OPINION
MUSGRAVE,
Senior Judge:
This opinion presumes familiarity with
Carpenter Technology Corp. v. United States,
Slip Op. 02-77 (July 30, 2002). Previously, the Court found that “banding” to account for the absence of complete variable cost of manufacturing (“VCOM”) information was not unlawful
per se
for the purpose of matching U.S. and foreign market sales, however the matter was remanded for
clarification of the statement that banding had been undertaken “in order to obtain more identical matches.” Commerce was also requested to clarify and, as necessary, reconcile Carpenter’s allegation that it had applied different standards to Viraj Impoexpo Ltd. (“Viraj”) and Panchmahal Steel Ltd. (“Panchmahal”).
See
Slip Op. 02-77 at 17-18.
Remand has resulted in
de minimis
antidumping duties. For U.S. sales that did not have an identical foreign market match due to incomplete VCOM information, Commerce determined that the weighted average of the dumping margin calculated for matched sales is “a reasonable approximation of dumping attributable to Viraj’s unmatched sales because it is based on Viraj’s own sales data and the Department has found no facts on the record to suggest that such a use would be distortive[ and] ... it is consistent with Department practice.”
Remand Results
at 4, referencing
Porcelain-on-steel Cooking Ware from Mexico: Final Results of Antidumping Duty Administrative Review,
58 Fed. Reg. 43327, 43329 (Aug. 16, 1993). Regarding resort to “banding” to compensate for the incomplete VCOM information, Commerce explains that
the facts of the record support the conclusion that
banding
is a reasonable alternative to the difference in merchandise analysis. Specifically, Viraj’s direct materials costs evidence cost differences between two size ranges. Commerce used
banding
according to these two size ranges in the absence of [complete] VCOM data. Therefore, where VCOM data was not available due to confusion over reporting requirements, rather than lack of cooperation, the use of
banding
as non-adverse facts otherwise available for Viraj was reasonable.
In addition, Commerce reviewed the record with respect to the “disparate” treatment between Viraj and Panchmahal. Based on this review, Commerce concluded that such “disparity” is attributable to Panchmahal’s failure to cooperate (i.e., Panchmahal’s refusal to provide information in the manner in which it was requested by the Department).] ] Furthermore, the Department noted that Panchmahal never reported cost data that took size into account for its comparison market, thus precluding the Department from banding its sales or deriving any information to make differences in merchandise adjustments to normal value. Unlike Panchmahal, and as affirmed by this Court, Viraj was a cooperative respondent.
* *
*
. . . [W]hile the deficiencies [in Viraj’s and Panchmahal’s responses] may relate to similar cost information, the Department determined that the reasons for the deficiencies were different.
i[I\.e.,
Viraj was confused as to the reporting
requirements while Panchmahal simply refused to report the requested data.
See Final Results
at 4-7, 11-13.) The Act clearly authorizes the Department to treat respondents differently based on their level of cooperation.
See
19 U.S.C. 1677e(b). As such, the “disparate” treatment of Viraj and Panchmahal does not violate the statute.
Remand Determination
at 3, 4-5 (internal citations omitted; highlighting in original).
The government argues the remand results should be sustained in their entirety. Viraj has provided no comment. Carpenter Technology Corporation (“Carpenter”) argues for a different model-matching methodology and for a different margin to assign to unmatched sales as facts otherwise available. Specifically, Carpenter again takes issue with the fact that Commerce reached different results with respect to Panchmahal and Viraj. It argues that on remand the correct interpretation of the remand order was for Commerce to
address in a substantive manner any difference in the respondents’ data submissions that supports use of different methodologies as to the two respondents. To simply state that the different methodologies are appropriate because Panchmahal failed to cooperate in not providing cost information to enable accurate product matching, when Commerce itself recognizes that Viraj also did not provide cost information to enable size-specific product matching, is a distinction without any meaning here. Because the two respondents both failed to provide the requested data to Commerce that would have enabled the agency to make size-specific product matches, Commerce has no reasonable basis for penalizing Panchmahal for this failure and treating Viraj much more favorably.
In its decision to apply an adverse facts available rate to Panchmahal, the Department highlighted the significance of the size variable, noting that Panchmahal’s “exclusion of the size characteristic in its reported control numbers” was contrary to the instructions expressed in the original questionnaire, and that it was a “necessity” to report costs on a size basis.
Decision Memorandum,
at 6. Yet, despite the similar admission that Viraj’s costs were reported “irrespective of size altogether” (and, therefore, unusable), Commerce treated Viraj much more favorably by applying a “banding” methodology and by using non-adverse facts available.
Given this, the only way to correctly respond to the Court’s remand instructions was for Commerce to acknowledge that the similar failings of both companies should have similar consequences. As the Court has said, “[i]t was incumbent upon
Commerce to apply its rationale to all respondents similarly situated.”
