PAM, S.P.A. v. United States

Opinion & Order CARMAN, Judge. This matter is before this Court on motion for judgment upon the agency record. Plaintiffs, PAM, S.p.A. (“PAM”) and JCM, Ltd. (“JCM”), 1 appeal the final results of the sixth administrative review of the anti-dumping order on certain pasta from Italy. See Certain Pasta from Italy and Determination not to Revoke in Part, 69 Fed. Reg. 6,255, 6,257 (Dep’t Commerce Feb. 10, 2004) (notice of final results and determination not to revoke in part) (“Final Results ”). Specifically, Plaintiffs challenge Commerce’s decision to apply a dumping margin of 45.49 percent to PAM, based on total adverse facts available. As discussed below, this Court finds that Commerce’s decision to apply adverse facts available to PAM is supported by *1363 substantial evidence and is otherwise in accordance with law. However, this Court finds that Commerce’s application of the 45.49 percent rate to PAM is not supported by substantial evidence. This Court therefore grants in part Plaintiffs’ motion for judgment on the agency record and remands the Final Results to Commerce for further proceedings consistent with this Opinion and Order. Procedural History Commerce published the Final Results of the sixth administrative review of the antidumping order on certain pasta from Italy in February 2003. Plaintiffs timely appealed the Final Results on two grounds. First, Plaintiffs argued that the results of the administrative review were void as to PAM because the domestic industry petitioners 2 failed to provide notice to PAM that they requested an administrative review. 3 Second, Plaintiffs argued that Commerce was not justified in applying adverse facts available to PAM. This Court in PAM, S.p.A. v. United States, 29 CIT -, 395 F.Supp.2d 1337, 1345 (2005), rev’d, 463 F.3d 1345 (Fed.Cir.2006) (“PAM I ”), held that the Final Results were void ab initio as to PAM due to the petitioners’ failure to provide notice of the administrative review. On appeal, the Court of Appeals for the Federal Circuit (“CAFC”) reversed, holding that PAM was not prejudiced by the lack of notice and remanding the case to this Court for further proceedings on the merits. PAM, S.p.A. v. United States, 463 F.3d 1345, 1349 (Fed.Cir.2006). Plaintiffs now challenge Commerce’s application of adverse facts available to PAM. Factual Background 4 On August 27, 2002, Commerce initiated the sixth administrative review of the anti-dumping order on certain pasta from Italy, covering the period of review July 1, 2001, through June 30, 2002. Certain Pasta from Italy, 67 Fed.Reg. 55,000, 55,002 (Dep’t Commerce Aug. 27, 2002) (notice of initiation). Soon thereafter, Commerce sent out questionnaires to the respondents, including PAM. The questionnaires requested sales and production information from the respondents and set a deadline of October 7, 2002 for responses to be filed. (Letter from James Terpstra to Salvatore Lubrano 3 (Aug. 29, 2002), Pub. R. Doc. 19.) PAM notified Commerce via letter that Petitioners had not properly served PAM with their requests for review, and as a result, PAM was not notified of Commerce’s administrative review in a timely manner. (Letter from David J. Craven to the Honorable Donald Evans 1 (Sept. 3, 2002), Pub. R. Doc. 33.) PAM requested and was granted a series of extensions of time to file responses to the initial questionnaire and two supplemental questionnaires. (See Letters granting extensions to PAM, Pub. Docs. 53, 179, 202 & 245.) PAM submitted its completed questionnaires by the extended deadlines. At the time Commerce conducted a verification of PAM’s questionnaire responses, the agency discovered that PAM had not *1364 reported a large number of sales made in the home market. (Verification of the Sales Response of PAM in the 01/02 Administrative Review of the Antidumping Duty Order of Certain Pasta from Italy 3 (Jul. 28, 2003) (“PAM Verification Rep.”), Pub. R. Doc. 305.) When asked about the missing sales, PAM explained that some of the sales were omitted due to a computer programming error and that another portion of the sales were not reported because they were made outside the ordinary course of trade. (Id. at 17-18.) Regarding the computer programming error, PAM explained that it designated invoices with one of three codes: “FT, for regular invoices; FA, for invoices issued for merchandise to be shipped; and FP, for invoices that are issued by PAM for merchandise sold from a non-PAM warehouse.” 5 (Id. at 17.) To extract sales data from PAM’s accounting system, PAM used the same computer programs in this administrative review as it had used in a prior administrative review. Because PAM had not used FP invoices in the home market during the period covered by the prior administrative review, the computer programs were not coded to extract invoices with the FP designation. As a result, the external warehouse sales were omitted from PAM’s home market sales database. (Id.) The other portion of unreported invoices reflected sales to a single customer, AG. E.A., “an Italian governmental] agency [that] supplies pasta to charitable organizations in Italy.” 