of fittings is an important cause of the serious injury to the domestic fittings industry and not less important than any other cause, and therefore have made an affirmative determination."5521
5515 (original footnote) CR and PR at Table TUBULAR-C-6.
5516 (original footnote) Joint Respondents Prehearing Injury Brief on Product 22, Carbon Steel Flanges, Fittings, and Tool Joints, at 53, 59.
5517 (original footnote) CR and PR at Table TUBULAR-C-6.
5518 (original footnote) Joint Respondents Prehearing Injury Brief on Product 22, Carbon Steel Flanges, Fittings, and Tool Joints, at 51, 53.
5519 (original footnote) Joint Respondents Prehearing Injury Brief on Product 22, Carbon Steel Flanges, Fittings, and Tool Joints, at 58.
5520 (original footnote) Joint Respondents Prehearing Injury Brief on Product 22, Carbon Steel Flanges, Fittings, and Tool Joints, at 52.
5521 USITC Report, Vol. I, pp. 177-178
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 872
Factors considered by the USITC
Increased capacity
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.3(b)(vii) supra.
Analysis by the Panel
The Panel considers that the USITC acknowledged that domestic capacity increases played a role in causing the injury that was suffered by the domestic industry. In particular, the USITC stated that: "Thus, the increase in capacity would not be expected to place substantial pressure on domestic prices." In our view, this statement implies that increases in capacity would be expected to place some pressure on domestic prices, even if not "substantial".
That the USITC considered pricing to be important is evident from the following statement contained in its analysis of the conditions of competition for FFTJ:
"Purchasers of tubular products indicated that price was a key factor in their purchasing decisions, behind only quality.5522 Moreover, nearly all purchasers indicated that imported and domestic fittings and flanges made to the same grade and specification may be used in the same applications. We find that such broad interchangeability indicates that price plays an important role in the market."5523
In addition, in the same section of its report, the USITC stated that:
"The increasing presence of imports, in at least some cases at substantial underselling margins, prevented the industry from recouping increased costs through higher prices; instead, prices fell somewhat over the period. Accordingly, we find that imports are a substantial cause of serious injury.5524"5525
It is clear to the Panel from the foregoing that the USITC considered that downward pressure on prices played an important role in causing the injury that was suffered by the domestic industry. Accordingly, it can be deduced from the foregoing that the USITC conceded that increases in capacity lead, at least in part, to downward pressure on domestic prices, which, in turn, impacted upon the state of the domestic industry. Indeed, the Panel considers that downward pressure was exerted by increases in capacity on prices, regardless of how one interprets "substantial" (the adjective used by
5522 (original footnote) CR and PR at Table TUBULAR-53. Domestic producers of fittings and flanges and a distributor of fittings testified that price was an important consideration in customer purchasing decisions. Tr. at 2516 (Zidell); Tr. at 2518 (Graham); Tr. at 2520 (Ketchum); Tr. at 2523 (Bernobich).
5523 See para. 10.505.
5524 (original footnote) Domestic producers cited increased imports as the cause of injury to the domestic industry. In the questionnaire sent to fittings producers, the Commission asked recipients to identify the factors, from a list of 13, including imports, that are adversely impacting the domestic industry. Recipients were given the option of identifying more than one factor. Of those responding, 16 producers identified
imports, and one identified the general economic downturn. No other factors were identified. Persons testifying at the public hearing also cited imports. One company official asserted that declining sales volumes and profits caused by imports have forced his firm to shelve plans for capital investment, severely impairing the firm's competitiveness and efficiency. Tr. at 2517 (Zidell).
5525 See para. 10.505.
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 873
the USITC). The Panel is of the view that all relevant "other factors" – even those with limited injurious effects on the domestic industry – must, together with other relevant factors, be identified, distinguished and assessed with a view to reaching an overall conclusion that increased imports have a genuine and substantial relationship of cause and effect with the injury suffered by the relevant domestic producers.
