An Giang Fisheries Import & Export Joint Stock Co. v. United States

Kelly, Judge: This action is before the court on a motion for judgment on the agency record challenging various aspects of the U.S. Department of Commerce's ("Department" or "Commerce") final determination in the eleventh administrative review of the antidumping duty ("ADD") order covering certain frozen fish fillets from the Socialist Republic of Vietnam ("Vietnam"). SeeCertain Frozen Fish Fillets From [Vietnam], 81 Fed. Reg. 17,435 (Dep't Commerce Mar. 29, 2016) (final results and partial rescission of [ADD] administrative review; 2013-2014) ("Final Results"), and accompanying Certain Frozen Fish Fillets from [Vietnam]: Issues and Decision Memorandum for the Final Results of the Eleventh [ADD] Administrative Review; 2013-2014, A-552-801, (Mar. 18, 2016), ECF No. 20-3 ("Final Decision Memo"); see also Certain Frozen Fish Fillets From [Vietnam], 68 Fed. Reg. 47,909 (Dep't Commerce Aug. 12, 2003) (notice of [ADD] order) ("ADD Order"). Plaintiffs, An Giang Fisheries Import and Export Joint Stock Company, Cuu Long Fish Joint Stock Company, C.P. Vietnam Corporation, GODACO Seafood Joint Stock Company, International Development and Investment Corporation, Seafood Joint Stock Company No. 4-Branch Dong Tam Fisheries Processing Company, Thuan An Production Trading and Services Co., Ltd., and Viet Phu Foods and Fish Corporation (collectively, "Plaintiffs"), commenced this action 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).1 See Summons, Apr. 28, 2016, ECF *1365No. 1. Plaintiffs challenge several determinations made by Commerce in the Final Results as unsupported by substantial evidence or otherwise not in accordance with law. See Mem. Law Supp. Pls.' Rule 56.2 Mot. J. Upon Agency R. at 1-2, 5-54, Nov. 4, 2016, ECF No. 31 ("Pls.' Br."); see also 19 U.S.C. § 1516a(b)(1)(B)(i). First, Plaintiffs challenge Commerce's decision to apply facts otherwise available to calculate Hung Vuong Group's ("HVG") and Thuan An Production Trading & Services Co., Ltd.'s ("TAFISHCO") dumping margins. Pls.' Br. at 5-31. Second, Plaintiffs challenge Commerce's calculation of HVG's farming factors as part of its application of facts otherwise available. Id. at 1, 31-32. Third, Plaintiffs challenge Commerce's decision to assign partial facts otherwise available with an adverse inference ("AFA") to TAFISHCO.2 Id. at 1, 33-36.3 Fourth, Plaintiffs challenge Commerce's selection of surrogate value ("SV") data sources to value various factors of production ("FOP") used to produce the subject merchandise, including fish feed, fingerlings, packing tape, water, and fish waste by-product. Pls.' Br. at 1-2, 36-54. For the reasons set forth below, the court sustains Commerce's SV selections for fish feed, fingerlings, water, fish waste by-product, and packing tape. The court also sustains Commerce's application of facts otherwise available to HVG and TAFISHCO, and Commerce's application of partial AFA to products produced by tollers for TAFISHCO. However, the court remands Commerce's calculation of HVG's farming factors. BACKGROUND Commerce initiated this eleventh ADD administrative review covering subject imports entered during the period of review ("POR"), August 1, 2013 through July 31, 2014. See Initiation of Antidumping and Countervailing Duty Administrative Reviews, 79 Fed. Reg. 58,729, 58,731 -32 (Dep't Commerce Sept. 30, 2014) (initiation of [ADD] administrative review in Certain Frozen Fish Fillets from [Vietnam], A-552-801). Commerce subsequently selected Vinh Hoan Corporation ("Vinh Hoan") and HVG as mandatory respondents in this review. See [Certain Frozen Fish Fillets from Vietnam:] Selection of Respondents *1366for Indiv. Review at 1, 4-7, A-552-801, PD 33, bar code 3240494-01 (Nov. 7, 2014).4 However, on November 25, 2014, Commerce rescinded the administrative review as it pertained to Vinh Hoan, and on December 1, 2014 selected TAFISHCO as a mandatory respondent. See [Certain Frozen Fish Fillets from Vietnam:] Second Selection of Resp't for Indiv. Review, A-552-801, PD 67, bar code 3244597-01 (Dec. 1, 2014). Commerce published its preliminary results on September 14, 2015. See Certain Frozen Fish Fillets From [Vietnam], 80 Fed. Reg. 55,092 (Dep't of Commerce Sept. 14, 2015) (preliminary results and partial rescission of the [ADD] administrative review; 2013-2014) ("Prelim. Results"), and accompanying Certain Frozen Fish Fillets from [Vietnam]: Decision Memorandum for the Preliminary Results of the 2013-2014 [ADD] Administrative Review, A-552-801, PD 260, bar code 3301537-01 (Aug. 31, 2015) ("Prelim. Decision Memo"). Commerce preliminarily calculated weighted-average dumping margins of $0.36 per kilogram for HVG,5 $0.84 per kilogram for TAFISHCO, and $0.60 per kilogram for the "Separate Rate Companies," which includes Plaintiffs C.P. Vietnam Corporation, Cuu Long Fish Joint Stock Company, GODACO Seafood Joint Stock Company, International Development and Investment Corporation, Seafood Joint Stock Company No. 4-Branch Dong Tam Fisheries Processing Company, and Viet Phu Foods and Fish Corporation. See Prelim. Results, 80 Fed. Reg. at 55,094 ; see also Prelim. Decision Memo at 7 (providing a list of companies who submitted separate rate applications). In its final determination, Commerce calculated final weighted-average dumping margins of $0.41 per kilogram for HVG, $0.97 per kilogram for TAFISHCO, and the Vietnam-wide rate of $2.39 per kilogram for other exporters who did not qualify for a separate rate.6 Final Results, 81 Fed. Reg. at 17,436 -37. JURISDICTION AND STANDARD OF REVIEW The court has jurisdiction pursuant to section 19 U.S.C. § 1516a(a)(2)(B)(iii) and 28 U.S.C. § 1581(c) (2012), which grant the Court authority to review actions contesting the final determination in an administrative review of an antidumping duty order. "The court shall hold unlawful any determination, finding, or conclusion found ... to be unsupported by substantial evidence on the record, or otherwise not in accordance with law." 19 U.S.C. § 1516a(b)(1)(B)(i). *1367DISCUSSION I. Application of Facts Otherwise Available to HVG and TAFISHCO Plaintiffs challenge Commerce's decision to apply facts otherwise available to HVG and TAFISHCO as unsupported by substantial evidence and not in accordance with law. Pls.' Br. at 5-31. Plaintiffs claim that Commerce's request for control-number ("CONNUM")7 specific information was unreasonable because it placed an impossible burden on HVG and TAFISHCO, id. at 8-23, and that the reporting methodologies used by HVG and TAFISHCO to calculate the FOP databases did not produce distortive results. Id. at 23-29. Defendant responds that Commerce's decision to apply facts otherwise available is supported by substantial evidence and is in accordance with law because HVG and TAFISHCO knew that they were to report on a CONNUM-specific basis, and were given several opportunities to comply. Def.'s Resp. Pls.' Mots. J. Upon Agency R. at 14-24, June 9, 2017, ECF No. 47 ("Def.'s Resp. Br."). Further, Defendant argues that Commerce's determination that FOPs reported by HVG and TAFISHCO were distorted is supported by substantial evidence and is in accordance with law. Id. at 26-31. For the reasons that follow, Commerce's use of facts otherwise available to calculate HVG's and TAFISHCO's normal value is supported by record evidence and is in accordance with law. Pursuant to 19 U.S.C. § 1677e(a)(1) and (2), Commerce has the power to use facts otherwise available in making its determinations when it lacks necessary information on the record. See 19 U.S.C. § 1677e(a). However, prior to applying facts otherwise available, Commerce must explain why the information it does have is insufficient and provide, where practicable, the non-complying party an opportunity to comply. 