6 (Id. at 18.) PAM explained that it did not report the sales to AG.E.A. because they were made outside the ordinary course of trade. 7 PAM believed the AG.E.A. sales to be outside the ordinary course of trade because the pasta sold to AG.E.A. was not for commercial use and was produced in larger quantities at lower cost than pasta sold commercially. (Id.) Altogether, Commerce estimated that PAM failed to report almost two-thirds of its home market sales. 8 (Issues & Decision Mem. 13, 18.) Commerce determined that its “ability to calculate PAM’s dumping margin using the data reported by PAM [was] severely compromised” because “[s]uch a small sample may not provide a reasonable approximation of PAM’s actual sales practice in the home market”. Certain Pasta from Italy, 68 Fed.Reg. 47,020, 47,026 (Aug. 7, 2003) (notice of prelim *1365 inary results of the sixth administrative review) (“Preliminary Results ”). Commerce further determined that PAM failed to cooperate with Commerce’s requests for information, warranting application of adverse facts available. (Issues & Decision Mem. for the Final Results of the Sixth Antidumping Duty Administrative Rev. of Certain Pasta from Italy 18-14 (Feb. 3, 2004) (“Issues & Decision Mem.”), Nonconfidential App. of PAM for J. upon an Agency R Ex. 1.) Commerce determined that PAM [did] not act[ ] to the best of its ability in failing to report approximately two-thirds of its home market sales in this review, because, (1) [Commerce] issued clear instructions requiring this information in its initial questionnaire; (2) PAM had the opportunity to provide the information in responding to two supplemental questionnaires, all of the deadlines of which were extended at PAM’s request by [Commerce]; (3) [Commerce] had instructed PAM to report all sales, including those claimed to be outside the ordinary course of trade, and (4) PAM has successfully participated in previous reviews. Additionally, the fact that [Commerce] was readily able to obtain general information regarding the existence of such sales at verification supports our determination that PAM did not act to the best of its ability in reporting its home market sales. (Id. at 18.) Commerce assigned PAM a dumping margin of 45.49 percent ad valo-rem based on adverse facts available. 9 Final Results, 69 Fed.Reg. at 6,257. Jurisdiction This Court has exclusive jurisdiction to review final results of administrative reviews pursuant to 28 U.S.C. § 1581(c) (2000). 10 Standard of Review When reviewing Commerce’s final results of administrative reviews the United States Court of International Trade must sustain Commerce’s determinations, findings, or conclusions unless they are “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(l)(B)(i) (2000). “More specifically, when reviewing whether Commerce’s actions are unsupported by substantial evidence, the Court assesses whether the agency action is ‘unreasonable’ given the record as a whole.” Mittal Steel Galati S.A v. United States, 31 CIT -, 491 F.Supp.2d 1273, 1275 (2007) (May 14, 2007) (citing Nippon Steel *1366 Corp. v. United States, 458 F.3d 1345, 1350-51 (Fed.Cir.2006)). Discussion I. Commerce’s Use of Adverse Facts Available. A. Parties’ Contentions Plaintiffs 11 argue that Commerce erred in applying adverse facts available to PAM because the omissions in PAM’s questionnaire responses “were not of a nature to justify the application of adverse facts available.” (Pr. Br. of PI. for J. upon the Agency R. Pursuant to R. 56.2 (“PAM’s Br.”) 24.) Plaintiffs explain that the omissions stemmed from two errors. First, “PAM omitted the AG.E.A. sales on the advice of counsel” because the sales were outside the normal course of trade. (Id.) PAM contends “that the failure to report the AG.E.A sales was harmless error” because Commerce “concluded] that these sales were not for commercial purposes, and were to a government agency for purposes of charity.” (Id. at 30 (citing PAM Verification Rep. 3).) The second error, omitting the external warehouse sales, was the result of a “minor clerical error in the computer program” used to extract sales information from the company’s database. 12 (Id. at 24.) PAM blames its failure to discover this error on prior counsel. PAM contends that prior counsel “failed to respond to the portion of [Commerce’s] questionnaire requiring PAM to submit a complete reconciliation of its reported sales to its accounting system.” 13 (Id. at 30.) PAM argues that the company would have detected the omission of the external warehouse sales had a reconciliation occurred, but “prior counsel never called upon PAM to prepare any reconciliations, and never reviewed such issues with PAM.” (Id. at 31.) PAM argues that “its errors were attributable to counsel rather than to PAM itself, as the company had done everything within its capacity to comply with the requirements of [Commerce] as relayed by its attorney.” 