Full opinion text
OPINION RESTANI, Judge: This matter is before the court on plaintiffs’ motion for judgment on the agency record, challenging the United States International Trade Commission’s (“ITC” or “the Commission”) negative material injury determination with respect to certain hot-rolled steel products from Belgium, Brazil, Canada, France, Germany, Japan, Korea and the Netherlands that were found to be subsidized or sold at less than fair value (“LTFV”). Certain Flat-Rolled Carbon Steel Prods, from Argentina, Australia, Austria, Belgium, Brazil, Canada, Finland, France, Germany, Italy, Japan, Korea, Mexico, the Netherlands, New Zealand, Poland, Romania, Spain, Sweden, and the United Kingdom, USITC Pub. 2664, Inv. Nos. 701-TA-319-332, 334, 336-342, 344, and 347-353, and Inv. Nos. 731-TA-573-579, 581-592, 594-597, 599-609, and 612-619 (Aug. 1993) (final determ.) (“Final Det. ”); 58 Fed.Reg. 43,905, 43,906-07 (USITC 1993). In that determination, a majority of the ITC Commissioners found that the United States industry was not materially injured or threatened with material injury by reason of LTFV sales of hot-rolled carbon steel. Plaintiffs U.S. Steel Group, AK Steel Corp., Bethlehem Steel Corp., Geneva Steel, Gulf States Steel Inc. of Alabama, Inland Steel Industries, Inc., LTV Steel Company, Inc., Laclede Steel Company, National Steel Corp., and WCI Steel, Inc. (collectively “Petitioners”) are U.S. manufacturers of hot-rolled carbon steel products. Hot-rolled steel is employed in the construction, automotive, machinery and equipment industries. U.S. Int’l Trade Comm., Certain Flab-Rolled Carbon-Steel Products: Final Report to the Commission at 1-34 (1993) (“Pub. Staff Rpt.”). Most hot-rolled products are either used internally or transferred to an affiliated company for use in the manufacture of cold-rolled sheet and strip, or welded pipe. Pub. Staff Rpt. at 1-34; Final Det. at 15. At oral argument, petitioners stated that a percentage of hobrolled steel is also employed as substrate in the manufacture of corrosion-resistant steel. Approximately two-thirds of the domestic hot-rolled steel produced is “captively consumed” in the production of “downstream” products. Final Det. at 15 n. 37. The remainder represents hot-rolled steel sold to the open market. See id. at 15. On August 18, 1993, ITC gave notice of its final negative antidumping duty determination. ITC found that: a/ LTFV imports from Belgium, Brazil, Canada, France, Germany, Japan, Korea and the Netherlands did not cause or threaten material injury to the U.S. industry, and b/ subsidized imports from Belgium, Brazil, France, Germany and Korea did not cause or threaten material injury. Id. at 9; 58 Fed.Reg. at 43,906-07. Petitioners contest these negative determinations. A. ITC’s Determination The Commission found by a vote of 5-1 that imports from Brazil, Germany, France, Korea, and Japan did not cause or threaten material injury. All Commissioners voted in the negative regarding imports from Belgium, and by a vote of 4-2 the majority found imports from Canada and the Netherlands did not cause or threaten material injury. Dissenting views as to the threat finding were written by Commissioners Newquist and Rohr. The Commission reasoned that a negative determination was appropriate for all countries subject to investigation because there was a lack of causal nexus between the subject imports and the condition of the industry. Final Det. at 52. The Commission found the domestic like product to consist of hot-rolled carbon steel flat products, including plate in coils and hot-rolled floor plate in coils. Id. at 14. During the period of investigation, the domestic industry’s market share, by quantity, declined by one percent, and remained in excess of 93 percent. Id. at 52; Pub. Staff Rpt. at 1-144 tbl. 103. The Commission also found market penetration of the subject imports to be low. Final Det. at 46-47. The Commission was unable to find convincing evidence that there was a relation between declines in domestic industry performance and minor increases in the volume of imports. Id. at 52-53. The Commission noted that the large percentage of domestic industry production was captively consumed and concluded that the industry was affected only minimally by the subject imports. Id. at 21, 53. The Commission also determined that there was a lack of price effects by reason of the subject imports, partly because the product market was not highly price sensitive, nor were imports greatly substitutable with domestic products. Id. at 47-48. Although prices for domestic and imported products decreased somewhat during the period, import prices fluctuated and fell less than did domestic prices, imports generally oversold domestic products, and no lost sales or revenue allegations could be confirmed. Id. at 48-49. In addition, the majority concluded that most Korean imports were shipped from defendant-intervenor Pohang Iron & Steel Company (“POSCO”) to USS-POSCO Industries (“UPI”), an affiliated domestic producer. Thus, the remaining volume of imports sold on the merchant market was very small. Id. at 54-55. Some of these merchant market Korean products fell into niche product categories, thus substitutability was somewhat limited. Id. at 55. The majority found a lack of causal nexus between Korean imports as a whole and any material injury to the domestic industry. Id. The Commission also determined that negligible imports from the non-cumulated subject countries were insignificant in absolute volume, and as a share of domestic consumption. Id. at 56. In part on the basis of data demonstrating significant overselling, the Commission found these imports had no significant effect on domestic prices. Id. at 56-57. With regard to threat of material injury, the Commission exercised its discretion not to cumulate, making an individual negative finding for each subject country. Id. at 59. The majority determined that for Canada, an increase in market penetration over the period was offset by a significant increase in home market and third country shipments. Id. at 62. Although production capacity had increased, the majority noted that projections indicated a decline in 1993. Id. The majority stated that although there was the possibility of product shifting from corrosion-resistant products to hot-rolled products, the high cost of investment in a corrosion-resistant production line was a large incentive to continue to make as much higher value-added product as possible. Id. at 63. Thus, the majority concluded that the potential for product shifting was not significant, and that Canadian imports did not pose a threat. Id. at 63, 64. Lastly, the Commission found that imports from the Netherlands did not threaten material injury. Despite increases in volume over the period, the absolute volume and market share of Dutch imports were found to be too low to indicate a likelihood that imports would rise to injurious levels. Id. at 70. Levels of capacity utilization were found to be high. Id. Although the Commission recognized the possibility of product shifting here also, it found the record lacked concrete evidence to warrant an affirmative finding for threat based on this factor. Id. at 71. Standard of Review On a motion for judgment on the agency record, the scope of review of ITC’s determination is whether it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(l)(B) (1988). Discussion To determine whether material injury has occurred because of LTFV or subsidized imports under investigation, ITC is required to consider the volume of imports, their effect on domestic prices, the impact of imports on domestic production of the merchandise at issue, and other relevant economic factors. 