19 U.S.C. § 1677m(d). In non-market economies ("NME"), where Commerce constructs the normal value based on the FOPs used to produce the subject merchandise, Commerce will resort to facts otherwise available if there is insufficient information on the record to value an FOP.8 Final Decision Memo at 9-10. In its antidumping questionnaire, Commerce provides instructions to respondents on how to report FOPs, and prioritizes CONNUM-specific reporting.9 Id. at 10; see *1368also [ Antidumping ] Questionnaire at D-2, PD 39, bar code 3240511-01 (Nov. 7, 2014) (reproducing, in relevant part, an unmarked copy of Section D of Commerce's Antidumping Questionnaire, addressing CONNUM) ("Antidumping Questionnaire"). Using CONNUMs, Commerce "defines the key physical characteristics of the subject merchandise as those that are commercially meaningful in the U.S. marketplace, and [have an] impact [on] costs of production." Id. (citing e.g., Large Residential Washers from the People's Republic of China [ ("PRC") ], 81 Fed. Reg. 1,398, 1,399 (Dep't Commerce Jan. 12, 2016) (initiation of less-than-fair-value investigation); Stainless Steel Wire Rod from Sweden, 73 Fed. Reg. 12,950 (Dep't Commerce Mar. 11, 2008) (final results of [ADD] review) and accompanying Issues and Decision Memorandum for the Final Results of the Administrative Review of Stainless Steel Wire Rod from Sweden at 8-14, A-401-806, (Mar. 5, 2008), available at http://ia.ita.doc.gov/frn/summary/sweden/E8-4824-1.pdf (last visited Feb. 8, 2018) ). In the final determination, Commerce found that HVG and TAFISHCO failed to report FOPs on a CONNUM-specific basis and failed to accurately report water soaking levels of the fillets sold in the United States. Final Decision Memo at 13-15. As a result, Commerce disregarded the FOP databases submitted by TAFISHCO and HVG, and instead used data prepared by tollers for TAFISHCO that was specific to the subject merchandise. Id. at 15 (citing Prelim. Decision Memo at 21); see id. at 13-15. Commerce explained that "different product forms, i.e. [,] whole fish versus fish fillet, have vastly different yields that distort the FOP data when they are included in the FOP denominator as if they have the same yield."10 Final Decision Memo at 14 (citation omitted). As a result of HVG and TAFISHCO's refusal to account for the yield differences in their reported FOPs, Commerce was unable to calculate the level of distortion created by the inclusion of non-subject merchandise. Final Decision Memo at 14. As to HVG's and TAFISHCO's failure to report an accurate water weight, Commerce learned during this POR "that HVG and [TAFISHCO] apply very different soaking formulae based on the requirements of different markets." Final Decision Memo at 14 (citing [Petitioners'] Comments Re: Prelim. [ADD] Margin Calculations for the HVG Respondents & TAFISHCO at 21-27, PD 230, bar code 3297374-01 (Aug. 10, 2015) ). The variance in soaking formula resulted in frozen subject merchandise bound for the United States being soaked at a lower percentage than frozen merchandise bound for other markets. Id. As a result, Commerce determined that it could not make an apples-to-apples comparison between FOPs and sales. More specifically, by expanding the denominator to include products that were soaked to a greater degree, i.e., by adding water to the denominator of all reported FOPs, HVG and [TAFISHCO] underreported all of their FOPs for subject merchandise [s]old to the United States. *1369Final Decision Memo at 14. Respondents knew that they were to report what percentage of the subject-merchandise's weight, if any, was the result of adding, for example, ice, water, or glazing to the product. Id. However, as Commerce explains, neither TAFISHCO nor HVG accounted for added water weight in their submissions, although the data was available to them. Id. at 15. Plaintiffs acknowledge that the actions of soaking and tumbling subject merchandise result in added weight gain. Pls.' Br. at 25-26. However, Plaintiffs fail to explain why their failure to account for the variance in the amount of water added to the fillets based on the market the subject merchandise is bound for is not distortive.11 Plaintiffs also make several arguments challenging Commerce's request for CONNUM-specific reporting. See Pls.' Br. at 6-23, 27-31. Plaintiffs argue that CONNUM-specific reporting has never been required of it in the past, see Pls.' Br. at 6-9, that Commerce's request for CONNUM-specific data was fundamentally unfair, id. at 9-23, and that in previous reviews Commerce has concluded that respondents' reporting methodologies produce "reliable" FOPs. Id. at 27-29; see also id. at 23. Plaintiffs' reliance on Commerce's past practice is misplaced because Commerce did put respondents such as HVG and TAFISHCO on notice of future enforcement of the CONNUM-specific reporting requirement as early as the eighth administrative review of the ADD Order. See Certain Frozen Fish Fillets from [Vietnam]: Issues and Decision Memorandum for the Final Results of the Eighth Administrative Review and Aligned New Shipper Reviews at 43-44, A-552-801, (Mar. 13, 2013), available at http://ia.ita.doc.gov/frn/summary/vietnam/2013-06550-1.pdf (last visited Feb. 8, 2018) ("Eighth Admin. Review IDM"). During the ninth review, Commerce accepted CONNUM-incompliant data and excused the respondents' non-compliance out of notice concerns.12 See Certain Frozen Fish Fillets from [Vietnam]: Issues and Decision Memorandum for the Final Results of the Ninth Administrative Review and Aligned New Shipper Review at 73-74, A-552-801, (Mar. 28, 2014), available at http://ia.ita.doc.gov/frn/summary/vietnam/2014-07714-1.pdf (last visited Feb. 8, 2018) ("Ninth Admin. Review IDM").13 This is the eleventh administrative review *1370of the ADD Order. See Final Results, 81 Fed. Reg. at 17,435. Given the advance notice afforded to respondents, the court cannot find that Commerce's request for CONNUM-specific reporting, here, was unreasonable or a practice reversal.14 Plaintiffs also claim that Commerce's identification of product form as a physical characteristic reflected in the CONNUM is commercially irrelevant. See Pls.' Br. at 9-10. It is reasonably discernible that Commerce found that product form is relevant because different product forms have different yields and therefore the lack of CONNUM-specific reporting leads to distortions. See Final Decision Memo at 15 (explaining that the tollers' data was free from distortions caused by the mixing of product forms that have different yields).15 Plaintiffs contend that, despite HVG's and TAFISHCO's best efforts, it was not possible for them to comply with Commerce's data request because they do not track sales and FOPs on the basis of the product characteristics identified by CONNUM.16 See Pls.' Br. at 17-23. In the final *1371determination, Commerce rejected the impossibility argument, explaining that HVG and TAFISHCO can still track CONNUM-specific information in the manner the Department requested, even if the documents are not maintained in such a way in the respondents' "normal course of business[.]" Final Decision Memo at 13. Commerce's decision to require CONNUM-specific reporting and to apply facts otherwise available to HVG and TAFISHCO is reasonable. II. Commerce's Application of Facts Available to HVG's Farming Factors Plaintiffs also challenge Commerce's application of facts otherwise available for HVG's Farming Factors by adjusting the FOP denominator as both unsupported by substantial evidence and not in accordance with law. Pls.' Br. at 31-32. Further, Plaintiffs argue that, even if Commerce correctly adjusted the farming FOP denominator, Commerce should have adjusted the numerator as well. Id. at 32. Plaintiffs claim that Commerce's adjustment to the denominator, without changing the numerator, wrongfully inflates HVG's farming FOPs. See id. Defendant claims that Plaintiffs failed to exhaust their argument as to the