14 (Id.) Therefore, PAM “submits that the application of adverse facts available is unlawful because PAM *1367 used its best efforts to answer to questionnaire.” (Id. at 33.) In contrast, the Government and Defendants-Intervenor argue that Commerce properly applied adverse facts available to PAM. 15 First, the Government stresses that Commerce appropriately resorted to facts otherwise available in making its determination because PAM failed to report around two-thirds of its home market sales. (Def.’s Mem. in Opp’n to Pis.’ R. 56.2 Mots, for J. upon the Agency R. (“Def.’s Mem.”) 20.) Because of the significant omissions from PAM’s home market sales data, “Commerce was unable to verify whether PAM’s reported home market sales reflected the entire universe of its sales.” (Id. at 22-23 (emphasis omitted).) As a result, “Commerce had no choice but to resort to total facts available as a result of PAM’s reporting failure.” (Id. at 24-25.) Second, the Government argues that Commerce was justified in applying adverse inferences to the facts available because PAM did not cooperate to the best of its ability in responding to Commerce’s questionnaires. (Id. at 25.) The Government notes that PAM, as “an experienced producer” that “had participated in previous reviews of the antidumping order at issue,” was aware of the Commerce’s reporting requirements. (Id. at 26-27.) The Government also contends that PAM had the ability to comply with Commerce’s requests for information, as evidenced by “Commerce’s discovery of information at verification regarding the existence of the unreported sales.” (Id. at 27.) The Government contends that PAM’s excuses for failing to report two-thirds of its home market sales do not establish that PAM acted to the best of its ability. The Government submits that PAM’s argument that it did not need to report its AG.E.A. sales as they were outside the ordinary course of trade runs counter to Commerce’s instructions to report “all sales, including those sales which you believe are outside the ordinary course of trade.” (Id. (quoting Questionnaire Instructions G-7; Pub. R. Doc. 31).) Likewise, the Government contends, the inadvertent coding error that led to the omission of PAM’s external warehouse sales is not a valid excuse; “inadvertent errors are insufficient to demonstrate that one is acting to the best of its ability.” (Id. at 28.) Furthermore, the Government contends that PAM cannot shield itself by blaming prior counsel for the deficiencies in PAM’s submissions because PAM certified the accuracy and completeness of its responses. (Id.) Therefore, the Government declares that Commerce’s decision to apply adverse facts available is supported by substantial evidence on the record and is in accordance with law. B. Analysis 19 U.S.C. § 1677e governs the use of facts available. Section 1677e provides: (a) In general If— if: ‡ # ‡ (2) an interested party or any other person— * * * * * * (B) fails to provide [requested] information by the deadline for submission ... [or] H: ‡ ‡ ‡ ‡ (D) provides such information but the information cannot be verified *1368 [Commerce] shall ... use the facts otherwise available in reaching the applicable determination. (b) Adverse inferences If [Commerce] ... finds that an interested party has failed to cooperate by not acting to the best of its ability to comply for a request for information ... [Commerce] may use an inference that is adverse to the interests of that party in selecting from among the facts otherwise available. First, this Court finds that Commerce’s decision to resort to facts available is supported by substantial evidence on the record and is otherwise in accordance with law. Commerce is justified in applying facts available when a respondent fails to provide Commerce with information that it needs to make a determination, or the information the respondent provides cannot be verified. 19 U.S.C. § 1677e(a)(2)(B) & (D) (2000); see also Nippon Steel, 337 F.3d at 1381 (“There mere failure of a respondent to furnish the requested information — for any reason — requires Commerce to resort to other sources of information to complete the factual record on which it makes its determination.”). Commerce determined that PAM’s home market sales could not be verified because PAM failed to report roughly two-thirds of its home market sales. 16 Due to the significant omissions, Commerce’s “ability to calculate PAM’s dumping margin using the data reported by PAM has been severely compromised. Such a small sample may not provide a reasonable approximation of PAM’s actual sales practice in the home market.” Preliminary Results, 68 Fed.Reg. at 47,026. The Government explains that “a complete record of sales is indispensable for an accurate measure of dumping.” (Def.’s Mem. 23 (emphasis added)) Allowing respondents to report only a portion of their sales might lead to manipulation of dumping margins as respondent’s “selectively report sales that are the most favorable to their dumping margin.” (IcL) Second, this Court finds that Commerce’s decision to apply adverse inferences to the facts available is similarly supported by substantial evidence on the record and is otherwise in accordance with law. Commerce may apply adverse facts available if it “finds that [a respondent] has failed to cooperate by not acting to the best of its ability to comply with a request for information from [Commerce].” 