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

179 F. Supp. 3d 1256, 38 I.T.R.D. (BNA) 1221, 2016 Ct. Intl. Trade LEXIS 57

Opinion

OPINION AND ORDER

Kelly, Judge

This consolidated action comes before the court on USCIT Rule 56.2 motions for judgment on the agency record, challenging the U.S. Department of Commerce’s (“Department” or “Commerce”) determination in the ninth administrative review of the antidumping duty order covering certain frozen fish fillets from the Socialist Republic of Vietnam (“Vietnam”). See Certain Frozen Fish Fillets From the Socialist Republic of Vietnam, 79 Fed. Reg. 19,053 (Dep’t Commerce Apr. 7, 2014) (final results of antidumping duty administrative review and new shipper review; 2011-2012) (“Final Results”), as amended, 79 Fed. Reg. 37,714 (Dep’t Commerce July 2, 2014) (“Amended Final Results”); see also Certain Frozen Fish Fillets from the Socialist Republic of Vietnam: Issues and Decision Memorandum for the Final Results of the Ninth Administrative Review and Aligned Shipper Review, A-552-801, (Mar. 28, 2014), available at http://ia.ita.doc.gov/frn/ summary/vietnam/2014-07714-l.pdf (last visited May 26, 2016) (“Final Decision Memo”).

An Giang Fisheries Import and Export Joint Stock Company, Asia Commerce Fisheries Joint Stock Company, Cuu Long Fish Joint Stock Company, Hiep Thanh Seafood Joint Stock Company, International Development and Investment Corporation, NTSF Seafoods Joint Stock Company, QVD Food Company Ltd., Southern Fishery Industries Company, Ltd., and Vinh Hoan Corporation (collectively “Agifish”) commenced this action pursuant to section 516A of the Tariff Act of 1930, as amended, 19 U.S.C. § 1516a (2012).¹ See Summons, May 5, 2014, ECF No. 1. The court consolidated Agifish’s challenge with actions filed by: (1) Catfish Farmers of America, an association of U.S. catfish processors and growers, and individual U.S. catfish processors, America’s Catch, Alabama Catfish Inc., Heartland Catfish Company, Magnolia Processing, Inc., and Simmons Farm Raised Catfish, Inc. (collectively “CFA”); (2) Binh An Sea food Joint Stock Company (“Binh An”); and (3) Vietnam Association of Seafood Exporters and Producers, a trade association comprised of Vietnamese seafood producers and/or exporters (“VASEP”). See Order, Aug. 26, 2014, ECF No. 32. Each of the above named parties filed Rule 56.2 motions challenging Commerce’s final determination. .See An Giang Fisheries Import and Export Joint Stock Company, Asia Commerce Fisheries Joint Stock Company, Cuu Long Fish Joint Stock Company, Hiep Thanh Seafood Joint Stock Company, International Development and Investment Corporation, NTSF Seafoods Joint Stock Company, QVD Food Company, Ltd., Southern Fishery Industries Company Ltd., and Vinh Hoan Corporation’s Rule 56.2 Mot. J. Upon Agency R., Jan. 27, 2015, ECF No. 48; Pis. Catfish Farmers of America, et al.’s Rule 56.2 Mot. J. Upon Agency R., Jan. 27, 2015, ECF No. 49; Consolidated PI. Binh An Seafood Joint Stock Co. Ltd.’s Mot. J. Agency R., Jan. 27, 2015, ECF No. 47; PL’s Rule 56.2 Mot. J. Upon Agency R., Jan. 27, 2015, ECF No. 46. In addition, CFA filed a response, as a defendant-intervenor, in opposition to motions by Agifish, VASEP, and Binh An. See Def.-Intervenors’ Resp. Opp’n Pis.’ Rule 56.2 Mots. J. Agency R., July 29,2015, ECF No. 79.

BACKGROUND

On September 26, 2012, Commerce initiated this ninth antidumping duty (“AD”) administrative review covering subject imports entered during the period of review (“POR”), August 1, 2011 through July 31, 2012. See Initiation of Antidumping and Countervailing Duty Administrative Reviews and Request for Revocation in Part, 77 Fed. Reg. 59,168, 56,169 (Dep’t Commerce Sept. 26,2012).

In its preliminary determination, Commerce selected Indonesia as the primary surrogate country for obtaining surrogate values (“SV”) for respondents’ factors of production (“FOP”) used to produce the subject merchandise. See Certain Frozen Fish Fillets From the Socialist Republic of Vietnam, 78 Fed. Reg. 55,676 (Dep’t Commerce Sept. 11, 2013) (preliminary results of the antidumping duty administrative review and new shipper review; 2011-2012) (“Prelim. Results”); see also Certain Frozen Fish Fillets from the Socialist Republic of Vietnam: Decision Memorandum for the Preliminary Results of the 2011-2012 Antidumping Duty Administrative Review and New Shipper Review at 18, A-552-801, (Sept. 3, 2013), available at http://ia.ita.doc. gov/frn/summary/vietnam/2013-22123-l.pdf (last visited May 26, 2016) (“Prelim. Decision Memo”). Commerce selected Vinh Hoan Corporation (“Vinh Hoan”) and Hung Vuong Group (“HVG”) as mandatory respondents, Prelim. Results, 78 Fed. Reg. at 55,676, initially assigning them weighted-average dumping margins of $0.42 per kilogranr.and $2.15 per kilogram, respectively. Id. at 55,677. Commerce assigned the average of these rates, $0.99 per kilogram, to the separate rate respondents that were not individually examined. Id. Commerce also assigned a weighted-average dumping margin of $2.11 per kilogram to those exporters subject to the order who did not rebut the presumption of government control (the “Vietnam-Wide Rate”). Id.

In its final determination, Commerce continued to select Indonesia as the primary surrogate country for valuing respondents’ FOPs. Final Decision Memo at 7, 16. Commerce calculated final weighted-average dumping margins of $0.03 per kilogram for Vinh Hoan and $1.20 for HVG, from which Commerce assigned a rate of $0.42 per kilogram to the separate rate respondents. Final Results, 79 Fed. Reg. at 19,055. Commerce published amended final results on July 2, 2014, which altered the weighted-average dumping margin assigned to mandatory respondent Vinh Hoan to $0.00 per kilogram. Amended Final Results, 79 Fed. Reg. at 37,714. Commerce also changed the weighted-average dumping margin assigned to separate rate respondents to $1.20 per kilogram.² Id. The Vietnam-Wide Rate remained unaltered. Id.

Agifish, VASEP, and Binh An challenge Commerce’s selection of Indonesia as the primary surrogate country. See Mem, Law Supp. Pis.’ Rule 56.2 Mot. J. Upon Agency R. 10-15, Jan. 27, 2015, ECF No. 48 (“Agifish Br.”); Mem. Law Supp. PL’s Rule 56.2 Mot. J. Upon Agency R. 7-85, Jan. 27, 2015, ECF No. 46 (“VASEP Br.”); Mem. Law Supp. Consolidated Pl.’s Binh An Seafood Joint Stock Co., Ltd.’s Mot. J. Upon Agency R. 12-18, Jan. 27, 2015, ECF No. 47-1 (“Binh An Br.”). Agifish also challenges Commerce’s selection of SV data sources to value various FOPs used to produce the subject merchandise, including whole live fish, fingerlings, various fish farming FOPs, various byproducts, labor, and various expenses related to shipping, brokerage and handling. See Agifish Br. 16-45. Agifish challenges Commerce’s decision to apply a “cap” or constructed value (“CV”), to obtain a SV for Vinh Hoan’s fish oil byproduct rather than using Indonesian import data as is. See Agifish Br. 45-49. Agifish also challenges Commerce’s consideration of arguments raised by CFA in its rebuttal case brief relating to SV data sources for certain FOPs. See Agifish Br. 49-51. CFA. challenges Commerce’s SV data selection .to value respondents’ .panga-sius feed FOP. See Am. Mem. Supp. Pis. Catfish Farmers of America, et al.’s Rule 56.2 Mot. J. Upon Agency R. 26-38, Jan. 29, 2015, ECF No. 54 (“CFA Br.”). Finally, CFA challenges Commerce’s determination to decline to adjust Vinh Hoan’s margin calculation to exclude glazing weight.³ Defendant United States (“Defendant”), responds that the court should sustain Commerce’s surrogate country selection and SV data selections. See Def.’s Resp. Consolidated Pis.’ Rule 56.2 Mot. J. Upon Agency R., July 29, 2015, ECF No. 75 (“Def.’s Resp. Br.”). However, Defendant requests a voluntary remand to revisit its normal value (“NV”) calculation with respect to glazing for Vinh Hoan. Id. at 92.

The court sustains Commerce’s selection of Indonesia as the primary surrogate country. The court also sustains Commerce’s determination to consider arguments relating to SV data selection raised by CFA in its rebuttal brief before the agency. The court sustains Commerce’s SV data selections for the following FOPs: (1) whole live fish; (2) fingerlings; (3) fish feed; (4) sawdust; (5) labor; (6) fish waste byproducts; (7) fresh broken fillets; (8) brokerage and handling; and (9) truck freight. However, the court remands Commerce’s SV data selections for rice husk and Vinh Hoan’s fish oil byproduct for further consideration and explanation. Lastly, the court grants Defendant’s voluntary remand request for Commerce to. reconsider its NV calculation with respect to glazing for Vinh Hoan.

JURISDICTION AND STANDARD OF REVIEW

The court has jurisdiction pursuant to 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)(l)(B)(i).

DISCUSSION

I. Primary Surrogate Country Selection

A. Factual Background

Since Commerce treats Vietnam as a nonmarket economy (“NME”), Commerce initiated its process of selecting a surrogate market economy country to value the FOPs used to produce the subject imports by requesting that its Office of Policy (“OP”) issue a list of potential surrogate countries. See Letter to All Interested Parties re: 9th Administrative Review of Frozen Fish Fillets from the Socialist Republic of Vietnam: Surrogate Country List, PD 42, bar code 3105149-01 (Nov. 8, 2012). The list included the following six countries: Bangladesh, Pakistan, Nicaragua, India, Bolivia, and the Philippines. See id. at Attach. I (“OP List”). After considering comments from interested parties, Commerce selected Indonesia, which was not among the countries on the OP List, as the primary surrogate country for its preliminary results. See Prelim. Results, 78 Fed. Reg. at 55,677; see' also Prelim. Decision Memo at 18.

in its preliminary determination, Commerce noted the per capita gross national income (“GNI”) values of the countries on the OP List, and found that all countries on the OP List are at an equal level of economic development. Prelim. Decision Memo at 13. Commerce found Indonesia’s per capita GNI places it at “a higher and, thus, less comparable level of economic development than that represented by the six countries on the initial surrogate country candidate list, but still comparable to that of Vietnam.”⁴ Id at 14. Commerce found Indonesia, and all of the countries on the OP List except Bolivia, to be significant producers of comparable merchandise. Id. at 15. Commerce relied upon data quality considerations to preliminarily select Indonesia as the primary surrogate country to value respondents’ FOPs. See id. at 16-18.

