Ad Hoc Shrimp Trade Action Committee v. United States

OPINION and ORDER POGUE, Senior Judge: This action arises from the seventh administrative review by the U.S. Department of Commerce (“Commerce”) of the antidumping duty (“AD”) order on certain frozen warmwater shrimp from the People’s Republic of China (“PRC” or “China”). 1 In this review, Commerce determined to revoke the order with respect to respondent Zhanjiang Regal Integrated Marine Resources Company, Limited (“Regal”). 2 Appealing Commerce’s determination, Plaintiff Ad Hoc Shrimp Trade Action Committee (“AHSTAC”) — an association of domestic warmwater shrimp producers that participated in this review 3 ■ — ■ claims that Commerce’s revocation was in error. 4 Specifically, AHSTAC challenges (1) Commerce’s reliance, in concluding that Regal was eligible for company-specific revocation, on data and analysis that were previously held not to have been based on a reasonable reading of the record evidence because, inter alia, the agency arbitrarily ignored economic comparability in its evaluation of factor of production data; and (2) Commerce’s determination to disregard discrepancies between Regal’s verified sales data and the entry information provided by U.S. importers in concluding that the continued application of the AD order to Regal’s merchandise was not necessary to offset dumping. 5 The court has jurisdiction pursuant to Section 516A(a)(2)(B)(iii) of the Tariff Act of 1930, as amended, 19 U.S.C. § 1516a(a)(2)(B)(iii) (2012), 6 and 28 U.S.C. § 1581(c) (2012). As explained below, Commerce’s reliance, without reconsideration or additional explanation, on data and analysis from the fifth review of this AD order — despite this Court’s prior holding that these same determinations were not based on a reasonable reading of record evidence, and despite material differences between the *1331 record of that proceeding and this revocation inquiry — is remanded for reconsideration. On the other hand, Commerce’s decision to disregard any discrepancy between Regal’s verified sales data and the entry information provided by importers was reasonable, and is therefore sustained. BACKGROUND Antidumping duty orders are imposed on imported merchandise that is sold at prices below normal value (i.e., “dumped”), where “normal value” is usually the price at which like products are sold in the exporting country or, for merchandise originating in non-market economies (“NMEs”), a value calculated. using-appropriate surrogate market economy data. 7 Such orders are regularly reviewed by Commerce, such that the agency determines producer/exporter-specific dumping margins, covering discrete (typically one-year) time periods, by making contemporaneous normal value to export price comparisons. 8 Pursuant to a regulation in effect at the time of the administrative review at issue here, Commerce was authorized to revoke the AD order with respect to particular exporters/producers after considering whether (A) such an exporter/producer had “sold the merchandise at not less than normal value for a period of at least three consecutive years”; (B) such exporter/producer (if previously determined to have sold the merchandise at less than normal value) “agrees in writing to its immediate reinstatement in the order, as long as any exporter or producer, is subject to the order, if [Commerce] concludes that the exporter or producer, subsequent to the revocation, sold the subject merchandise at less than normal value”; and (C) whether “the continued application of the anti-dumping duty order is otherwise necessary to offset dumping.” 9 Pursuant to this regulation, Regal requested company-specific revocation, citing its zero percent dumping margins in the fourth and fifth administrative reviews (and its expected zero percent dumping-margins in the sixth and seventh reviews), and certifying in writing its agreement to its immediate reinstatement under the order should Commerce determine in the future that Regal is selling subject merchandise to the United States at prices below normal value. 10 By the time of Commerce’s decision regarding this revocation request, Regal had been individually examined in the sixth and seventh reviews, *1332 and received zero percent dumping margins in both proceedings. 11 Regal was not, however, individually examined in the fifth review, 12 in which it was assigned a zero percent dumping margin based on its individually-calculated zero percent rate in the previous (fourth) review. 13 Because Regal was not individually examined in the fifth review, Commerce requested from Regal information and sales data from the time period covered by that review, “to confirm that Regal did not dump during that time,” 14 and hence to confirm that Regal did not dump for three consecutive years, as required for revocation eligibility under the regulation. 15 Finding that Regal’s fifth review sales data confirmed that Regal did not sell subject merchandise at less than the normal values calculated during that proceeding, Commerce concluded that Regal satisfied this regulatory requirement. 16 As explained below, AHSTAC now challenges this finding in so far as it relies, without additional analysis, on comparison values.from the fifth review that were held by this' Court to require reconsideration. 17 AHSTAC’s first challenge is directed at Commerce’s decision to compare Regal’s sales data for the period covered by the fifth administrative review 18 with the nor *1333 mal values calculated during that proceeding. 19 Because Commerce considers China to be a non-market economy, these normal values were based on “the value of the factors of production utilized in producing the merchandise,” including “an amount for general expenses and profit plus the cost of containers, coverings, and other expenses” (collectively, the “FOPs”), in a surrogate market economy country. 20 Commerce’s selection of the primary surrogate country for the fifth review period was successfully challenged by AHSTAC in that original proceeding. 21 Now AHS-TAC again challenges this same determination, as reiterated in the context of Commerce’s examination of Regal’s fifth review prices, as part of the agency’s evaluation of Regal’s revocation request. 22 Specifically, in the original fifth review, Commerce used Indian data to value the FOPs for its normal value calculation. 23 AHSTAC challenged this decision, arguing that record data from Thailand, rather than India, provided the best available information for the normal value calculation. 