19 U.S.C. § 1677e(b). As the CAFC has stated: Whether a respondent has lived up to [the requirement to act to the best of its ability] is assessed by determining whether [the] respondent has put forth its maximum effort to provide Commerce with full and complete answers to all inquiries in an investigation. While that standard does not require perfection, it does not condone inattentiveness, carelessness, or inadequate record keeping. NSK, Ltd. v. United States, 481 F.3d 1355, 1361 (Fed.Cir.2007) (internal citations and quotations omitted) (first bracket added; second bracket in original). Here, Commerce determined that PAM had not put forth its maximum effort to *1369 provide Commerce with the information it needed to determine PAM’s dumping margin because PAM neglected to report two-thirds of its home market sales. Commerce based its determination on the following: (1) Commerce “issued clear instructions” regarding the information it requested from PAM; (2) PAM was provided several opportunities to report its complete home market sales prior to verification; (3) PAM had “successfully participated in previous administrative reviews”; and (4) Commerce was “readily able to obtain general information regarding the existence of [the omitted] sales at verification.” (Issues & Decision Mem. 18.) PAM’s attempts to explain the omitted home market sales information are not persuasive. First, PAM contends that it did not provide AG.E.A. sales because they were made outside the ordinary course of business. Yet, Commerce explicitly asked for all home market sales, including those made outside the ordinary course of business. (Questionnaire Instructions G~7, Pub. R. Doc. 31.) Regardless of whether the AG.E.A. sales were in fact made outside the ordinary course of business, Commerce was clear that PAM was required to report them. 17 PAM failed to cooperate by not complying with Commerce’s explicit instructions. Second, PAM contends that the omission of the external warehouse sales was the result of a minor clerical error in the computer program used to extract the home market sales information from PAM’s accounting system. However, PAM was provided with several opportunities to report its complete home market sales prior to verification. Specifically, Commerce issued an initial questionnaire and two supplemental questionnaires to PAM and granted PAM’s requests for extensions of time to respond to each of the questionnaires. Further, Commerce was “readily able” to detect at verification that PAM had omitted a large amount of home market sales. (Issues & Decision Mem. 18.) Because the omitted sales were so readily discoverable, this Court finds reasonable Commerce’s position that PAM did not act to the best of its ability in reporting the sales to Commerce. “While that standard [of acting to the best of one’s ability] does not require perfection, it does not condone inattentiveness, carelessness, or inadequate record keeping.” NSK, 481 F.3d at 1361. Finally, PAM contends that its omissions should be excused as they are attributable to prior counsel, and not to PAM itself. This Court reiterates the answer Commerce gave when PAM raised this argument at the administrative level: “PAM cannot blame the acknowledged deficiencies in its responses on previous counsel, its software house, or its consultants, because PAM certified the accuracy of these submissions.” ■ (Issue & Decision Mem. 16 (emphasis added).) Commerce’s regulations require a representative of the company participating in an administrative review or investigation to certify that he has read the attached submission, and that to the best of his knowledge, the information contained in the submission is complete and accurate. See 19 C.F.R. § 351.303(g)(1) (2006). Furthermore, this argument is particularly unpersuasive when made by a respondent like PAM, that has successfully participated in previ *1370 ous administrative reviews and thus should presumably be aware of Commerce’s reporting requirements. In addition, it makes sense not to distinguish between a respondent’s and its counsel’s actions because counsel acts on behalf of the respondent. If this Court were to sanction a “bad counsel” defense, it might create an incentive to hire ineffective counsel. In sum, this Court affirms Commerce’s decision to apply adverse facts available to PAM. II. Commerce’s Application of the 45.49 Percent Dumping Margin to PAM. A. Parties’ Contentions PAM contends that the 45.49 percent dumping margin Commerce applied to PAM “was corroborated for Barilla, but it was never corroborated for PAM.” (Reply Br. of PAM 15.) Instead, PAM argues, the rate is punitive and not reasonably related to PAM’s activity. (PAM’s Br. 34.) PAM contends that the rate is punitive because PAM has “established itself as an exporter with a [dumping] margin in the range of 4 to 5 percent.” 