19 U.S.C. § 1677(7)(B) (1988). Regarding impact of imports, relevant economic factors include, but are not limited to, the following: (I) actual and potential decline in output, sales, market share, profits, productivity, return on investments, and utilization of capacity, (II) factors affecting domestic prices, (III) actual and potential negative effects on cash flow, inventories, employment, wages, growth, ability to raise capital, and investment, and (IV) actual and potential negative effects on the existing development and production efforts of the domestic industry, including efforts to develop a derivative or more advanced version of the like product. Id. § 1677(7)(C)(iii). To find whether threat of material injury caused by LTFV imports exists, ITC must review ten factors, including foreign production capacity, market penetration, price suppression or depression, inventories of subject merchandise, underutilized foreign production capacity, potential for product shifting, and actual or potential negative effects on the industry’s existing development and production efforts. Id. § 1677(7)(F)(i). Petitioners argue that because of numerous errors in ITC’s final negative material injury and threat determinations this matter should be remanded. Petitioners specifically contend that 1/ the Commission improperly included in its definition of the domestic industry the production of hot-rolled steel captively consumed in the production of downstream products; 2/ the Commission failed to cumulate Korean imports in its analysis of material injury; 3/ the Commission misapplied the negligibility exception and failed to cumulate imports from Belgium, Brazil, Germany, Japan, and South Africa; 4/ Commissioner Brunsdale misapplied the negligibility exception, basing it solely on absolute magnitude of imports; 5/ the causation analyses of Commissioners Brunsdale and Crawford were improper and contrary to law; 6/ the causation analysis of Commissioner Watson was vague and contrary to law; 7/ the analyses of price sensitivity and substitutability by the Commission and Commissioners Watson, Brunsdale and Crawford, in particular, were inconsistent and unsupported by the record; 8/ the Commission’s finding of no adverse price effects was based upon flawed methodology; 9/ the Commission’s finding of lack of causal nexus, between marketplace developments after imposition of the preliminary bond and material injury, was incorrect; 10/ the Commission’s negative threat determinations regarding imports from Germany, Korea and the Netherlands were incorrect as there was evidence of a risk of product shifting; and 11/ the Commission’s negative threat determinations for Canada and the Netherlands were flawed because they were based on unreliable projections and were not supported by substantial evidence. The court will discuss in turn each of petitioners’ contentions, and the opposing views of defendant and defendant-intervenors. A. Captive production The court and the parties agree that the captive production issue is one of major importance to the disposition of this matter, if not the most important issue. Because the petitioners continually adjust their argument, as various versions fail, this issue is somewhat difficult to discuss. The court will attempt to address all permutations of the argument. Petitioners clearly assert that the Commission incorrectly addressed hot-rolled captive production in its assessment of the hot-rolled industry. Petitioners argue that the Commission’s approach is contrary to the methodology applied in prior determinations for the steel industry, in which captive production was excluded. Petitioners also contend that inclusion of captive production is contrary to general agency practice. Because of this alleged error, according to petitioners, hot-rolled captive production has been counted multiple times across the steel investigations, in cold-rolled, corrosion-resistant and plate production data, in addition to the hot-rolled industry analysis. Petitioners insist that the Commission should have applied a semi-finished product analysis to the hot-rolled captive production, thus treating such hot-rolled product as “work-in-progress” to be excluded from assessments of apparent consumption, market share, and financial performance in the hot-rolled industry investigation. Final Det. at 15. Petitioners originally filed four individual steel cases for the hot-rolled, cold-rolled, corrosion-resistant, and plate industries. In the context of a like product definition, defendant indicates that petitioners could have, but did not, argue that captively produced hot-rolled steel was part of a downstream like product, was a separate like product, or that all flat-rolled steel constituted one like product. Instead, petitioners’ argument was made in the context of an objection to ITC’s data collection methodology. Defendant contends that in past practice captive production has been included in U.S. industry data for other products, and its inclusion here is not a departure from such methodology. Consistent with the petition filed, the Commission considered hot-rolled production to constitute one like product, regardless of end use. The Commission rejected petitioners’ proposal to exclude captive production from its analysis, despite petitioners’ argument that inclusion would require the Commission to manipulate analysis of statutory factors for each industry to reflect a larger steel “industry,” due to the “integrated nature of various producers of various products.” Id. at 16-17. The Commission explained its change in methodology from prior steel industry determinations by noting that previously, it had not considered, and no party had raised, the issue of captive production in the steel industry. Id. at 18 n. 50; see Citrosuco Paulista, S.A v. United States, 12 CIT 1196, 1209-10, 704 F.Supp. 1075, 1088 (1988) (upholding definition of like product and finding earlier determinations not binding for purposes of definition, in part because parties did not argue issue previously). First, the statute defines like product as a product which is “like, or in the absence of like, most similar in characteristics and uses with, the article subject to an investigation.” 19 U.S.C. § 1677(10) (1988). The statute frames the definition of an industry in terms of like product, as the industry is “the domestic producers as a whole of a like product, or those producers whose collective output of the like product constitutes a major proportion of the total domestic production of that product.” 