numerator, Def.'s Resp. Br. at 32-33, and argue that Commerce's determination to modify the denominator is necessary to ensure a proper comparison of FOPs and U.S. prices. Def.'s Resp. Br. at 32 (citing Final Decision Memo at 17). For the reasons detailed below, the court remands Commerce's decision to adjust HVG's farming FOP denominator without making a parallel adjustment to the numerator. The basis for Commerce's decision to apply facts otherwise available to HVG's farming factors, see Final Decision Memo at 16-17, and Plaintiffs' challenge to that decision are identical to the basis of, and Plaintiffs' challenge to, Commerce's demand for CONNUM-specific reporting. See Pls.' Br. at 31-32. Specifically, Commerce explained that respondents' reported data was distorted because they included non-subject merchandise in the denominator that had different yields and fillets that have greater weight gain due to soaking in preservatives. Final Decision Memo at 17; see id. at 13-15. Commerce's decision to apply facts otherwise available with respect to HVG's farming factors is supported by substantial evidence. In applying facts otherwise available, Commerce resorted to using data from tollers. Final Decision Memo at 16-17. However, the processing FOPs in the tollers' data are reported on a subject merchandise basis, while the farming FOPs, as reported by HVG, are reported on a whole live fish harvested basis. See id. at 17. To remedy the unparalleled denominators, Commerce converted HVG's denominator, which was based on whole live fish harvested, by what Commerce called, the "shank equivalent conversion factor." Id. Commerce explained the conversion factor as "simply the whole live fish to subject merchandise (shank fillets) FOP that is used for the processing factors [.]"17 Id. In using the conversion factor, Commerce arrived at what it called the "shank equivalent of the total harvested fish." See id. at 16 (citation omitted). Commerce then added the farming FOPs from all farming *1372activities at HVG, Agifish, and Europe JSC, and used the total as the numerator. Id. at 16 (citations omitted). Commerce, however, fails to explain why adjusting the denominator to represent the "shank-equivalent" does not require adjusting the numerator to reflect the "shank-equivalent" farming FOP. Defendant argues that although HVG now argues that Commerce's changes to the denominator would require a modification to the numerator, HVG did not actually raise the argument at the agency level, and therefore, HVG's argument has not been exhausted. Def.'s Resp. Br. at 32 (citing [Respondents'] Case Br. at 20-21, PD 344, bar code 3441848-01 (Fed. 12, 2016) ("Respondents' Agency Case Br.").18 However, Plaintiffs did adequately present this argument below, arguing specifically that [t]he reason the adjustment methodology does not make since [sic] is because it incorrectly assumes that the reported farming FOPs were only consumed to raise the [ [ ] ] kg of the so-called shank equivalent live fish. However, to the contrary, the reported farming FOP were consumed to raise the total POR harvested fish-and not just the shank equivalent fish. Accordingly, the appropriate denominator for Agifish's farming FOP is the total harvested fish quantity of [ [ ] ] kg. Respondents' Agency Case Br. at 26 (emphasis omitted). In their brief before the agency, Plaintiffs specifically recognize that the farming FOPs in the numerator are overstated, when compared to Commerce's adjusted denominator, because the numerator contains data for all farming factors consumed in raising the total number of harvested fish, across all of HVG's companies, during the POR. See id. Defendant's only response to Plaintiffs' argument is that Plaintiffs fail to support their position that Commerce's analysis leads to distortions or that adjusting the numerator would resolve such distortions. Def.'s Resp. Br. at 31-32. Therefore, the issue is remanded to Commerce for further explanation or consideration. III. Application of Partial AFA to TAFISHCO Plaintiffs challenge Commerce's decision to apply as partial AFA the Vietnam-wide rate of $2.39/kg to calculate TAFISHCO's dumping margin as contrary to law and unsupported by substantial evidence. Pls.' Br. at 33-36; see also Final Results, 81 Fed. Reg. at 17,437. Defendant supports Commerce's decisions by arguing that although TAFISHCO attempted to secure the cooperation of its tollers, it did not induce cooperation, and as a result, "the uncooperative tollers, who accounted for a significant percentage of [TAFISHCO's] total production, could benefit from [TAFISHCO's] cooperation." Def.'s Resp. Br. at 38 (citation omitted); see id. at 36-38. Commerce's determination to apply partial AFA to TAFISHCO was reasonable. In the final determination, Commerce applied AFA to TAFISHCO's unaffiliated Toller A19 and Toller B,20 and partial AFA to TAFISHCO itself.21 Final Decision Memo at 23; see also Verification: [Letters to the DOC Regarding Efforts to Contact *1373TAFISHCO's Tollers], PD 278, bar code 3305863-01 (Sept. 14, 2015) (identifying the uncooperative tollers) ("TAFISHCO Sept. 14, 2015 Letter Re: Tollers"). Commerce acknowledged that TAFISHCO was cooperative. Final Decision Memo at 23. However, at verification, Commerce "learned that during the POR, when cash deposit rates for frozen fish fillet producers from Vietnam went up," TAFISHCO retained its low cash deposit rate, and was therefore in a "good position" to export frozen fish fillets. Final Decision Memo at 23 (citing Verification of the Sales and [FOPs] Resp. of [TAFISHCO] in the 2013-2014 Administrative Review of Certain Frozen Fish Fillets from [Vietnam] at 7, PD 336, bar code 3437882-01 (Feb. 2, 2016) ). Commerce then relied on this trend to support its conclusion that TAFISHCO should have induced Tollers A and B to cooperate as its own cooperation with Commerce benefited the tollers. Id. Considering this information, together with Commerce's finding that the uncooperative tollers' involvement in the total production was significant, Commerce decided that partial AFA should be applied to TAFISHCO. Id. at 22-23. Pursuant to 19 U.S.C. § 1677e(a), (b) Commerce may use facts otherwise available and may subsequently apply an adverse inference to those facts if it "finds that in interested party has failed to cooperate by not acting to the best of its ability[.]" See 19 U.S.C. § 1677e(a), (b). "The statute does not provide an express definition of 'the best of its ability.' " Nippon Steel Corp. v. United States, 337 F.3d 1373, 1382 (Fed. Cir. 2003). However, as the Nippon Steel court explained, a respondent acts to "the best of its ability" when it "do[es] the maximum it is able to do." Id. Commerce may apply an adverse inference, notwithstanding the respondent's motivation or intent for noncompliance, id. at 1383, and even if doing so may "have collateral consequences for a cooperating party." Mueller Comercial de Mexico, S. de R.L. de C.V. v. United States, 753 F.3d 1227, 1236 (Fed. Cir. 2014). Plaintiffs do not dispute that Tollers A and B failed to cooperate. See Pls' Br. at 34-35 (providing a history of correspondence between TAFISHCO and the two unaffiliated tollers to secure the tollers' cooperation, ultimately culminating in failure). Plaintiffs, however, claim that record evidence demonstrates their very best efforts in attaining the tollers' cooperation. Id. (citing TAFISHCO Sept. 14, 2015 Letter Re: Tollers; Verification of TAFISHCO and Its Tollers [Letter to DOC Re: Toller [ [ ] ] ], PD 288, bar code 3307959-01 (Sept. 22, 2015); Verification of TAFISHCO Toller [ [ ] ] [Letter to the DOC], PD 304, bar code 3308756-01 (Sept. 25, 2015) ). Plaintiffs therefore argue that Commerce should not have applied partial AFA given that TAFISHCO's did exert its "best efforts" to secure cooperation, Pls.' Br. at 35, and 1) did obtain the cooperation of two substitute