18 (Id. at 35.) In fact, PAM offers, if faets-available margins are excluded, the overall dumping margins for all Italian exporters across all periods of review “have been in low single digits.” (Id.) PAM also argues that the 45.49 percent rate bears no relation to PAM’s own activity during the period of review. PAM insists that Commerce need not have resorted to applying to PAM a dumping margin calculated for another respondent, Barilla, because PAM had placed sales and cost information on the record sufficient for Commerce to calculate a dumping margin specific to PAM. (Id. at 36.) PAM contends that Commerce could have constructed an adverse facts available rate for PAM given the information PAM had placed on the record: a verified U.S. sales database, the (incomplete) home market sales data, home market sales adjustments, and PAM’s cost of production data. (Id.) PAM suggests that Commerce could have come up with a margin rationally related to PAM. One possibility of an adverse facts available rate for PAM would be “to take the average of PAM’s previous rates, namely, 4.57%, and then double the rate, yielding 9.14%.” (Id. at 36-37.) The Government contends that Commerce’s application of a 45.49 percent adverse facts available dumping margin is supported by substantial evidence on the record. First, the Government notes that Commerce has discretion “to select among an enumeration of secondary sources as a basis for its adverse factual inferences.” (Def.’s Mem. 31) (quoting F.LLI De Ceceo Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir.2000).) The Government contends that “Commerce’s selection of the highest calculated rate from the proceeding to deter future uncooperative behavior by PAM is a reasonable exercise of its discretion and is fully supported by precedent.” (Id. at 32.) “Further,” the Government argues, “Commerce corroborated PAM’s selected rate with contemporaneous information” by “examining individual transactions of other respondents during the period of review.” (Id. at 33.) The Government concludes that “[b]ecause PAM did not cooperate to the best of its ability, PAM is [now] prohibited from controlling the selection of a *1371 more favorable adverse facts available rate.” {Id. at 34.) B. Analysis In the case of an uncooperative respondent, Commerce has “discretion to choose which sources and facts it will rely on to support an adverse inference.” F.LLI De Ceceo, 216 F.3d at 1032; accord Ta Chen Stainless Steel Pipe, Inc. v. United States, 298 F.3d 1330, 1338-39 (Fed.Cir.2002). Commerce may base an adverse facts available rate on information derived from: (1) the petition, (2) a final determination in [an anti-dumping duty] investigation, (3) any previous [administrative] review ..., or (4) any other information placed on the record. 19 U.S.C. § 1677e(b). However, when Commerce “relies on secondary information rather than information obtained in the course of an investigation or review” as the basis for adverse facts available, Commerce is required to “corroborate [, to the extent practicable,] that information from independent sources that are reasonably at [its] disposal.” 19 U.S.C. § 1677e(c). The Court of Appeals for the Federal Circuit explained that Congress included the corroboration requirement to ensure that an adverse facts available rate is “a reasonably accurate estimate of the respondent’s actual rate, albeit with some built-in increase as a deterrent to non-compliance.” F.LLI De Ceceo, 216 F.3d at 1032. It is not within Commerce’s discretion “to select unreasonably high rates with no relationship to the respondent’s actual dumping margin.” Id. “The statute does not prescribe any methodology for corroborating secondary information,” Mittal Steel, 491 F.Supp.2d at 1278, but legislative history explains that “[c]orroborate means that the agencies will satisfy themselves that the secondary information to be used has probative value,” Uruguay Round Agreements Act Statement of Administrative Action, H R. Rep. No. 103-316, vol. 1 at 870 (1994), reprinted in 1994 U.S.C.C.A.N. 3773, 4199. “Commerce assesses the probative value of secondary information by examining the reliability and relevance of the information to be used.” Mittal Steel, 491 F.Supp.2d at 1278 (emphasis added); (accord Issues & Decision Mem. 22.) This Court finds Commerce’s determination that it adequately corroborated the 45.49 percent margin to be unsupported by substantial evidence on the record. Commerce attempted to corroborate the margin by establishing that the “45.49 percent rate is still relevant to the level of dumping during the [period of review].” (Issues & Decision Mem. 22.) Commerce did so by finding “individual sales transactions of other respondents ... at or above 45.49 