19 U.S.C. § 1677(4)(A) (1988). The methodology applied here by the Commission, that is, analysis of the entire production of like product, previously has been used both in instances where the captively consumed goods were a component input of another entirely different product, or were a semi-finished or upstream version of another product. See, e.g., Fresh Garlic from China, USITC Pub. 2755, Inv. No. 731-TA-683, at I-5-I-6 (Mar. 1994) (prelim.) (finding one like product for all forms of fresh garlic, regardless of intended use, corresponding with scope investigation); Tungsten Ore Concentrates from the People’s Republic of China, USITC Pub. 2447, Inv. No. 731-TA-497, at 5, 7-8 (Nov. 1991) (final) (determining one like product category contained three grades of product that were interchangeable, possessed similar production process and overlapped in channels of distribution); and Polyethylene Terephthalate Film, Sheet, and Strip from Japan and the Republic of Korea, USITC Pub. 2383, Inv. No. 731-TA-458 and 459, at 14, 19 (May 1991) (final) (determining one like product existed due to similarity in physical characteristics, product perceptions, channels of distribution, and lack of statutory basis to support exclusion of captive production). Second, despite this statutory structure and agency practice, at oral argument petitioners contended that the way to count data for each like product group was to count the hob-rolled steel at the point at which it was to be consumed to manufacture a finished product. Tr. of Oral Argument at 82 (Sept. 28-29, 1994) (“Tr.”). This approach, if applied evenhandedly, would most notably exclude from the foreign hot-rolled data the substantial amount of Korean hot-band imports manufactured by POSCO. If petitioners had wanted this data for both foreign and domestic hot-rolled steel destined for cold-rolled production, to be excluded from the hot-rolled investigation, they had ample opportunity to express this view in previous stages of the investigation. The court notes that neither did petitioners challenge the like product determination to seek exclusion of captive production on the basis just articulated, nor did they ask Commerce to define a class so that the corresponding like product would exclude captive production. Petitioners explain that prior to the determination, they had no reason to expect that domestic captive production would be treated differently from past steel determinations. Nonetheless, they do not now clearly ask for a different like product definition. Third, petitioners now seem to request alternatively that the U.S. industry be defined separately from like product, and that the steel be counted only when it leaves the domestic producer’s plant. The eourt cannot grant the petitioners’ request. Once the like product has been determined, the definition of the industry follows. See § 1677(4)(A); Asociacion Colombiana de Exportadores de Flores v. United States, 12 CIT 634, 636, 693 F.Supp. 1165, 1167 (1988) (finding identification of like product necessary to determine which industry is examined for injury or threat of injury). Corporate boundaries are not contemplated as dividing lines within the definitions of either like product or the industry. See §§ 1677(4)(A), (10). The eourt will assume arguendo that the petitioners have asserted and continue to assert a “like product” argument. Even so, the Commission did not err in following its general methodology of treating all of hot-rolled steel as one like product. See, e.g., Smith-Corona Group v. United States, 713 F.2d 1568, 1571 (Fed.Cir.1983), cert. denied, 465 U.S. 1022, 104 S.Ct. 1274, 79 L.Ed.2d 679 (1984) (concluding agency has broad discretion in choice of methodology); Torrington Co. v. United States, 16 CIT 220, 230, 790 F.Supp. 1161, 1172 (1992), aff'd, 991 F.2d 809 (Fed.Cir.1993) (same). The court sees no clear way to draw a line between hot-rolled steel destined for in-house downstream steel products and hot-rolled steel that is destined, for example, for automobiles to be manufactured in-house. See, e.g., Cambridge Lee Indus., Inc. v. United States, 13 CIT 1052, 1054-55, 728 F.Supp. 748, 750 (1989) (holding lack of sufficiently clear dividing line existed to separate domestic brass into different like products); Fresh Garlic from China, USITC Pub. 2755, at I-5-I-6 (“We look for clear dividing lines between possible like products.”); Certain Compact Ductile Iron Waterworks Fittings and Accessories Thereof from the People’s Republic of China, USITC Pub. 2671, Inv. No. 731-TA-621, at 6 (Aug. 1993) (final) (same). As indicated, and as acknowledged at oral argument, petitioners want POSCO imports to be included for comparison to the domestic hot-rolled like product, thus any reasonable request for a revised like product definition to segregate captive production would have operated to the disadvantage of petitioners’ claimed objective. Finally, the Commission’s like product decision and its decision to include captive production did not lead to a skewed method of data collection. The Commission distinguished between data reported for merchant sales and for internal transfers, as to apparent consumption, shipments and income and loss experience. Pub.Staff Rpt. at I-39-I-40 tbls. 10-11; I-71-I-74 tbls. 26, 28-30; Def.’s App., vol. IV, List 2, Doc. 216 at I — 105—1—106 tbl. 27 (“Conf.Staff Rpt.”). Some firms reported data for employment, wages and productivity on an allocated basis, based on merchant market sales, and on a portion of upstream processes, and others reported this data on a nonalloeated basis. Pub.Staff Rpt., at I-58-I-60 & tbls. 18-19. The Commission, however, also articulated its methodology and indicated it had applied a consistent approach to estimating profitability of internal transfers eaptively produced, for all firms for which data was collected. Id. at I-64-I-65. Data on production capability, inventories, property, plant and equipment, capital expenditures, and research and development were reported on an aggregate basis because the Commission was unable to allocate the data between the merchant market and captive production. Id. at I-50-I-51 & tbl. 15, 1-56-1-57 & tbl. 17, 1-65. Petitioners have not demonstrated to the court that the Commission abused its discretion, or ignored substantial evidence, in making its like product determination or in selecting a methodology for data collection consistent with the manner in which the petitions were filed. B. Cumulation If certain statutory requirements are met, cumulation of imports from two or more countries “subject to investigation” is mandated for the determination of present material injury. 19 U.S.C. § 1677(7)(C)(iv) (Supp. V 1993). The Commissioners have discretion not to cumulate if a country’s imports are found to be negligible and to have no discernible impact upon the domestic industry. Id. § 1677(7)(C)(v). 1. Imports from Korea Petitioners find error in the Commission’s negative present material injury determination because Korean imports were excluded from cumulative analysis. Petitioners urge that both the POSCO imports to supply UPI, and the Korean imports sold to the merchant market, should have been cumulated with other imports. Petitioners allege that there was evidence on the record that non-Korean imports competed significantly with Korean hot-rolled products, contrary to the finding of the majority. See Final Det at 38-39. The majority determined however that because the volume of these competing merchant market imports was so small, both absolutely and as a share of total Korean imports, such imports did not by themselves result in a reasonable overlap of competition. Id. at 39. Regarding POSCO imports to supply UPI, the majority found that “these imports are in certain niche categories in which there were no other imports from subject countries during the period examined.” Id. Because competition with other imports was lacking, the majority chose not to cumulate POSCO imports. Id. The majority also noted that there existed “no unaffiliated domestic producer ... currently