The Panel notes that the USITC considered trends in domestic industry capacity during the period of investigation, noting that domestic producer's capacity increased by 7.4% over the period of investigation. In particular, it stated that:
"Domestic producers' capacity increased by 7.4 percent over the period examined, somewhat less than the growth rate in consumption. Domestic capacity reached its highest level of the period examined in 1999, and declined by 5.2 percent in 2000, and by an additional 4.6 percent in interim 2001 compared to interim 2000.5526"
Despite the fact that the USITC acknowledged the role played by this factor in causing injury to the industry, it appeared to dismiss it in its non-attribution analysis. In our view, in dismissing increased capacity in its non-attribution analysis, the USITC did not, through a reasoned and adequate explanation, separate, distinguish and assess the nature and extent of the injurious effects caused by increased capacity so that the injury caused by this factor, together with other factors, was not attributed to increased imports.
Purchaser consolidation
Claims and arguments of the parties
The arguments of the parties are set out in Sections VII.H.3(b)(vii) supra.
Analysis by the Panel
The Panel also considers that the USITC acknowledged that purchaser consolidation played a role in the injury that was suffered by the domestic industry. In particular, the USITC stated that: "[i]n general, purchaser consolidation would be expected to place some pressure on domestic prices. However, any consolidation would not explain the reduction in domestic production, shipments, employment, and other non-price indicators that occurred during the period examined." In our view, although this statement indicates that the USITC did not consider that purchaser consolidation would explain declines in all injury factors, it nevertheless implicitly accepted that purchaser consolidation would place "some" pressure on domestic prices. For the reasons explained below, we consider that this effectively amounts to an acknowledgement that purchaser consolidation played a role in causing injury to the industry.
As mentioned above, it is clear to the Panel that the USITC considered that downward pressure on prices played an important role in causing the injury that was suffered by the domestic industry. Also pointed out above, the USITC stated that purchaser consolidation exerted downward pressure on prices. Therefore, following the USITC's logic, we consider that there is a link between purchaser consolidation and injury suffered by the domestic industry.
However, despite this link, the USITC dismissed this factor in its non-attribution analysis on the basis of the assertion that "any consolidation would not explain the reduction in domestic
5526 (original footnote) CR and PR at Table TUBULAR-C-6.
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 874
production, shipments, employment, and other non-price indicators that occurred during the period examined." Other than this statement, the USITC did not provide any explanation of why "consolidation would not explain the reduction in domestic production, shipments, employment, and other non-price indicators that occurred during the period examined." Accordingly, in the Panel's view, the USITC's explanation of its analysis of purchaser consolidation was not adequately reasoned. Further, it is our view that in failing to adequately explain this factor, the USITC failed to meet its obligation to establish explicitly, through a reasoned and adequate explanation, that the injury caused by purchaser consolidation, together with other factors, was properly separated and distinguished and not attributed to increased imports.
Conclusions
The Panel considers that, with respect to FFTJ, the USITC failed to comply with its non- attribution obligation contained in the second sentence of Article 4.2(b). In particular, we consider that the USITC failed to properly separate, distinguish and assess the nature and extent of the injurious effects of factors other than increased imports that were causing injury to the domestic industry on the basis of a reasoned and adequate explanation. This, to us, is clear from the fact that the USITC dismissed a number of factors (namely, increased capacity and purchaser consolidation) in its non-attribution analysis even though it effectively acknowledged that those factors were causing injury to the industry.
Overall conclusion on USITC's determination of a causal link
Notwithstanding the fact that the USITC did not provide an adequate and reasoned explanation of how the facts supported its finding of coincidence, the Panel was of the view that clear coincidence existed between the upward trend in imports and the downward trend in injury factors. The Panel proceeded to review the USITC's examination of the condition of competition and concluded that the USITC provided a compelling explanation that indicated, subject to fulfilment of the non-attribution requirement, a causal link existed between increased imports of FFTJs and serious injury to the relevant domestic producers. Further, we found that the USITC's non-attribution analysis failed to separate, distinguish and assess the nature and extent injurious effects of purchaser consolidation and increased capacity so that the injury caused by these factors, together with other factors, was not attributed to increased imports. Thus, the USITC did not provide a reasoned and adequate explanation supporting a determination that there was a genuine and substantial relationship of cause and effect between increased imports and serious injury to the relevant domestic producers.
Therefore, the Panel concludes that the USITC's finding that a causal link existed between imports of FFTJ and injury caused to the relevant domestic producers is inconsistent with Articles 4.2(b), 2.1 and 3.1 of the Agreement on Safeguards.