tollers; 2) "the cooperating tollers provided over 80 percent of the product that TAFISHCO had toll processed during the [POR]"; and 3) Tollers A and B were not affiliated with TAFISHCO. Id. at 36.22 *1374Commerce recognizes TAFISHCO's efforts to contact the tollers and obtain their cooperation. Final Decision Memo at 23. However, Commerce notes, TAFIHSCO could have done more to induce cooperation, "for example, [by] refusing to do business with Tollers A and B, who they had an ongoing business relationship with, unless they cooperated." Id. at 23 (citations omitted). The court in Mueller found that it was appropriate for Commerce to use deterrence as a justification for applying AFA to calculate a cooperating party's dumping rate margin, "as long as the application of [deterrence] policies is reasonable on the particular facts and the predominant interest in accuracy is properly taken into account as well." Mueller, 753 F.3d at 1233. Here, as in Mueller, Commerce found that uncooperative tollers could escape their own high cash deposit by selling through TAFISHCO, the cooperating party, and further that the cooperating party could have, but did not, use its influence to induce cooperation. See Final Decision Memo at 23; see also Mueller, 753 F.3d at 1235. Therefore, Commerce's determination is sustained. IV. Commerce's Analysis of Specific Surrogate Values Plaintiffs challenge Commerce's SV data selections for fish feed, fingerlings, packing tape, water, and fish waste by-products. See Pls.' Br. at 36-54. Defendant refutes all of these challenges and argues that Commerce's final determination should be sustained in all respects. See Def.'s Resp. Br. at 39-53. For the reasons that follow, the court sustains Commerce's SV selections for fish feed, fingerlings, water, fish waste by-product, and packing tape. A. Legal Framework In antidumping proceedings involving non-market economies ("NME"), Commerce generally calculates normal value using the FOPs used to produce the subject merchandise and other costs and expenses. 19 U.S.C. § 1677b(c)(1). Commerce will value respondents' FOPs using the "best available information regarding the values of such factors in a market economy country or countries considered to be appropriate by [Commerce]." 19 U.S.C. § 1677b(c)(1)(B). To the extent possible, Commerce uses FOPs from market economy countries that are: "(A) at a level of economic development comparable to that of the nonmarket economy country, and (B) significant producers of comparable merchandise." 19 U.S.C. § 1677b(c)(4). Commerce's regulatory preference is to "value all factors in a single surrogate country." 19 C.F.R. § 351.408(c)(2) (2014).23 *1375Commerce's methodology for selecting the best available information evaluates data sources based upon their: (1) specificity to the input; (2) tax and import duty exclusivity; (3) contemporaneity with the period of review; (3) representativeness of a broad market average; and (5) public availability. See Import Admin., U.S. Dep't Commerce, Non-Market Economy Surrogate Country Selection Process, Policy Bulletin 04.1 (Mar. 1, 2004), available at http://ia.ita.doc.gov/policy/bull04-1.html (last visited Feb. 8, 2018); Final Decision Memo at 34. Commerce uses the same methodology to calculate SV of by-products generated during the production process, and that offset production costs incurred by a respondent. See Final Decision Memo at 46-47, 50; see also Tianjin Magnesium Int'l Co., v. United States, 34 CIT 980, 993, 722 F.Supp.2d 1322, 1336 (2010) ; Guangdong Chems. Imp. & Exp. Corp. v. United States, 30 CIT 1412, 1422-23, 460 F.Supp.2d 1365, 1373-74 (2006). Commerce's practice for selecting the best available information to value individual FOPs favors selecting a data source that satisfies the breadth of its selection criteria, where possible. Final Decision Memo at 50 (citing e.g., Fifth Administrative Review of Certain Frozen Warmwater Shrimp from the [PRC]: Issues and Decision Memorandum for the Final Results at 5-10, A-570-893, (Aug. 12, 2011), available at http://ia.ita.doc.gov/frn/summary/prc/2011-21259-1.pdf (last visited Feb. 8, 2018) ). B. Fish Feed Plaintiffs challenge Commerce's decision to value fish feed using prices for floating feed only, claiming that by doing so Commerce failed to use the most specific value and its decision was not in accordance with law. Pls.' Br. at 36-39. Defendant argues that record evidence supports Commerce's decision to use Indonesian prices for floating fish feed, and that "HVG cannot now claim that Commerce's determination was not specific to its input," as HVG "failed to build the record" demonstrating its use of any other type of fish feed. Def.'s Resp. Br. at 42 (citations omitted). The court agrees with the Defendant. In the final determination, Commerce valued HVG's pangasius feed using floating feed prices because Commerce found those prices specific to fish feed consumed during the POR. Final Decision Memo at 35. Commerce found that HVG consumed floating pellets during the POR, based on usage guidance HVG submitted, HVG's website, and statements HVG made in its brief before the agency. Id. (citing [An Giang] Suppl. Section A, C, & D Questionnaire Resp. at Ex. 11, PD 159, bar code 3275379-02 (May 7, 2015) ("An Giang Questionnaire Resp. Re: Fish Feed"); [Petitioners'] Submission of Info. to Rebut, Clarify, and Correct Info. in Respondents' Questionnaire Resp. at Ex. 8, PD 190, bar code 3282538-06 (June 5, 2015) ("Petitioners' Rebuttal Factual Info."); Respondents' Rebuttal Agency Case Br. at 3) ). Commerce found that there was no record evidence to suggest that HVG used sinking fish feed anytime during the POR, and that it would not consider fish feed consumption data provided in past reviews "as the record of each review stands on its own." Id. (citing Issues and Decision Memorandum for the Final Results of the [CVD] Administrative Review of Certain Kitchen Appliance Shelving and Racks from the [PRC] at 29, A-570-942, (Apr. 4, 2012), available at http://ia.ita.doc.gov/frn/summary/prc/2012-8727-1.pdf (last visited Feb. 8, 2018) ). No party challenged the Indonesian price quotes for floating fish feed as not contemporaneous, publicly available, representative of a broad market average, and tax and duty exclusive.24 Id. *1376Plaintiffs acknowledge that the record supports the determination that the respondents used floating fish feed, but not the conclusion "that the respondents only used floating fish feed during the [POR]." Pls.' Br. at 38 (emphasis in original).25 It is the respondents' burden to populate the record with all relevant information.26 See QVD Food Co. v. United States, 658 F.3d 1318, 1324 (Fed. Cir. 2011). Here, the record before Commerce showed that during the POR, HVG consumed floating fish feed and not any other type or combination of feed. Final Decision Memo at 35; see also An Giang Questionnaire Resp. Re: Fish Feed at Ex. 11; Petitioners' Rebuttal Factual Info. at Ex. 8. Therefore, it was reasonable for Commerce to consider floating fish feed prices as the best available information. C. Fingerlings Plaintiffs challenge Commerce's use of the 2012 affidavit from Dr. Djumbuh Rukmono, an official from the Indonesian Ministry of Marine Affairs and Fisheries. See Pls.' Br. at 40-42; [Petitioners'] Surrogate Country Comments & Submission of Proposed Factor Values at Ex. I-10B, PD 175, bar code 3278422-05 (May 26, 2015) ("2012 Rukmono Affidavit"). Plaintiffs claim that the 2012 Rukmono Affidavit contains "significant errors in the conversion of fish lengths-to-weights" and is therefore "not based on record evidence[.]" Pls.' Br. at 40. Defendant argues that HVG failed to exhaust this argument below and should be precluded from raising it now. Def.'s Resp. Br. at 43 (citing Respondents' Agency Case Br. at 28-29). Defendant-Intervenors' join in opposition. Def.