percent.” {Id.) However, Commerce’s factual finding cannot be generalized absent information regarding whether those individual transactions represent a significant portion of the transactions that occurred during the period of review. See World Finer Foods, Inc. v. United States, 24 CIT 541, 547 (2000) {“World Finer Foods /”) (pointing out the same shortcoming in Commerce’s corroboration of an adverse facts available rate). During the sixth administrative review, overall dumping margins for cooperating respondents ranged from 0.24 percent to 7.23 percent. See Final Results, 69 Fed.Reg. at 6,257. Therefore, to this Court it appears that the transactions at or above 45.49 percent are aberrant. Even if Commerce was able to find a few transactions with dumping margins at or above 45.49 percent, it is not “reasonable to conclude that the 45.49 percent rate is still rele *1372 vant to the level of dumping during the [period of review]” if high-margin transactions are the exception rather than the rule. (Issues & Decision Mem. 22.) Furthermore, Commerce did not explain how other respondents’ transaction specific margins were related PAM’s dumping activity during the period of review. It is not sufficient for Commerce to establish that the margin it selects is related to the overall level of dumping during the period of review. Rather, Commerce must select an adverse facts available margin that is a “reasonably accurate estimate of the respondent’s actual rate, albeit with some built-in increase as a deterrent to non-compliance.” F.LLI De Ceceo, 216 F.3d at 1032 (emphasis added); accord Shandong Huarong Gen. Group Corp. v. United States, 31 CIT -, 2007 WL 54067, *2 (Jan. 9, 2007) (corroboration “requires that an assigned rate relate to the company to which it is assigned”) (citation omitted). For instance, courts have affirmed Commerce’s selection of adverse facts available margins where Commerce corroborated the margin with respect to a respondent’s own transaction specific margins, either from the period of review at issue, Ta Chen, 298 F.3d at 1339-40, or a previous period of review, Mittal Steel, 491 F.Supp.2d at 1279. Similarly, Commerce adequately ' corroborated where it compared the adverse facts available margin selected to the highest previously calculated margin for that respondent. Shandong Huarong, 2007 WL 54067, *3. The Shandong Huarong court found that the rate was corroborated because “Commerce’s chosen rate [was] not dramatically different from [the] rates that the [respondent] previously received.” Id. at *5. Conversely, the courts have remanded for lack of corroboration adverse facts available rates where Commerce did not establish a link between the respondent and the rate selected. See, e.g., F.LLI De Ceceo, 216 F.3d at 1032; Gerber Food (Yunnan) Co., Ltd. v. United States, 31 CIT -, 491 F.Supp.2d 1326, 1345 (2007); World Finer Foods I, 24 CIT at 547-48; Ferro Union, Inc. v. United States, 23 CIT 178, 44 F.Supp.2d 1310 (1999). Here, the adverse facts available margin at issue has no apparent connection to PAM. Commerce calculated the margin for another uncooperative Italian respondent, Barilla, in the first administrative review. World Finer Foods II, 24 CIT at 1235. Commerce used a Barilla domestic price list as a proxy for Barilla’s home market sales and U.S. import statistics as a proxy for Barilla’s U.S. sales. Id. The court acknowledged that the margin was only Commerce’s “best guess” 19 of Barilla’s margin. Id. at 1238, 120 F.Supp.2d 1131. Yet, the court accepted the rate as adequately corroborated because Commerce lacked alternative sources 'of information, as Barilla entirely neglected to submit sales and cost data to Commerce. World Finer Foods I, 24 CIT at 546 n. 9, 2000 WL 897752. In contrast, the instant record contains PAM-specific information, including dumping margins from previous administrative reviews, a verified U.S. sales database,, an incomplete home market sales database, home market sales adjustments, and cost of production data. PAM’s dumping margins in the two previous administrative reviews in which it par *1373 ticipated were 5.04 percent and 4.10 percent. (Issues & Decision Mem. 28.) While this Court recognizes that previous dumping margins are not necessarily indicative of current margins as “each review stands on its own,” (id.) there is a large disparity between PAM’s previously calculated dumping margins and the adverse facts available rate selected in this administrative review. Absent findings that the rate is a “reasonably accurate estimate” of PAM’s dumping activity during the period of review, the 45.49 percent margin appears “punitive [and] aberrational.” See F.LLI De Ceceo, 216 F.3d at 1032 (not within Commerce’s discretion to select adverse facts available rate that is punitive or aberrational). That said, this Court recognizes that Commerce has “discretion to choose which sources and facts it will rely on to support an adverse inference,” F.LLI De Ceceo, 216 F.3d at 1032, and therefore declines