24 Responding to AHSTAC’s challenge, this Court remanded Commerce’s determination to use Indian surrogate data in the fifth review, “[b]ecause Commerce’s stated reasoning regarding the surrogate country selection in this review does not comport with a reasonable reading of the record.” 25 On remand, however, the question was rendered moot when the sole individually-examined respondent was found to be non-credible and uncooperative, and accordingly was assigned, based on adverse facts available, 26 a rate derived *1334 from the domestic industry’s petition to impose- this AD order. 27 “As a result, [Commerce] did not ... reexamine the issue of surrogate country selection” in the fifth review, and then subsequently continued to rely on the same surrogate FOP values in examining Regal’s fifth review pricing as part of its revocation analysis. 28 AHSTAC now challenges Commerce’s determination to continue to rely, without any additional consideration or explanation, on surrogate FOP values that were previously held to have been inadequate when considered in light of other record evidence. 29 In addition, AHSTAC challenges Commerce’s determination that “the continued application of the order is not otherwise necessary to offset dumping.” 30 Specifically, AHSTAC argues that a discrepancy between the volume of entries identified by U.S. importers as Regal’s subject merchandise and the volume of such shipments revealed in Regal’s own data reflects a “chronic concern with respect to Regal’s subject merchandise being incorrectly entered ... throughout the three-year revocation period.” 31 AHSTAC claims that *1335 this discrepancy is evidence indicating that the continued application of the order with respect to Regal remains necessary. 32 Following a statement of the relevant standard of review, each challenge is addressed in turn. STANDARD OF REVIEW The court upholds Commerce’s antidumping determinations if they are in accordance with law and supported by substantial evidence. 19 U.S.C. § 1516a(b)(l)(B)(i). Where, as here, the antidumping statute does not directly address the question before the agency, the court will defer to Commerce’s construction of its authority if it is reasonable. Timken Co. v. United States, 354 F.3d 1334, 1342 (Fed.Cir.2004) (relying on Chevron U.S.A Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984)). Substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion,” Consol. Edison Co. of N.Y. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 83 L.Ed. 126 (1938), and the substantial evidence standard of review “can be translated roughly to mean ‘is [the determination] unreasonable?’ ” Nippon Steel Corp. v. United States, 458 F.3d 1345, 1351 (Fed.Cir.2006) (citation omitted, alteration in the original); On-Line Careline, Inc. v. America Online, Inc., 229 F.3d 1080, 1085 (Fed.Cir.2000) (“The substantial evidence standard requires the reviewing court to ask whether a reasonable person might find that the evidentiary record supports the agency’s conclusion.”) (citations omitted). Importantly, “[t]he sub-stantiality of the evidence must take into account whatever in the record fairly detracts from its weight.” Univ. Camera Corp. v. NLRB, 340 U.S. 474, 488, 71 S.Ct. 456, 95 L.Ed. 456 (1951). Moreover, an agency acts arbitrarily, and therefore unreasonably, when it “entirely fail[s] to consider an important aspect of the problem” or “offer[s] an explanation for its decision that runs counter to the evidence before [it].” Motor Vehicle Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983). 33 DISCUSSION I. Commerce’s Choice of Surrogate Factor of Production Values for the Fifth Revieiv Period Is Not Supported by Substantial Evidence. First, AHSTAC challenges Commerce’s decision — in examining Regal’s fifth review period sales to evaluate Regal’s revocation request — to continue to rely on the same surrogate market economy data that were previously held to be inadequate when read in light of the other record evidence. 34 *1336 Specifically, Commerce’s choice of surrogate market economy values in the fifth review was based on the agency’s selection of India as the primary surrogate market economy country for China. 35 Because Commerce’s selection of an appropriate surrogate market economy must be such that the chosen dataset provides the “best available information” for approximating the NME producers’ experience, 36 Commerce chooses a primary surrogate country that is economically comparable to the NME country 37 (measured in terms of the countries’ comparative per capita gross national income (“GNI”) 38 ), is a significant producer of comparable merchandise, 39 and provides publicly-available, reliable, and relevant data. 40 In China Shrimp AR5, this Court held that Commerce acted arbitrarily in the fifth- review by disregarding “the concern that India’s per capita GNI was nearly a third of China’s [during the relevant time period], whereas Thailand’s per capita GNI was nearly identical thereto,” 41 despite the record evidence that the quality of the available datasets from these two potential surrogates was nearly indistinguishable. 42 Defendant argues that, in revisiting the issue in the context of Regal’s revocation request, Commerce recognized “that the Court had previously remanded Commerce’s fifth review primary surrogate country selection,” and that Commerce therefore “reconsidered its surrogate country selection for the limited purpose of evaluating Regal’s revocation request.” 43 But in fact the agency itself explicitly states that Commerce did not reconsider this matter. Specifically, Commerce explained that because it “ultimately was not required to respond to the surrogate country issue,” Commerce “did not ... reexamine the issue of surrogate country selection’’ and “[tjherefore ... continues to find India to be a reliable source for [surrogate values for the calculation of normal value for the period covered by the fifth *1337 review].” 44 Disregarding the court’s holding in China Shrimp AR5, Commerce did not consider or weigh the effect of the significant divergence between India and Thailand’s respective economic comparability to China when determining, based on reasoning reiterated from the fifth review, that while the record provided adequate surrogate FOP datasets from both potential surrogates, the Indian dataset provided the best available information. 45 Defendant emphasizes Commerce’s position that “[w]ithin a given [GNI] range, differences in per capita GNI between the countries do not imply any difference in level of economic development,” 46 and argues that “given that minor GNI differences do not correlate to differences in countries’ levels of economic development, Commerce’s methodology of considering data quality in choosing among the countries that are at a comparable level of economic development and significant producers of subject merchandise is reasonable.” 