-Intervenors' *1377Resp. Br. at 23-24. The court agrees with Defendant and Defendant-Intervenors and sustains Commerce's choice to use the 2012 Rukmono Affidavit to value respondents' fingerlings as reasonable. If a party fails to exhaust available administrative remedies before the agency, " 'judicial review of Commerce's actions is inappropriate.' " See Consol. Bearings Co. v. United States, 348 F.3d 997, 1003 (Fed. Cir. 2003) (citation omitted). This Court has generally taken a " 'strict view' of the requirement that parties exhaust their administrative remedies before the Department of Commerce in trade cases." See Corus Staal BV v. United States, 502 F.3d 1370, 1379 (Fed. Cir. 2007) (citations omitted). In its brief to the agency, HVG did not contest the quality of the data in the 2012 Rukmono Affidavit, but only asked that Commerce instead use the 2013/14 affidavit from the same source for the final results, as it was contemporaneous with the POR. See Respondents' Agency Case Br. at 28-29; see generally [Petitioners' Second SV] Submission at Ex. I-2, PD 209, bar code 3292359-01 (July 20, 2015) (reproducing the 2013/2014 Rukmono Affidavit). It is only now, before this court, that Plaintiffs claim the data in the 2012 Rukmono Affidavit wrongfully computes prices per fingerling length. Pls.' Br. at 40-42, Exs. 1, 2. Absent exceptional circumstances, it would be inconsistent with the purposes of the exhaustion doctrine to require Commerce to explain a challenge to its findings that was not raised at the administrative level. See Consol. Bearings, 348 F.3d at 1003 ; Corus Staal, 502 F.3d at 1379. Therefore, the court will not address Plaintiffs' argument that the 2012 Rukmono Affidavit contains "significant errors in the conversion" which "yield[ ] an extremely wide range of per kilogram prices for the same input[.]" Pls.' Br. at 40; id. at 41. Accordingly, Commerce's determination regarding fingerlings is sustained. D. Water Plaintiffs challenge Commerce's decision to value the respondents' water input using the value for treated water from Pam Jaya, an Indonesian water utility company. Pls.' Br. at 46-50. Plaintiffs argue that record evidence demonstrates that "[both Indonesian and Vietnamese subject merchandise] producers do not purchase ready-to-use water for their production activities." Id. at 46 (emphasis omitted); see also [Respondents'] Surrogate Values Submission at Ex. 32-3, PD 170, bar code 3278288-03 (May 22, 2015) ("Soetrisno Letter") (containing a copy of a letter from the Director of Aquaculture Fisheries, Coco Kokarkin Soetrisno, providing information regarding water usage practices of patin farmers and producers in Indonesia).27 Instead, Plaintiffs argue, record evidence supports the conclusion that subject merchandise producers pump raw river water, and then treat it with chemicals. Pls.' Br. at 47. Defendant argues that Commerce's decision to rely on the Pam Jaya data is supported by substantial evidence. Def.'s Resp. Br. at 47-50. The court agrees with the Defendant, and Commerce's determination as to water is sustained. In the final determination, Commerce explains that its "practice is to value all inputs consumed in the production of the merchandise under consideration [,]" even if the FOP is obtained at no cost. Final Decision Memo. at 39 (citing Issues and Decision Memorandum for the Final Results *1378of the [ADD] New Shipper Review: Certain Non-Frozen Apple Juice Concentrate from the [PRC] at 2, A-570-855, (Dec. 20, 2010), available at http://ia.ita.doc.gov/frn/summary/prc/2010-32675-1.pdf (last visited Feb. 8, 2018) ). Commerce valued respondents' water using Pam Jaya data, which Commerce explained was most specific to the input and fit Commerce's preference to "valu[e] FOPs within a single surrogate country." Final Decision Memo at 39; see also 19 C.F.R. § 351.408(c)(2). Further, Commerce explained that it relied on the Pam Jaya data because record evidence did not show that surrogate companies in Indonesia, like PT Dharma Samudera Fishing Industries Tbk, specifically "pumped water from wells or from a nearby river for free." Final Decision Memo at 39. Commerce considered, but was not persuaded by, the Soetrisno Letter, which speaks of general practices of patin farmers and producers in Indonesia but does not identify individual companies who are able to pump water for free, either from a well or river. See Final Decision Memo at 39; see also Soetrisno Letter. It is reasonably discernible that Commerce did not find the Soetrisno Letter compelling because it addresses general practices within Indonesia and did not constitute substantial evidence that surrogate companies in Indonesia pump water at no cost. Final Decision Memo at 39. Therefore, Commerce's decision to value water on the basis of the Pam Jaya data is supported by substantial evidence. E. Fish Waste By-Product Plaintiffs challenge, as flawed, Commerce's decision to value fish waste by-product using data Commerce obtained in 2015 from the company Adib Food Supplies, which provides prices for the period of 2013-2014 ("2013/14 Adib price quotes"). Pls.' Br. at 50-54. Plaintiffs argue that this data is not a price quote, but rather historical data. Id. at 51-52. Defendant states that Commerce's reliance on the 2013/14 Adib price quotes constitutes the best available information because it is publicly available, tax exclusive, reflects pick-up prices, and breaks out the contents of food waste in a manner similar to TAFISHCO's practices. Def.'s Resp. Br. at 50-51. On the record presented, Commerce's choice is reasonable and is therefore sustained. Commerce explained that the 2013/14 Adib price quotes are more specific because they cover the full range of fish waste by-products sold by TAFISHCO. See Final Decision Memo at 51. In the final determination, Commerce states that TAFISHCO defines its food waste as "including head, bone, blood, [and] skin." Id. (quoting [TAFISHCO] Suppl. Section A, C & D Questionnaire Resp.-Part I at 17, PD 155, bar code 3275296-01 (May 7, 2015) ("TAFISHCO Resp. Re: Fish Waste Components") ). The 2013/14 Adib price quotes broke out the component parts of fish waste similarly to TAFISHCO, and "list[ed] prices for head, bones and skin." Id. (citing Petitioners['] Second Surrogate Submission-Part 1 at Exs. I-7, I-8, PD 208, bar code 3292359-02 (July 20, 2015) ). In comparison, "[t]he 2015 Adib/Alam Jaya price quotes are for fish bait and fish waste, and do not break out the individual components head, bone, blood, skin." Id. Furthermore, the 2015 Adib/Alam Jaya price quotes encompass fish waste that was "further processed and packed for international trade, whereas [TAFISHCO's] fish waste [was] not." Id. Plaintiffs argue that the 2013/14 Adib price quotes are "post facto-prepared historical price lists," Pls.' Br. at 51, and not commercial offers to sell. Id. at 51-52. However, it is reasonably discernable that Commerce agreed with the petitioners that record evidence demonstrated the existence *1379of bona fide sales.28 See Final Decision Memo at 51 (providing Commerce's reasoning for why the 2013/14 price quotes provided reliable data); see also id. at 49 (citing Petitioners' Rebuttal Br. Pertaining to [SVs], PD 359, bar code 3444238-01 (Feb. 22, 2016) ). The court will not reweigh the evidence. Plaintiffs also argue that "even if Commerce's data source is appropriately used, Commerce incorrectly applied it" to TAFISHCO. Pls.' Br. at 52. Specifically, Plaintiffs challenge what Commerce determined would be classifiable as fish waste by-product. Plaintiffs claim that Commerce improperly limited TAFISHCO's fish waste to head, bone, skin, and blood. Id. However, Commerce's determination is supported by record evidence. Plaintiffs do not point to record evidence demonstrating what components other than head, bone, skin and blood went into TAFISCHO's fish waste by-product. In fact, on the record there