Def.-Int.’s Comments at 3-4. Carpenter complains that Commerce did not follow its policy of calculating costs consistent with model matching criteria developed at the outset of an investigation or review
with respect to Viraj although it applied that policy with respect to Panchmahal. Carpenter argues that banding is not a “neutral” use of facts otherwise available because Commerce here “made product comparisons without size-specific data by creating bands of merchandise so broad that a certain amount of sales were thus destined to be ‘comparable’ to each other[,] ... a ‘gift’ to Viraj, while Panchmahal was penalized for a similar deficiency in its data.”
Id.
at 4-5.
Commerce has correctly interpreted the order of remand. In its initial briefs, Carpenter challenged Commerce’s determination on Viraj’s cooperativeness by arguing
inter alia
that deeming Viraj cooperative and Panchmahal uncooperative could not be reconciled.
Free access — add to your briefcase to read the full text and ask questions with AI
OPINION
MUSGRAVE,
Senior Judge:
This opinion presumes familiarity with
Carpenter Technology Corp. v. United States,
Slip Op. 02-77 (July 30, 2002). Previously, the Court found that “banding” to account for the absence of complete variable cost of manufacturing (“VCOM”) information was not unlawful
per se
for the purpose of matching U.S. and foreign market sales, however the matter was remanded for
clarification of the statement that banding had been undertaken “in order to obtain more identical matches.” Commerce was also requested to clarify and, as necessary, reconcile Carpenter’s allegation that it had applied different standards to Viraj Impoexpo Ltd. (“Viraj”) and Panchmahal Steel Ltd. (“Panchmahal”).
See
Slip Op. 02-77 at 17-18.
Remand has resulted in
de minimis
antidumping duties. For U.S. sales that did not have an identical foreign market match due to incomplete VCOM information, Commerce determined that the weighted average of the dumping margin calculated for matched sales is “a reasonable approximation of dumping attributable to Viraj’s unmatched sales because it is based on Viraj’s own sales data and the Department has found no facts on the record to suggest that such a use would be distortive[ and] ... it is consistent with Department practice.”
Remand Results
at 4, referencing
Porcelain-on-steel Cooking Ware from Mexico: Final Results of Antidumping Duty Administrative Review,
58 Fed. Reg. 43327, 43329 (Aug. 16, 1993). Regarding resort to “banding” to compensate for the incomplete VCOM information, Commerce explains that
the facts of the record support the conclusion that
banding
is a reasonable alternative to the difference in merchandise analysis. Specifically, Viraj’s direct materials costs evidence cost differences between two size ranges. Commerce used
banding
according to these two size ranges in the absence of [complete] VCOM data. Therefore, where VCOM data was not available due to confusion over reporting requirements, rather than lack of cooperation, the use of
banding
as non-adverse facts otherwise available for Viraj was reasonable.
In addition, Commerce reviewed the record with respect to the “disparate” treatment between Viraj and Panchmahal. Based on this review, Commerce concluded that such “disparity” is attributable to Panchmahal’s failure to cooperate (i.e., Panchmahal’s refusal to provide information in the manner in which it was requested by the Department).] ] Furthermore, the Department noted that Panchmahal never reported cost data that took size into account for its comparison market, thus precluding the Department from banding its sales or deriving any information to make differences in merchandise adjustments to normal value. Unlike Panchmahal, and as affirmed by this Court, Viraj was a cooperative respondent.
* *
*
. . . [W]hile the deficiencies [in Viraj’s and Panchmahal’s responses] may relate to similar cost information, the Department determined that the reasons for the deficiencies were different.
i[I\.e.,
Viraj was confused as to the reporting
requirements while Panchmahal simply refused to report the requested data.
See Final Results
at 4-7, 11-13.) The Act clearly authorizes the Department to treat respondents differently based on their level of cooperation.
See
19 U.S.C. 1677e(b). As such, the “disparate” treatment of Viraj and Panchmahal does not violate the statute.
Remand Determination
at 3, 4-5 (internal citations omitted; highlighting in original).
The government argues the remand results should be sustained in their entirety. Viraj has provided no comment. Carpenter Technology Corporation (“Carpenter”) argues for a different model-matching methodology and for a different margin to assign to unmatched sales as facts otherwise available. Specifically, Carpenter again takes issue with the fact that Commerce reached different results with respect to Panchmahal and Viraj. It argues that on remand the correct interpretation of the remand order was for Commerce to
address in a substantive manner any difference in the respondents’ data submissions that supports use of different methodologies as to the two respondents. To simply state that the different methodologies are appropriate because Panchmahal failed to cooperate in not providing cost information to enable accurate product matching, when Commerce itself recognizes that Viraj also did not provide cost information to enable size-specific product matching, is a distinction without any meaning here. Because the two respondents both failed to provide the requested data to Commerce that would have enabled the agency to make size-specific product matches, Commerce has no reasonable basis for penalizing Panchmahal for this failure and treating Viraj much more favorably.