Stainless steel bar
The Panel notes at the outset that it has focused in this section on the arguments made by the complainants that, for us, raised the most problematic aspects of the USITC's determinations on causation, that is, those aspects that more obviously entailed violations of the Agreement on Safeguards. Since we will dispose of the complainants' claims in our review below, the Panel sees no need to deal with the other arguments.
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 875
(i) Coincidence and conditions of competition
USITC findings
The USITC's findings read as follows:
"We find that the increased imports of stainless bar are an important cause, and a cause not less than any other cause, of serious injury to the domestic industry. Accordingly, we find that increased imports of stainless bar are a substantial cause of serious injury to the domestic stainless bar industry.
Conditions of Competition
We have taken into account a number of factors that affect the competitiveness of domestic and imported stainless bar in the US market, including factors related to the product itself, the degree of substitutability between the domestic and imported articles, changes in world capacity and production, and market conditions. These factors affect prices and other considerations taken into account by purchasers in determining whether to purchase domestically-produced or imported articles.
First, demand for stainless bar fluctuated somewhat but grew overall during the five full-years of the period of investigation. Apparent US consumption of stainless bar increased from 276.6 thousand short tons in 1996 to 294.4 thousand short tons in 1997 but then declined to 280.3 thousand short tons in 1998 and to 265.5 thousand short tons in 1999. In 2000, however, apparent consumption of bar increased by 22.2 percent, growing to 324.2 thousand short tons.5527 This level of consumption was 17.2 percent larger than in 1996.5528 As the overall economy declined in 2001, apparent consumption of bar declined by 13 percent between interim 2000 and interim 2001.5529
Second, stainless steel bar is used in the aerospace, automotive, chemical processing, dairy, food processing, pharmaceutical equipment, marine application, and other fluid handling industries.5530 The large majority of market participants indicate that there are no known substitutes for stainless bar.5531
Third, although fourteen domestic firms reported producing stainless steel bar in 20005532, four firms accounted for the large majority of domestic production of stainless bar in 2000: Carpenter/Talley, Crucible Specialty Metals, AvestaPolarit, and Slater Steels Corp.5533 The domestic bar industry became more concentrated during the period of investigation. In 1997, Carpenter Technology, the *** domestic
5527 (original footnote) CR and PR at Tables STAINLESS-67 and STAINLESS-C-4. 5528 (original footnote) CR and PR at Tables STAINLESS-67 and STAINLESS-C-4. 5529 (original footnote) CR and PR at Tables STAINLESS-67 and STAINLESS-C-4. 5530 (original footnote) CR at STAINLESS-2, PR at STAINLESS-1.
5531 (original footnote) EC-Y-046 at Table STAINLESS-6.
5532 (original footnote) CR and PR at Table STAINLESS-1.
5533 (original footnote) In 2000, these four firms accounted for *** percent of reported domestic production of stainless bar. CR and PR at Table STAINLESS-1.
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 876
producer of stainless bar in 20005534, purchased Talley, the *** largest producer in 2000.5535 In addition, Empire Specialty Steel, the *** largest bar producer in 2000, shut down its stainless operations in June 2001.5536
The industry's aggregate capacity level increased during the period of investigation, growing by 5.5 percent from 1996 to 2000.5537 Capacity was 2.2 percent higher in interim 2001 than in interim 2000.5538 Capacity utilization declined from 63.0 percent in 1996 to 52.1 percent in 1999 but increased to 55.8 percent in 2000.5539 Industry capacity utilization then declined from 59.5 percent to 49.6 percent between interim 2000 and 2001.5540
Fourth, price is an important factor in purchasing decisions for stainless bar. Although quality was generally ranked by the majority of responding purchasers as the most important factor in the purchasing decision for stainless bar, the large majority of purchasers reported price as being one of the three most important factors in the purchase decision.5541
Fifth, like many stainless steel products, the price of stainless bar is directly affected by the price of nickel.5542 To account for fluctuations in the cost of nickel, stainless steel producers impose a surcharge on the price of their stainless bar products whenever the price of nickel reaches a certain level.5543 Generally, after declining during the first three years of the period of investigation, nickel prices increased significantly throughout 1999 and the first half of 2000. Nickel prices fell thereafter, declining through interim 2001.5544 The price of domestic stainless bar followed this trend somewhat during the period of investigation, with average unit values of domestic bar shipments and sales declining through the end 1999, recovering in 2000, and then declining in interim 2001.5545
Sixth, during the period of investigation, there were imports of stainless bar from over 40 countries, although not every country exported stainless bar to the United States in every year.5546 The quantity of imports of stainless bar from sources other than Mexico increased by 54 percent from 1996 to 2000 but fell by 17 percent
5534 (original footnote) Carpenter accounted for *** percent of reported domestic production of stainless bar in 2000. CR and PR at Table STAINLESS-1.