able to supply UPI with more than a small percentage of its requirements.” Id. In that context, the majority found that, [the] discussion of imports from Korea destined for UPI does not turn on the fact that there exists a contract (however labeled) for the supply of UPI. The Commission is expressly not making a determination that any type of contractual supply relationship precludes a finding of competition. Id. at 39 n. 218. Petitioners first argue that the majority departed from the general rule for cumulation, because it assessed not only whether a reasonable overlap of competition existed between Korean and other subject imports, but also because it considered the contractual relationship between POSCO and UPI in deciding not to cumulate. This reading of the determination blurs the cumulation analysis the majority made clear. To establish whether a reasonable overlap of competition exists, the Commission applies a four-factor test, assessing 1/ the degree of fungibility among imports and with the domestic like product, 2/ the presence of sales or offers to sell in the same geographic markets, 3/ the existence of common or similar channels of distribution, and 4/ whether the products at issue were simultaneously present in the market. United Eng’g & Forging v. United States, 15 CIT 561, 582, 779 F.Supp. 1375, 1393 (1991); Fundicao Tupy S.A. v. United States, 12 CIT 6, 10-11, 678 F.Supp. 898, 902, aff'd, 859 F.2d 915 (Fed.Cir.1988). These factors are not exhaustive, and no single factor is determinative. Wieland Werke, AG v. United States, 13 CIT 561, 563, 718 F.Supp. 50, 52 (1989). In its determination, the majority initially found that POSCO hot bands imported by UPI were in niche categories and .did not compete with other imports, on the basis of import data broken down by product category. See Final Det. at 39; Conf.Staff Rpt. I-235-1-236 tbl. 98, F-22-F-23 tbl. F-2. This data indicates that a significant majority of Korean imports fell into 2 product categories, for which no other .country imports were received. The majority thus relied upon product-specific characteristics in finding POSCO imports did not compete with other subject imports. The majority next stated that because of this finding, it was- not required to reach the issue of competition with the domestic product. Final Det. at 39. The majority went on to note, however, that no unaffiliated domestic producer was currently able to supply more than a small amount of UPI’s requirements, concluding that any theoretical domestic competition was too speculative to support a reasonable overlap finding. Id. at 39 & n. 217. The majority expressly stated its finding was not based upon the presumption that any contractual supply relationship precludes a finding of competition. Id. at 39 n. 218. Three of the Commissioners in the majority, Commissioners Rohr, Crawford and Nuzum, adopted this analysis without modification. Two Commissioners, in their additional views, indicate that they relied not only upon product characteristics, but also upon the nature of the POSCO-UPI venture. Commissioner Watson reiterated the majority’s findings, but added that on the basis of “the specific requirements of UPI and the relative lack of fungibility between the POSCO shipments to UPI and the other subject imports,” there was no reasonable overlap of competition. Id. at 73. In addition, Commissioner Watson found no reasonable overlap of competition with domestic products, on the bases that no unaffiliated domestic producer could supply UPI with more than a small portion of its requirements, id., and that “the relationship among UPI, USX and POSCO is unique.” Id. at 74. Commissioner Watson noted that his determination regarding reasonable overlap of competition was based upon this unique relationship, but also stated the determination might have been otherwise had there been evidence of another domestic producer “ready, willing and able to supply UPI with hot band.” Id. at 74 n. 9. Commissioner Brunsdale also decided not to cumulate Korean imports, in part on the basis that imports to UPI were in certain niche categories where there were no competing imports. Id. at 311. Commissioner Brunsdale found that because “UPI is affiliated only with USX and POSCO and not with any other importer,____ [i]t is not economic to purchase this steel from other import sources.... [Thus,] there is no competition between POSCO and other importers.” Id. While long-term contractual relationships cannot be used to shield unfairly traded imports, any error that may have existed in this regard is harmless, as at least three Commissioners in the majority were clear about avoiding this pitfall, and Commissioner Watson tempered his conclusion in a manner which indicates that he did not find the existing contractual relationship a bar to cumulation. Petitioners next argue strenuously that the majority incorrectly found an absence of overlap between POSCO imports and domestic product on the basis of alleged lack of capability to produce the good. Petitioners note that the specifications UPI requires are not especially unique. Petitioners also assert that domestic manufacturers possess the capability to produce, although they may not do so currently, the quantity required. As to non-Korean imports, petitioners contend that the capability to produce is what the Commission should have considered, rather than actual production. This argument is flawed for two reasons. First, UPI at no time argued, and the Commission did not find, that domestic producers are unable to meet UPI specifications. Rather, UPI argued, and the Commission found, that no domestic producer could satisfy all of UPI’s production requirements: quantity, ability to supply long-term, and product specifications. Evidence on the record indicates that in the hot-rolled market, and for UPI specifically, the condition of the market is such that those companies that captively consume hot-rolled steel require large quantities over a long term to operate efficiently. See, e.g., Def.-Ints.’ Post Oral Argument Subm., Ex. 8, at 624, 663-64, 693 (test, of Leonard H. Chuderewiez); and id. Ex. 8, at 669-60 (test, of Robert Crandall). Second, petitioners have miseharacterized the ability to produce as it relates to the reasonable overlap of competition requirement. Petitioners rely upon Certain Forged Steel Crankshafts from the Federal Republic of Germany and the United Kingdom, USITC Pub. 2014, Inv. Nos. 731-TA-351 and 353 (Sept. 1987) (final) and Thermostatically Controlled Appliance Plugs and Internal Probe Thermostats Therefor from Canada, Japan, Malaysia, and Taiwan, USITC Pub. 2152, Inv. No. 701-TA-292 and Inv. Nos. 731-TA-400, 402-404 (Jan. 1989) (final), in support. Yet these determinations stand for the proposition that actual production, and not mere theoretical ability to produce, establishes competition. In Crankshafts, the products at issue were forged steel crankshafts in a specific weight range, used in vehicle engines. USITC Pub. 2014 at 8. While the crankshafts were produced to particular specifications, the Commission found those of similar design to be fungible, as they were used interchangeably by purchasers. Id. at 15-16. The Commission noted the long-term contract nature of the market, id. at 21, and that bids from producers were requested and supplied. Id. at A-43-A-46. Similarly, in Probe Thermostats, the imported and domestic products were produced to particular specifications, but were found to be interchangeable. USITC Pub. 2152 at 18-19. The manufacturers in Probe Thermostats also submitted bids, id. at A-35-A-37, thus they were ready, willing and able to meet requests for supply. Here, there is no evidence in the record that domestic steel manufacturers were ready to meet UPI’s time and quantity requirements, nor had bids been submitted indicating such readiness. Thus, petitioners’ position is not analogous to that of the producers in Crankshafts and Probe Thermostats. Regarding overlap of competition between Korean and other imported products, petitioners contend that evidence exists concerning Japan’s ability to produce high quality hot-rolled products that satisfy UPI requirements. Petitioners cite to the Japanese Respondents’ Joint Prehearing Brief, wherein are described the same technologies that petitioners allege POSCO uses to make high quality hot-rolled products. See App. to Pis.’ Mem.Supp.Mot.JAgency R., Tab 5, at 49, 52 (Japanese Resp’ts.’ Joint Prehearing Br.). Petitioners argue that there is evidence that POSCO and a U.S. producer possess the same hot-rolled equipment, purchased from a Japanese manufacturer. See Pis.’ Post Oral Argument Subm., sec. IV (Korea), Tab 1, at 3-4 (Aff. of Gary E. Hoff). Petitioners assert that from this it is reasonable to infer that a Japanese respondent would possess such equipment as well, and thus be capable of producing competing imports. Mere capability of production (which is by no means clear) does not necessarily mean that a company will produce a particular niche product. Furthermore, defendant states that, contrary to petitioners’ assertions, the Commission investigated equipment used by domestic producers, in supplemental questionnaires. Defendant concedes it did not inquire specifically into equipment owned by foreign producers, but petitioners did not propose that this question be incorporated in the questionnaire. Although an agency’s failure to collect pertinent data may constitute an abuse of discretion, Wieland Werke, 13 CIT at 574, 718 F.Supp. at 60, in these circumstances petitioners have not established that the Commission failed to investigate Korean imports adequately. Petitioners also maintain that the majority incorrectly found absence of overlap of competition between POSCO imports and other imports, in mistaken reliance upon importer questionnaire responses that allegedly covered less than 27% of total hot-rolled imports by volume in any year of the investigation. Petitioners conclude that the data does not support a finding that none of the subject imports met UPI specifications. Defendant and defendant-intervenors respond that coverage of the questionnaires was more than adequate, and that petitioners’ characterization is grossly inaccurate. Information in the Final Staff Report indicates that a large number of firms provided questionnaire data for these investigations overall. Specifically, of the questionnaires sent to the 351 firms who imported steel and the 77 domestic producers, responses came from 190 importers and 38 producers, accounting for over 70% of all importers of the subject product, and 96% of all 1992 open-market shipments by U.S. producers of hot-rolled product. Pub.Staff Rpt. at 1-38 n. 92, I-41-I-42 & n. 96, 1-45 & nn. 100-01. Also, the Commission sought data from importers for selected hot-rolled products, to be used in its pricing analysis or in review of niche product arguments. Id. at 45 n. 102. This selection was the subject of debate among the parties and the Commission, with petitioners having some success in persuading the Commission of its views on the subject. At oral argument, defendant-intervenors submitted charts to demonstrate that total 1992 import shipment data in importer questionnaire responses excéeded 75% of total shipments. Def.-Ints.’ Post Oral Argument Subm., Ex. 5, at Exs. A-F. Defendant-intervenors indicated that petitioners’ assertion was formulated on the basis of the wrong questionnaire responses, wherein importers had indicated shipments made from inventory only. Rather, defendant-intervenors argued, the coverage should be calculated from the data reported in response to the question requesting estimated 1992 total shipment volume by customer type. See id. Ex. 5, at Exs. D-F. Defendant-intervenors are correct. The court notes that the Commission is not required to gather 100% coverage in the questionnaire responses before it can make a determination. The applicable standard is “not whether the Commission might have •obtained additional information, but whether the determination is supported by substantial evidence on the record and according to law.” Hannibal Indus., Inc. v. United States, 13 CIT 202, 208, 710 F.Supp. 332, 337 (1989). Thus, the court will only remand determinations where the Commission has failed “ ‘to seek necessary information.’ ” Trent Tube Div. v. United States, 14 CIT 780, 789, 752 F.Supp. 468, 476 (1990) (quoting Hannibal, 13 CIT at 207, 710 F.Supp. at 336) (“Trent Tube I ”), aff'd sub nom. Trent Tube Div. v. Avesta Sandvik Tube AB, 975 F.2d 807 (Fed. Cir.1992). In Trent Tube I, the court rejected the argument that receipt of 23 out of 35 producers’’ questionnaires, some of which were incomplete, was insufficient to support an affirmative injury finding. Id. at 788-89, 752 F.Supp. at 475-76. Here, the questionnaire responses contained data for the two niche product categories for which UPI imported the highest volumes of POSCO products. See Apps. to Def.-Ints.’ Mem.Opp’n to Pis.’ Mot.JAgency R., App. 3, at 30 (sample importers’ questionnaire). On the basis of the summaries of this data, see Conf.Staff Rpt. at I-235-I-236 tbl. 98 F-22-F-23 tbl. F-l, the Commission found that POSCO products did not compete with other subject country imports. Lastly, petitioners argue that the majority was inconsistent in its finding that the small volume of non-UPI Korean imports, less than 10% of total Korean hot band imports, did not satisfy the reasonable overlap of competition requirement. See Final Det. at 38-39. Petitioners point to the even smaller volumes of imports from South Africa and Belgium that the Commission found satisfied the competition requirement. See Pub.Staff Rpt. at 1-135 tbl. 94. Korean imports shipped to the merchant market totalled significantly less than 100,-000 tons in 1991 and in 1992. While this volume exceeds those reported for South Africa and Belgium, petitioners’ argument fails because the majority properly found the portion of competing non-UPI Korean imports to be a very small part of overall Korean imports. Final Det. at 39. The Commission also found the volume of merchant market Korean imports to be less than the volume of any other country’s imports that were cumulated. Id. at 39 n. 215. In the case of South Africa and Belgium, although absolute import volumes were small, the majority of these imports competed, representing a reasonable overlap of competition for the imports as a whole. Id. at 34, 42 & n. 234, 44. To conclude that Korean imports in general competed and thus should be cumulated, based upon the small percentage sold in the merchant market, would not be reasonable. For the foregoing reasons, the court finds ITC’s determination not to cumulate Korean imports is supported by substantial evidence and is in accordance with law. 2. The negligibility exception In its present material injury analysis, the Commission did not cumulate imports from Belgium, Brazil, Germany, Japan and South Africa, finding such imports were negligible and did not cause any discernible adverse impact. Petitioners challenge the manner in which the Commission applies the negligibility provision of the cumulation statute. 