BRYSON, Circuit Judge.
This case concerns the fourth administrative review of an antidumping duty order on imports of frozen fish fillets from Vietnam. The fish that are the subject of the order are of the genus pangasius ("pangas fish"), which compete with domestic catfish in the retail fish market in this country. The period of review covered August 2006 through July 2007 and resulted in the assessment of duties against several Vietnamese manufacturers, including appellant QVD Food Company, Ltd.
The Department of Commerce calculates antidumping duty margins by comparing the "normal value" of the goods in question with their actual or constructed export price. 19 U.S.C. § 1677b(a). If the normal value exceeds the export price, Commerce imposes an antidumping duty on imports equivalent to the percentage difference between those two values, i.e., the dumping margin. Id. §§ 1673, 1677(35)(A).
Commerce treats Vietnam as a "nonmarket economy" country. See 19 U.S.C. § 1677(18). Because it is not always possible to determine the normal value of goods from nonmarket economy countries in the manner outlined in 19 U.S.C. § 1677b(a)(1), Congress allows Commerce to value the factors of production for such goods by looking to the best available information from appropriate market economy countries, referred to as "surrogate countries." See id. § 1677b(c)(1); Dorbest Ltd. v. United States, 604 F.3d 1363, 1367 (Fed.Cir.2010). For the fourth administrative review of the antidumping order in this case, Commerce chose Bangladesh as the primary surrogate market economy country to use in valuing the factors of production. Commerce used information gathered from producers in Bangladesh to determine the value of whole pangas fish.
In the preliminary results of the fourth administrative review, Commerce valued whole pangas fish at 45 Takas/kilogram ("Tk/kg"). That figure was based on the financial statement of Bangladeshi pangas fish producer Gachihata for the 2006-2007 fiscal year ("FY2006-07"). Commerce determined that Gachihata's financial statement for that year contained the best available information in the record as to the price of the fish. Commerce used the FY2006-07 financial statements of two other Bangladeshi seafood producers to calculate financial ratios for "general expenses" to ascribe to Vietnamese exporters. See 19 C.F.R. § 351.408(c)(4). Based on those values, Commerce preliminarily found that QVD should be assigned a de minimis dumping margin.
After publishing its preliminary results on September 8, 2008, Commerce gave all parties until September 28, 2008, to place information relevant to the valuation of factors of production in the administrative record. See 19 C.F.R. § 351.301(c)(3)(ii). Commerce also gave the parties an opportunity to file comments and respond to comments submitted by others. Commerce was statutorily required to issue final results of the fourth administrative review within six months of the publication
In February 2009, both QVD and appellees Catfish Farmers of America et al. ("CFA") submitted written comments addressing the preliminary results. CFA argued that the FY2006-07 Gachihata financial statement was an unreliable source for valuing whole pangas fish. CFA pointed in particular to the Gachihata Board of Directors' Report for FY2006-07, which painted a grim picture of the company's financial condition. The Directors' Report described Gachihata as being "in dire need of fund[s]" and "dying day by day." QVD argued in a rebuttal brief that the FY2006-07 Gachihata data continued to constitute the best available information in the record for valuing whole pangas fish. Following the briefing, Commerce held an administrative hearing at which the parties addressed the issue of fish valuation.
On March 3, 2009, six days before the deadline to issue final results, Commerce placed in the administrative record a 2007 report by the United Nations Food and Agriculture Organization ("the FAO report"). When offering the report into the record, Commerce explained that it might be an appropriate source of information for valuing whole pangas fish. The report cited an average price of 42 Tk/kg based on a survey of 60 Bangladeshi fish farms between October 2005 and February 2006. Commerce invited the parties to comment within two days on "the appropriateness of basing the surrogate value of the whole live fish input" on the FAO report.
QVD and CFA both commented on the FAO report two days later. QVD argued that Commerce should adopt the whole pangas fish price quoted in the report as the best available information as to the surrogate value of the fish. CFA, on the other hand, argued that the FAO report should be removed from the administrative record because Commerce had not given the parties the same amount of time to comment that would normally be available after new factual information was placed in the record. CFA also argued that there were significant questions as to the reliability of the FAO report, noting several difficulties encountered by the surveyors. CFA further contended that the data in the report may have been based on anecdotal evidence and that it was unclear whether the reported prices were contemporaneous with the survey that was conducted for the report.