47 But this argument ignores this Court’s repeated holdings that where, as here, adequate data is available from more than one country that is both at a level of economic development compárable to the NME and a significant producer of comparable merchandise, Commerce must weigh the relative merits of such potential surrogates’ datasets in a way that does not arbitrarily discount the accuracy-enhancing value of sourcing surrogate data from a market economy whose economic develop *1338 ment is as close as possible to that of the NME, and in that regard may provide the “best available information.” 48 Here, the record of the fifth review, like that of the AD proceeding at issue in Vinh Hoan, 49 clearly reveals the basis for the court’s concern. During the fifth review, Commerce found that both India hnd Thailand fell within a range of GNI values comparable to the per capita GNI of China, that both of these potential surrogates were significant producers of comparable merchandise, and that “[t]here exist[ed] on the record sufficient, publicly available surrogate factor information for the majority of FOPs from both India and Thailand” that was “of roughly equal specificity,” and that otherwise satisfied the agency’s usual data-quality standards. 50 But in deciding which of these two datasets would provide the “best available” information, - Commerce (both in the original fifth review and in examining Regal’s fifth review pricing as part of its revocation analysis) categorically and formulaically disregarded the evidence that the Indian data came from a country whose per capita GNI was barely a third of China’s, whereas the Thai data was from an economy whose per capita GNI was virtually identical to China’s. 51 Commerce’s refusal to account for the accuracy-enhancing value of relative GNI *1339 proximity when evaluating the relative merits of alternative satisfactory datasets, to determine which set constitutes the best available surrogate value information, is arbitrary and, therefore, unreasonable. 52 Commerce’s own comparability metric implies that, all other considerations being roughly equal, surrogate data from a country whose GNI is nearly identical to that of the NME would be more likely to better approximate the values that would prevail within the NME itself, if the latter were a market economy, than would data from a country whose per capita GNI diverges from that of the NME by multiple orders of magnitude. 53 Commerce maintains, as it did in China Shrimp AR5, that significant “differences of quality of data sources” adequately support the agency’s selection of Indian rather than Thai surrogate values to determine the normal comparison values in the fifth review, notwithstanding the Thai economy’s far greater comparability to that of China. 54 Specifically, Commerce relies on its reasoning from the fifth review that “the data from India were superior to that from Thailand [because,] of the ten FOPs, three had a more specific Indian [Harmonized Tariff Schedule (‘HTS’) ] number while seven had equally specific Indian and Thai HTS numbers,” and because the Indian financial statement on record was from a producer that, like Regal, was “a shrimp farmer as well as shrimp processor.” 55 But by Commerce’s usual data evaluation standards, the record contained adequate and suitable data from both India and Thailand, because Commerce was unable to make a distinction between the two countries based on the datasets’ specificity to the input in question, exclusivity of taxes and import duties, contemporaneity with the period of review, or public availability. 56 *1340 Thus, to differentiate between the two satisfactory datasets, Commerce focused on minute, seemingly hair-splitting differences. Because the Indian and Thai data generally were of such similar quality that Commerce was unable to distinguish them using its usual standards, the agency compared Indian and Thai information for valuing shrimp larvae, the critical input used by the mandatory respondent in the fifth review to produce the subject merchandise. Here again Commerce found that the Indian and Thai information for valuing shrimp larvae was of very similar quality, but the Thai data were specific to black tiger shrimp, whereas the Indian data did not specify a species. Based on this distinction, - Commerce concluded that because the sole mandatory respondent • neither produced nor sold black tiger shrimp, the Indian . shrimp larvae data were superior (because, unlike the Thai data, they did not specify the species of shrimp to which they pertained). 57 But unlike the sole mandatory respondent in the fifth review, Regal did not use shrimp larvae to produce the subject merchandise, 58 so the relative quality of Indian and Thai larvae data is not relevant. As AHS-TAC points out, Regal used a different set of FOPs from those used by the respondent in the original fifth review and, unlike that respondent, Regal used broodstock rather than shrimp larvae as its critical input for producing the subject merchandise. 59 And as Commerce itself concluded, the better surrogate value data for Regal’s broodstock came from Thailand, not India. 60 Thus Commerce’s reasoning that its original analysis in the fifth review supports its conclusion here that the Indian data is superior to the Thai data because the former more closely matches Regal’s own FOPs and production process 61 is not supported by a reasonable reading of the evidence. Not only is this reasoning belied by Commerce’s own finding in the original fifth review that the Indian and Thai data were of “roughly equal specificity,” this reasoning is moreover no longer applicable, because Regal’s FOPs are substantially different from those originally considered by Commerce in the fifth review, and particularly because the key fac *1341 tor of production on which Commerce based its data-evaluation in the original fifth review (shrimp larvae) was not used by Regal at all. Commerce’s alternative reasoning — that the Indian data provided the best available surrogate information for Regal’s fifth-review normal value calculation because the Indian financial statement on record more closely approximated Regal’s experience 62 —is also not supported by substantial evidence. This is because Commerce did not account for or in any way address the additional evidence submitted by AHSTAC in support of its argument that the Thai financial statement on the record is from a company that, like the Indian company, and like Regal, is also an integrated producer (i.e., a shrimp farmer as well as shrimp processor). 