5535 (original footnote) Talley accounted for *** percent of reported domestic production of stainless bar in 2000. CR and PR at Table STAINLESS-1.
5536 (original footnote) Empire Specialty Steel, Inc. Questionnaire Response at August 6, 2001 Attachment.
5537 (original footnote) CR and PR at Tables STAINLESS-18 & STAINLESS-C-4.
5538 (original footnote) CR and PR at Tables STAINLESS-18 & STAINLESS-C-4. 5539 (original footnote) CR and PR at Tables STAINLESS-18 & STAINLESS-C-4. 5540 (original footnote) CR and PR at Tables STAINLESS-18 & STAINLESS-C-4. 5541 (original footnote) INV-Y-212 at 95.
5542 (original footnote) CR at STAINLESS-95-96, PR at STAINLESS-70-71. 5543 (original footnote) CR at STAINLESS-95-96, PR at STAINLESS-70-71. 5544 (original footnote) CR at STAINLESS-95-96, PR at STAINLESS-70-71.
5545 (original footnote) CR and PR at Tables STAINLESS-18, STAINLESS-30, & STAINLESS-C-4.
5546 (original footnote) INV-Y-180 at Table G25 – Stainless Bar and Light Shapes.
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 877
between interim 2000 and interim 2001.5547 The record indicates that domestic and imported stainless bar are comparable in most respects.5548
The aggregate capacity of foreign producers of stainless bar in countries other than Mexico increased by 10.5 percent during the period examined. The capacity utilization of these producers increased from 74.2 percent in 1996 to 82.3 percent in 1998, declined to 77.2 percent in 1999, and then increased to 87.1 percent in 2000. Aggregate foreign capacity utilization increased from 89.2 percent to 90 percent in interim 2001.5549
Seventh, antidumping duty orders were imposed on imports of stainless bar from Brazil, India, Japan, and Spain in 1995.5550 Antidumping duty orders were imposed against imports of stainless steel angle from Japan, Korea, and Spain in May 2001.5551
Analysis
We find first that the import increases between 1996 and 2000 had a serious adverse impact on the production levels, shipments, commercial sales and market share of the domestic industry. As we described above, the quantity and market share of imports both increased considerably during the period of investigation, with the quantity of imports increasing by 53.8 percent during the period from 1996 to 2000 and import market share increasing by 11 percentage points during that period as well.5552 Despite the fact that these import increases occurred during a period of growing demand, the industry's production volumes, shipment levels and sales revenues all declined significantly as a result of increases in import volume during the period.5553
In particular, the industry's production levels fell by 10 thousand short tons (or 5.5 percent) during the period between 1996 and 20005554, its net commercial sales fell by *** short tons (or *** percent) during that period5555, and the value of its net commercial sales declined by *** percent during the period.5556 As a result of these production and sales declines, the industry's capacity utilization rates fell considerably as well, dropping from 63.0 percent in 1996 to 55.8 percent in 2000.5557
5547 (original footnote) CR and PR at Tables STAINLESS-6 & STAINLESS-C-4.
5548 (original footnote) EC-Y-046 at Table STAINLESS-24; see generally EC-Y-046 at 14-28.
5549 (original footnote) CR and PR at Table STAINLESS-45. 5550 (original footnote) CR and PR at Table OVERVIEW-1. 5551 (original footnote) CR and PR at Table OVERVIEW-1.
5552 (original footnote) CR and PR at Tables STAINLESS-67 & STAINLESS-C-4.
5553 (original footnote) CR and PR at Tables STAINLESS-18, STAINLESS-30, & STAINLESS-C-4.