On March 10, 2009, two business days after the parties submitted comments, Commerce publicly announced the final results of the fourth administrative review. Commerce valued whole pangas fish using an inflation-adjusted value from the FY2000-01 Gachihata financial statement. The FY2000-01 statement priced pangas fish at 68 Tk/kg, and Commerce adjusted that price upward for inflation to 98 Tk/kg. Commerce calculated the inflator by dividing the average Bangladeshi consumer price index ("CPI") for the months encompassing the period of review by the average CPI for the months encompassing Gachihata's 2000-01 fiscal year. Commerce did not alter the financial ratios used in the preliminary results. As a result of those calculations, Commerce assigned QVD an antidumping duty margin of 0.52%.
Commerce explained the differences between the preliminary and final results in a decision memorandum. It stated that it had attempted to locate new data relevant to whole pangas fish pricing after the parties expressed concerns about each of the potential data sources in the record. The last-minute introduction of the FAO report into the administrative record was the result of that research effort. The decision memorandum went on to explain that, while the FAO report might be deserving
Turning to the FY2006-07 Gachihata financial statement, Commerce determined that the Gachihata Directors' Report for that year "cast considerable doubt on the reliability of using [that statement] as the basis for calculating a whole fish input surrogate value." Commerce raised several concerns about the data in the FY2006-07 statement and concluded that it was an unreliable source for valuing whole pangas fish. Once the FY2006-07 statement was deemed unreliable, Commerce determined that the market price quoted in the FY2000-01 Gachihata financial statement was the best available information in the record to value whole pangas fish. Commerce's decision to use the price in the FY2000-01 statement echoed its decision to use that same price in the initial antidumping duty determination and the first two administrative reviews. Commerce had chosen to use the FY2006-07 Gachihata financial statement data for the third administrative review, but the Directors' Report was not on file for that review.
After the final results were announced, QVD alleged that Commerce had committed "ministerial errors," including miscalculating the financial ratio for general expenses ascribed to Vietnamese exporters. Specifically, QVD argued that Commerce had double-counted certain general expenses reported by the Bangladeshi surrogate companies because QVD had reported those expenses separately as sales expenses. Commerce refused to change its financial ratio calculation because it did not view QVD's alleged errors as "ministerial" in nature.
Both QVD and CFA appealed Commerce's final results to the Court of International Trade. QVD contested Commerce's determination as to the surrogate value of whole pangas fish and the general expense ratio assigned to QVD. In a preliminary order, the court determined that it would not consider any argument by QVD that referenced administrative proceedings after the final results of the fourth administrative review, and it sustained Commerce's decision not to rely on the FAO report with regard to the valuation of whole pangas fish. In a subsequent opinion, the court sustained Commerce's valuation of whole pangas fish in its entirety. The court agreed with Commerce that the Directors' Report presented a "very grim and unsettling picture of Gachihata's financial condition" in FY2006-07 and provided reasonable grounds for Commerce to favor the FY2000-01 Gachihata financial statement data over the more contemporaneous data. The court found that Commerce had made a reasonable choice given "imperfect alternatives." The court also agreed with Commerce that QVD's challenge to the general expense ratio was "a substantive challenge to Commerce's assignment of certain expenses to the surrogate ratio calculations" rather than a ministerial error and therefore was raised too late.
On appeal, QVD argues that Commerce erred in treating the FY2000-01 Gachihata financial statement as the best available information for valuing whole pangas fish in the fourth administrative
With respect to Commerce's decision not to consider the merits of the FAO report, QVD notes that the agency did not identify any specific flaw in the FAO report and that Commerce found the report potentially useful enough to be considered in future proceedings. QVD argues that Commerce cannot refuse to consider information having strong indicia of reliability simply because future investigation might alter Commerce's view about the quality of that information.