63 Specifically, Commerce reasoned in the original fifth review that the evidence was insufficient to conclude that the Thai company was an integrated producer, because the certification submitted by AHSTAC in support of this claim did not clearly indicate that it applied to this specific company. 64 But in this revocation proceeding, AHSTAC submitted additional evidence in support of its claim that this specific Thai company was indeed certified as a hatchery, farm, and processing plant. 65 Commerce completely ignored this evidence, instead relying entirely on its original fifth review analysis. 66 But most importantly, Commerce completely (and categorically) ignored the biggest difference in quality between the two datasets, which is that the Thai data was from a market economy that very nearly mirrored China’s level of economic development (by Commerce’s own metric, which “uses per capita income to measure [countries’ economic] comparability” 67 ), whereas the Indian data reflected values present in an ■ economy whose per capita GNI was multiple orders of magnitude lower than China’s. In doing so, Commerce arbitrari *1342 ly ignored an important aspect of the issue. 68 Accordingly, the agency’s reliance in this revocation proceeding upon its original fifth review analysis of surrogate dataset alternatives is not supported by substantial evidence, and must therefore be remanded for reconsideration. II. Regal’s Import Volume Discrepancy AHSTAC additionally challenges Commerce’s determination that “the continued application of the order [as to Regal] is not otherwise necessary to offset dumping,” 69 claiming that a discrepancy between the volume of entries identified by Regal’s U.S. importers as merchandise subject to the AD order and the volume of such shipments revealed in Regal’s own data indicates that the continued application of the order with respect to Regal remains necessary. 70 But as Commerce explains, the agency conducted an on-site verification of Regal as part of this revocation proceeding, during which Commerce reviewed and analyzed Regal’s sales data for the periods covered by the fifth, sixth, and seventh reviews. 71 Commerce “completed [quantity and value] reconciliations and completeness tests for [those periods],” found no discrepancies, and therefore concluded that “there is no basis to find that Regal’s reported data is inaccú-rate.” 72 AHSTAC concedes that Commerce verified the accuracy of Regal’s sales data, and does not challenge that finding. 73 Thus, consistent with Commerce’s prior findings in this regard, 74 to the extent that the record reveals a discrepancy between the volume of subject merchandise exported by Regal and that reported as such by U.S. importers, the inaccuracies are in the information submitted to U.S. Customs and Border Protection (“Customs”) by the importers, not in the data used by Commerce to determine that Regal did not export to the United States at dumped *1343 prices during the relevant time periods. Because Commerce is concerned solely with the latter inquiry, and because the agency adequately verified that the necessary data for that determination was accurate, Commerce reasonably concluded that this discrepancy did not affect the accuracy of Regal’s verified sales information, upon which Commerce based its determination that the continued application of the order as to Regal was not necessary to offset dumping. 75 Because Regal’s own information was verified as accurate, the discrepancy in reported import volume has no bearing on the accuracy of the dumping determination, and is a matter that is more appropriately addressed to Customs’ enforcement authority. 76 Accordingly, Commerce’s determination that any discrepancy between Regal’s verified sales data and entry data reported by U.S. importers to Customs does not impugn the accuracy of Commerce’s dumping determinations, or its consequent determination that the continued application of the order as to Regal is not necessary to offset dumping, is sustained. CONCLUSION For all of the foregoing reasons, Commerce’s company-specific revocation of this antidumping duty order as to Regal is remanded solely for reconsideration of the surrogate data used to determine normal value for Regái s price comparisons during the period covered by the fifth administrative review. Commerce shall have until July 17, 2015, to complete and file its remand results. Plaintiff shall have until July 31, 2015, to file its comments, and the agency shall then have until August 10, 2015, to respond. It is SO ORDERED. Footnotes [1] See Certain Frozen Warmwater Shrimp from the People’s Republic of China, 78 Fed.Reg. 56,209 (Dep't Commerce Sept. 12, 2013) (final results of administrative review; 2011—2012) (“AR7 Final Results ") and accompanying Issues & Decision Mem., A-570-893, ARP 11-12 (Sept. 12, 2013) (“AR7 I & D Mem."). [2] AR7 Final Results, 78 Fed.Reg. at 56,210. [3] See Compl., ECF No. 2, at ¶ 7. [4] See [Conf. & Pub.] Mem. of L. in Supp. of [AHSTAC]'s USCIT Rule 56.2 Mot. for J. on the Agency R., ECF Nos. 41 (conf. version) & 42 (pub. version) (“PL's Br.”). [5] Id. [6] Further citations to the Tariff Act of 1930, as amended, are to the relevant provisions of Title 19 of the U.S.Code, 2012 edition. [7] See 19 U.S.C. §§ 1673, 1677b(a)(l)(B)(i), 1677b(c). [8] See id. at § 1675(a). [9] 19 C.F.R. § 351.222(b)(2)(i)(A)-(C) (2012). This regulatory provision was subsequently revoked for administrative reviews initiated on or after June 20, 2012. Modification to Regulation Concerning the Revocation of Anti-dumping and Countervailing Duty Orders, 77 Fed.Reg. 29,875, 29,876 (Dep’t Commerce May 21, 2012). As the review at issue here was initiated on April 30, 2012, the regulation was still in effect. See Initiation of Antidump-ing and Countervailing Duty Administrative Reviews and Request for Revocation in Part, 77 Fed.Reg. 25,401, 25,403 (Dep’t Commerce Apr. 30, 2012). [10] See AR7 I & D Mem. cmt. 2 at 6; Req. for Admin. Review & Revocation, Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 11-12 (Feb. 28, 2012), reproduced in [Conf. & Pub.] App. to Def.’s Resp. in Opp'n to Pl.’s Mot. for J. Upon the Agency R., ECF Nos. 55 (conf. version) & 56 (pub. version) ("Deft's App.”) at Tab 1 (‘Revocation Req.") at 2-3 & Attach. 