5554 (original footnote) CR and PR at Tables STAINLESS-18 & STAINLESS-C-4. 5555 (original footnote) CR and PR at Tables STAINLESS-30 & STAINLESS-C-4. 5556 (original footnote) CR and PR at Tables STAINLESS-30 & STAINLESS-C-4.
5557 (original footnote) In this regard, purchasers in the market reported that there was a moderately high level of substitutability between the imported and domestic merchandise, suggesting that the volume increase on the part of imports came directly out of domestic market share.
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 878
Moreover, the industry's share of the market also fell considerably, dropping from
64.6 percent in 1996 to 59.8 percent in 1999 and then to 53.5 percent in 2000.5558
In fact, the declines in the industry's production, shipment and market share levels occurred despite the fact that the industry added significant amounts of capacity during a period of reasonably strong growth in demand for stainless bar. Even with this increased capacity, the industry was unable to take advantage of the growth in demand for stainless bar as imports obtained an increasingly larger share of the domestic market for bar over the period of investigation. In particular, while apparent consumption of stainless bar grew by 48 thousand short tons between 1996 and 2000, the quantity of imports grew at a more accelerated rate, increasing by nearly 53 thousand short tons during this same period. This growth in imports effectively foreclosed the domestic industry from participating in the growth in demand during the period of investigation. In sum, the import increases that occurred during the period clearly had a serious adverse impact on the production volumes, sales levels, sales revenues, and market share of the industry during the period.
The record also indicates that imports affected domestic prices of stainless bar negatively during the period of investigation. The record in this investigation shows that most purchasers consider domestic and imported stainless bar to be comparable in most respects5559, indicating that there is a high degree of substitutability between the products. Moreover, the record of this investigation also indicates that price is an important part of the purchasing decision.5560 Finally, we note that imports undersold the domestic merchandise throughout the period of investigation in 47 of 53 possible quarterly comparisons at underselling margins of up to 51 percent.5561
We find that this underselling depressed and suppressed domestic prices during the period of investigation. Although the price of stainless bar is expected by market participants to track the price of nickel, the net sales revenues of the domestic stainless bar industry failed to keep pace with movements in the cost of nickel during the second half of the period of investigation, particularly during the latter half of 1999 and 2000, when the price of nickel increased substantially.5562 While the average unit value of the industry's net commercial sales increased in 2000 and interim 2001, the industry's cost of goods sold rose from *** percent of its net sales revenues in 1998 to *** percent of its net commercial values in 1999, *** percent of net commercial sales in 2000, and *** percent in interim 2001. As a result of these
5558 (original footnote) CR and PR at Tables STAINLESS-67 & STAINLESS-C-4. Indeed, the most significant adverse impact of imports in quantity terms occurred during the last full-year of the period of investigation, when apparent consumption of stainless bar grew by 22.1 percent and import quantities grew by
41.3 percent. In that year, the industry lost 6.3 percentage points of market share and experienced the most significant declines in its capacity utilization rates of the entire period of investigation. CR and PR at Tables STAINLESS-18, STAINLESS-30, STAINLESS-67, & STAINLESS-C-4.
5559 (original footnote) INV-Y-212 at 95.
5560 (original footnote) INV-Y-212 at 95.
5561 (original footnote) CR and PR at Tables STAINLESS-87, STAINLESS-99, & Figure
STAINLESS-9. These consistent underselling figures are supported by an examination of the average unit value for domestic and imported merchandise, which also show imports being priced at consistently lower levels than domestic merchandise during the period. CR and PR at Table STAINLESS-C-4.
5562 (original footnote) CR and PR at 95-96, PR at STAINLESS-70-71 & Tables STAINLESS-6, STAINLESS-18, STAINLESS-30, & STAINLESS-C-4.
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 879
decreasing margins between the industry's cost of goods sold and its net sales values, the industry's operating income levels declined from a profit of *** percent in 1998 to a loss of *** percent in 1999, recovered only slightly to a minimal profit of *** percent in 2000, and then fell to a loss of *** percent in interim 2001.5563 Moreover, the overall declines in the industry's operating levels in the last two-and-a-half years of the period occurred when imports were at their highest market share levels during the period5564 and when imports were consistently underselling the domestic merchandise.5565 Therefore, we find that consistent and significant price underselling by imports during the latter half of the period of investigation suppressed and depressed domestic prices to a serious degree, despite the fact that nickel prices and the industry's average unit values also increased significantly during this period.