Thereafter, the parties continued to submit additional SV data and comments on the issue of surrogate country selection. See Catfish Farmers of America Surrogate Value Data, PD 290-294, bar codes 3167326-01-05 (Dec. 6, 2013) (“CFA SV Data”); Agifish, Vinh Hoan Post-Preliminary Surrogate Value Data, PD 295-296, bar codes 3167265-01-02 (Dec. 9, 2013) (“Respondents’ Post-Prelim. SV Data”); Vinh Hoan, Agifish, VASEP Supplemental Surrogate Value Submission, PD 298, bar code 3167703-01 (Dec. 11, 2013); Respondents’ Rebuttal Surrogate Value Data, PD 305-309, bar codes 3172989-01-05 (Jan. 10, 2014); Catfish Farmers of America Rebuttal Surrogate Value Data, PD 310-311, bar codes 3173134-01-02 (Jan. 10, 2014) (“CFA Rebuttal SV Data”). After the submission of case briefs, Commerce issued its final determination maintaining its selection of Indonesia as the primary surrogate country. See Final Decision Memo at 15-16.

B. Legal Framework

In NME AD proceedings, Commerce generally calculates NV using the best available information to value respondents’ FOPs and other costs and expenses “in a market economy country or countries considered to-be appropriate by [Commerce].” 19 U.S.C. § 1677b(c)(l). 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). The statute does not define economic comparability, nor does it provide a methodology or criteria for evaluating the other statutory surrogate country selection standards.

Although Commerce may use multiple surrogate countries to calculate NV, Commerce’s regulatory preference is to “value all factors in a single'-surrogate country.” 19 C.F.R. § 351.408(e)(2) (2013).⁵ To implement this preference: (1) the OP assembles “a list of potential surrogate countries that are at a comparable level of economic development to the NME country,” whose per capita GNIs fall within a range of comparability to the GNI of the NME country;⁶ (2) Commerce identifies countries from the OP List' that produce comparable merchandise; (3) Commerce “determines whether any of the countries which produce comparable merchandise are ‘significant’ producers of that comparable merchandise”; and, (4) if more than one country satisfies steps (1)—(3), Commerce will select the country with the best factors data. See 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 26, 2016) (“Policy Bulletin 04.1”).

According to Commerce’s practice, the surrogate countries on the list “are not ranked and should be considered equivalent in terms of economic comparability.” Id. at 2. If more than one country is economically comparable to the NME country and a significant producer of comparable merchandise,. Commerce selects the country with the best factors data as the primary surrogate country to value FOPs. Id. at 4. Commerce selects the country with the best factors data based upon the data’s: (1) specificity to the input; (2) tax and import duty exclusivity; (3) contemporaneity with the period of review; and (4) public availability. Id.

C. Analysis

1. Economic Comparability

VASEP, Agifish, and Binh An argue that Commerce selected a primary surrogate country that is not economically comparable to Vietnam. See VASEP Br. 13-20; Agifish Br. 12-13; Binh An Br. 12-13. Defendant argues that Commerce reasonably determined that Indonesia is economically comparable to Vietnam despite the fact it is not among the countries on the OP List. See Def.’s Resp. Br. 19-26. The court agrees with Defendant.

It is not unreasonable for Commerce to consider countries proposed by interested parties that are not included on the OP List so long as Commerce satisfies its statutory mandate of choosing an economically comparable country to the extent possible. Commerce emphasized that the OP List advised parties that it is not an exhaustive list, and noted that there are other countries that could reasonably be viewed as “being at Vietnam’s level of economic development.” See Prelim. Decision .Memo at 13; see also OP List at 2. Commerce considered the GNI information placed on the record relating to the countries on the OP List as well as for Indonesia and acknowledged that Indonesia was at a “higher and, thus, less eompa-rable level of economic development than that represented by the six countries on the initial surrogate country candidate list, but still comparable to that of Vietnam.”⁷ Final Decision Memo at 6-7 (quoting Prelim. Decision Memo at 14). The court cannot say that Commerce unreasonably concluded that Indonesia was at a level of economic development comparable to Vietnam.⁸

VASEP, Agifish and Binh Ah argue that Commerce’s selection of a potential surrogate country not on the OP List amounts to an impermissible expansion of the statutory economic comparability requirement. See VASEP Br. 13-14; Agifish Br. 12-13; Binh An’s Br. 12-13. VASEP argues that, in the ninth administrative review, the OP consciously removed Indonesia because of its “galloping GNI that placed Indonesia outside of the GNI bookends established by the Department’s OP memoranda.”⁹ VASEP Br. 15. VASEP’s argument is speculative and.unsupported in the record. VASEP, Agifish and Binh An cite no authority that prevents Commerce from se lecting off-list potential surrogate countries whose GNI fall outside what VASEP characterizes as the bookends set by the OP List.¹⁰ VASEP ignores the notion that the size of the OP List and its range of per capita GNIs may vary from review to review and that the size of the list is also a function of how broadly or narrowly Commerce defines its initial list. See OP List; see also Prelim. Decision Memo at 13. VASEP cites to nothing in the statute or in Commerce’s practice that requires economic comparability to be defined by the GNI range of the potential surrogate countries on the OP List,¹¹ None of the cases cited by VASEP support the notion that Commerce’s OP List sets the outer limits of economic comparability.¹²

2. Significant Producer of Comparable Merchandise

Commerce reasonably concluded that Indonesia is a significant producer of comparable merchandise because export information from Fisheries Statistics, an online data source published by the Food.and Agriculture Organization of the United Nations, indicates that Indonesia, along with Bangladesh, India, Nicaragua, Pakistan, and the Philippines, are exporters of fish fillets. ■ No party challenges . Com merce’s finding that Bangladesh, India, Indonesia, Nicaragua, and Pakistan are significant producers of comparable merchandise. However, Agifish challenges Commerce’s finding that the Philippines, is a significant producer of comparable merchandise, citing Commerce’s finding in its preliminary results that the Philippines produced only 72 metric tons (“mt”) of pangasius fish in 2011. Agifish Br. 13. Since Commerce did not select the Philippines as its primary surrogate country, the court need not review Commerce’s determination that the Philippines was a significant producer of comparable merchandise because the court sustains Commerce’s surrogate country selection of Indonesia.

3. Data Quality

Agifish, VASEP, and Binh An argue that Commerce’s finding that Indonesian data was the best available data for valuing respondents’ FOPs is not supported by substantial evidence. See Agifish Br. 14-16, VASEP’s Br. 28-35, Binh An’s Br. 14-18. Commerce reasonably determined that Indonesian data for valuing respondents’ FOPs is superior to the other SV data alternatives.

The record supports Commerce’s conclusion that Indonesian data is superior to both Philippine and Bangladeshi data to value respondents’ whole live fish FOP. First, Commerce found that Indonesian Aquaculture Statistics (“IAS”) data is the only of the three sources that meets the full breadth of Commerce’s criteria for SV data selection. Final Decision Memo at 15. Second, Commerce relied upon the declining quality of the Bangladeshi Department of Agriculture Marketing (“DAM”) data relative to that submitted in past reviews. Id. at 15, 21-23. Third, Commerce found IAS data is the most representative of a broad market average because the Philippine Fishery Statistics (“FS”) data has much lower production volumes (490,000 mt for IAS versus 72 mt for Philippine FS), id. at 20, and the Bangladeshi DAM data is missing data for the largest panga-sius producing district, Mymensingh, which accounted for two-thirds of all pan-gasius produced during the POR.¹³ Id. at 21 (citing Issues and Decision Memorandum for the Final. Results of the Eighth Administrative Review of and Aligned New Shipper Reviews for Certain Frozen Fish Fillets from the Socialist Republic of Vietnam at 20-22, A-552-801, (Mar. 13, 2013), available at ■http://enforcement. trade.gov/frn/summary/VIETNAM/2013-06550-l.pdf (last visited May 26, 2016)). Commerce also found that the DAM data only represents 25 of 68 districts in Bangladesh and 730 data points representing 39,000 mt,. whereas IAS data contains data for 31 .of 33 districts in Indonesia representing 490,000 mt.¹⁴ Id. at 21-22.

■ Commerce also concluded that the DAM data is not reliable because of inadequate vetting procedures and missed statistical anomalies.¹⁵ Id. at 26. Commerce concluded that the DAM data is less specific because Commerce credited record evidence indicating that the DAM data contained dead fish over other record evidence that it did not.¹⁶ Id. at 23-24. Commerce weighed the record evidence and concluded that the inclusion of dead fish in the DAM data distorts the SV for this FOP.¹⁷ Id. at 24. VASEP points to no evidence that detracts from these conclusions.

VASEP argues that other studies on the record corroborate the DAM data, which undermines Commerce’s findings about the unreliability of the DAM. data and its lack of broad coverage. See VASEP Br. 33. However, as discussed in more detail below, Commerce was not persuaded that these sources adequately filled gaps in the DAM data. The court cannot say that Commerce’s conclusion that Bangladeshi DAM data suffers from reliability issues is unreasonable in light of the record evidence supporting its finding.

Commerce also noted that, because of increased integration in the Vietnamese pangasius industry, SV data sources of FOPs other than whole fish, including fingerlings and fish feed, are significant con-sidei-ations in surrogate country selection and the data on the record for these FOPs supports selecting Indonesia as the primary surrogate country.¹⁸ Final Decision Memo at 14-15. Commerce pointed to rec ord evidence that the Indonesian sources for these FOPs satisfy the breadth of its criteria while the competing Bangladeshi sources fail to meet several data selection criteria. See id. at 29-35; see also Prelim. Decision Memo at 16-18: Commerce also noted its preference for valuing respondents’ FOPs using import statistics given that they generally meet Commerce’s data selection criteria. Final Decision Memo at 14. Commerce is able to value 80 percent of respondents’ FOPs (34 of 43) using contemporaneous import statistics from Indonesia, whereas Commerce stated that no party has submitted contemporaneous import statistics from Bangladesh. Id. at 14-15.

VASEP’s • argument that Bangladeshi data offers the best available SV data sources for respondents’ farming factors asks the court to reweigh the evidence. See VASEP Br. 34. At no point does VA-SEP point to any record evidence that detracts from the reasonableness of Commerce’s determination that Indonesian SV data sources are superior on the whole to Bangladeshi data sources. Therefore, Commerce reasonably concluded that Indonesian data is the best available information for the widest range of FOPs in this administrative review.¹⁹

4. Conclusion

For the reasons discussed above, Commerce’s selection of Indonesia as the primary surrogate country is reasonable. Commerce reasonably found that Indonesia was economically comparable to Vietnam, a significant producer of comparable merchandise, and offered the best available information for valuing respondents’ FOPs. While Commerce acknowledged that Indonesia is less economically comparable to Vietnam as compared to the countries on the OP List, Commerce reasonably concluded that data considerations, particularly those relating to the primary input: for the subject merchandise—whole live fish—“outweigh the fact that Indonesia is not at the same level of economic development as Vietnam.” Final Decision Memo at 7. Therefore, Commerce’s primary surrogate country selection is sustained.