Commerce responds that it was reasonable not to rely on the FAO report to value whole pangas fish given the time constraints on its decisional process. The statutory deadline to issue final results was less than a week away when the FAO report was placed in the record. Commerce argues that the concerns raised by CFA in its comments were sufficient to call into question the reliability of the report, and that without more time to assess those comments, the agency reasonably decided to rely on other data in the record for valuation purposes. Commerce further contends that its decision to use the FAO report in a subsequent new shipper review is irrelevant to the question whether its time-driven decision in the fourth administrative review was reasonable.
We review Commerce's antidumping determinations for substantial evidence. 19 U.S.C. § 1516a(b)(1)(B)(i). Our inquiry "takes into account the entire record, which includes evidence that supports and detracts from the conclusion reached." Sango Int'l L.P. v. United States, 567 F.3d 1356, 1362 (Fed.Cir.2009). However, Commerce has broad discretion to determine the best available information for an antidumping review, given that the term "best available information" is not defined by statute. Ad Hoc Shrimp Trade Action Comm. v. United States, 618 F.3d 1316, 1322 (Fed.Cir.2010); see Nation Ford Chem. Co. v. United States, 166 F.3d 1373, 1377 (Fed.Cir.1999) (Commerce is afforded "wide discretion" to value factors of production in nonmarket economies). We hold that Commerce did not abuse its discretion by declining to use the FAO report to value whole pangas fish in the fourth administrative review.
QVD and CFA timely filed several pieces of information with Commerce both before and after the publication of the preliminary results, and both parties discussed the appropriate source for valuing whole pangas fish in briefs and at an agency hearing. Most of the submitted information had already been considered in the previous administrative review, where Commerce had the benefit of extensive comments by the parties to this case. In earlier reviews, Commerce had been comfortable using certain of Gachihata's financial statements as the best available information for valuing whole pangas fish. Those sources were well vetted.
However, at a hearing two weeks before the final results were due, interested parties continued to raise concerns about the contemporaneity of the FY2000-01 statement and the overall reliability of the
In conducting an administrative review, Commerce must consider all information timely filed by interested parties, 19 C.F.R. § 351.301(c)(3)(ii), and it must provide all parties with an opportunity to comment on that information, 19 U.S.C. § 1677m(g). The agency also must give parties an opportunity to submit case briefs and rebuttal briefs addressing any information submitted, 19 C.F.R. § 351.309, and it must hold a hearing when requested, 19 U.S.C. § 1675(e); 19 C.F.R. § 351.310(d)(1). While Commerce may not have been legally required to solicit comments from interested parties on information that the agency itself placed on the record, it was permissible for Commerce to decline to rely on that information given the agency's lack of a full opportunity to assess the reliability of that evidence in light of the parties' comments.
Moreover, QVD is in an awkward position to argue that Commerce abused its discretion by not relying on evidence that QVD itself failed to introduce into the record. The FAO report is dated 2007. There is no suggestion that the report was not publicly available in late 2008, when Commerce invited QVD and other interested parties to submit relevant factual information for valuing factors of production. Although Commerce has authority to place documents in the administrative record that it deems relevant, "the burden of creating an adequate record lies with [interested parties] and not with Commerce." Tianjin Mach. Imp. & Exp. Corp. v. United States, 806 F.Supp. 1008, 1015 (Ct. Int'l Trade 1992); see NTN Bearing Corp. of Am. v. United States, 997 F.2d 1453, 1458-59 (Fed.Cir.1993). QVD clearly would not be in a position to contest Commerce's refusal to consider the FAO report had QVD itself attempted to introduce it into the record a week before the deadline for final results. See 19 C.F.R. § 351.302(d) (stating that Commerce will not consider untimely filed materials); see also id. § 351.104(a)(2). QVD's argument therefore turns on the fact that it was Commerce, not QVD, that placed the FAO report in the record in the first instance. But because Commerce could have declined to place the report in the record due to concerns about the short time available to assess the report's reliability, we see no unfairness or impropriety in Commerce's decision to submit the report into the record for the parties' comments and then, based on those same concerns, to decide not to rely on the report. In this setting, the difference between deciding not to place the report in the record and deciding not to rely on it after placing it in the record is a technical distinction that should not be given dispositive weight.