19 U.S.C. § 1677(7)(C)(v). The court finds no legal error in the interpretation or application of the negligibility provision by any of the Commissioners, even though the determinations are not uniform on the issue. a. Commissioner Brunsdale’s interpretation From the petitioners’ point of view, it is Commissioner Brunsdale who appears to engage in the most objectionable application of the statute. Commissioner Brunsdale focuses on the percentage of the market held by imports from a particular country. Except where additional factors indicate another result is appropriate, if a country’s imports account for less than 1% (or sometimes less than 1.5%) of the U.S. market, generally Commissioner Brunsdale does not cumulate imports. Cf. Torrington, 16 CIT at 228, 790 F.Supp. at 1171 (finding imports from countries with less than 1% market share, absent extraordinary circumstances, and in market not very sensitive to price, were negligible). Commissioner Brunsdale’s approach is somewhat troublesome, because it leads to charges of a bright-line test, which is not permitted by the statute. Id. at 228, 790 F.Supp. at 1171. The results in these cases and Commissioner Brunsdale’s adoption of the majority opinion in this matter, however, lead the court to conclude that she considered the full record in reaching her determination herein. On the other hand, her application of the statute appears to be at the most expansive boundary that may still be said to be within the discretion allowed the Commission on this issue. In another case, where imports were more pervasive, more diffuse by country source or more fungible, such a mode of analysis might lead to a conflict with the basic purpose of the cumulation provision. See H.R.Rep. No. 40, 1st Sess., pt. 1, at 131 (1987) (indicating negligibility exception is not to be applied in manner that subverts purpose of cumulation provision). All other Commissioners apply the negligibility statute in a less broad-brush manner, and have not erred as to the interpretation of law in this regard. b. Burdens of proof Petitioners raise another complaint regarding the law as to application of the negligibility provision. They allege that the Commission is not applying the proper burdens of proof to the resolution of the negligibility dispute. This argument is based in large part on the decision in Creswell Trading Co. v. United States, 15 F.3d 1054 (Fed. Cir.1994) (“Creswell III”), which discussed the burdens applicable to the resolution of an issue before the United States Department of Commerce in a countervailing duty investigation. The original question before this court, in Creswell Trading Co. v. United States, 16 CIT 37, 37, 783 F.Supp. 1418, 1419 (1992) (“Creswell I ”), was whether Commerce properly determined that a price reimbursement scheme was countervailable. Under the program, users of domestically-produced steel, pig iron, received a payment from the Indian government upon export of the finished product. Id. at 38, 783 F.Supp. at 1419. The payment equalled the difference between the higher-cost domestic product and the cost of the raw material on the international market. Id., 783 F.Supp. at 1419. Plaintiffs argued that this program was not a countervailable subsidy, and relied upon Item (d) of the Illustrative List of Export Subsidies annexed to the Agreement on Interpretation and Application of Articles VI, XVI, and XXIII of the General Agreement on Tariffs and Trade, which defines a subsidy as follows: (d) The delivery by governments or their agencies of imported or domestic products or services for use in the production of exported goods, on terms or conditions more favourable than for delivery of like or directly competitive products or services for use in the production of goods for domestic consumption, if (in the case of products) such terms or conditions are more favourable than those commercially available on world markets to their exporters. Id. at 39, 783 F.Supp. at 1420 (quoting H.R.Doc. No. 153, 96th Cong., 1st Sess. pt. 1 at 295 (1979)). Plaintiffs maintained that the program did not provide payments that exceeded the difference between the prices of goods in the domestic market and on the world market. See id. at 38, 783 F.Supp. at 1419-20. Commerce determined that this fact was irrelevant and that such a program was a subsidy under U.S. law, whether or not the rebate only offset the difference between domestic and world market prices. Id., 783 F.Supp. at 1419-20. This court instructed Commerce to determine whether the program met the condition of the “if’ clause set forth in Item (d). Id. at 40, 783 F.Supp. at 1421. On remand, Commerce based its determination upon information that had been submitted during the original proceeding. Creswell III, 15 F.3d at 1059. While this information tended to demonstrate that the Indian program met the conditions of Item (d), it was not presented clearly, causing Commerce to make an error in its calculations. Id. at 1058-59. Commerce rejected any attempts by the respondent to correct the error, and simply concluded that the record evidence was “ambiguous” and that it did not support the claim. Id. at 1059. This court upheld the determination, emphasizing that the Indian government failed to carry its burden to establish a singular “world-market price” during the relevant period. Creswell Trading Co. v. United States, 16 CIT 776, 779, 797 F.Supp. 1038, 1040-1 (1992) (“Creswell II”). In Creswell III, the Federal Circuit held that where the respondent produced evidence establishing a world price, Commerce could not reject the evidence without some other evidence showing that it was inaccurate. 15 F.3d at 1061-62. The court found that Commerce had the burden of proof to establish by a preponderance of the evidence the statutory condition established by the “if’ clause of Item (d). Id. at 1060. It further held that Commerce had satisfied its burden of production by establishing the existence of the rebate program, but that the burden had shifted back to Commerce when respondents produced evidence that the services or products were not provided on terms or conditions more favorable than available on world markets. Id. at 1060-61. The court then concluded that Commerce had not met its burden of demonstrating that respondents’ information was either inaccurate or insufficient. Id. at 1061-62. Although the Federal Circuit did not rely upon any specific statutory or precedential authority, the court finds Creswell III may be understood more clearly if viewed in light of the requirements of the Administrative Procedure Act (“APA”) and several other decisions in the Federal Circuit. In the first sentence of § 556(d) of the APA, the general rule applicable to burdens of proof is that: “[ejxcept as otherwise provided by statute, the proponent of a rule or order has the burden of proof.” 