II. Consideration of Arguments Raised By CFA for the First Time in its Rebuttal Case Brief

Agifish argues that Commerce abused its discretion by declining to strike non-responsive portions of CFA’s rebuttal brief from the administrative record.²⁰ See Agifish Br. 49-51. The court concludes that Commerce did not abuse its discretion in not rejecting portions of CFA’s rebuttal brief.

On January 22, 2014, Agifish, Vinh Hoan, VASÉP, and certain of its pangasius fillet exporting members, filed their initial case brief before Commerce objecting to Commerce’s SV data sources for various FOPs in its preliminary determination. See Respondents’ Case Brief, PD 314, bar code 3175397-01 (Jan. 22, 2014) (“Respondents’ Case Brief’). On February 22, 2014, CFA submitted a rebuttal case brief to Commerce supporting the SV data selections Commerce made in its preliminary determination. See Rebuttal Brief on Behalf of Catfish Farmers of America; America’s Catch; Alabama Catfish Inc. dba Harvest Select Catfish, Inc.; Heartland Catfish Company; Pride of the Pond; and Simmons Farm Raised Catfish, Inc., CD 232, bar code 3180220-01 (Feb. 10, 2014) (“CFA Rebuttal Case Brief’).

On February 26, 2014, Agifish, VASEP, and Vinh Hoan filed a letter requesting that Commerce strike the pages from CFA’s rebuttal case brief relating to its arguments concerning alternative SV data selections for various FOPs including fingerlings, labor, fish oil, fish meal, and fish waste. See Letter from Respondents Requesting Hearing and to Strike Petitioners’ Arguments, PD 331, bar code 3184050-01 (Feb. 26, 2014) (“Letter Requesting Strike”). They' argued that CFA could not raise affirmative arguments for Commerce’s final determination that had not been raised in Respondents’ Case Brief. Id. In response, CFA filed a letter objecting that their rebuttal case brief rebutted “Respondents’ positions as to what constitutes the best available surrogate values on the record if the Department should accept Respondents’ premise that the values used in the preliminary results are not the; best values.” Letter from Petitioners Responding to Respondents’ Request to Strike at 2, PD 334, bar code 3184738-01 (Feb. 28,2014).

On March 7, 2014, Commerce denied the request to strike portions of CFA’s rebuttal case brief. Commerce Letter to Interested Parties Rebuttal Brief Rejection Request, PD 338, bar code 3186776-01 (Mar. 7, 2014) (“Commerce Denial Letter”). Commerce found that the .arguments in CFA’s rebuttal case brief are “limited to the arguments .., raise[d] in [the] case brief. ... The fact that [CFA’s] rebuttal brief addresses other record evidence relating to the same issues ... does not cause it to run afoul of the Department’s regulation.” Id. (citing 19 C.F.R. § 351.309(d)(2)).

Commerce’s regulations governing the submission of administrative case briefs provide that “[t]he case brief must present all arguments that continue in the submitter’s view to be relevant to [Commerce’s] final determination or final results.” 19 C.F.R. § 351.309(c)(2). Further, Commerce’s regulations regarding rebuttal briefs provide that “[t]he rebuttal brief may respond only to arguments raised in case briefs and should identify the arguments to which it is responding.” Id. § 351.309(d)(2). Here, Commerce declined to reject CFA’s rebuttal case brief because it found that the arguments CFA raised wei’e all responsive to arguments raised by Agifish, VASEP, and Vinh Hoan in their initial case brief. See Commerce Denial Letter. This. determination was not an abuse of discretion.

Agifish argues that Commerce’s determination is unfair, and therefore an abuse of discretion, because it denies them an opportunity to rebut the arguments raised for the first time in CFA’s rebuttal case brief. However, Agifish did not request an opportunity to respond to petitioners’ arguments. See Letter Requesting Strike. Moreover, Agifish continues to challenge Commerce’s selection of SV data sources for all of these FOPs before the. court, which the court reviews for substantial evidence. Defendant is not arguing that these arguments should be precluded by exhaustion. If the court deems Commerce’s determinations unsupported by substantial evidence, then the determination would be remanded to the agency for further explanation in light of the detracting record evidence cited by Agifish. Therefore, Commerce’s determination to decline to strike CFA’s arguments in its rebuttal case brief did not deny Agifish, VASEP, and Vinh Hoan a right to judicial relief.

III. Commerce’s Analysis of Specific Surrogate Values

Agifish and ■ Binh An challenge Commerce’s SV selections for whole live fish, fingerlings, rice husk, labor, sawdust, fish waste byproducts, fresh broken meat, trucking, brokerage and handling expenses, and fish oil.²¹ See Agifish Br. 16-49; Binh An Br. 18-19. CFA challenges Commerce’s SV selection for pangasius feed as unsupported by substantial evidence and contrary to law. See CFA Br. 26-38. Defendant- refutes all these challenges; See Def.’s Resp. Br. 31-91 The court sustains Commerce’s SV data selections for whole live fish, fingerlings, fish feed, sawdust, labor, fish waste, fish belly, and fish skin byproducts, fresh broken fillets, brokerage and handling, and truck freight. However, the court remands Commerce’s SV. data selections for rice husk and fish oil for further consideration and explanation.

A. Whole Live Fish

Agifish challenges the selection of IAS data to value respondents’ whole live fish FOP.- See Agifish Br. 16-23. Defendant counters that the record supports Commerce’s determination that there are serious flaws with regard to the specificity and reliability of the Bangladeshi. DAM Data for whole live pangasius fish. See Defl’s Resp. Br. 33-42. Commerce’s determination that IAS data is the best available data for valuing whole live fish is supported by substantial evidence.

Commerce found that both IAS data and Bangladeshi DAM data are equally contemporaneous, publicly available, and tax and duty exclusive. Final Decision Memo at 20. Commerce, however, found that IAS data is more representative of a broad market average, more specific and more reliable than Bangladeshi DAM or Philippine FS data.²² See id. at 20-27. As to specificity,-Commerce concluded the inclusion of dead fish in the DAM Data “distorts the SV for whole, live fish, rendering it not as specific as Indonesian AS.” Id at 24. Commerce also found that, although Philippines FS data is species-specific, the data is not specific to whole live pangasius because it includes prices for fish that may be further processed, which causes price distortions. Id. at 23.

Commerce reasonably found that Bangladeshi DAM data is not reliable, while it found IAS and Philippine FS data are. Id. at 26-27. Commerce concluded that the DAM data is unreliablé because the DAM data shows statistical anomalies and inadequate vetting procedures.²³ Id at 26. Commerce also recognized data anomalies because prices from a particular district in Bangladesh share one-third of the POR average and those of another district are one-tenth of that average. Id. In contrast, Commerce found no similar reliability concerns with 'IAS data because IAS data is collected in stages at the household, village, and municipal level úsing random sampling and because the data is' córrected and revised when necessary. Id. at 27.

Agifish and VASEP argue that other official data sources corroborate the DAM data, which they argue undermines Commerce’s reliability and broad market average findings. Agifish' Br. 20-22; VASEP Br. 32-33. Agifish and VASEP argue these sources- corroborate the DAM price surveys and fill the gaps in the DAM data. Agifish Br. 20; VASEP Br. 33. Agifish also argues that numerous sources explain the apparent gaps in data. Agifish Br. 19. Commerce considered, but was not persuaded by the sources relied upon to explain the gaps in the DAM data and to corroborate the survey prices. Final Decision Memo at 22-23. Commerce concluded that the sources failed to explain why DAM data coverage decreased as Bangladeshi production increased. Id. at 22. Moreover, the sources did not explain why there had “been no data for any pangasius sales for two years in Bangladesh’s largest pangasius producing district.” Id. Commerce therefore found the DAM data unreliable and unrepresentative, and the cóurt will not reweigh the evidence. Therefore,' Commerce’s selection of IAS data is supported by substantial evidence because Commerce reasonably concluded that IAS data best satisfied the data selection criteria.

B. Fingerlings

Agifish challenges Commerce’s selection of the affidavit of Dr. Djumbuh Rukmono (“Rukmono Affidavit”), an official from the Indonesian Directorate General of Aquaculture, to value fingerlings as unreliable. See Agifish Br. 23-28. Commerce’s selection of the Rukmono Affidavit is supported by substantial evidence because Commerce reasonably concluded that the Rukmono Affidavit meets the breadth of its SV data selection criteria.²⁴

Agifish claims that the Rukmono Affidavit is not rehable because Commerce has not addressed record evidence that reveals significant errors converting fish lengths to weights. Agifish Br. 24. Specifically, Agifish argues that a comparison of the prices generated by the affidavits oh the record provided by Dr. Rukmono and Dr. Soetrisno, another official from the same Indonesian ministry, show too wide a range of prices per kilogram to be accurate. Id. In support of this argument, Agifish provides a chart showing conversions from price per piece to price per kilogram using these data sources, which it argues yield prices ranging from over $1,000.00 per kilogram for 0.5 to 1.0 inch fingerlings to $1.44 per kilogram for 5.0 to 6.0 inch fingerlings. Id. at 25, Ex.l. The price range alone, without some benchmark from an economically comparable country to compare it to, is insufficient to render the data source aberrational. In the absence of such record evidence, Commerce’s conclusion is not unreasonable.

Agifish also argues that Commerce has failed to address inconsistencies between the Rukmono Affidavit and the Soetrisno Affidavit in . per kilogram prices for the same size bands. Id. at 25. However, Agifish’s argument, regarding the reliability of the Rukmono Affidavit rests, on comparing prices in that source to those from an unreliable source.²⁵ An unreliable source cannot logically be used to question the reliability of another source. Without other relevant and reliable data points, Agifish’s complaint that the prices in the Rukmono Affidavit are aberrational is speculative and unpersuasive.

Agifish also argues that the Rukmono Affidavit is unreliable because it requires Commerce to make assumptions regarding fingerling sizes consumed by respondents. Agifish Br. 26-27. Agifish argues these assumptions distort respondents’ fingerling values, as‘highlighted by the fact that Commerce assigned different values to the fingerlings consumed by each mandatory respondent where no record evidence supports doing so. Id. at 27. However, Commerce’s assignment of different values to the fingerlings of each mandatory respondent results from the fact that Vinh Hoan’s affiliate, Van Due Tien Giang (“VDTG”), did not record the size of the fingerlings it purchased, see Surrogate Values for the Final Results at 2-3, PD 345, bar code 3192913-01 (Mar. 28, 2014) (“Final SV Memo”), not from inaccuracies in the Ruk-mono Affidavit.²⁶ Agifish offers no record evidence that undermines the accuracy of the values assigned to either mandatory respondent. Moreover, Agifish does not cite any record evidence indicating that the mandatory respondents consumed fingerlings of the same size, so the fact that Commerce used different values for fingerlings consumed by the two mandatory respondents does not indicate the Rukmono Affidavit is unreliable or that using its conversion data results in distorted values for fingerlings. Therefore, Commerce’s reliance on the Rukmono Affidavit to value fingerlings is reasonable.