QVD points out that in a new shipper review concluded in June 2009, Commerce found the FAO study to be a more reliable data source for whole pangas pricing than the FY2000-01 Gachihata financial statement. Judicial review of antidumping duty administrative proceedings is normally limited to the record before the agency in the particular review proceeding at issue and does not extend to subsequent
For the final results of the subsequent new shipper review, Commerce chose to rely on the FAO report to value whole pangas fish. In its decision in that review, Commerce emphasized that the parties had ample time to submit comments on the FAO report and that the agency had fully considered all the comments. In contrast to the fourth administrative review, interested parties were given several weeks to submit comments, case briefs, and rebuttal addressing the FAO report. Commerce's decision to use the FAO report as the best available information for the subsequent new shipper review was based on all the record evidence in that proceeding, which included those comments and arguments regarding the integrity and relevance of the FAO data. The fact that Commerce later determined, after more than three months of study, that the FAO report provided better information than the FY2000-01 Gachihata financial statement does not undermine Commerce's decision not to rely on that report in the fourth administrative review, when Commerce had only a week to consider it.
Apart from the exclusion of the FAO report, QVD argues that Commerce's decision to rely on the FY2000-01 Gachihata financial statement rather than on the more contemporaneous FY2006-07 statement was not supported by substantial evidence. QVD asserts that Commerce's reasons for rejecting the FY2006-07 statement are not rationally related to the whole pangas fish price quoted in that statement. In response, Commerce points out that new evidence justified its decision to conduct a reevaluation of the data in the FY2006-07 statement and ultimately to reject that statement as a source of surrogate valuation information.
We agree with the Court of International Trade that it was reasonable for Commerce to reexamine the FY2006-07 financial statement in its entirety in light of the Gachihata Directors' Report for that year, which made clear that Gachihata was in serious financial trouble during that period. For example, Gachihata's sales of pangas fish were listed as 115.5 metric tons in FY2000-01 but only 6.0 metric tons in FY2006-07, a drop of nearly 95 percent. The Directors' Report described the company as in "crisis" and in "dire need of fund[s] to restart its production" at the close of FY2006-07. The Report further noted that the company "suffered heavily due to operati[ng] loss[es]." Gachihata's situation was so bleak that the Board of Directors offered to resign for the good of the company. As the Court of International Trade pointed out, in light of those facts Commerce might have found it difficult to defend the reasonableness of relying on the FY2006-07 statement.
In the memorandum explaining its decision, Commerce raised specific concerns with the FY2006-07 Gachihata financial statement as to (1) the continued deterioration of Gachihata's financial condition;
QVD argues that a 2007 report from Gachihata's auditors allays any concern that the whole pangas price quoted in the FY2006-07 financial statement was inaccurate. The auditors' report states that Gachihata maintained proper books and that the company's balance sheet and income statement were in agreement with account books. Those statements, however, do not necessarily dispel doubts about the reliability of financial statements generated by a company on the verge of financial collapse. QVD also cites to a document entitled "Notes to the Accounts" for FY2006-07. That document states that "[t]he amount of revenue was fixed at fair value" and "[t]he quantity of goods delivered were measured reliably." QVD's position appears to be that Commerce should have trusted the amounts quoted in the FY2006-07 statement because Gachihata stated that its accounting was accurate. However, there does not appear to be any reason for Commerce to have placed more trust in the "Notes to the Accounts" document than the FY2006-07 financial statement itself.
Finally, QVD contends that Commerce should not have inflated the whole pangas fish price from 68 Tk/kg to 98 Tk/kg, because other information in the record suggested that the price had dropped in absolute terms between FY2000-01 and the period of review. In addition to the FAO report and Gachihata's FY2006-07 financial statement, discussed above, that information consisted of Gachihata's financial statements from FY2001-02, FY2002-03, and FY2003-04; a 2004 study by the Asian Development Bank ("ADB study"); and a 2007 price quote from Bangladesh Catfish Ltd. ("BCL quote").