to endorse PAM’s suggested adverse facts available rate of 9.14 percent, calculated as twice the average of PAM’s previous dumping margins. On remand, Commerce must select an adverse facts available rate that is corroborated in accordance with the requirements of 19 U.S.C. § 1677e(c). Conclusion This court affirms in part and remands in part the Final Results. This Court concludes that Commerce’s determination that PAM failed to cooperate by acting to the best of its ability to comply with Commerce’s request for information is supported by substantial evidence on the record and is otherwise in accordance with law. This Court thereby affirms Commerce’s decision to apply adverse facts available to PAM. This Court concludes, however, that the specific adverse facts available rate Commerce selected, 45.49 percent ad valorem, is not supported by substantial evidence on the record because Commerce did not corroborate that the rate is a rationally related to PAM’s actual dumping activity during the period of review. This Court therefore remands the Final Results for Commerce to explain and recalculate an adverse facts available rate that is corroborated in accordance with 19 U.S.C. § 1677e(c). It is hereby ORDERED that Plaintiffs’ motions for judgment on the agency record are partially granted; it is further ORDERED that the Final Results of the sixth administrative review of the anti-dumping duty order on certain pasta from Italy are remanded to Commerce to explain and recalculate the adverse facts available rate for PAM; it is further ORDERED that Commerce shall assign to PAM an adverse facts available rate that satisfies the requirements of 19 U.S.C. § 1677e as discussed in this Opinion and Order, particularly the corroboration requirement set forth therein; it is further ORDERED that Commerce shall file with this Court the remand results no later than September 10, 2007; that Plaintiffs may file comments with this Court indicating whether they are satisfied or dissatisfied with the remand results no later than October 9, 2007; and that Defendant and Defendants-Intervenor may file responses to Plaintiffs’ comments no later than November 5, 2007. SO ORDERED. Footnotes [1] PAM is an Italian pasta manufacturer subject to the antidumping order on certain pasta from Italy. JCM imports pasta from PAM. [2] A number of the petitioners, A Zarega’s and Sons, American Italian Pasta Company, New World Pasta Company, and Dakota Growers Pasta Company, intervened in this action as Defendants-Intervenor. [3] Commerce’s regulations require a party that requests an administrative review to serve notice on "each exporter or producer specified in the request ... by the end of the anniversary month [of the antidumping order] or within ten days of filing the request for the review, whichever is later.” 19 C.F.R. § 351.303(f)(3)(h) (2006). [4] This Court’s prior opinion in this matter contains a recitation of the facts. See PAM I, 395 F.Supp.2d at 1339-40. This Opinion and Order includes only those facts relevant to deciding the motion for judgment on the agency record. [5] This Court refers to the FP-invoice sales as "external warehouse sales” throughout the Opinion and Order. [6] PAM actually made these sales to a private company, which served as an intermediary for AG.E.A. Because the name of the company is confidential, and because AG.E.A. was the ultimate customer, this Court will refer to AG.E.A. as the customer. At verification, PAM explained the commercial relationship between PAM, the private company, and AG. E.A. as follows: AG.E.A. has a supply of French wheat. The AG.E.A. offers to give this wheat to whoever will provide them with the largest quantity of pasta. [The private company] wants the wheat, but does not produce pasta, so [it] contracts] with PAM [to buy pasta from PAM] and then gives the pasta to AG.E.A. in exchange for wheat. (PAM Verification Rep. 18.) [7] Commerce generally excludes sales that are not made "in the usual commercial quantities and in the ordinary course of trade,” 19 U.S.C. § 1677b(a)(l)(B) (2000), because sales made in unusual quantities or outside the ordinary course of trade might not be made at market prices. [8] PAM reported home market sales of approximately ten thousand tons. (Pr. Br. of PL for J. upon the Agency R. Pursuant to R. 56.2 ("PAM’s Br.”) 24.) PAM failed to report approximately twenty-one thousand tons, of which its external warehouse sales accounted for five thousand tons and the AG.E.A. sales accounted for sixteen thousand tons. (Id.) [9] The 45.49 percent margin is the highest dumping margin given to any respondent at any point in the proceedings that has been upheld by the courts. Commerce calculated this dumping margin as an adverse facts available rate for another uncooperative respondent, Barilla, in the first administrative review. (Issues & Decision Mem. 21.) Commerce used a domestic price list of Barilla's as a proxy for home market price and U.S. import statistics as a proxy for U.S. price. World