C. Fish Feed

CFA’s challenge to Commerce’s selection of an article-in the Atrobos Aqua publication to value respondents’ fish feed, see CFA Br. 26-38, must-fail. Commerce found the article published in Atrobos Aqua is publicly available, contemporaneous, tax and duty exclusive, and representative of a broad market average. Final Decision Memo at 35. Commerce acknowledged that the article does not mention protein content of the feed, but it nonetheless found the data source specific because the article lists prices for pangasius feed. Id. Commerce also noted that the Atrobos Article, which includes data from Indonesian producers, would “most likely be inclusive of the [Indonesian price quote] data submitted by the parties, as these companies are Indonesian producers of pangasi-us feed.”. Id.

While Commerce concluded that the Rukmono Affidavit’s prices for fish feed are publicly available and tax and duty exclusive, it also found this source neither contemporaneous nor representative of a broad market average. Id at 34. Commerce also concluded that the Rukmono Affidavit’s prices had broad geographic coverage but lacked broad temporal coverage because they only cover two months of data. Id. Commerce concluded that the Bangladeshi Aquaculture Study on the record is non-contemporaneous because, although published in 2011, the data is gathered in September and October of 2007. Id. Like the Rukmono Affidavit, these statistics have broad geographic coverage, but lack broad temporal coverage because they only include two months of data. Id. Because Commerce found that the Atrobos Aqua article is the only source on the record that fully satisfied the breadth of its data selection criteria, Commerce’s data selection to value fish feed is reasonable.

CFA argues that protein content information is necessary to find a SV data source for fish feed specific because respondents reported purchasing feed of a specific protein content.²⁷ CFA Br, 29-31, However, while CFA points to record evi dence that feed prices differ depending on protein content, see id. CFA does not argue that any of the feed values on the record differ so significantly that they are aberrational. Therefore, the fact the article does not specify protein content of the fish feed does not render Commerce’s specificity finding unreasonable.²⁸ CFA also argues the Atrobos Aqua article is neither representative of a broad market average nor reliable because it provides no information on its methodologies or how its prices reflect sales throughout Indonesia. CFA Br. 32-33. In the absence of record evidence showing the source is unreliable or not representative of a broad market average, the court concludes Commerce’s findings are reasonable.²⁹

D. Rice Husk

Agifish argues that the import data in HTS category 1213.00 results in an aberrational value for rice husk. Agifish Br. 30. Commerce must reconsider its choice of Indonesian import data under HTS category 1213.00 to value rice husk because Commerce has failed to adequately address Agifish’s arguments that this data source is not specific and contains aberrational prices.

Agifish argues that the SV derived from this import data resulted in a value of $0.66 per kilogram, which it argues is much too high when compared to Indonesian values for rice for the same POR. Id. Agifish points out that rice husk is simply a waste material from the hulling of rice. Id. Agifish separately points to values for rice placed on the record ranging from $0.22 per kilogram to $0.27 per kilogram for domestic rice prices in Indonesia during the POR. Id. (citing Respondents’ Post-Prelim. SV Data at Ex. 11F). Without data on the record to suggest that rice husk can trade for more than rice, Commerce must explain on remand why it is its SV for rice husk, a waste byproduct of rice, is not abberational.

Commerce acknowledged that rice husk is a byproduct of rice and “should not be at the same price level as the main input.” Final Decision Memo at 40. Nonetheless, Commerce explains that it could not deter mine whether the domestic rice prices are specific to the type of rice that generated the specific husks used by respondents. Id. at 37. Commerce’s explanation is non-responsive. It is unclear to the court how the type of rice the rice husk is derived from matters given the use of the rice husk, he., as an energy source to heat the waste byproducts to extract the oil.

Commerce points to no information indicating that rice husk derived from rice of one type has a significantly different price than rice of another type.³⁰ On remand, it must do so or otherwise explain why the SV generated for rice husk, a fuel source for heating fish waste to produce fish oil and fish meal byproducts, is not aberrational in light of record evidence cited by Agifish. Commerce’s focus on specificity and the HTS category description, which covers “Cereal Straw and Husks, unprepared, Whether or Not Chopped, Ground, Pressed, or in the Form of Pellets,” does not address the evidence that detracts from its finding that the data obtained from this import data is aberrational. The HTS description, by its terms, also includes items other than rice husk.

Agifish also argues that the aberrational nature of Indonesian import data is significantly highlighted by the fact that the import statistics in this HTS category reflect imports from one country in a single month.’ Agifish Br. 29, 31. The court agrees. Commerce has not explained why data reflecting imports over only one month, October of 2012, and from only one country, Australia, are representative of a broad market average. Commerce’s explanation that the quantity of 21,403 kilograms is commercially significant, see Final Decision Memo at 37, does not explain why the data reflects a broad base either temporally or geographically in this case.³¹ On remand, Commerce must reconsider its reliance on the import data to value rice husk or explain why it is reasonable to conclude this data is representative of a broad market average and non-aberrational.

E. Sawdust

Agifísh challenges Commerce’s selection of Indonesian import data under HTS 4401.30 to value sawdust as not specific and aberrational. See Agifish Br. 34-36. Defendant counters that Commerce reasonably found the category is specific, non-aberrational and otherwise meets its SV data selection criteria. See Def.’s Resp. Br. 57-64. Commerce’s selection of Indonesian import data to value respondents’ sawdust FOP is supported by substantial evidence.

Commerce found that Indonesian import data under HTS 4401.30, which covers “Sawdust and Wood Waste and Scrap,” is specific to the “pressed sawdust” used by respondents because the HTS category, by its terms, covers the input. Final Decision Memo at 57 (citing Vinh Hoan Section D Questionnaire Response at 40, PD 122, bar code 3131530-01 (Apr. 22, 2013) (“Vinh Hoan Sec. D Resp.”)). Commerce also concluded that the Indonesian import data meets its other data selection criteria. Id at 57-58. Commerce dismissed the argument that the Indonesian import data is aberrational because only non-comparable benchmark data was submitted to the record.³² Id. at 58. Commerce reasonably concluded these benchmarks are unsuitable because they are not comparable to sawdust.

Commerce found that the Pt. Serba Dowel price quote from Indonesia is specific, publicly available, and tax and duty free but not contemporaneous or representative of a broad market average. Final Decision Memo at 58. Commerce also found this price quote unreliable because it is not “addressed to anyone and it is un clear how the Respondents acquired the price quote .” Id. Commerce reasonably concluded that the- Indonesian import data satisfies the breadth of its SV data source criteria, so it selected Indonesian import data under HTS 4401.30 to value sawdust.

Agifísh argues that-the Indonesian import data is not specific because, although the HTS category includes sawdust, it also éovers “wood waste and scrap.” Agifísh Br. 35. However, Agifish offers no evidence that the HTS import data significantly consisted of non-specific merchandise whose prices differed significantly from the pressed sawdust used by respondents. Although Agifish argues that the range of prices included in this import data is too wide to be sawdust and not a higher value-added product, id., its evidence is misleading. ³³ If a tiny subset of the total imports in the category from only one country are excluded; the range of prices shrinks to $1.10 per kilogram to $1.73 per kilogram. Defendant argues that it would be “absurd to disqualify data for 81,964 kg of imports because 17 of those kg were likely too expensive to be sawdust.” Def.’s Resp. Br. 62. The court agrees.³⁴

F. Labor

Agifish challenges Commerce’s use of Indonesian ILOSTAT Chapter 5 “Fishing, Operation of Fish Hatcheries and Fish Farms; Service Activities Incidental to Fishing” data to value labor. Agifish Br. 32-34. Commerce’s use of ILOSTAT data is reasonable because the data met all of Commerce’s data selection criteria and Commerce reasonably relied upon its practice of using industry-specific labor data from the primary surrogate country.

Commerce found the ILOSTAT data more specific than Indonesian International Labor Organization (“ILO”) Chapter 5B data because the ILOSTAT data covers the fishing industry whereas ILO Chapter 5B data covers a broader spectrum of industries. Final Decision Memo at 47. Consistent with its practice, Commerce preferred the Indonesian ILOSTAT data because it is the most industry-specific data from the primary surrogate country.³⁵ Id. at 44. Commerce did not select the Mymensingh District pangasius farms human labor costs data (“Mymen-singh Data”) or Bangladeshi labor cost by pangasius farming system/feeding practice (“Bangladeshi Labor Data”) because they are not from the primary surrogate country and because they suffer from numerous deficiencies.³⁶ H. at 45. It is reasonable for Commerce to prefer data from the primary surrogate country where such data satisfies its selection criteria because doing so creates consistency with the values for other FOPs, and therefore, also enhances accuracy. Id. at 44-45.

Agifish argues that specificity is a special concern for valuing labor, which Commerce failed to consider adequately. Agifish Br. 33 (citing Dorbest Ltd. v. United States, 604 F.3d 1363 (Fed.Cir.2010); Allied Pac. Food (Dalian) Co. v. United States, 32 CIT 1328, 587 F.Supp.2d 1330 (2008)). The cases Agifish argues support its argument are inapposite. Both Dorbest and Allied Pac, invalidated Commerce’s use of its regression based labor methodology that contradicted the statutory command given to it by Congress. See Dorbest, 604 F.3d at 1372-73; Allied Pac., 32 CIT at 1364, 587 F.Supp.2d at 1361. Here, Commerce used its present labor selection methodology and selected an industry-specific labor data source from the primary surrogate country in accordance with its practice. Agifish does not argue that Commerce’s practice is unreasonable generally or as applied in this review.

G. Fish Waste Byproducts

Agifish argues the average of price quotes from two Philippine pangasius processors, Vitarich Corporation (“Vita-rich”) and Bluebay Aquaculture Inc. (“Bluebay”), used by Commerce to value respondents’ fish waste, fish belly and fish skin byproducts is unreliable and aberrational. Agifish Br. 37-38. Agifish’s arguments are unpersuasive, and Commerce’s selection of the Philippine price quotes is reasonable.

Commerce concluded that the Vitarich and Bluebay price quotes are specific because together these price quotes cover the full range of fish waste byproducts sold by respondents. Final Decision Memo at 54 (citing Vinh Hoan Sec. D Resp. at 41; HVG Section D Questionnaire Response at 21, Ex. D.8, PD 120-121, bar codes 3131496-01-02 (Apr. 22, 2013); CFA SV Data at Ex. 13)). Commerce acknowledged that the price quotes are not as representative of a broad market average as other data sources, but it found them to be publicly available and tax and duty, exclusive. Id. Commerce also acknowledged the price quotes are not contemporaneous since the Vitarich quote is from April of 2010 and the Bluebay quote is from July of 2013. Id. Commerce reasonably concluded that in this case the. superior specificity of these price quotes outweighs their lack of con temporaneity and inferior representativeness.