Commerce found each of those sources of information less reliable than the FY2000-01 Gachihata financial statement. We have already discussed the FAO report and Gachihata's FY2006-07 financial statement. QVD contends that even if the FAO report and the FY2006-07 Gachihata financial statement are not the best available information for valuing whole pangas fish, those sources should have been considered when Commerce determined whether to inflate the value from the FY2000-01 statement. The problem with QVD's position is that Commerce not only found that those sources were not the best available information but also that it would
With respect to Gachihata's financial statements from FY2001-02, FY2002-03, and FY2003-04, Commerce had found those statements unreliable in previous administrative reviews because the independent auditor's notes to those statements called into question Gachihata's internal control procedures and valuation of assets. Commerce determined that the auditor's notes were sufficient to cast reasonable doubt on the reliability and accuracy of the overall statements. QVD did not challenge those findings of unreliability in the fourth administrative review. Instead, in its rebuttal brief before Commerce, QVD used the absence of the same comments by the auditor to support its argument for the reliability of the FY2006-07 statement.
In the third administrative review, Commerce refused to rely on the BCL price quote because the record contained no information about how or when that quote was obtained or whether the company that supplied the quote produced pangas fish in commercial volumes. QVD does not suggest that any of those problems were remedied when it resubmitted the same price quote for the record of the fourth administrative review. Similarly, Commerce declined to rely on the ADB study in the second and third administrative reviews because that study did not provide a market price reflecting actual pangas fish transactions and because there was no evidence of the timeframe of the survey that led to the price quoted in the study. QVD did not place additional information in the record in the fourth administrative review that would have prompted Commerce to reconsider the reliability of the ADB study.
The dissent reproduces a table from QVD's brief that contains each of the potential sources for pangas fish prices placed in the record. The dissent argues that the inflation-adjusted price of 98 Tk/kg is unreasonable because it is outside the range of prices in those data sources. As we have discussed, however, Commerce had reasonable grounds for not relying on any of those data sources other than the FY2000-01 Gachihata financial statement to value pangas fish. Two of the sources listed in the table (the FAO study and the FY2006-07 Gachihata financial statement) are sources that we have already examined in detail. As for the remaining sources, Commerce did not consider any of those sources for reasons that it explained in previous administrative reviews, and QVD has not challenged those explanations in this proceeding.
In sum, QVD has not come forward with any evidence that undermines Commerce's findings that all of the possible data sources in the record for pricing whole pangas fish, with the exception of the FY2000-01 Gachihata financial statement, were not reliable sources to use in the fourth administrative review. This was not a situation in which Commerce ignored several reliable sources in making its determination. Commerce had substantial evidentiary support for its finding that only the FY2000-01 statement was reliable enough to serve as the basis for valuing whole pangas fish in the fourth administrative review. Having selected the data from FY2000-01, Commerce reasonably chose to inflate that value using a standard methodology that QVD has not challenged. We therefore agree with the Court of International Trade that Commerce's valuation of whole pangas fish is supported by substantial evidence.
QVD also appeals from the trial court's rejection of its challenge to Commerce's financial ratio calculations. QVD notes that Commerce's regulations prohibit double-counting of adjustments made to normal value in antidumping proceedings, and it contends that its objection to that double-counting should have been sustained as a "ministerial error" even though QVD did not raise that issue until after the final results were announced. Although Commerce refused to address the double-counting claim on the ground that it was not a "ministerial error" and therefore could not be raised after the final results were announced, QVD argues that the error is cognizable as a "ministerial error" and should be corrected.
The term "ministerial error" is defined to mean "errors in addition, subtraction, or other arithmetic function, clerical errors resulting from inaccurate copying, duplication, or the like, and any other type of unintentional error which the administering authority considers ministerial." 19 U.S.C. § 1675(h); see 19 C.F.R. § 351.224(f). Commerce concluded that "QVD's arguments with respect to [the general expense ratio] are methodological in nature because QVD essentially disagrees with the Department's assignment of certain expenses to the calculation of the ... ratio." The trial court viewed that statement as clear evidence that Commerce had intended to characterize the disputed expenses as "general expenses." We agree. QVD's argument is not that Commerce inadvertently included certain items in its list of general expenses, but that Commerce was wrong to think that it could do so. Because the error alleged by QVD is not an arithmetic or clerical error or similar inadvertent mistake, it does not fall within the statutory definition of "ministerial error."