1. At the time of Regal’s request for revocation, the seventh review had not yet been initiated, and the results of the sixth review had not yet been finalized, although Regal had been preliminarily assigned a zero percent margin in the sixth'review. Revocation Req., ECF Nos. 55 & 56 at Tab 1, at 3. [11] Where it is not practicable to make individual weighted average dumping margin determinations for each known exporter and producer of the subject merchandise for whom review was requested, Commerce may limit its individualized examination to a smaller number of companies, 19 U.S.C. § 1677f-1(c)(2), and assign to the remaining respondents the “all-others” rate (calculated in accordance with 19 U.S.C. § 1673d(c)(5)) or, where appropriate, the NME countrywide rate. See Jiangsu Jiasheng Photovoltaic Tech. Co. v. United States, - CIT -, 28 F.Supp.3d 1317, 1339-40 n. 107 (2014). Regal was individually examined in the sixth and seventh administrative reviews of this order. Certain Frozen Warmwater Shrimp from the People’s Republic of China, 77 Fed. Reg. 12,801, 12,801 (Dep’t Commerce Mar. 2, 2012) (preliminary results, partial rescission, extension of time limits for the final results, and intent to revoke, in part, of the sixth antidumping duty administrative review) (explaining that Commerce selected Regal for individual examination in the sixth' review) (unchanged in 77 Fed.Reg. 53,856 (Dep’t Commerce Sept. 4, 2012) (final results, partial rescission of sixth antidumping duty administrative review and determination not to revoke in part) ("AR6 Final Results ")); Decision Mem. for Prelim. Results, Partial Rescission of Antidumping Duty Admin. Review, Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 11-12 (Mar. 12, 2013) at 3 (explaining that Commerce selected Regal for individual examination in the seventh review) (adopted in 78 Fed.Reg. 15,696, 15,696 n. 1 (Dep’t Commerce Mar. 12, 2013) (preliminary results of administrative review; 2011-2012) (unchanged in AR7 Final Results, 78 Fed.Reg. 56,209)). [12] Certain Frozen Warmwater Shrimp from the People’s Republic of China, 76 Fed.Reg. 8338, 8341 (Dep’t Commerce Feb. 14, 2011) (preliminary results and preliminary partial rescission of fifth antidumping duty administrative review) (explaining that Commerce selected only one company for individual examination in the fifth review, which was not Regal) (unchanged in 76 Fed.Reg. 51,940 (Dep’t Commerce Aug. 19, 2011) (final results and partial rescission of [fifth] antidumping duty administrative review) ("AR5 Final Results ”)). [13] See AR5 Final Results, 76 Fed.Reg. at 51,-942. [14] AR71 & D Mem. cmt. 2 at 6. [15] See 19 C.F.R. § 351.222(b)(2)(i)(A). [16] [Commerce’s] Post-Prelim. Analysis for [Regal] and [Another Resp’t], Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 11-12 (May 20, 2013), reproduced in Def.’s App., ECF Nos. 55 & 56 at Tab 7 (“Regal Post-Prelim. Mem.") at 6 (unchanged in the AR7 Final Results, 78 Fed.Reg. at 56,210). [17] PL’s Br„ ECF Nos. 41 & 42, at 13-22. [18] The period of review for that proceeding was February 1, 2009, through January 31, *1333 2010. AR5 Final Results, 76 Fed.Reg. at 51,-940. [19] See Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 3. [20] See 19 U.S.C. § 1677b(c)(l); Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 3. Although the statute permits Commerce to source its data from multiple surrogate market economies, see 19 U.S.C. § 1677b(c)(1), Commerce normally values all factors of production using data from a single surrogate market economy country (the “primary surrogate country”). See 19 C.F.R. § 351.408(c)(2); Clearon Corp. v. United States, Slip Op. 13-22, 2013 WL 646390, at *6 (CIT Feb. 20, 2013) (noting that Commerce's “preference for the use of a 'single surrogate country’ ” is reasonable because, “as Commerce points out, deriving the surrogate data from one surrogate country limits the amount of distortion introduced into its calculations”). [21] See Ad Hoc Shrimp Trade Action Comm. v. United States, - CIT -, 882 F.Supp.2d 1366, 1376 (2012) ("China Shrimp AR5"). [22] Pl's Br., ECF Nos. 41 & 42, at 13-22. [23] Issues & Decision Mem., Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 09-10 (Aug. 12, 2011) (adopted in AR5 Final Results, 76 Fed. Reg. at 51,940) ("AR5 I & D Mem.’’) cmt. 2 at 10. [24] China Shrimp AR5, CIT at -, 882 F.Supp.2d at 1372; see 19 U.S.C. § 1677b(c)(l) (requiring that Commerce use the “best available information” in selecting appropriate surrogate FOP values when calculating the normal value of NME merchandise). [25] China Shrimp AR5, CIT at-, 882 F.Supp.2d at 1376. [26] See 19 U.S.C. § 1677e(b) ("If [Commerce] finds that an interested party has failed to cooperate by not acting to the best of its ability to comply with a request for information from [Commerce], [Commerce], in reaching the applicable determination under this subtitle, may use an inference that is adverse to the interests of that party in selecting from among the facts otherwise available. Such adverse inference may include reliance on information derived from — (1) the petition, (2) a final determination in the investigation under this subtitle, (3) any previous review under section 1675 of this title or determination under section 1675b of this title, or (4) any other information placed on the record.”). [27] See Final Results of Redetermination Pursuant to Ct. Remand, Ct. No. 11-00335, ECF No. 74, at 2 ("[BJecause'we have found [the sole individually-examined respondent] to be part of the PRC-wide entity, which is receiving [a rate based entirely on adverse facts available], there are no calculated margins for this period of review ... and it is [therefore] unnecessary to select a surrogate country in which to value a respondent’s factors of production ....”); id. at 24 ("[T]he PRC-wide rate ... represents a rate calculated in the petition in the [original investigation into whether sales of subject merchandise were being made at less than fair value]....”); Ad Hoc Shrimp Trade Action Comm. v. United States, — CIT -, 925 F.Supp.2d 1315 (2013) (affirming Commerce's determination to rely on adverse facts available in re-calculating the individually-investigated respondent's rate, but remanding the rate for adequate corroboration); Ad Hoc Shrimp Trade Action Comm. v. United States, -CIT-, 992 F.Supp.2d 1285 (2014) (sustaining the agency's revisited corroboration of the PRC-wide rate). [28] Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 3 ("The [Court of International Trade] remanded the final results [of the fifth review] to [Commerce] for further, consideration .... However, pursuant to the [redeter-mination on remand], [Commerce] ultimately was not required to respond to the surrogate country issue in the remand as the question was rendered moot after