Agifish asks the court to reweigh the evidence to find the Vitarich price quote unreliable and aberrational. See Agifish Br. 37. Commerce found the Vita-rich price quote consistent with other benchmarks on the record, including price quote data submitted by respondents. Final Decision Memo at 55 (citing CFA Rebuttal Case Br. at 195-196). Commerce accorded little weight to an affidavit questioning the reliability of the price quote.³⁷ See id. at 55; see also Post-Prelim Direct Surrogate Value Submission at Ex. 2, PD 295-96, bar codes 3167265-01-02 (Sept. 6, 2013). Commerce also found any doubts as to the reliability of the Vitarich and Blue-bay price quotes are outweighed by then-superior specificity relative to other data sources, public availability, and tax and duty exclusivity. Final Decision Memo at 55.

In contrast, Commerce found that the Indonesian import data under HTS 0511.90, which includes “Products & Dead Fish, Molluscs Etc., Inedible Nesoi,” does not list the products sold by respondents and includes non-specific material in the form of molluscs and other unspecified sea creatures.³⁸ Id. Commerce also relied upon the fact that the range of Indonesian import prices is too wide to be specific and non-aberrational.³⁹ Id.

Commerce also reasonably determined the Philippine price quotes are superior to the Asian Seafood price quote from Bangladesh, which includes a narrower range of fish waste byproducts, he., only waste, belly and skin, and does not contain individualized prices for multiple kinds of fish waste products. Id. Commerce also found it is unclear whether these price quotes are tax and duty exclusive. Id. Commerce, therefore, reasonably selected the Philip pine price quotes to value respondents’ fish waste byproducts.

H. Fresh Broken Fillets

Agifish challenges Commerce’s use of Vitarich and Bluebay price quotes to value respondents’ fresh broken fillet byproduct. Agifish Br. 39. Agifish reiterates the same arguments made with respect to the reliability of these price quotes to value respondents’ fish waste-byproducts. Id. Commerce’s SV data selection for fresh broken fillet is reasonable.

Commerce concluded that the .Vitarich and Bluebay price quotes are specific because they cover the exact byproduct sold by respondents, he., fish trimmings. Final Decision Memo at 56. Commerce recognized these price quotes are not as representative of a broad market average as other sources, but found the quotes are publicly available, tax and duty exclusive, and reliable for the same reasons discussed above. Id. Finally, as it had in its fish waste analysis, Commerce found that in this ease the superior specificity of the price quotes outweighed other deficiencies.⁴⁰ Id. Commerce weighed the reliability of record evidence and reasonably concluded the price quotes are reliable.

Agifish argues that Commerce lacked substantial evidence to conclude that fresh broken fillets are a waste byproduct. PL’s Reply Opp’n Resps. Mot. J. Agency R. 12, Nov. 6, 2015, EOF No. 104. However, Vinh Hoan described this product as derived from the trimming step. See Vinh Hoan Sec. D Resp. at 41-42. Agifish points to no record evidence to suggest that trimmings and fresh broken meat are not the same or to support its contention that its fish meat is a high value-added product.

I. Fish Oil Byproduct

Agifish challenges Commerce’s determination to “cap” the value of fish oil by constructing a value. Agifish Br. 45-49. Agifish argues that Commerce’s use of the word “cap” is a misnomer because Commerce simply calculated a CV using the inputs consumed in Vinh Hoan’s byproduct production process with no reference to the import data under Indonesian HTS category 1504.20, which includes “Fish Fats & Oils & Their Fractions Exc Liver, Refined Or Not, Not Chemically Mod.” Id. at 48-49. Defendant responds that Commerce’s determination to apply a CV as a cap to value Vinh Hoan’s’fish oil byproduct is reasonable because: (1) the' value of its fish oil input exceeded the value of the main input if the raw’import data were used; and (2) the HTS 'category, by its terms, includes refined fish oil, and record evidence indicates the fish oil produced by Vinh Hoan is unrefined. Def.’s Resp. Br. 89-90. Commerce’s purported “cap” is in fact a rejection of the import data in favor of a- CV. Commerce has not explained the reasonableness of its decision to favor a CV over any other SV data source on the record.⁴¹ The court therefore remands this issue to Commerce for further consideration and explanation,

Commerce has broad discretion in deciding what constitutes the best available information because the term is not defined in the statute. See QVD Food Co. v. United States, 658 F.3d 1318, 1323 (Fed.Cir.2011). However, Commerce must ground its selection of the best available information in the overall purpose of the AD statute, calculating accurate dumping margins. See CS Wind Vietnam Co. v. United States, 38 CIT -, -, 971 F.Supp.2d 1271, 1277 (2014) (citing Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed.Cir.1990)); see also Parkdale Int’l v. United States, 475 F.3d 1375, 1380 (Fed.Cir.2007).

Commerce found that Indonesian import data under HTS 1504.20 is publicly available, broadly representative, tax and duty exclusive, contemporaneous, and specific. See Final Decision Memo at 83. Commerce found that the import data is specific to Vinh Hoan’s “unrefined” fish oil because “unrefined” fish oil is one of the items listed in the HTS heading. Id. at 82. Commerce concluded that Vinh Hoan’s fish oil is unrefined based upon Vinh Hoan’s descriptions of its fish oil, Commerce’s observations at verification from the eighth administrative review, as well as an affidavit Commerce found reliable from the Secretary General of the Association of Indonesian Catfish Entrepreneurs indicating that fish oil from pangasius does not contain free fatty acid and Omega-3 oil. Id. at 81 (citing CFA Rebuttal SV Data at Exs. 4, 10; CFA Rebuttal Surrogate Value Data at Ex. 10, PD 310-311, bar codes 3173134-01-02 (Jan. 10, 2014)).

However, Commerce also concluded that the import data in the HTS is overly broad because it includes refined, as well as unrefined fish oil. See id. at 82 (citing Petitioner’s Submission of Rebuttal Factual Information at Ex. 7, PD 125-130, bar codes 3133680-01-06 (May 2, 2013)). Commerce found the presence of refined fish oil prices within the import category resulted in a SV of $1.73 per kilogram, which it found too high relative to the SV of the subject merchandise from which this byproduct is derived, he., $1.63 per kilogram. Id. at 82. Commerce “capped” the price at the CV of the FOPs and ratios to make fish oil rather than using import data from the HTS category. Id.

Although Commerce purports to be following its practice of choosing the best SV data source, it has actually taken a different approach. In its final determination, Commerce stated that it

will continue to value fish oil using the Indonesian GTA import data under HTS 1504.20.9000 because it is the most specific of the available Indonesian HTS categories on the record and, by its terms, encompasses “unrefined” fish oil. Moreover, the GTA data is contemporaneous with the POR. And, as stated above, the Department previously found GTA data to be publicly available, free of taxes and duties, and representative of broad market averages. However, because of the concerns articulated in the immediately-preceding paragraph, the Department will “cap” the price of HTS 1504.20.9000 at the calculated CV of the FOPs and ratios used by Vinh Hoan to make fish oil, i.e., fish wage, labor and energy, plus surrogate ratios, to ensure that it is a fully-loaded fish oil value.

Id. at 83 (citations omitted). Commerce acknowledges that it has .constructed a value for fish oil rather than relying upon data from the Indonesian HTS category. See Oral Arg. 02:14:50-2:15:05, Mar. 17, 2016, ECF No. 125.⁴² Therefore, the “cap” value bore no relation to the import data Commerce purported to consider the best available information for valuing fish oil. It is unclear why Commerce continues to invoke the Indonesian GTA import data when the value it has chosen has no relationship to it. Commerce’s explanation of why the HTS import data is superior to the other SV data sources on the record is thus irrelevant to its calculation of a SV for Vinh Hoan’s fish oil byproduct. Although the court cannot say Commerce unreasonably concluded that Vinh Hoan’s fish oil is unrefined fish oil (a low value-added product), see Final Decision Memo at 81, Commerce has not explained why it is reasonable to depart from its normal methodology of choosing the best SV data source to value respondents’ byproduct. In fact, Commerce seems to assert that it is following its normal methodology.

Commerce explained that using the SV drawn from Indonesian import statistics without manipulation would have yielded an illogically high value for Vinh Hoan’s unrefined fish oil, see id. at 82, but it does not even try to explain how the cap best ameliorates these issues. Commerce may have good reason to go beyond its stated methodology and construct a value instead of choosing the best available SV data source on the record to value fish oil. If so, Commerce needs to state what it is doing and explain why this alternative methodology is reasonable so that the court may review Commerce’s methodology and determination. For these reasons, the court remands Commerce’s fish oil determination for further consideration and explanation.

J. Brokerage and Handling Expenses

Agifish challenges Commerce’s selection of the World Bank’s “Doing Business 2012: Indonesia” report to value respondent’s brokerage and handling because the report is not specific to prices for brokerage and handling of pangasius fillets. See Agifish Br. 43-44. Defendant responds that Commerce reasonably determined the report is- the best available information for valuing this FOP. See Def.’s Resp. Br. 83-84. Commerce’s selection is reasonable.

Commerce found that the “Doing Business 2012: Indonesia” report is publicly available and representative of a broad market average because it is a broad-based survey of costs in the Indonesian market published by the World Bank. Final Decision Memo at 64. Commerce also concluded that the source is contemporaneous with the POR and tax and duty exclusive. Id, With regard to specificity, Commerce acknowledged that, while the price quote issued by Indonesian company PT Jayantara Setia Sejahtera (“Jayan-tara”), is “arguably more specific than Doing Business,” the price quote does not meet any of the other SV criteria. Id. Commerce also noted that it does not choose price quotes where a broad market average source is available that meets its SV data selection criteria. Id. at 64-65, Commerce reasonably preferred the Indo nesian “Doing Business” report to the same report from Bangladesh- because it favors a source from the primary surrogate country where it otherwise satisfies its SV data selection criteria. Id. at -64.

Agifish argues the report is not specific because it does not reflect brokerage and handling prices for the pangasius industry. Agifish Br. 43. Agifish argues this is the ease because the report contains data only for leading industries, and the pangasius industry is not a leading industry in Indonesia. Id. Commerce acknowledged the Indonesian data is less specific relative to price quote data, but nonetheless concluded it is contemporaneous, representative of a broad market average, tax and duty free, and publicly available.⁴³ Final Decision Memo at 64. Commerce’s finding .is therefore reasonable.

K. Truck Freight

Agifish raises precisely the same specificity objections with regard to Commerce’s selection of the World Bank’s “Doing Business 2012: Indonesia” source to value respondents’ truck freight FOP as it raised in connection with the same source to value respondents’ brokerage and handling. See Agifish Br. 40-41. Commerce’s selection is supported by substantial evidence.

Commerce found the “Doing Business 2012: Indonesia” report contemporaneous with the POR because it is from 2012. Final Decision Memo at 60. Commerce also found this report publicly available and representative of a broad market average because it is a broad-based survey of costs in the Indonesian market with an official nature published by the World Bank. Id. Commerce concluded not only that this report is specific to respondents’ trucking expenses, but also that it is as specific as the Jayantara price quote. Id. Commerce questioned the reliability of the Indonesian price quote on the record because it covers from a secondary source, not the company itself. Id. As with brokerage and handling, Commerce concluded that the Jayantara price quote data is not representative of a broad market average. Id. Commerce found the Indonesian price quote to be specific, but no more so than the Indonesian “Doing Business” report because, although the price quote would be specific to the trucking expenses associated with respondents’ production of panga-sius fillets, it would not be specific to the many other FOPs consumed in their production process. Id. Commerce found the “Bangladeshi Statistics Yearbook” is no more specific than the “Doing Business” report, and non-contemporaneous since it is from 2005. Id. Commerce’s specificity finding as well as its SV data selection for truck freight are reasonable.