Finer Foods, Inc. v. United States, 24 CIT 1235, 1235 (2000) ("World Finer Foods II ”). Using the proxies, Commerce calculated individual dumping margins for three different categories of pasta. The individual dumping margins were 39.63, 60.06, and 63.36 percent. Id. at 1236, 120 F.Supp.2d 1131. While Commerce selected the highest of the three calculations, 63.36 percent, as Barilla’s adverse facts available rate, the court reduced the adverse facts available rate to 45.49 percent, a simple average of the three individual dumping margins. Id. at 1238, 120 F.Supp.2d 1131. [10] 28 U.S.C. § 1581(c) provides that "[t]he Court of International Trade shall have exclusive jurisdiction of any civil action commenced under section 516A of the Tariff Act of 1930[, codified as amended at 19 U.S.C. § 1516a].” Section 516A of the Tariff Act of 1930 provides for judicial review of, inter alia, final results of administrative reviews of anti-dumping duty orders. See 19 U.S.C. § 1516a(a)(2)(B)(iii) (2000). [11] Both PAM and JCM filed memoranda in support of their respective motions for judgment upon the agency record. Because the arguments of Plaintiffs track each other, this Court refers only to PAM’s briefs except in instances where Plaintiffs differ. [12] The computer program used to extract data from the company's accounting system did not pick up on any of the external warehouse sales because those sales were coded differently than PAM’s other home market sales. (See PAM Verification Rep. 17.) [13] It appears that PAM’s prior counsel submitted to Commerce on March 31, 2004, a purported reconciliation of sales quantities with production quantities. The document showed that the total "quantity [of pasta] produced [by PAM] was some 90 thousand tons, while the quantity [reported by PAM in its questionnaire as] sold was some 22 thousand tons.... [Prior] counsel simply presented these figures to [Commerce], with no explanation whatsoever.” (PAM's Br. 31.) This Court understands PAM’s argument to be that the submission filed by prior counsel purporting to be the reconciliation did not actually reconcile the sales data with the production data. [14] JCM argues that Commerce, when evaluating whether PAM cooperated to the best of its ability, should have taken into consideration the difficulty PAM faced in responding to Commerce's questionnaire. "Unlike companies with U.S. affiliates, PAM could not draw upon U.S. based personnel to help with the task of translating, computerizing, collating!,] and reconciling its confidential business information.” (Principal Br. of PL JCM in Supp. of Mot. Pursuant to R. 56.2 for J. upon the Agency R. ("JCM Br.”) 34-35.) Furthermore, ”[t]he Record shows that [Commerce] was informed that [PAM] had a very small staff, including only two people fluent in English.” (Id.) JCM contends that PAM "did, indeed, do its best, given its resources.” (Id. at 37.) [15] Because many of the arguments raised by Defendants-Intervenor track those made by the Government this Court will refer only to the Government’s briefs. [16] PAM reported home market sales of around ten thousand tons. (PAM's Br. at 24) PAM did not report twenty-one thousand tons of home market sales (AG.E.A. sales of sixteen thousand tons and external warehouse sales of five thousand tons). It should be noted that PAM alleges that the AG.E.A. sales are properly excludable from PAM’s home market sales because they were made outside the ordinary course of trade. (See PAM’s Br. 30.) However, Commerce's instructions to PAM were clear that PAM "must report all sales, including those sales which you believe are outside the ordinary course of trade." (Questionnaire Instructions G7 (emphasis added), Pub. R. Doc. 31.) [17] Commerce's requirement that respondents report all sales — even those sales the respondent believes to be excludable — is reasonable. Commerce is thus able to verify that the sales alleged to be excludable were in fact made outside the ordinary course of trade. Commerce would not be able to verify the circumstances of the sales and to determine whether those sales should be excluded if the respondent failed to report these sales in the first instance. [18] PAM states that it participated in two reviews before the instant review. In the third review period (1998/99), PAM received a 5.04 percent rate. In the fourth review period, PAM received a rate of 4.10 percent. In fact, in the seventh review period, i.e., the period immediately following the period on appeal, PAM has now preliminarily received a rate of 4.79 percent, following a full verification. (PAM's Br. 34-35 (internal citations omitted).) [19] Because Commerce's own "investigation revealed widespread discounting” by Italian pasta manufacturers off the price lists, World Finer Foods II, 24 CIT at 1237, Barilla’s price list probably did not accurately capture the company’s home market sales price. As a result, the dumping margins calculated using the price list were likely inflated. See id. at 1238, 120 F.Supp.2d 1131.