is evidence showing that TAFISHCO classified items such as "nuggets, stomach, bladder (maw), fish oil and fish meal separately" from its fish waste. Final Decision Memo at 51 (citing TAFISHCO Resp. Re: Fish Waste Components at 17). Commerce's decision to rely on known components is reasonable. Commerce then looked at what data was available on the record for head, bone, skin and blood. See [SVs] for the Final Results [Memo] at 1-2, PD 373, bar code 3451526-01 (Mar. 18, 2016) ("Final SV Memo"). Commerce explained that because there were "no surrogate values on the record for fish blood," it did not include fish blood in its SV calculations. Id. at 1. As for fish bone, Commerce explained that the fish bone prices "may [have been] overstated due [to] the inclusion of [fin and meat from the belly]."Id. at 2 (citation omitted). Commerce was therefore left with a finite number of fish waste components, i.e., fish head and fish skin, for which it could reasonably discern prices. Id. at 1-2. Therefore, to value fish waste by-product, Commerce valued the fish wastes reported by [TAFISHCO] using prices for fish head and fish skin.... [and subsequently] weighted these prices by record information found in the verification reports which indicate that the amount of fish skin produced from a fillet, is approximately ten percent of the fish waste produced. Final SV Memo at 2. Plaintiffs do not raise an argument disputing the weight afforded to fish skin. Commerce's valuation of fish waste by-product using 2013/14 Adib price quotes is therefore supported by substantial evidence. F. Packing Tape Plaintiffs challenge Commerce's selection of Indonesian Global Trade Atlas ("GTA") import data under the Harmonized Tariff Schedule ("HTS") 3919.10 to value packing tape as unsupported by substantial evidence, and as contrary to law. Pls.' Br. at 42-46. Specifically, Plaintiffs *1380claim that HTS 3919.10 "is not specific to the input being valued, and results in an aberrational value." Pls.' Br. at 43 (citing Respondents' Agency Case Br. at 30-31). Defendant argues that Commerce's decision to use HTS 3919.10, which covers "Plates, Sheets, Film, Foil, Tape and Other Flat Shapes of Plastics, Self-adhesive, in Rolls Not Over 20 cm (8 In.) Wide," fulfilled the SV criteria. Def.'s Resp. Br. at 45-46. In the final determination, Commerce continued to value packing tape using Indonesian GTA import data under HTS 3919.10, see Final Decision Memo at 45, with one adjustment-it removed the average unit value ("AUV") from Switzerland because it was "significantly higher as compared to other countries' data on the record." Id. at 46; see also Final SV Memo at 2. With the Swiss data removed, Commerce explained that HTS 3919.10 was specific to respondents' input, publicly available, representative of a broad market average, contemporaneous, and free of taxes and duties.29 Final Decision Memo at 45-46.30 In the final determination, Commerce explained that respondents did not meet their burden of demonstrating that HTS 3919.10 contained aberrational data. Final Decision Memo at 46. Plaintiffs point to the fact that the average unit values ("AUVs") in the import data vary greatly "from a high of IDR 21,946,194/kg (Switzerland) to a low of IDR 32,738/kg (Malaysia)." Pls.' Br. at 44. Commerce looked at packing tape data from other countries on the surrogate country list and compared it to HTS 3919.10. Final Decision Memo at 46. It found that the price difference was "not so large as to demonstrate an aberration with the current POR's data for HTS 3919.10." Id. In their brief before this court, Plaintiffs simply cite to benchmark data they put on the record, but do not explain how it undermines Commerce's determination. Pls.' Br. at 44, 45-56. Considering the discretion Commerce is allotted in evaluating the data before it, and Commerce's determination that alternate data sources did not demonstrate that HTS 3919.10 contained aberrational data, Commerce's decision is reasonable. CONCLUSION The court remands Commerce's application of facts otherwise available to HVG's farming factors. The court sustains the Final Results in all other respects. In accordance with the foregoing, it is ORDERED that Commerce's calculation of HVG's farming factors is remanded *1381for further consideration consistent with this opinion; and it is further ORDERED that Commerce shall file its remand redetermination with the court within 60 days of this date; and it is further ORDERED that the parties shall have 30 days thereafter to file comments on the remand redetermination; and it is further ORDERED that the parties shall have 30 days to file their replies to comments on the remand redetermination. Footnotes [1] 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. [2] Although 19 U.S.C. § 1677e(a) -(b) and 19 C.F.R. § 351.308(a)-(c) (2014) each separately provide for the use of facts otherwise available and the subsequent application of adverse inferences to those facts, Commerce sometimes uses the shorthand "adverse facts available" or "AFA" to refer to its use of such facts otherwise available with an adverse inference. See, e.g., Final Decision Memo at 20-23. Commerce assigned, as partial AFA, the Vietnam-wide rate of $2.39/kg to products produced by two of TAFISHCO's unaffiliated tollers who did not cooperate; this had the effect of increasing TAFISHCO's dumping margin. See Final Results, 81 Fed. Reg. at 17,437 ; Final Decision Memo at 23. [3] Respondent in the underlying review Can Tho Import-Export Joint Stock Company commenced the action Can Tho Import-Export Joint Stock Company v. United States, Court No. 16-00071 ("Court No. 16-00071") to challenge Commerce's determination that it is not entitled to a separate rate. See Compl. at ¶¶ 1, 14, 20, May 4, 2016, ECF No. 8, Can Tho Import-Export Joint Stock Company v. United States, Court No. 16-00071. Court No. 16-00071 was later consolidated under An Giang Fisheries Import and Export Joint Stock Company v. United States, Consol. Court No. 16-00072 ("Consol. Court No. 16-00072"). See Scheduling Order and Order on Consolidation, Aug. 3, 2016, ECF No. 22. On November 14, 2017, after a period of briefing, this court ordered that Court No. 16-00071 be severed from Consol. Court No. 16-00072, and that Court No. 16-00071 be stayed pending the final disposition of all challenges in An Giang Fisheries Import and Export Joint Stock Company et al. v. United States, Consol. Court No. 15-00044, and all challenges in the present action. Mem. & Order at 12-13, Nov. 14, 2017, ECF No. 85. [4] On July 5, 2016, Defendant submitted indices to the confidential and public administrative records, which can be found at ECF Nos. 20-4 and 20-5, respectively. See Administrative Record, July 5, 2016, ECF Nos. 20-4-5. All further references to documents from the administrative records are identified by the numbers assigned by Commerce in these administrative records. [5] HVG includes An Giang Fisheries Import and Export Joint Stock Company, as well as other exporters of subject merchandise. Prelim. Results, 80 Fed. Reg. at 55,094 n.19 ; Final Decision Memo at 1 n.2. [6] Commerce also calculated a weighted-average dumping margin of $0.69 per kilogram for Plaintiffs C.P. Vietnam Corporation, Cuu Long Fish Joint Stock Company, GODACO Seafood Joint Stock Company, Seafood Joint Stock Company No. 4-Branch Dong Tam Fisheries Processing Company, and Viet Phu Foods and Fish Corporation. Final Results, 81 Fed. Reg. at 17,437. In the final determination Commerce denied Plaintiff International Development and Investment Corporation a separate rate, Final Decision Memo at 4, and that company was, therefore, assigned the Vietnam-wide rate of $2.39 per kilogram. Final Results, 81 Fed. Reg. at 17,437. [7] "CONNUM" are control-numbers created by Commerce and specific to the subject merchandise under review. They are unique because they