adverse facts available were applied to the sole respondent. As a result, [Commerce] did not perform an anti-dumping calculation or reexamine the issue of surrogate country selection. Therefore, [Commerce], consistent with the determination made in the ARS Pinal Results, continues to find India to be a reliable source for [surrogate FOP values for the period covered by the fifth review] ....") (citation omitted). [29] See Pl.’s Br., ECF Nos. 41 & 42, at 13-22. [30] Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 6 (applying 19 C.F.R. § 351.222(b)(2)(i)(C)) (unchanged in the AR7 Final Results, 78 Fed.Reg. at 56,210); see Pl's Br., ECF Nos. 41 & 42, at 22-35 (challenging this determination). [31] Pl's Br., ECF Nos. 41 & 42, at 31. This discrepancy was first noted in the third administrative review of this order, in which ' Commerce found that "certain importers improperly classified [some of Regal's] subject entries as non-dutiable.” Issues & Decision Mem., Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 07-08 (Aug. 28, 2009) (adopted in 74 Fed.Reg. 46,565 (Dep't Commerce Sept. 10, 2009) (final results and partial rescission of [third] antidumping duty administrative review)) cmt. 7 at 23. In the fourth review, responding to a court-ordered remand to reexamine the agency's reliance on importer-provided entry volume data to select Regal for individual examination as one of the largest exporters of subject merchandise by volume, see Ad Hoc Shrimp Trade Action Comm. v. United States, - CIT -, 791 F.Supp.2d 1327, 1330-34 (2011), Commerce verified that no such discrepancy of entry volumes existed during the period covered by that review, see Ad Hoc Shrimp Trade Action Comm. v. United States, — CIT-, 828 F.Supp.2d *1335 1345, 1351 (2012). The discrepancy again appeared, however, during the sixth review, when Commerce again found that Regal's reported U.S. sales quantity differed from that reported by U.S. importers of Regal's subject merchandise. See Issues & Decision Mem., Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 10-11 (Aug. 27, 2012) (adopted in AR6 Final Results, 77 Fed.Reg. at 53,858) (“ARÓ I & D Mem.”) cmt. 7 at 36. And while conceding that Commerce verified the accuracy of Regal’s own sales data for the period covered by the seventh review, AHSTAC contends that the discrepancy continued into that period as well. Pl.’s Br.. ECF Nos. 41 & 42. at 29. [32] See Pl.’s Br., ECF Nos. 41 & 42, at 32-35. [33] Although the Court in State Farm was discussing the "arbitrary or capricious” (rather than the "substantial evidence”) standard of review, this reasoning is also relevant here because an agency determination that is arbitrary is ipso facto unreasonable. See, e.g., Ward v. Sternes, 334 F.3d 696, 704 (7th Cir.2003) (noting that "a decision [that] is so inadequately supported by the record as to be arbitrary [is] therefore objectively unreasonable”) (quotation marks and citations omitted). . [34] Pl’s Br., ECF Nos. 41 & 42. at 13-22. [35] See AR51 & D Mem. cmt. 2 at 10. [36] See 19 U.S.C. § 1677b(c)(l); AR7 I & D Mem. cmt. 2 at 10 (emphasizing,the importance of surrogate data’s resemblance to the NME producers’ experience). [37] 19 U.S.C. § 1677b(c)(4)(A). [38] Antidumping Methodologies in Proceedings Involving Non-Market Economy Countries: Surrogate Country Selection and Separate Rates, 72 Fed.Reg. 13,246, 13,247 (Dep’t Commerce Mar. 21, 2007) ("Surrogate Country Selection Policy ") (explaining that “[Commerce] uses per capita income to measure [economic] comparability”). [39] 19 U.S.C. § 1677b(c)(4)(B). [40] See 19 U.S.C. § 1677b(c)(1); Import Admin., U.S. Dep’t Commerce, Non-Market Economy Surrogate Country Selection Process, Policy Bulletin 04.1 (2004), available at http:// enforcement.trade.gov/policy/bull04-l.html (last visited May 11, 2015) (“Commerce Policy 4.1 ”) ("[D]ata quality is a critical consideration affecting surrogate country selection. After all, a country that perfectly meets the requirements of economic comparability and significant producer is not of much use as a primary surrogate if crucial factor price data from that country are inadequate or unavailable.”). [41] China Shrimp AR5, - CIT at-, 882 F.Supp.2d at 1376. [42] See id. at 1375 (“Commerce found that the Indian and Thai data were so similar in quality that Commerce was unable to make a distinction between the two countries based on the datasets’ specificity to the input in question, exclusivity of taxes and import duties, contemporaneity with the period of investigation or review, or public availability — i.e., based on its usual data-evaluation standards.”) (citing AR5 I & D Mem. cmt. 2 at 7). [43] Def.’s Resp. in Opp'n to Pl.’s Mot. for J. Upon the Agency R., ECF Nos. 53 (conf. version) & 54 (pub. version) ("Def.’s Br.”) at 11 (citing AR71 & D Mem. cmt. 2 at 7-10; Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 2-3). [44] Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 3 (emphasis added) (unchanged in AR7 I & D Mem. cmt. 2 at 7 (relying on and citing explanation contained in Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 3)); see also AR7 I & D Mem. cmt. 2 at 9-10 (relying on the agency's original analysis in the fifth review, without any indication of reconsideration). [45] Compare AR7 I & D Mem. cmt. 2 at 9 ("[Commerce] does not believe that the [anti-dumping statute] requires it to compare relative GNI of the [potential surrogate] countries in its analysis.”); id. at 7 ("[Commerce] continue[s] to regard both Thailand and India as being at the same level of economic development as the PRC.”) (emphasis added), and AR5 I & D Mem. cmt. 2 at 6-7 (relied on in AR7 I & D Mem. cmt. 2 at 7) ("[When selecting surrogate market economy countries for the normal value calculation in NME cases, Commerce] creates a list of possible surrogate countries that are to be treated as equally comparable in evaluating their suitability for use as a surrogate countryf, regardless pi any differences among the potential surrogates in terms of their relative GNI proximity to the per capita GNI of the NME country] .... [Consistent with [this policy], [Commerce] continues to find that [India and Thailand] are equally economically comparable to the PRC for purposes of [surrogate value] calculations.”) (emphasis added), with China Shrimp AR5, — CIT at-, 882 F.Supp.2d at 1375 ("An unexplained and conclusory blanket policy of simply ignoring relative GNI comparability within a particular range of GNI values does not amount to a reasonable reading of the evidence in support of a surrogate selection where more than one potential surrogate within that