In its decision to apply an adverse facts available rate to Panchmahal, the Department highlighted the significance of the size variable, noting that Panchmahal’s “exclusion of the size characteristic in its reported control numbers” was contrary to the instructions expressed in the original questionnaire, and that it was a “necessity” to report costs on a size basis.
Decision Memorandum,
at 6. Yet, despite the similar admission that Viraj’s costs were reported “irrespective of size altogether” (and, therefore, unusable), Commerce treated Viraj much more favorably by applying a “banding” methodology and by using non-adverse facts available.
Given this, the only way to correctly respond to the Court’s remand instructions was for Commerce to acknowledge that the similar failings of both companies should have similar consequences. As the Court has said, “[i]t was incumbent upon
Commerce to apply its rationale to all respondents similarly situated.”
Def.-Int.’s Comments at 3-4. Carpenter complains that Commerce did not follow its policy of calculating costs consistent with model matching criteria developed at the outset of an investigation or review
with respect to Viraj although it applied that policy with respect to Panchmahal. Carpenter argues that banding is not a “neutral” use of facts otherwise available because Commerce here “made product comparisons without size-specific data by creating bands of merchandise so broad that a certain amount of sales were thus destined to be ‘comparable’ to each other[,] ... a ‘gift’ to Viraj, while Panchmahal was penalized for a similar deficiency in its data.”
Id.
at 4-5.
Commerce has correctly interpreted the order of remand. In its initial briefs, Carpenter challenged Commerce’s determination on Viraj’s cooperativeness by arguing
inter alia
that deeming Viraj cooperative and Panchmahal uncooperative could not be reconciled. Viraj’s cooperation could be sustained on the basis of independent record evidence. Panchmahal is not a party to this proceeding, but the treatment of its circumstances also appeared relevant to Carpenter’s allegation of a results-oriented determination. Rather than undertake
arguendo
examination of the respective requests for data and responses from those respondents, the Court considered it appropriate to remand the issue to Commerce for clarification and, as necessary, reconciliation. The remand results adequately explain Commerce’s reasoning for its treatment of Viraj as distinct from Panchmahal for purposes of this proceeding.
Carpenter also continues to insist that “[b]anding for either party has the effect of
obviating,
rather than correcting, differences in product characteristics and the attendant matching problems caused by missing cost data.”
Id.
at 5. Carpenter alleges that through banding “a
double
benefit was conferred, as the product matches that were so dissimilar that they required constructed value even after the products were redefined by two universal ‘bands’ produced a margin that was based on already artificially constructed ‘identical matches.’ ”
Id.
(italics in original).
Carpenter’s other arguments essentially restate arguments the Court rejected in ruling on the initial motions for judgment on the agency record,
see
Slip Op. 02-77 at 16-17, and these arguments are no more persuasive at this time. As noted in the prior opinion, differences in merchandise must generally exceed 20 percent before merchandise is presumed not comparable. Carpenter’s double benefit theory, one of compounding, does not demonstrate, as a matter of fact, that banding produced distorted results, that products within each band are so dissimilar in size from their counterpart as to be incomparable, or that Commerce could not have been other than “satisfied” that a size difmer adjustment was absolutely necessary. Commerce may depart from policy if it provides a reasonable explanation for doing so,
e.g., Industria de Fundicao Tupy v. United States,
20 CIT 875, 876, 936 F.Supp 1009, 1015 (1996), and it has done so here.
Carpenter lastly complains that it was unclear why the Court observed that the 3.87 percent non-adverse less-than-fair-value margin determined against Grand Foundry in 1994 appeared to be a “tenuous fit” to use for Viraj’s unmatched sales. The Court’s observation was based upon: (1) the age of that rate; (2) the apparent lack of connection between that respondent and the one at bar; (3) whether there is other evidence in the record that Commerce may reasonably conclude is more probative of Viraj’s present circumstances than the older non-adverse LTVF rate
(e.g.,
the zero percent margin determined against Viraj in 1997); and (4) the Court’s conclusion with respect to the government’s argument in opposition thereto. The Court did not direct Commerce to use a particular rate, only that selected data have to evince a “rational relationship between data chosen and the matter to which they apply.” Slip Op. 02-77 at 21 n.12, quoting
Manifattura Emmepi S.p.A. v. United States,
16 CIT 619, 624, 799 F.Supp. 110, 115 (1992).
Conclusion
Commerce’s remand results comply with the prior opinion and order and will be sustained.