IV. Commerce’s Normal Value Calculation for Vinh Hoan

CFA challenges Commerce’s determination not to adjust Vinh Hoan’s NV to exclude glazing weight from the ratio reflecting total quantity of FOPs consumed per kilogram of finished product. See CFA Br. 5-26. Defendant requests a voluntary remand for Commerce to reconsider its comparison calculation. See Def.’s Resp. Br. 92. The court grants Defendant’s request for voluntary remand for Commerce to reconsider its margin calculation.

The court has discretion to grant a remand request when Commerce wishes to reconsider its previous position without confessing error. See SKF USA Inc. v. United States, 254 F.3d 1022, 1029 (Fed.Cir.2001). “Generally, a request for a voluntary remand due to substantial and legitimate agency concerns should be granted.” Timken Co. v. United States, 38 CIT —, -, Slip Op. 14-51, at *5 (May 2, 2014) (citing SKF USA Inc., 254 F.3d at 1029). Commerce has substantial and legitimate concerns if (1) there is a compelling justification for remand, (2) that justification is not outweighed by the need for finality, and (3) the scope of the requested remand is appropriate. Ad Hoc Shrimp Trade Action Comm. v. United States, 37 CIT -, -, 882 F.Supp.2d 1377, 1381 (2013).

Here, Defendant’s request for voluntary remand satisfies the test, and Commerce’s concerns are substantial and legitimate. A remand request for Commerce to correct a potentially erroneous calculation of a dumping margin is a compelling justification. Baroque Timber Industries (Zhongshan) Co. v. United States, 37 CIT -, -, 925 F.Supp.2d 1332, 1339 (2013) (citing Parkdale Int’l, 475 F.3d at 1380). The need for finality here does not outweigh the need to calculate accurate margins, especially in the context of an administrative review. Id. at - - -, 925 F.Supp.2d at 1338-39. No party has objected to Defendant’s request for voluntary remand. Lastly, the scope of Defendant’s remand request, he., for Commerce tó reconsider its export price and NV comparisons for Vinh Hoán and to reopen the record if necessary; is appropriate for this issue. Accordingly, Defendant’s remand request is based on a substantial and legitimate concern. Therefore, the court grants Defendant’s request for voluntary remand to reconsider its margin calculations in light of the concerns raised by CFA.

CONCLUSION AND ORDER

In accordance with the foregoing, it is hereby

ORDERED that Commerce’s Final Determination is sustained with respect to the selection of Indonesia as the primary surrogate country, consideration of arguments raised in Catfish Farmers of America et al.’s rebuttal case brief, and selection of surrogate value data sources to value whole live fish, fingerlings, fish feed, sawdust, labor, fish waste byproducts, fresh broken fillets, brokerage and handling, and truck freight; and it is further

ORDERED that this action is remanded to Commerce to clarify or reconsider, as appropriate, the selection of surrogate value data sdurces to value rice husk and Vinh Hoan’s fish oil byproduct; and it is further

ORDERED that Defendant’s request for voluntary remand is granted for Commerce to reconsider, as appropriate, Vinh Hoan’s margin calculation, which compared combined sales and expense data for glazed and unglazed fillets; 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 Í5 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

. Where Commerce assigns a zero rate to one of its mandatory respondents, its practice is to exclude the zero rate from its averaging when assigning a rate to the separate rate respondents. See Prelim. Decision Memo at 10-11; see also 19 U.S.C. § 1673d(c)(5)(A).

3

. Glazing of frozen fish refers to coating the finished fillet with water and then freezing it. See CFA Br. 6 (citing, e.g., Certain Frozen Fish Fillets from the Socialist Republic of Vietnam, 68 Fed. Reg. 37,116 (Dep't Commerce June 23, 2003) (notice of final anti-dumping duty determination of sales at less than fair value and affirmative critical circumstances) and accompanying Issues and Decision Memorandum for the Antidumping Duty Investigation of Certain Frozen Fish Fillets from the Socialist Republic of Vietnam at 4 n.7, A-552-801, (June 16, 2003), available at http://ia.ita.doc.gov/frn/summary/viemam/03-15794-l.pdf (last visited May 26, 2016)),

4

. Indonesia's 2011 per capita GNI was $2,940. See Catfish Farmers of America Sur- . rogate Comments and Factor Values at Ex. 2-B, PD 145-151, bar codes 3137290-01-07 (May 29, 2013). Vietnam’s 2011 per capita GNI was $1,260. See OP List. The per capita GNIs on the OP List ranged from $770 (Bangladesh) to $2,210 (Philippines). See id. Commerce did not reference Indonesia’s GNI in its preliminary or final determinations, but rather, only referenced CFA’s characterization of Indonesia’s per capita GNI as approximately twice Vietnam’s. See Prelim. Decision Memo at 12.

5

. Further citations to Title 19 of the Code of Federal Regulations are to the 2013 edition.

6

. Although Commerce’s regulations provide that it uses per capita gross domestic product (“GOF’’) as thé measure of economic comparability, Commerce began relying on per cap-ita GNI as opposed to per capita GDP in 2007. Antidumping Methodologies in Proceedings Involving Non-Market Economy Countries: Surrogate Country Selection and Separate Rates, 72 Fed. Reg. 13,246, 13,246 n.2 (Dep’t Commerce Mar. 21, 2007) (request for comment); see also 19 C.F.R. § 351.408(b). No party has challenged the use of GNI to determine economic comparability as contrary to Commerce's regulation. Nonetheless, Commerce's use of GNI to determine economic comparability has been considered reasonable. See, e.g., Clearon Corp. v. United States, 38 CIT -, Slip Op. 14-88, at *9-10 (July 24, 2014) (finding Commerce's reliance on GNI reasonable and in accordance with law).

7

. The court recognizes that Policy Bulletin 04.1 provides that, where no country from the OP List is a significant producer of comparable merchandise or insufficient data (with respect to quantity and quality) is available, Commerce is to request a second list of potential surrogate countries from the OP, and then follow the country selection procedure. See Policy Bulletin 04.1 at 4. However, the OP List put parties on notice that Commerce's practice is to consider other countries not on the OP List if there is adequate record data to do so. See OP List at 1.

8

. The court is mindful of the fact that Vinh Hoan Corp. v. United States, 39 CIT -, 49 F.Supp.3d 1285 (2015) (“Vinh Hoan”), required Commerce to compare the relative economic comparability of countries on its OP List oh the record of that case. Vinh Hoan, 39 CIT at -, 49 F.Supp.3d at 1302. However, in Vinh Hoan, the court found that the record did not demonstrate that Indonesian whole fish data was so superior to Bangladeshi whole fish data that weighing relative GNIs "would not improve Commerce's selection of the best available information.” Id. at -, 49 F.Supp.3d at 1304-05 (citing Ad Hoc Shrimp Trade Action Comm. v. United States, 36 CIT -, - - -, 882 F.Supp.2d 1366, 1374-75 (2012) (“Ad Hoc Shrimp”)). Here, data considerations for respondents’ whole live fish and other key inputs is not so close that Commerce needed to engage in a comparison of relative GNI proximity to support its surrogate country selection. Therefore, VASEP’s argument that Vinh Hoan and Ad Hoc Shrimp control here is misplaced. See Pl.-Intervenor VASEP’s Reply Brief 4-5, Nov. 6, 2015, ECF No. 103 (citing Vinh Hoan, 39 CIT at -, 49 F.Supp.3d at 1303).

9

.Commerce also noted that “Indonesia and Vietnam’s relative levels of per capita GNI remained the same over the past several review periods.” Final Decision Memo at 5 n.17 (citing Petitioner's Rebuttal Brief at 16, CD 232, bar code 3180220-01 (Feb. 10, 2014)).

10

. The fact that Commerce has elected to implement a practice of considering five or , six potential surrogate countries does not necessarily set the limits of economic comparability. The OP generates the list of potential surrogate countries based upon their per cap-ita GNI in relation to the NME country under consideration. Policy Bulletin 04.1 at 2. Commerce explained that its rationale for selecting a fixed number of countries for the OP List is primarily administrative feasibility:

This list is, of course, not exhaustive; there are certainly other countries that could be reasonably viewed as being at Vietnam’s level of economic development. Of course, the number of such countries is potentially large depending upon how broadly the term ''level" in the Act is considered and could be very large, e.g., over 50 countries under and expansive consideration of "level.” It is not administratively feasible for the Department to manage such a long initial list of potential surrogate countries, so the Department considers "level” relatively narrowly and limits the initial list to .five or six countries.

. Prelim, Decision Memo at 13.

11

. VASEP argues that Commerce's economic comparability analysis is contrary to law because it reversed the order of Commerce’s surrogate country selection process, as outlined in its Policy Bulletin 04.1. VASEP Br, 16. This argument relies on VASEP’s flawed premise that ,the OP List sets the limits of economic comparability.

12

. VASEP and Binh An cite several cases they argue support the notion that Commerce may not select a country outside the range of GNIs on its OP List. See VASEP Br. 14-20; Binh An Br. 10-11. None of the holdings from these cases support any such premise. See Dorbest Ltd. v. United States, 604 F.3d 1363, 1371-72 (Fed.Cir.2010) (holding Commerce must select data to value FOPs from economically comparable countries except where such data is not available or irretrievably tainted by some statistical flaw); Dupont Teijin Films v. United States, 38 CIT -, 997 F.Supp.2d 1338, 1345 (2014) (upholding Commerce’s revised surrogate country selection of South Africa, which was among the countries on the OP List from the same review, after consideration of all data placed on the record); Jiaxing Bro. Fastener Co. v. United States, 38 CIT -, 961 F.Supp.2d 1323, 1331-32 (2014) (holding it reasonable for Commerce to decline to select potential surrogate country not included on the OP List); Foshan Shunde Yongjian Housewares & Hardwares Co. v. United States, 37 CIT -, 896 F.Supp.2d 1313 (2013) (holding that the exclusion of India from the surrogate country list was not unreasonable in light of the fact that it had a lower GNI than the countries on Commerce's surrogate pountry list); Dupont Teijin Films v. United States, 37 CIT -, 896 F.Supp.2d 1302, 1307-08 (2013) (holding Commerce had not supported its conclusion that India was economically comparable to the NME in question because it failed to explain its refusal to consider'2009 GNI data and a surrogate country list from another review placed on the record by a party),

13

. VASEP argues Commerce's finding that DAM prices do not represent a broad market average is unsupported by the record because the DAM data represents actual spot prices, segregated by size for most major producing districts. VASEP Br. 29-30. However, the fact that the DAM data may represent actual prices does not ameliorate the fact that the DAM database, does not include data from Mymensingh, the largest pangasius producing region in Bangladesh. VASEP also argues the record contains price data from Mymensingh in the form of hardcopy official DAM pricing circulars. Id, at 31-32. There is no mention of these hardcopy circulars in VASEPs case brief before the agency. See Vinh Hoan/Agifish/VA-SEP Case Brief at 14-18, PD 314, bar code 3175397-01 (Jan. 22, 2014). 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). 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. Therefore, the court need not address VASEP's argument that hardcopy circulars remedy any gaps in the online data. VASEP did argue before the agency that record evidence explained the gaps in data, but the agency did not find those explanations persuasive, see Final Decision Memo 22, and the court declines to reweigh the evidence.