GNI range is a substantial producer of comparable merchandise for which adequate data is publicly available. Rather, in such situations, Commerce must explain why its chosen surrogate's superiority in one of the three eligibility criteria outweighs another potential surrogate's superiority in one or more of the remaining criteria.”) (citing Amanda Foods (Vietnam) Ltd. v. United States, 33 CIT 1407, 1413, 647 F.Supp.2d 1368, 1376 (2009)); id. at 1376 ("Contrary to. the Government's assertions, ... this record is not so clear as to lead to the conclusion that [the] difference in data quality [between the Indian and Thai surrogate value datasets] necessarily outweighed the concern that India’s per cap-ita GNI was nearly a third of China’s, whereas Thailand’s per capita GNI was nearly identical thereto.”) (quotation marks and citation omitted). [46] AR71 & D Mem. cmt. 2 at 8; see Def.’s Br., ECF Nos. 53 & 54, at 15-16. [47] Def.’s Br., ECF Nos. 53 & 54, at 16 (footnote omitted). [48] See 19 U.S.C. § 1677b(c)(1); China Shrimp AR5,-CIT at-, 882 F.Supp.2d at 1375-76; Amanda Foods, 33 CIT at 1413, 647 F.Supp.2d at 1376; Vinh Hoan Corp. v. United States, - CIT-, 49 F.Supp.3d 1285, 1302-06 (2015). Although Commerce cites to Fujian Lianfu Forestry Co. v. United States, 33 CIT 1056, 638 F.Supp.2d 1325 (2009), for the broad proposition that the agency may select India as the primary surrogate for China "even though there were other economically-comparable countries with GNIs closer to the GNI of China,” AR71 & D Mem. cmt. 2 at 8 & n. 36, the record in that case (unlike the record here, or in Vinh Hoan, for example) revealed that Commerce’s surrogate country selection was based on significant and substantial differences between the alternative datasets. Compare China Shrimp AR5, — CIT at-, 882 F.Supp.2d at 1375-76, and Vinh Hoan, -CIT at-, 49 F.Supp.3d at 1304 (each discussing the unusual level of scrutiny Commerce resorted to in those cases to distinguish between multiple suitable data-sets), with Fujian, 33 CIT at 1079-80, 638 F.Supp.2d at 1350-51 & n. 11 (explaining Commerce’s findings in that case that the "Indian data provided - [significantly] more comprehensive coverage of, and were more specific to, the inputs used in the production [of the subject merchandise]”; noting that no alternative data was available for several important inputs; and emphasizing that the plaintiff in that case did not "contest Commerce’s finding that Indian data provide[d] greater coverage than Philippine data for valuing inputs specific to the production [of the subject merchandise]” or "Commerce’s finding that Indian data provided more specific input values”). Thus the question of whether it is reasonable for Commerce to ignore relative economic comparability when evaluating datasets of otherwise similar quality was not before the court in Fujian. See Fujian, 33 CIT at 1075-80, 638 F.Supp.2d at 1347-51 (rejecting the plaintiff's argument that India’s per capita GNI differed so greatly from China's that it should not have been a surrogate candidate at all; noting that the plaintiff had conceded the clear superiority of the Indian data in terms of its specificity and comprehensiveness; and not addressing the issue of how to distinguish between datasets of very similar quality). [49] See Vinh Hoan, - CIT at -, 49 F.Supp.3d at 1304-05 (noting that “Commerce indicated the unusual level of scrutiny it would need to apply to distinguish between otherwise usable data sets” but nevertheless denied that "weighing the relative GNIs of the countries [may] improve [its] selection of the best available information”). [50] AR51 & D Mem. cmt. 2 at 7. [51] See AR71 & D Mem. cmt. 2 at 8 (reproducing the GNI data, showing that during the relevant time period, "the PRC had a GNI of $2,940, India had a GNI of $1,070 ... [and] Thailand had a GNI of $2,840”) (citation omitted). The accuracy of these GNI values is not in dispute. ' [52] China Shrimp AR5, - CIT at -, 882 F.Supp.2d at 1375-76; Amanda Foods, 33 CIT at 1413, 647 F.Supp.2d at 1376; Vinh Hoan,-CIT at-, 49 F.Supp.3d at 1302-06. [53] See Surrogate Country Selection Policy, 72 Fed.Reg. at 13,247 (noting that "the closest country to [an NMEJ’s level of economic development” is the country whose per capita GNI most closely approximates that of the NME); id. (implying a spectrum of economic comparability by stating that Commerce is not obligated to choose the "country [that] is the most economically comparable to the NME” when "us[ing] per capita income to measure comparability”). Here, the very existence of this dispute implies a meaningful difference in the outcome of the normal value calculation, depending on whether the Thai or the Indian surrogate values are used, and it is unreasonable for Commerce to imply that this difference (which is at least partially due to the GNI disparity between the two countries from which the competing data is sourced) has no bearing on the relative accuracy of the resulting normal value calculations, or on the question of which dataset constitutes the best available information for this purpose. [54] See AR7 I & D Mem. cmt. 2 at 10; China Shrimp AR5,-CIT at-, 882 F.Supp.2d at 1375-76. [55] AR7 I & D Mem. cmt. 2 at 10 (relying on AR5 I & D Mem. cmt. 2 at 7, 10); see AR51 & D Mem. cmt. 2 at 10 ("[B]ecause the Indian shrimp larvae [i.e., the critical input used by the sole individually-investigated ('mandatory') respondent in the fifth review] [surrogate value] source fulfills more of [Commerce]'s [surrogate value] selection criteria, and the Indian surrogate company, Falcon Marine, is more reliable than the surrogate financial data from Thailand [due to the absence of convincing evidence that the Thai company, Seafresh, is an integrated producer that farms as well as processes shrimp], we will continue to use India as the primary surrogate country for the valuation of FOPs and surrogate financial ratios.”). [56] China Shrimp AR5, - CIT at-, 882 F.Supp.2d at 1375 (citing AR5 I & D Mem. cmt. 2 at 7); see Commerce Policy Bulletin 04.1 (“In assessing [the quality and suitability of surrogate value data in NME cases], it is [Commercefs stated practice to use investigation or review period-wide price averages, prices specific to the input in question, prices *1340 that are net of taxes and import duties, prices that are contemporaneous with the period of investigation or review, and publicly available data.”). [57] China Shrimp AR5, - CIT at-, 882 F.Supp.2d at 1375-76 (quotation and alteration marks omitted) (citing AR5 I & D Mem. cmt. 2 at 8). [58] PI. [AHSTACJ’s [Conf. & Pub.] Reply Mem. in Supp. of