In sum, we find that increased quantities of imports of stainless bar during the period were a substantial cause of the declines in the industry's trade and financial condition during the period. …"5566
Claims and arguments of the parties
The arguments of the parties are set out in Section VII.H.2(g) supra.
Analysis by the Panel
At the outset, the Panel notes that the USITC undertook a coincidence analysis for stainless steel bar and concluded that coincidence existed. Accordingly, we will consider whether these findings provide a reasoned and adequate explanation of how the facts support this finding.
The Panel again recalls that coincidence, when examined, needs to be established between the movements or trends in imports and the movements or trends in injury factors. Applying our standard of review, the Panel has considered coincidence between a number of the injury factors mentioned in Article 4.2(a), including those referred to by the USITC, and imports on the basis of facts that were available to the USITC in making its determination.
First, with regard to import trends and production trends, the Panel notes that there does not appear to be any coincidence. In particular, production declined between 1997 and 1999, when import levels declined during the same period. Similarly, as imports increased from 1999 to 2000, so too did production. Finally, as imports decreased at the very end of the period of investigation, so too did production.5567
5563 (original footnote) CR and PR at Table STAINLESS-30 & STAINLESS-C-4.
5564 (original footnote) CR and PR at Table STAINLESS-67 & STAINLESS-C-4.
5565 (original footnote) CR and PR at Tables STAINLESS-86-87 & STAINLESS-Figures 9-10.
5566 USITC Report, Vol. I, pp. 208-212.
5567 The data represented in the graph below are contained in the USITC Report, in particular in Table STAINLESS-6 at STAINLESS-11; Table STAINLESS-18 at STAINLESS-24; Table STAINLESS-C-4.
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 880
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
50,000
1996 1997 1998 1999 2000 2001
Years Imports
Imports (semi-annual)
Production
Production (semi-annual)
There appears to be a similar disconnect between import trends and trends in employment as that detected in relation to production. In particular, employment declined between 1997 and 1999. During the same period, import levels declined. Similarly, as imports increased from 1999 to 2000, so too did employment. Finally, as imports decreased at the very end of the period of investigation, so too did employment.5568
250,000 | 2,500 | ||||
225,000 | |||||
200,000 | 2,250 | ||||
175,000 | |||||
150,000 | 2,000 | ||||
125,000 | |||||
100,000 | 1,750 | ||||
75,000 | |||||
50,000 | 1,500 | ||||
1996 | 1997 | 1998 | 1999 | 2000 | 2001 |
Imports
Imports (semi-annual) Employment
Employment (semi-annual)
We also discern no coincidence between import trends and productivity trends. In particular, productivity progressively climbed from 1997 onwards. Apparently, this occurred independently of trends in imports, which declined between 1997 and 1999 and then increased between 1999 and 2000.
5568 The data represented in the graph below are contained in the USITC Report, in particular in Table STAINLESS-6 at STAINLESS-11; Table STAINLESS-18 at STAINLESS-24; Table STAINLESS-C-4.
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 881
Finally, as imports decreased at the very end of the period of investigation, so too did productivity, albeit slightly.5569
250,000 50
225,000 45
40
200,000
35
175,000 30
150,000 25
125,000 20
15
100,000
10
75,000 5
50,000 0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Productivity
Productivity (semi-annual)
Similarly, there does not appear to be any coincidence between increases in imports and capacity utilization. In particular, capacity utilization declined between 1997 and 1999 despite the fact that imports also declined during that period. Capacity utilization increased (albeit slightly) between 1999 and 2000, during which time imports also increased. Finally, as imports decreased at the very end of the period of investigation, so too did capacity utilization.5570
5569 The data represented in the two graphs below are contained in the USITC Report, in particular in Table STAINLESS-6 at STAINLESS-11; Table STAINLESS-18 at STAINLESS-24; Table STAINLESS-C-4.
5570 The data represented in the graph below are contained in the USITC Report, in particular in Table STAINLESS-6 at STAINLESS-15; Table STAINLES-18 at STAINLESS-24; Table STAINLESS-C-4.