14

. Agifish argues it is

highly improbable that [Bangladesh] may degrade so suddenly and rapidly from being the 'best' .country source for surrogate value data to approximate the subject industry's production experience, to the point that data quality concerns 'compel' disregarding its price information and going off the [surrogate [c]ountry [l]ist.

Agifish Br. 15. Commerce noted that in the seventh administrative review, the DAM data represented 31 of 68 districts in Bangladesh and 767 price observations, reflecting 115,000 mt. Final Decision Memo at 21. Commerce also observed that Mymensingh, the largest pangasius producing district in Bangladesh had been reported in the DAM data in that review. Id. In the eighth administrative review, Commerce noted that the DAM data represented 27 of 68 districts and 491 data points, reflecting 52,000 mt. Id. No data from Mymensingh had been reported in the DAM data in the eighth administrative review. Id. Agifish points to no record evidence that detracts from Commerce's finding regarding the declining coverage of the DAM data to support its bare assertion. Commerce's conclusion is reasonable,

15

. Commerce noted that DAM admitted its online data contained anomalous data for at least eighteen months. Final Decision Memo at 26. Commerce found IAS data did not present the same reliability concerns. IcL at 17 (citing CFA SV Data at Ex. 6),,27.

16

. VASEP attempts to undermine the affidavits submitted by petitioners that Commerce credited as supporting its finding that DAM data included dead fish, arguing that they were based on second-hand hearsay evidence. VASEP Br. 31. Commerce's conclusion is reasonable and the court declines to reweigh the evidence.

17

. In contrast, Commerce found that IAS data is specific to whole fish and that specific steps are taken to ensure the data does not include dead fish based upon an, affidavit submitted from the director of IAS from 2011. Final Decision Memo at 24-25. Commerce also concluded that the record evidence that other species of pangasius may be included in IAS data does not distort prices because of evidence indicating that pangasius jambal sell at identical prices and that prices for fish • grown in aquaculture methods other than those used by respondents are not significantly different than those grown using aquaculture techniques used by respondents. IcL at 25.

18

. Binh An argues that Commerce unreasonably considered non-fish FOPs equally in selecting a primary surrogate country when in past reviews it selected a surrogate country based upon whole live fish and financial ratios alone. Binh An Br. 16. But Commerce reasonably explained that the increased integration in the Vietnamese pangasius industry leads it to consider a wider range of FOPs in this review than in past reviews. Final Decision Memo at 14. The court does not find that Commerce based its surrogate country selection on minor FOPs. Rather, Commerce reasonably relied principally upon its data selection for whole live fish and farming FOPs.

19

. VASEP and Binh An argue that Commerce failed to give proper weight to specificity in its surrogate country selection process, which they argue trumps contemporaneity and other considerations in • evaluating the quality of data sources. VASEP Br. 34-35; Binh An Br. 18-19 (citing Taian Ziyang Food Co. v. United States, 35 CIT -, 783 F.Supp.2d 1292 (2011)). However, in Taian Ziyang, the court held only that Commerce’s analysis of " 'the best available information’ is not a straightforward exercise in arithmetic,” and that Commerce gave too little weight to specificity in that case because its analysis of specificity was conclusory. Taian Ziyang, 35 CIT at -, 783 F.Supp.2d at 1330. The court did not hold that product specificity is the most important consideration in selecting a SV data source as VASEP contends. See id. The court held Commerce did not adequately explain its decision to favor contemporaneity over the specificity evidence on the record. See id. No party argues that Commerce favored contemporaneity at the expense of specificity in this review.

20

. Agifish does not frame its argument in the context of a standard of review. Agifish does not argue that Commerce lacked authority to interpret its regulation to consider petitioner’s alternative argument that it should consider another SV source a rebuttal argument. Rather, Agifish argues that allowing petitioners to raise its arguments for the first time deprived Agifish of the opportunity to address the information sources and arguments they relied upon, Agifish Br. 50-51. Therefore, Agifish is arguing that Commerce should not have interpreted its regulation in the manner it has because it is unfair. See id. The court views this argument within an abuse of discretion standard of review.

21

. Binh An adopts the arguments of the other respondents regarding the quality of data of the FOPs. Binh An Br, 18-19.

22

. As the court already discussed, Commerce also relied upon its conclusion that the coverage of the DAM data source had declined over the past three periods of review. Final Decision Memo at 15, 21-22, Commerce attached great significance to this decline because it found that, while production of pangasius rose dramatically over the POR, the DAM data covered a declining share of Bangladesh's production. Id at 22. As also already discussed, Commerce was dissatisfied that the DAM data did not include data from Mymen-singh, the largest pangasius producing district, during the POR. Id. at 21. Additionally, Commerce concluded that the Philippines FS data source does not represent as robust a source because it represents a lower volume of production. Id. In contrast, Commerce found that IAS data contains data for 31 of 33 districts, 154 data points, arid represented approximately 490,000 mt of pangasius production. Id. at 21.

23

. Commerce cited an affidavit of a DAM official admitting there can be difficulties with the software filter used to catch anomalies and errors in support of its finding, and Commerce furtiier noted these findings are corroborated by other affidavits on the record recounting interviews that indicate DAM does not regularly vet its data for errors. Final Decision Memo at 26. (citing Petitioners’ Rebuttal Surrogate Value Data at Ex. 23, PD 165-171, bar codes 3140449-01-07 (June 14, 2013); Petitioners’ Surrogate Value Data at Ex, 3, PD 289-294, bar codes 3167326-01-05 (Dec, 6, 2013)). Commerce also cited several affidavits on the record from DAM officials admitting that data is incorrect or anomalous. Id. (citing Respondents’ Rebuttal Surrogate Value Data at Ex. LA, PD 305-309, bar codes 3172989-01-05 (Jan. 10,'2014); Agifisb/Vinh Hoan Surrogate Value Data at Ex, 14A, PD 141-144, bar codes 3137305-01-04 (May 24, 2013)).

24

. Commerce concluded that the Rukmono Affidavit provides fingerling prices that are publicly available, contemporaneous, tax and duty exclusive, and representative of a broad market average. Final Decision Memo at 29. ■ Commerce also concluded that the Rukmono Affidavit is the most specific because it provides eight size bands of fingerlings. Id.

25

. Commerce considered the Soetrisno Affidavit unreliable because the affidavit does not indicate whether the size variable referred to length, width, or height. Final Decision Memo at 30. Agifish points to no record evidence undermining Commerce’s conclusion concerning unreliability of the Soetrisno Affidavit.

26

.Hung Vuong Group ("HVG”), one of the mandatory respondents, and its Saffiliates reported purchasing fingerlings on a per kilogram basis without indicating the length of those fingerlings. See Final SV Memo at 2-3. To obtain a length of fingerlings, Commerce divided the number of fingerlings purchased by the number of kilograms purchased. Id. Then, using the Rukmono Affidavit, Commerce obtained a size-specific value. Id. For HVG, Commerce determined that the size closest to those purchased by HVG are the 5.0-6.0 inch range, which the Rukmono Affidavit valued at 39,500 Indonesian rupiah per piece.

VDTG, an affiliate of Vinh Hoan, the other mandatory respondent, did not record the size of fingerlings it purchased, either by weight or by length. Id Lacking any information concerning the size of VDTG's fingerlings, Commerce assumed VDTG purchased fingerlings of average value from the list of values in the Rukmono Affidavit. Id. at 3. Commerce then determined this average value corresponds to the 2.5-3.0 inch fingerling size range contained in the Rukmono Affidavit. Id. Commerce converted the average value to a per kilogram value using a conversion factor of 588 fingerlings per kilogram contained in the Rukmono Affidavit, resulting in a value of 150,013 Indonesian rupiah per piece for Vinh Hoan’s fingerlings. Ld

27

, CFA argues product specificity is the primary consideration in assessing the best available information. CFA Br. 28-29 (citing Jinan Yipin Corp. v. United States, 35 CIT —, -, 800 F.Supp.2d 1226, 1304 (2011); Taian Ziyang, 35 CIT at -, 783 F.Supp.2d at 1330). As already discussed, Taian Ziyang does not hold that product specificity is the most important consideration. See also Jinan Yipin, 35 CIT at -, 800 F.Supp.2d at 1304 (holding that Commerce’s specificity finding was unsupported where it provided no explanation in support). CFA also argues that the Court of Appeals for the Federal Circuit has given specificity greater weight than the other SV data selection criteria, particularly when certain characteristics of the input can strongly influence price. CFA Br. 29 (citing Qingdao Sea-Line Trading Co. v. United States, 766 F.3d 1378, 1386 (Fed.Cir.2014)). Yet, Qingdao affirmed Commerce’s determination to assign greater weight to product size over contemporaneity in valuing garlic. See Qingdao Sea-Line Trading Co., 766 F.3d at 1386. The court did not hold that specificity is generally given greater weight than other SV data selection criteria, but rather,. recognized Commerce’s discretion in determining the best available information and its preference for selecting SV data sources that satisfy the breadth of its criteria. See id. .

28

. CFA also argues that Commerce’s determination in this review is arbitraiy in light of its decision to favor protein content when evaluating the same exact sources in the tenth administrative review. CFA Br. 30-31 (citing Certain Frozen Fish Fillets From the Socialist Republic of Vietnam, 80 Fed. Reg. 2,394 (Dep’t Commerce Jan. 16, 2015) (final results of AD administrative review; 2012-2013); Certain Frozen Fish Fillets from the Socialist Republic of Vietnam; Issues and Decision Memorandum for the Final Results of the Tenth Antidumping Duty Administrative Review; 2012-2013 at 39-40, A-552-801, (Jan. 7, 2015), available at http://ia.ita.doc.gov/frn/ summary/vietnam/2015-00649-l.pdf (last visited May 26, 2016)). However, the records in each segment of a proceeding are highly fact-specific to the period that is investigated, and, since the record of that proceeding is not before the court, the court declines to consider Commerce's determination in the tenth administrative review in assessing whether its determination here is supported by substantial evidence.

29

. Moreover, Defendant points out that the article evidences a discussion of national prices. Def.’s Resp. Br. 49-50 (citing Vinh Hoan/Agifish/VASEP Supplement to Post-Preliminary Direct Surrogate Value Data at Ex. 1, PD 298, bar code 3167703-01 (Dec. 11, 2013)).