5 U.S.C. § 556(d) (1988). This provision has been construed to “allocate[ ] ... the burden of going forward rather than the burden of ultimate persuasion.” Environmental Defense Fund, Inc. v. EPA, 548 F.2d 998, 1004 (D.C.Cir.1976), cert. denied, 431 U.S. 925, 97 S.Ct. 2199, 53 L.Ed.2d 239 (1977). The EDF court thus recognized what the fact of § 556(d) implies, that the burden of persuasion may rest on the challenger of a rule or order, where no statute or regulation has allocated the burden. See id. The EDF court found this analysis of burdens to be consistent with the “traditional approach that [the] burden [of going forward] normally falls on the party having knowledge of the facts involved.” Id. An analogous principle applies in ITA cases to such matters as a respondent’s claims to adjustments to its price data, and other matters clearly within a respondent’s knowledge. See, e.g., Zenith Elecs. Corp. v. United States, 988 F.2d 1573, 1583 (Fed.Cir.1993) (finding that to demonstrate market is non-principal, for purposes of fair market value calculations, “[t]he burden of production should belong to the party in possession of the necessary information”); Industrial Fasteners Group v. United States, 710 F.2d 1576, 1582 n. 10 (Fed.Cir.1983) (holding in tax rebate adjustment dispute that “[b]eeause [respondent] possessed (or could gather) the necessary facts, the burden was its (not ITA’s) to furnish that information”). The court finds that the Creswell III “burdens” analysis is inapplicable to the instant negligibility provision. Unlike Item (d), the cumulation statute does not articulate a particular factual requirement, which if not established, precludes applicability of the negligibility provision. The statute instead provides that- in making a determination of whether imports are negligible and have no “discemable adverse impact,” the Commission shall consider “all relevant economic factors regarding the imports,” including, “but not limited to,” volume, market share, nature of sales, and price sensitivity. 19 U.S.C. § 1677(7)(C)(v). The language of Creswell III is difficult to apply broadly. Both Commerce and ITC are investigating agencies and triers of fact, and the facts found are to be judged by the reviewing court on a substantial evidence standard. 19 U.S.C. § 1516a(b)(l)(B); see, e.g., Atlantic Sugar, Ltd. v. United States, 744 F.2d 1556, 1561 (Fed.Cir.1984) (finding where-evidence insubstantial, reviewing court must either reverse ITC determination or remand for. further fact-finding); Chung Ling Co. v. United States, 16 CIT 843, 846, 805 F.Supp. 56, 61 (1992) (acknowledging ITC’s prerogative as factfinder and discretion to make reasonable judgments and inferences in interpretation of evidence, reviewable under substantial evidence standard); Maine Potato Council v. United States, 9 CIT 293, 300, 613 F.Supp. 1237, 1243-44 (1985) (same). Neither Commerce nor ITC is a party plaintiff or defendant. The only way the court perceives to apply Creswell III outside of its particular fact pattern is to assume that its specific language applies to a situation in which the statute or the regulation on its face creates a presumption, by giving dispositive weight to a particular fact, as opposed to a wide-ranging multi-faceted inquiry. Creswell III cannot be applied to establish a generally applicable burden of proof that the Commission must satisfy, without running afoul of the statute and numerous cases discussing the Commission’s fact-finding role and the judicial standard of review of such facts. Having said that the appropriate mode of analysis of issues decided by the Commission is, in general, no different from the preCreswell III mode, for the sake of argument the court will attempt to address the issue of whether the Commission or respondents failed to sustain either a burden of proof or production, as petitioners alleged, concerning the negligibility analyses. According to petitioners’ theory, the Commission must cumulate imports from all countries subject to investigation unless the Commission establishes by a preponderance of the evidence that the imports from a particular country “are negligible and have no discernable adverse impact on the domestic industry.” 19 U.S.C. § 1677(7)(C)(v). Petitioners argue that the Commission (or respondents) either failed to make a prima facie case, or that one of them failed to produce evidence to rebut petitioners’ evidence of impact or to show that it was inaccurate. According to petitioners, any evidence of impact is enough to defeat a claim of no discernible impact. - First, the court rejécts petitioners’ claim that even one confirmed lost sale or instance of lost revenue is per se evidence of adverse impact. The determination of no adverse impact is a multi-faceted one, and the court will not construct a per se rule. Second, the Commission rejected evidence of certain lost sales or revenue for reasons of inaccuracy or failure to meet reasonable standards concerning such allegations. This is within its powers. See infra part F.l. Third, in its role as investigator, the Commission is allowed to select allegations to be investigated. See Negev Phosphates, Ltd. v. United States, 12 CIT 1074, 1092, 699 F.Supp. 938, 953 (1988) (holding ITC may, but is not required, to incorporate lost sales allegations in analysis of lost sales); Copperweld Corp. v. United States, 12 CIT 148, 169-70, 682 F.Supp. 552, 572 (1988) (holding no statutory provision requires ITC to use particular type of analysis to review lost sales). As a trier of fact, the Commission is allowed to weigh the evidence of lost sales and revenue against other evidence relating to volume and price effects in making its impact determination. These three reasons, as well as demonstrating why the Commission did not fail to comply with any part of the Creswell III rule of analysis, are reasons why the broadest language in Creswell III cannot be applied directly here, or indeed, to most questions that come before this court in unfair trade eases. The court will discuss separately whether each of the challenged negligibility determinations is supported by substantial evidence. c. Negligibility findings Petitioners assert that the Commission incorrectly found that imports from Brazil, Belgium, Germany, Japan and South Africa were negligible. In all instances, petitioners argue that on the basis of absolute volumes, these imports accounted for substantial volume and thus must have had some injurious impact on the industry. Petitioners assert that when totalled, these absolute volumes represent nearly one quarter of the imports under investigation. Petitioners’ contentions as to these individual countries stem from the view that “discernable adverse impact” may be measured by absolute volume and value, without including evaluation of market share and other price-related factors. The court in Torrington, 16 CIT at 227-28, 790 F.Supp. at 1170, was presented with similar arguments. In affirming a determination that found imports from 12 of 14 subject countries to be negligible, the Torrington court held that absolute volume of imports does not have independent significance for negligibility purposes. See id. at 229, 790 F.Supp. at 1171. Rather, to assess whether imports are negligible and have no discernable adverse impact, the Commission is required by statute to weigh a number of factors in addition to volume, including trends in imports such as price declines, share of apparent domestic consumption, evidence of underselling, substitutability as well as whether imports were isolated and sporadic. 