USCIT Rule 56.2 Mot. for J. on the Agency R., ECF Nos. 57 (conf. version) & 58 (pub. version) ("Pl.'s Reply”) at 8 (relying on [Commerce’s] Surrogate Factor Valuations for the [Seventh Admin. Review] Post-Prelim. Analysis for Regal in [the Fifth Review], Certain Frozen Warmwater Shrimp from the Peoples Republic of China, A-570-893, ARP 11-12 (May 20, 2013), reproduced in [Conf. & Pub.] App. to Mem. in Supp. of PL’s Reply Mem. in Supp. of its USCIT Rule 56.2 Mot. for J. on the Agency R., ECF Nos. 59 (conf. version) & 60 (pub. version) ("Pl's Reply App.”) at Tab 3 (“Regal AR5 SV Mem") at 3-4, Ex. 1). See Regal AR5 SV Mem., ECF Nos. 59 & 60 at Tab 3, at 3 (noting that "[b]rood-stock was not a reported [surrogate value] in the original [fifth] review” and concluding that Global Trade Atlas data from Thailand "is clearly an exact match to the FOP used by Regal during the [period of review]”). [59] Pl's Reply, ECF Nos. 57 & 58, at 7-8 (relying on Regal AR5 SV Mem., ECF Nos. 59 & 60 at Tab 3, at 3-4, Ex. 1). [60] Regal AR5 SV Mem., ECF Nos. 59 & 60 at Tab 3, at 3. Commerce also selected Thailand as the primary surrogate country, for the sixth and seventh reviews. See Def.’s Br., ECF Nos. 53 & 54, at 17-18; Pl’s Br., ECF Nos. 41 & 42, at 15. [61] AR7 I & D Mem. cmt. 2 at 10 (relying on AR5 I & D Mem. cmt. 2 at 7, 10). [62] Id. at 10 &n. 42. [63] See AR7 I & D Mem. cmt. 2 at 10 (relying on AR5 I & D Mem. cmt. 2 at 10); AR5 I & D Mem. cmt. 2 at 9-10 (concluding that the Indian company’s financial statement on record more closely approximated the NME respondent’s experience than the Thai company's financial statement because, like the respondent in that review, the Indian company was "a shrimp farmer as well as shrimp processor,” whereas the Thai company’s financial statement provided no "indication that [the company] farms and processes shrimp”) (citation omitted); AR7 I & D Mem. cmt. 2 at 10 n. 42 (noting that Regal, like the respondent in the original fifth review, uses an integrated production process that both farms and processes shrimp). [64] See AR5 I & D Mem. cmt. 2 at 9-10 (explaining that AHSTAC submitted evidence from "an organization that certifies shrimp hatcheries, farms, feed mills, and processing plants for 'Best Aquaculture Practices,’" which listed the Thai company within a set of companies receiving a group certification, but concluding that "[e]ven if [Commerce] were to assume that these facilities were somehow related,” it was unable to conclude that the Thai company was an integrated producer because this evidence concerned the "Seaf-resh Industry Group — Thailand, [whereas] the financial statement on the record of this proceeding is specifically for Seafresh Industry Public Company Ltd., not Seafresh Industry Group — Thailand”). [65] Pl.’s Reply, ECF Nos. 57 & 58, at 10 (citing [AHSTAC’s] Data on Surrogate Values for the Fifth Admin. Review (2009-2010), Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 11-12 (Feb. 4, 2013), reproduced in Pl.’s Reply App., ECF Nos. 59 & 60 at Tab 5, at 7, Ex. 6(b)). [66] See AR71 & D Mem. cmt. 2 at 10 (relying on AR51 & D Mem. cmt. 2). [67] Surrogate Country Selection Policy, 72 Fed. Reg. at 13,247; see also id. (noting that "the closest country to [an NME]'s level of economic development” is the country whose per capita GNI most closely approximates that of the NME). [68] China Shrimp AR5; - CIT at -, 882 F.Supp.2d at 1375-76; see also Vinh Hoan, -CIT at-, 49 F.Supp.3d at 1305 ("[T]he ultimate question is what is the best available information to value [an NME respondent’s] factors of production? Thus, Commerce must choose the [primary surrogate] country that furthers this goal. The analysis suggested by [China Shrimp AR5 ], and adopted here, is that Commerce must compare differences in economic comparability with differences in the other factors, including data quality, when the facts so require.”); id. at 1304 (explaining that the record of that case, like that of China Shrimp AR5, required comparison of potential surrogates' relative economic comparability to the NME country, because the record revealed that multiple countries’ datasets satisfied the agency's threshold suitability criteria and, like in China Shrimp AR5, Commerce was therefore compelled to resort to an "unusual level of scrutiny ... to distinguish between otherwise usable data sets”). [69] Regal Post-Prelim. Mem., ECF Nos. 55 & 56 at Tab 7, at 6 (applying 19 C.F.R. § 351.222(b)(2)(i)(C)) (unchanged in the AR7 Final Results, 78 Fed.Reg. at 56,210); Pl.’s Br., ECF Nos. 41 & 42, at 22-35. [70] See Pl.’s Br., ECF Nos. 41 & 42, at 32-35. [71] AR71 & D Mem. cmt. 6 at 14. [72] Id.; see Verification of the Sales and Factors Responses of [Regal], Certain Frozen Warmwater Shrimp from the People’s Republic of China, A-570-893, ARP 11-12 (June 21, 2013), reproduced in Def.’s App., ECF Nos. 55 & 56 at Tab 8 (pub. version) & ECF No. 55 at Tab 15 (conf. version). [73] See Pl.’s Br., ECF Nos. 41 & 42, at 29. [74] See AR6 I & D Mem. cmt. 7 at 38 ("Regal has cooperated with [Commerce] and provided all requested information by the applicable deadlinesf,] ... [and] no record evidence demonstrates that Regal attempted to misclas-sify entries of subject merchandise. Moreover, there is no information on the record that indicates Regal underreported its U.S. sales information.”) (citation omitted); supra note 31 (detailing relevant history). [75] See AR71 & D Mem. cmt. 6 at 14. [76] See Def.’s Br., ECF Nos. 53 & 54, at 31 ("Commerce verified the data submitted by Regal, ensured there were no discrepancies, and calculated a dumping margin based on the verified information that supported Regal’s revocation request.... To the extent [AHSTAC] wishes to pursue claims regarding importer misclassification, Commerce previously has explained that it generally refers such matters to [Customs].”) (quoting AR6 I & D Mem. cmt. 7 at 38 ("[Customs] is the U.S. government authority responsible for determining whether the importer has properly classified merchandise as subject or non-subject at time of entry.”) (citation omitted); AR5 I & D Mem. cmt. 1 at 4 (“[Cjomplaints of deliberate misclassification of entries or fraudulent activity regarding entries into the United States should be properly lodged with [Customs].”) (citing Globe Metallurgical Inc. v. United States, — CIT -, 722 F.Supp.2d 1372, 1381 (2010))); Globe Metallurgical,CIT at -, 722 F.Supp.2d at 1381 (noting that “Commerce’s recognition of [Customs]’s authority to investigate fraud, gross negligence, or negligence involving entries of merchandise” is consistent with 19 U.S.C. § 1592).