Even if the error alleged by QVD were ministerial in nature, Commerce's failure to correct it would not constitute an abuse of discretion because QVD did not raise the issue in a timely fashion. In Dorbest Ltd. v. United States, we held that a ministerial error made by Commerce that was reflected in its preliminary antidumping duty determination need not be corrected when no interested party pointed out the error in a timely manner. 604 F.3d at 1376-77. In the present case, any alleged error in Commerce's calculation of the general expense ratio in the final results was necessarily present in the preliminary results, because Commerce made no change to the financial ratio calculations. Commerce's regulations specify the procedure by which ministerial errors in the preliminary results of antidumping duty administrative reviews may be challenged. Commerce discloses any calculations made in the preliminary results to interested parties, 19 C.F.R. § 351.224(b), and interested parties must point out any ministerial errors in their case briefs, id. § 351.224(c)(1); see also id. § 351.309(c)(2) ("The case brief must present all arguments that continue in the submitter's view to be relevant to the Secretary's final determination or final results...."). In its case brief following the preliminary results, QVD alleged certain ministerial errors, but it failed to allege any error concerning financial ratios. Commerce's refusal to make a ministerial correction is not reversible error when the alleged mistake was discoverable during earlier proceedings but was not pointed out to Commerce during the time period specified by regulation. Dorbest, 604 F.3d at 1377. For that reason as well, we conclude that the trial court properly rejected QVD's challenge to the financial ratio calculations.
I respectfully dissent from the court's endorsement of Commerce's methodology. While any reasonable valuation methodology may be supported, one that is blatantly unreasonable warrants judicial alteration, not automatic affirmance. Commerce selected the Gachihata 2000/2001 pangas sales price of 68 takas per kilogram, adjusted to 98 takas for inflation, as the surrogate fair value for pangas whole fish for the period covered by the fourth administrative review, from August 1, 2006 to July 31, 2007. Not only is the 2000/2001 figure devoid of any evidentiary support as the 2006/2007 value, even by inference, but unchallenged evidence shows that it is conspicuously outside of any reasonable range of the actual value for the period of review.
Commerce selected the nation of Bangladesh, and the Bangladesh company Gachihata Aquaculture Farms, Ltd., as the surrogate for determining the fair sales value of pangas fish in Vietnam. There was no contradiction to the record that Gachihata had not sold pangas fish at 68 takas since 2001. For the third administrative review of the QVD dumping order, for the period August 1, 2005 to July 31, 2006, Commerce had used the Gachihata value of 45 takas for FY 2006/2007. Now, for the fourth administrative review, Commerce used the 2000/2001 price of 68 takas, which Commerce inflated to 98 takas. The following table, from QVD's brief, summarizes the pangas prices in the Commerce record:
Source Whole Fish Value2000/2001 Gachihata Annual Report 68 takas per kilogram 2001/2002 Gachihata Annual Report 50 takas per kilogram 2002/2003 Gachihata Annual Report 49.7 takas per kilogram 2003/2004 Gachihata Annual Report 48 takas per kilogram 2004 ADB Study 55 takas per kilogram 2006/2007 Gachihata Annual Report 45 takas per kilogram 2007 FAO Study 42 takas per kilogram 2007 Price Quote from Bangladesh Catfish Co., Ltd. 40 takas per kilogram
QVD Br. 10.
Even if Commerce were now concerned about the fidelity of the Gachihata data, for the government states that Gachihata was subject to economic difficulties, the record contains no support for selecting Gachihata's 2000/2001 sales price to measure market value in 2007. QVD proposes that at least two other values are more reasonable than the 2000/2001 value selected by Commerce. QVD points to the value of 45 takas from the 2006/2007 financial statement of Gachihata as used by Commerce in its third administrative review; or the 2007 FAO Study (the United Nations Food and Agriculture Organization) at 42 takas. The FAO Study, entitled "Economics of Aquaculture Feeding Practices in Selected Asian Countries," contains a section on farming pangasius in Bangladesh that states that the farmgate price was approximately 42 takas, derived from research of sixty separate fish farms. Commerce's belated rejection of the FAO Study is not credibly explained, for Commerce had placed it in the record and invited and received commentary from all concerned.
Substantial evidence does not support 98 takas as the surrogate fair value for the fourth administrative review of imported pangas whole fish. From my colleagues' endorsement of this inappropriate process, I respectfully dissent.