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 882
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
50,000
100
90
80
70
60
50
40
30
20
10
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Capacity utilization
Capacity utilization (semi-annual)
While our evaluation is that coincidence does not exist between, on the one hand, import trends and, on the other hand, trends in production, employment, productivity and capacity utilization, this does not necessarily mean that, overall, coincidence did not exist. In this regard, we note that we were unable to consider whether the facts indicated that coincidence existed between the import trends and trends in operating margin and net commercial sales, given that data relating to the latter two factors had been redacted from the USITC Report on the ground of confidentiality. Such facts may affect the overall conclusion as to the existence or otherwise of coincidence in trends in imports and injury factors. Accordingly, the Panel is unable to come to a definitive conclusion as to whether, overall, coincidence existed.
As stated previously, in cases where, as part of an overall demonstration of causal link, a coincidence analysis has been undertaken but does not demonstrate a causal link, the Panel will continue its review turning to the conditions of competition analysis to assess whether the USITC, nevertheless, managed to provide a compelling analysis that a genuine and substantial relationship between cause and effect existed. We note that the USITC analysed the conditions of competition in addition to undertaking a coincidence analysis.
The Panel notes that the USITC considered that "imports affected domestic prices of stainless bar negatively during the period of investigation." The USITC additionally stated that "imports undersold the domestic merchandise throughout the period of investigation in 47 of 53 possible quarterly comparisons at underselling margins of up to 51 percent." Finally, it found that "this underselling depressed and suppressed domestic prices during the period of investigation … . Therefore, we find that consistent and significant price underselling by imports during the latter half of the period of investigation suppressed and depressed domestic prices to a serious degree, despite the fact that nickel prices and the industry's average unit values also increased significantly during this period."
We note as a preliminary point that the relevant domestic prices have been redacted from the USITC record, on the ground of confidentiality The Panel agrees that, in some circumstances, Members have the obligation, pursuant to Article 3.2 of the Agreement on Safeguards, to confidentialize certain information although the competent authorities can base their determination on
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 883
such confidentialized information.5571 Such an obligation should not reduce Members' rights to take safeguard actions. In cases where information has been confidentialized, the Panel will examine whether the competent authority provided a reasoned and adequate explanation through means other than full disclosure of that data.5572
We note in this regard in relation to stainless steel bar that Table STAINLESS-99 summarizes the number of instances of underselling and provides a range of the margins of underselling that occurred for all of those instances. In particular, that Table indicates that there were 40 instances of underselling by non-NAFTA imports and that the range of underselling was between 0.1%-51.8%. We note also that this factual allegation – that there were 40 instances of underselling by non-NAFTA imports – is not contested by the complainants and it is contrary to our standard of review to reassess the quality of this evidence in the absence of any prima facie challenge. In our view, although relevant data was redacted from the USITC Report, the USITC nevertheless provided alternative information in Table STAINLESS-99 that sought to substitute the redacted data. In light of the foregoing, the Panel concludes that the facts that are available to us tend to support the USITC's conclusion that there was import underselling during the period of investigation.
We note that trends in import market share are illustrated in the graph below, which has been generated using USITC data.5573
250,000
225,000
200,000
175,000
150,000
125,000
100,000
75,000
50,000
100
90
80
70
60
50
40
30
20
10
0
1996 1997 1998 1999 2000 2001
Imports
Imports (semi-annual) Imports' market shares
Imports' market shares (semi-annual)
The Panel notes that the facts indicate that import market share increased quite significantly during the period of investigation, which would be consistent with a finding of import underselling. In particular, the USITC found that "consistent and significant price underselling by imports during the latter half of the period of investigation suppressed and depressed domestic prices to a serious degree". In our view, given the facts referred to above, the USITC provided a compelling explanation indicating that, subject to the fulfilment of the non-attribution requirement, a causal link existed between increased imports and serious injury.
5571 Appellate Body Report, Thailand – H-Beams, paras. 111, 112 and 119.
5572 See our discussions in paras. 10.272-10.275.
5573 The data represented in the graph below are contained in the USITC Report, in particular in Table STAINLESS-6 at STAINLESS-11; Table STAINLESS 67 at STAINLESS-55; Table STAINLESS-C-4.