identify the key physical characteristics that are commercially meaningful to the U.S. market and have an impact on sale price and cost of production of the subject merchandise. See Final Decision Memo at 10 (citing e.g., Large Residential Washers from the People's Republic of China [ ("PRC") ], 81 Fed. Reg. 1,398, 1,399 (Dep't Commerce Jan. 12, 2016) (initiation of less-than-fair-value investigation); Stainless Steel Wire Rod from Sweden, 73 Fed. Reg. 12,950 (Dep't Commerce Mar. 11, 2008) (final results of [ADD] review) and accompanying Issues and Decision Memorandum for the Final Results of the Administrative Review of Stainless Steel Wire Rod from Sweden at 2-14, A-401-806, (Mar. 5, 2008), available at http://ia.ita.doc.gov/frn/summary/sweden/E8-4824-1.pdf (last visited Feb. 8, 2018) ). The control numbers are provided to respondents in the questionnaires issued by Commerce. See, e.g., Suppl. Section A, C, & D Questionnaire [sent to HVG] at 3-4, PD 142, bar code 3269661-01 (Apr. 9, 2015) (listing the key physical characteristics for the subject merchandise at issue here) ("Suppl. Questionnaire Sent to HVG Apr. 2015"). [8] Commerce focuses on the FOPs utilized in production of the subject merchandise to meet its statutory obligations under 19 U.S.C. § 1677b(a) and (c)"to compare [normal values] to U.S. prices on an apples-to-apples basis[.]" Final Decision Memo at 10 (citing 19 U.S.C. § 1677b(a), (c) ). [9] When Commerce investigates companies operating within NME countries, it requires respondents to provide to Commerce the FOPs associated with each CONNUM. Final Decision Memo at 10-11 (citation omitted). For the subject merchandise here, the CONNUM included species, product form, product coating, product size, frozen form, preservatives, and net weight factor. See Suppl. Questionnaire Sent to HVG Apr. 2015 at 3-4. [10] Commerce's final determination references two related, but distinct, distortions caused by the respondents' failure to track FOPs by product form. See Final Decision Memo 13-15. The first distortion is the inclusion of non-subject merchandise in the denominator, which Commerce explains causes distortions because fillets and whole fish have very different yields. Id. at 13-14. The second distortion may arise because even within subject merchandise, different product forms (i.e., types of fillets) have different yields. Id. at 15. [11] Commerce presented HVG and TAFISHCO with two opportunities to comply and provide Commerce with the requested information. Final Decision Memo at 11. Respondents, however, provided an alternative methodology that failed to account for the water weight of the product, id. at 14-15, and continued to include non-subject merchandise in their FOP denominators. Id. at 13-14. [12] Specifically, in the final determination to the ninth administrative review, Commerce explained that it would not implement the requirement of CONNUM-specific reporting because respondents would not have known of the need to maintain their records on a CONNUM-specific basis until "eight months after the conclusion of the POR for [the ninth administrative] review." Ninth Admin. Review IDM at 74. However, Commerce went on to remark that, "[f]or all future reviews, the Department intends to require Vinh Hoan and other respondents to report its FOPs on a CONNUM-specific, product-specific, or at a minimum, glazed- and unglazed-specific basis." Id. [13] Plaintiffs also claim that despite "an exceedingly similar fact pattern on CONNUM-specificity [in the tenth administrative review of the ADD Order ], the issue [of CONNUM compliance] was not even raised by the parties (including Petitioners and Commerce) in the final results." Pls.' Br. at 8 (citing Certain Frozen Fish Fillets from [Vietnam]: Issues and Decision Memorandum for the Final Results of the Tenth [ADD] Administrative Review; 2012-2013, A-552-801, (Jan. 7, 2015), available at http://ia.ita.doc.gov/frn/summary/vietnam/2015-00649-1.pdf (last visited (Feb. 8, 2018) ). Notwithstanding the lack of discussion of CONNUM in the tenth administrative review of the ADD Order, Plaintiffs were notified of Commerce's preference for CONNUM-specific reporting and had enough time to come into compliance. [14] Plaintiffs rely on cases such as Skidmore v. Swift & Co., United States v. Mead Corp., and NSK Ltd. v. United States, to argue that because deference to an agency depends on the reasonableness of the justification for deviation, and here, Commerce reversed one of its practices, the agency's actions should not be afforded deference. Pls.' Br. at 8-9 (citing United States v. Mead Corp., 533 U.S. 218, 228, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001) ; Skidmore v. Swift & Co., 323 U.S. 134, 139, 65 S.Ct. 161, 89 L.Ed. 124 (1944) ; NSK Ltd. v. United States, 390 F.3d 1352, 1358 (Fed. Cir. 2004) ). These cases are unhelpful to Plaintiffs' position. Mead and Skidmore direct courts to a variety of factors to evaluate the level of deference an agency's interpretation of its own statute warrants. Mead, 533 U.S. at 228, 121 S.Ct. 2164 (footnotes and citation omitted); Skidmore, 323 U.S. at 139-140, 65 S.Ct. 161. Here, Commerce's request for CONNUM-specific data was consistent with its practice from past reviews of the ADD Order. See generally Eighth Admin. Review IDM at 43-44; Ninth Admin. Review IDM at 73-74. NSK is similarly inapplicable, because here, Commerce made clear which information was distortive, and provided HVG and TAFISHCO two opportunities to report compliant data. Final Decision Memo at 11 (citations omitted). [15] Plaintiffs also argue that "product size" is commercially irrelevant. Pls.' Br. at 9-13. In the final determination, Commerce agreed with respondents that product size "did not impact [HVG and TAFISHCO's] cost accounting," and specifically stated that it would not require CONNUM-specific data for product size. Final Decision Memo at 15. [16] Plaintiffs argue that it was impossible for it to report on a CONNUM-specific data given the timing of Commerce's request, i.e., the supplemental questionnaires were issued well beyond the POR, and requested "data that would have had to have been collected well more than a year prior." Pls.' Br. at 13; see id. at 13-17. However, Plaintiffs have been on notice since the eighth administrative review of the ADD Order. See Eighth Admin. Review IDM at 44. Further, at the outset of this review, Commerce requested CONNUM-specific reporting of respondents' FOPs and in two follow-up questionnaires. See Suppl. Questionnaire Sent to HVG Apr. 2015 at 5-6; Suppl. Section A, C, & D Questionnaire [Sent to HVG] at 4-5, PD 223, bar code 3295664-01 (Jul. 31, 2015); [TAFISHCO] Suppl. Section A, C, & D Questionnaire at 8-9, PD 143, bar code 3269666-01 (Apr. 9, 2015); Second Suppl. Sections C & D Questionnaire [Sent to TAFISHCO] at 4-7, bar code 3295673-01 (Aug. 3, 2015); see, e.g., Antidumping Questionnaire at D-2 (questionnaire issued to HVG). Plaintiffs' reliance on Certain Activated Carbon from the [PRC] for the proposition that Commerce has, in the past, recognized the unfairness of "impos[ing] a previously un-enforced CONNUM-specificity requirement," is likewise misplaced. Pls.' Br. at 14 (citing Certain Activated Carbon from the [PRC], 76 Fed. Reg. 23,978, 23,986 -87 (Dep't of Commerce Apr. 29, 2011) (preliminary results of the third [ADD] administrative review, and preliminary rescission in part) ). In Certain Activated Carbon from the [PRC], Commerce specifically noted that it would be unreasonable to expect respondents to change how they track certain data so it would be CONNUM-specific, when the notice for CONNUM-specific reporting did not reach respondents until eight-months into the third administrative review. Certain Activated Carbon from the [PRC], 76 Fed. Reg. at 23,986. [17] Commerce used a conversion factor reported by pangasius producers who are also tollers for