30

. At oral argument, Defendant cited the following documents it stated supported the ranges of prices for rice and rice husk on the record: (1) Respondents’ Case Brief at Ex. 16; (2) Catfish Farmers of America' Surrogate Comments and Factor Values at Ex. 32, PD 145-151, bar codes 3137290-01-07 (May 29, 2013); and (3) Respondents’ Rebuttal Surrogate Value Data at Ex. 31A, PD 172-174, bar codes 3140530-01-04 (June 17, 2013). See Oral Arg. 02:34:30-02:35:31, Mar. 17, 2016, ECF No. 125. However, Defendant acknowledged there is no data on the record about how certain rice husks correlate to certain kinds of rice. The only variation in rice husk price supported by documents cited by Defendant is whether the rice husk was pressed or unpressed.

31

. Commerce cites two proceedings it implies support the notion that it will not decline to use import data because it comes from a single country. Final Decision Memo at 38 (citing Persulfates from the People's Republic of China, 68 Fed. Reg. 68,030 (Dep't Commerce Dec. 5, 2003) (final results of antidumping duty administrative review) and accompanying Issues and Decision Memorandum for the 2001-2002 Antidumping Duty Administrative Review of Persulfates from the People's Republic of China at 6-11, A-570-847, (Nov. 28, 2003), available at http://enforcement.trade.gov/frn/summary/prc/ 03-30259-l.pdf ("Persulfates from China I&D”) (last visited May 26, 2016); Wooden Bedroom Furniture From the People’s Republic of China, 75 Fed. Reg. 50,992 (Dep’t Commerce Aug. 18, 2010) (final results and final rescission in part) and accompanying Issues and Decision Memorandum for the Final Results of the Administrative Review of the Antidumping Duty Order on Wooden Bedroom'Furniture from the People’s Republic of China at 42-44, A-570-890, (Aug. 11, 2010), available at http://enforcement.trade. gov/frn/summary/prc/2010-20499-1 .pdf ("Wooden Bedroom Furniture from China I&D”) (last visited May 26, 2016)). However, neither proceeding permits Commerce to depart from evaluating the representativeness of a SV data source or to conclude that a source is representative of a broad market average because it reflects commercially significant volumes of imports.

In the Persulfates from China I&D, Commerce discussed past proceedings where it rejected using financial statements based on consideration of various factors, including where surrogate values for the main input in the production of subject merchandise were unreliable because they varied significantly from other surrogate values on the record, were primarily based on imports for one country, and were based on a relatively low quantity of imports, among other circumstances. Persulfates from China I&D at 8-9. In the Wooden Bedroom Furniture from China I&D, Commerce discussed various inquiries it undertakes to assess whether import data from a potential surrogate country is aberrational, including comparing the import data to import data from other economically comparable countries and comparing the volumes of imports into the potential surrogate country to the volumes of imports into other economically comparable countries. Neither proceeding referenced by Commerce explains why Indonesian imports from Australia from a single month were representative of a broad market average even when read together with Commerce’s finding that Australian imports were commercially significant.

32

. Agifish argues that the SV generated from Indonesian import statistics under this HTS category is, on its face, aberrational because the value is $1.18 per kilogram while subject merchandise is valued at $1.61 per kilogram. Agifish Br. 36. Agifish argues it does not stand to reason that a kilogram of whole fish costs only $0.43 per kilogram more than a kilogram of sawdust. Id, Agifish also argues that Commerce’s SV for sawdust at $1.18 per kilogram is absurd given that it is approximately 1.7 times that of coal at $0.69 per kilogram. Id. However, Agifísh does not point to any record evidence correlating sawdust values with those of whole fish or coal. Moreover, given the wide range in weights of the commodities for which Agifish placed SV data on the record, the court cannot agree that the values obtained by Commerce are absurd on their face in the absence of such record evidence.

Agifish argued at oral argument that it makes no commercial sense for respondents to prefer sawdust at $1.18 per kilogram over coal at $0.69 per kilogram as a fuel source to generate heat for producing byproducts. See Oral Arg. 01:45:07-01:45:35, Mar. 17, 2016, ECF No. 125. Again, Agifish cites to no record evidence that coal would be a more practical fuel source in the context of. respondents' production process. Without such evidence, this argument is speculative.

33

. Agifish cites record evidence that tire average prices in Indonesian import data for sawdust ranged from a low of $1.10 per kilogram to a high of $29.24 per kilogram. Agifish Br. 35, Ex. 4 (derived from Preliminary Surrogate Value Source Documents at Att. II.D at Products Extract at 78-79, PD 263-270, bar codes 3153382-01-06 (Sept. 3, 2013)). However, to generate this wide range of average values, Agifish relies upon 17 kilograms of imports from Luxembourg valued at approximately $497 per kilogram. IcL at Ex. 4. The import .category,, as a whole, contained a total of 81,964 kilograms of imports. See Preliminary Surrogate Value Source Documents at Attach. II.D at Products Extract at 78-79, PD 263-270, bar codes 3153382-01-06 (Sept. 3, 2013).

34

. The court recognizes that Commerce does not rely upon this logic to discount Agifish’s argument in its final determination. The court would have preferred that Commerce address this argument specifically. However, given the small volume of imports relied upon by Agifish to make this argument, Commerce’s reasoning that these 17 kilograms of imports is insufficient to discount 81,964 kilograms of total imports in the category, is reasonably discernible.

35

.In evaluating Indonesian ILO Chapter 5B data, Commerce analyzed publicly availability, representativeness of a broad market average, and tax and duty exclusivity. See Final Decision Memo at 47. However, Commerce’s practice is to use industry-specific labor rates from the primary surrogate country to value respondents' labor. See id. at 44 (citing Anti-dumping Methodologies in Proceedings Involving Non-Market Economies: Valuing the Factor of Production: Labor, 76 Fed. Reg. 36,092, 36,093 (Dep’t Commerce June 21, 2011)).

36

. While Commerce found the Mymensingh Data specific to the subject merchandise, it found that the data source do not meet any of the other data selection criteria. Final Decision Memo at 46. Commerce noted that the record lacks information to support finding that the Mymensingh Data is publicly available and tax and duty exclusive. Id. Commerce found the Mymensingh Data does not represent a broad market average because it only provides data from one district in Bangladesh and is not contemporaneous with the POR because it offered data from 2008, Id.

Commerce found that the Bangladeshi Labor Data is publicly available, tax and duty exclusive, and a broad market average. Id, However, Commerce found that the Bangladeshi Labor Data is not specific or contemporaneous. Id. Commerce found the data is not specific because the labor rate is not cálculated on a per hour basis, as reported by respondents, and the rate “included labor components, such as unpaid family labor and payments made by food/tobacco.” Id. Commerce noted that the Bangladeshi Labor Data is not contemporaneous with the POR because it is from October 1, 2005 through February 15, 2006. Id. at 46-47.

37

. Agifish argues that Commerce could not reasonably have concluded the Vitarich price quote was reliable because: (1) it could not be duplicated; and (2) it was obtained by an attorney, not in the ordinary course of business. Agifish Br. 37-38. Commerce accorded the competing affidavit submitted by respondents little weight because “the record does not fully detail the efforts [taken] to replicate the price quote.” Final Decision Memo 55. Agifish’s challenges are based upon speculation as to the circumstances in which the price quote was given. It is not for the court to reweigh the credibility of record evidence before Commerce.

38

. Agifish argues that Commerce’s finding that the Indonesian import data under HTS 0511.90 is not specific is belied by its finding in the seventh administrative review. Agifish Br. 38. However, in that review, Commerce did not find the Indonesian import data specific, but rather, found it more specific relative to other data on the record and import ' data better satisfied its data selection criteria. See Certain Frozen Fish Fillets From the Socialist Republic of Vietnam, 77 Fed. Reg. 15,039 (Dep’t Commerce Mar. 14, 2012) (final results and partial rescission of the seventh AD administrative review) and accompanying Certain Frozen Fish Fillets from the Socialist Republic of Vietnam ("Vietnam”): Issues and Decision Memorandum for the Final Results of the Seventh Antidumping Duty Administrative Review at 18, A-552-801, (Mar. 7, 2012), available at http://ia.ita.doc.gov/frn/summary/ vietnam/2012-6201-l.pdf (last visited May 26, 2016).

39

.Although Commerce did not specifically note the magnitude of the disparity in import prices in its Final Decision Memo, it did reference CFA’s Rebuttal Brief, which showed that the price quotes ranged from $0.09 per kilogram to $0.37 per kilogram whereas the Indonesian Import data showed a price of $1.33. See Final Decision Memo at 55 (citing CFA Rebuttal Case Br. 196 (deriving information from CFA SV Data at Exs. 9-A, 9-C, 13); Catfish Farmers of America Surrogate Comments and Factor Values at Ex. 38-B, PD 145-151, bar codes 3137290-01-07 (May 29, 2013). It is reasonably discernible that Commerce considered this disparity in making its finding. '

40

. Commerce reasonably found Indonesian import data under HTS 0304.19, which includes “Fish Fillets, Fresh, or Chilled, Other” non-specific because it found that the category, by its terms, includes whole, unbroken fish fillets. Final Decision Memo at 56 (citing Respondents’ Post-Prelim. SV Data at Ex. 10.A,).

Agifish argues that Commerce's determinations in the seventh and eighth administrative reviews seriously detract from its finding that the Indonesian import data is not specific. Agifish Br. 39. However, in the seventh administrative review, Commerce found that HTS 0304.19 was less specific than another HTS category because it included species other than pangasius. Certain Frozen Fish Fillets from the Socialist Republic of Vietnam, 77 Fed. Reg. 15,039 (Dep’t Commerce Mar. 14, 2012) (final results and partial rescission of the seventh AD administrative review) and accompanying Certain Frozen Fish Fillets from the Socialist Republic of Vietnam (“Vietnam”): Issues and Decision Memorandum for the Final Results of the Seventh Antidumping Duty Administrative Review at 20, A-552-801, (Mar. 7, 2012), available at http://ia.ita.doc. gov/frn/summary/vietnam/2012-6201-1 .pdf (last visited May 26, 2016).

41

. Commerce acknowledged there are five potential SV data sources on the record from economically comparable countries to value fish oil other than Indonesian import data, including price quotes from Indian companies for "unrefined fish oil,". Final Decision Memo at 78, 84. Commerce did not explain why it is reasonable to depart from its SV methodology by constructing a value or why it could not manipulate an alternative data source to arrive at an accurate SV.

42

. Moreover, counsel for CFA also pointed out that every entry into Indonesia of fish oil during the POR was at a higher value than the $1.73 per kilogram cap Commerce derived for fish oil. See id. at 02:17:15-02:17:33.

43

. Agifish also argues that the superior specificity alone favors selecting the Jayantara price quote, but the court does not find Commerce's selection unreasonable despite the specificity concerns raised by Agifish given that the import data satisfied the balance of its SV data selection criteriá. Moreover, Agifish does not cite-record evidence that Commerce’s reliance upon this data source results in aberrationally high values for brokerage and handling.