19 U.S.C. § 1677(7)(C)(v). In its analysis of negligibility, the majority here reviewed these factors, and its conclusion to apply the negligibility exception was supported by substantial evidence. Three Commissioners found the market for hot-rolled products to be generally not price sensitive, and two Commissioners found the market to be moderately price sensitive. Final Det. at 32 nn. 161-62, 33 n. 163 and 315. The Commission noted that Belgian imports fell to statistically insignificant levels in 1992 after an increase to .2% in 1991. Id. at 41. The Commission determined that the sharp increase of Belgian imports in 1991 was aberrational and attributed it to response to short U.S. supply. Id. at 41 n. 231. In 16 of 19 instances, Belgian imports were found to have oversold the domestic product. Id. at 42 n. 235. Although these imports were not isolated and sporadic, the Commission determined that they were negligible, based on the level of overselling and the extremely low and decreasing market share. Id. at 41-42. Similarly for Brazil, the Commission observed that during the period of investigation Brazilian imports held a very small share of the market, never exceeding .3%. Id. at 42. In 21 of 35 instances, these imports oversold domestic products. Id. at n. 242. Although there was evidence of greater substitutability for Brazilian products, most of which were commercial grade, the Commission concluded imports were negligible because of the extremely low volume and market share. Id. at 42-43. German imports declined in volume by 35% over the period of investigation, and market share declined steadily, to .4% by 1992. Id. at 43; Pub.Staff Rpt. at 1-135 tbl. 94, 1-144 tbl. 103. In 62 of 66 pricing comparisons, German imports were priced higher than domestic product. Final Det. at 43 n. 248. Also, there were no confirmed lost sales or revenue allegations. Id. at 56. Again, the Commission found that the market share and volume data, as well as evidence of significant overselling, indicated German imports were negligible. Id. at 43. The market share of Japanese imports declined over the period as well, to .3% in 1992. Id. The Commission also found a pattern of consistent overselling, with no instances of underselling. Id. at 43-44 & n. 253. The Commission determined that competition was attenuated because for the niche products Japan produced, either there was evidence of consistent overselling, or quality differences limited substitutability between certain Japanese specialty products and domestic products. Id. at 44. On the basis of low volume, market share and consistent overselling, the Commission found Japanese imports to be negligible. Lastly, imports from South Africa only entered the U.S. market in 1992, accounting for a .1% market share. Id. The Commission determined that the low volume and market share for this limited time period warranted a finding of negligibility. In each of these instances, the Commission clearly reviewed several factors in addition to absolute volume, to reach its negligibility determinations. In no instance have petitioners relied upon anything other than one factor, absolute volume, to challenge the Commission’s reasoning. The court finds that for Belgium, Brazil, Germany, Japan and South Africa, the majority’s determinations were supported by substantial evidence and in accordance with law. C. Causation analysis Respondents argue that Commissioners Brunsdale, Crawford and Watson utilize causation analyses that are inconsistent with the statute, its legislative history and judicial precedent. The court disagrees. The statute provides that antidumping duties may not be imposed unless: the Commission determines that— (A) an industry in the United States— (i) is materially injured, or (ii) is threatened with material injury, or (B) the establishment of an industry-in the United States is materially retarded, by reason of imports of that merchandise. 19 U.S.C. § 1673(2) (1988). Material injury is defined as “harm which is not inconsequential, immaterial, or unimportant.” ' Id. § 1677(7)(A). In an apparent attempt to avoid potentially conflicting explanations of this statutory standard, Commissioner Watson simply quotes from § 1673 to describe the causation analysis he employs. Final Det. at 45 n. 266. Commissioner Crawford, as well as the remainder of the majority, take care to repeat the admonitions of various court opinions and the legislative history that instruct the Commission not to weigh causes. S.Rep. No. 249, 96th Cong., 1st Sess. 57 (1979) U.S.Code Cong. & Admin.News 1979, p. 381; Final Det. at 45 & n. 267; see generally id. at 318. What then is the source of contention regarding the legal standard? Respondents allege that Commissioner Watson has not spelled out his causation theory. While adhering to the statutory language, Commissioner Watson is very clear that he is searching for a sufficient causal link between imports and injury, a standard recognized in the legislative history. Final Det. at 45 n. 266 (quoting S.Rep. No. 249 at 75). Given Commissioner Watson’s acceptance of the instruction not to weigh various causes of injury and his close adherence to statutory language and legislative history, the court cannot say that Commissioner Watson is applying anything other than the proper standard. As to Commissioner Crawford, the problem appears to be that the Commissioner is not pleased with the “contributing cause” language used by some Commissioners and some court opinions. From this, petitioners infer that she is not complying with the court interpretation of the statute. If the Commissioner reads the “contributing cause” language to mean that where imports are a de minimis cause of injury, or a cause of de minimis injury, but are then combined with other causes of material injury so that an affirmative determination results, she is correct in rejecting the language. See id. at 45 n. 267. In actuality, the contributing cause language relates to a mode of analysis wherein one assesses the state of the industry, which is affected by numerous factors, determines if such industry is materially injured, and then attempts to determine if imports contribute in a non-de minimis way to such material injury. See, e.g., Encon Indus., Inc. v. United States, 16 CIT 840, 841, 1992 WL 245899 (Ct.Int’l Trade 1992) (finding that imports which contribute even minimally to material injury is sufficient); LMI-La Metalli Industriale, S.p.A. v. United States, 13 CIT 305, 321, 712 F.Supp. 959, 971 (1989) (same), aff'd in part and rev’d in part on other grounds, 912 F.2d 455 (Fed.Cir.1990); British Steel Corp. v. United States, 8 CIT 86, 97, 593 F.Supp. 405, 413 (1984) (finding test for causation to be whether imports are contributing to injury suffered by domestic industry). As Commissioner Crawford does not employ this mode of analysis, she has no reason to address contributing causes. Commissioner Crawford, as do Commissioners Watson and Brunsdale, attempts to determine in one step if the domestic industry is materially injured by reason of LTFV or subsidized imports. Thus, the problem appears to be a semantic one created by shifting modes of analysis. The language developed in earlier cases reviewing decisions based on a two-step method of analysis is not necessary if only a one-step analysis is involved. See, e.g., Gifford-Hill Cement Co. v. United States, 9 CIT 357, 368, 615 F.Supp. 577, 585-86 (1985) (stating in two-step analysis context: “[T]he Commission must rule in the affirmative if it finds even slight contribution from imports to material injury.”) (emphasis added); Maine Potato Council, 9 CIT at 300-01, 613 F.Supp.