Plaintiffs. Final Decision Memo at 16 (citations omitted). [18] Defendant's citation directs the reader to the section of respondents' brief addressing the CONNUM factors. However, respondents' brief to the agency includes a separate section addressing farming factors and Commerce's methodology in converting the denominator on a shank-equivalent basis. Respondents' Agency Case Br. at 25-27. [19] [ [ ] ] [20] [ [ ] ] [21] TAFISHCO's unaffiliated tollers are interested parties under 19 U.S.C. § 1677(9)(A). [22] Plaintiffs argue that Commerce's decision to apply partial AFA is contrary to Commerce's practice, citing two prior determinations in support of that claim. Pls.' Br. at 35-36 (citing Issues and Decision Memorandum for the Final Determination in the [ADD] Investigation of Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, from the [PRC] at 76-78, A-570-979, (Oct. 9, 2012), available at http://ia.ita.doc.gov/frn/summary/prc/2012-25580-1.pdf (last visited Feb. 8, 2018) ("Solar Cells IDM"); Small Diameter Graphite Electrodes from the [PRC]: Issues and Decision Memorandum for the Final Results of the First Administrative Review of the [ADD] Order at 14-17, A-570-929, (Sept. 6, 2011), available at http://ia.ita.doc.gov/frn/summary/prc/2011-23357-1.pdf (last visited Feb. 8, 2018) ("Electrodes IDM") ). However, in Solar Cells IDM, Commerce made specific findings as to the percentage the missing data represented and concluded that because "the non-reported production quantities [i.e., 0.67 percent] account for a small portion of the FOPs[,]" and there was other reliable data on the record, Solar Cells IDM at 77, a finding of AFA was not appropriate for the unreported data. Id. at 78. In Electrodes IDM, one unaffiliated toller refused to cooperate, "stating that it would not provide its proprietary tolling information to the [respondent in that case]." Electrodes IDM at 15. Nevertheless, Commerce found that it did not need to pursue the data being withheld, as "it had sufficient other data from another of the [respondent's] tollers for this tolling operation[.]" Id. The common thread between Solar Cells IDM and Electrodes IDM is Commerce's finding of other, sufficient data and the impact of the missing information. In comparison, here, Commerce was unable to verify the data provided by the tollers, see Final Decision Memo at 21, and the "uncooperative tollers ... represent a much higher percentage of total production than [the tollers] in Solar Cells [IDM]." Id. at 22-23 (citation omitted). [23] Further citations to the Code of Federal Regulations are to the 2014 edition. [24] Plaintiffs contend that, by relying on floating fish feed prices only, Commerce excluded data for three protein levels and therefore had to average other data in their place. Pls.' Br. at 39 (citing [SVs] for the Final Results [Memo] at 2, PD 373, bar code 3451526-01 (Mar. 18, 2016) ). Plaintiffs claim that the need to use the averaging method demonstrates that Commerce did not use the best available information. Id. However, in stating its preferred approach, i.e., using prices from a mix of sinking and floating fish feed, Plaintiffs fail to demonstrate that Commerce's preference for floating fish feed as more specific is unreasonable. [25] Plaintiffs also claim that Commerce abused its discretion by not specifically asking whether only floating fish feed was used. See Reply of [Pls.'] to Def. & Def.-Intervenor's Resps. Pls.' Mots. J. Upon Agency R. at 4, Sept. 8, 2017, ECF No. 56; Pls.' Br. at 38-39. Commerce abuses its discretion when its decision "is clearly unreasonable, arbitrary, or fanciful ... is based on an erroneous conclusion of law ... rests on clearly erroneous fact findings[,] or follows from a record that contains no evidence on which the [agency] could rationally base its decision." See Gerritsen v. Shirai, 979 F.2d 1524, 1529 (Fed. Cir. 1992) (citations omitted). Commerce asked for data on fish feed. Commerce viewed respondents' web site providing usage guidelines for floating fish feed. Final Decision Memo at 35 (citation omitted). No guidelines were provided for sinking feed. Therefore, it is not unreasonable for Commerce to assume that HVG only used floating fish feed. [26] Plaintiffs claim that "usage guidelines" represent recommended practices, and should not be used to infer that HVG "only used floating fish feed during the [POR]." Pls.' Br. at 38 (emphasis in original). To bolster their position, Plaintiffs point to record evidence "from the Ratna Sari Fish Farm & Poultry Shop that demonstrates that [the shop] sold both floating and sinking [pangasius] fish feed in Indonesia during the [POR]." Id. (emphasis omitted) (citing [Petitioners'] Surrogate Country Comments & Submission of Proposed Factor Values at Ex. I-10C, PD 175, bar code 3278422-05 (May 22, 2015) ). Plaintiffs also argue that Commerce could have, but did not ask "whether the feed actually consumed during the POR was all floating, all sinking, or a mix of the two[,]" in its questionnaires. Id. at 38. Plaintiffs' arguments are not persuasive because they misunderstand the crux of Commerce's reasoning for using floating feed fish price quotes only. Commerce used a more specific data set because HVG, during the administrative review, did not populate the record with evidence showing a use of fish feed, other than floating fish feed. Final Decision Memo at 35. [27] Patin and pangasius are used interchangeably by the parties to refer to the subject merchandise at issue here. [28] Plaintiffs also argue that Commerce only critiqued the July 2015 Adib price quote as being not specific and claim that critique was inapplicable to the April 2015 Adib price quote. Pls.' Br. at 52 (citing [An Giang & TAFISHCO] Final Direct [SV] Submission at Ex. 6-C, PD 204, bar code 3291951-02 (July 17, 2015) (reproducing a copy of the July 13, 2015 Adib price quote); [An Giang & TAFISHCO] Surrogate Values Submission at Ex. 11-D, PD 168, bar code 3278288-01 (May 22, 2015) (reproducing the Apr. 10, 2015 Adib price quote for fish waste) (also reproduced as Ex. 5 in Pls.' Br.) ). Plaintiffs do concede that the April 10, 2015 Adib price quote is not contemporaneous. Pls.' Br. at 52. In the final determination, Commerce explained that, "[a]ll other things being equal, [it] prefers contemporaneous [SV] information to non-contemporaneous [SV] information." Final Decision Memo at 51 (citation omitted). [29] Plaintiffs argue that Indonesian GTA data for HTS 3919 is more specific to the input used by HVG as "[t]his value is much more in line with actual packing tape prices in Indonesia[.]" Pls.' Br. at 45 (citing [An Giang &TAFISHCO] Surrogate Values Submission at Ex. 17-A, PD 169, bar code 3278288-02 (May 22, 2015) ). As Commerce explained in the final determination, HTS 3919 "encompasses HTS 3919.10, but is less specific [because] it covers larger rolls of adhesive plastics ... [and] contains the very [Swiss] data Respondents argue is aberrational." Final Decision Memo at 45. [30] Plaintiffs also argue that there was "better data on the record" in the form of price quotes generated by the Indonesian company, Adib Food Supplies. Pls.' Br. at 45-46. Plaintiffs claim that the Adib Food Supplies' price quotes are reliable, as the company is reputable and both Commerce and petitioners have relied upon its data for SV information. Id. at 46. In the final determination, Commerce addressed the Adib price quote, explaining that although input specific, publicly available and tax and duty free, "it reflects the experience of [a] single company and, thus, is not representative of a broad market average." Final Decision Memo at 46. Additionally, the quote "is not contemporaneous with the POR." Id. Comparatively, Commerce found Indonesian import data under HTS 3919.10 contemporaneous, publicly available, representative of a broad market average and tax and duty free. Id. at 45.