Full opinion text
OPINION AND ORDER Kelly, Judge: This action is before the court on US-CIT Rule 56.2 motions for judgment on the agency record challenging the U.S. Department of Commerce’s (“Department” or “Commerce”) determination in the first administrative review of the antidumping duty (“ADD”) order covering crystalline silicon photovoltaic cells, whether or not assembled into modules, from the People’s Republic of China (“China” or “the PRC”). See SolarWorld’s Mot. J. Agency R., Apr. 15, 2016, ECF No. 62; Consolidated PL Goal Zero LLC’s Rule 56.2 Mot. J. Agency R., Apr. 15, 2016, ECF No. 60-1; Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, From the [PRC], 80 Fed. Reg. 40,998 (Dep’t Commerce July 14, 2015) (final results of ADD administrative review and final determination of no shipments; 2012-2013) (“Final Results”) and accompanying Decision Memorandum for the Final Results of the 2012-2013 [ADD] Administrative Review of Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled into Modules, From the [PRC], Sept. 16, 2015, ECF No. 20-5 (“Final Decision Memo”). Plaintiff, SolarWorld Americas, Inc. (“SolarWorld”), commenced this action pursuant to section 516A of the Tariff Act of 1930, as amended, 19 U.S.C. § 1516a(a)(2)(B)(iii) (2012). See Summons, Aug. 12, 2015, ECF No. 1. The action was consolidated with an action filed by Goal Zero, LLC (“Goal Zero”), which challenges aspects of the same determination. See Order, Nov. 3, 2015, ECF No. 39. Goal Zero challenges on several grounds Commerce’s application of the China-wide rate of 249.96 percent to ERA Solar Co., Ltd. (“ERA Solar”), the exporter of subject merchandise imported by Goal Zero. See Consolidated PI. Goal Zero LLC’s Br. Supp. Rule 56.2 Mot. J. Agency R. Final Confidential Version, Apr. 15, 2016, ECF No. 59 (“Goal Zero Br.”). First, Goal Zero argues that Commerce’s presumption that Chinese companies are subject to government control is contrary to law. See id. at 10-19. Second, Goal Zero challenges the assignment of an ADD rate based on adverse facts available (“AFA”) to ERA Solar, arguing that Commerce improperly imputes non-cooperation to ERA Solar without record evidence. See id. at 19-25. Third, Goal Zero challenges as uncorroborated. the AFA rate assigned to the China-wide entity. See id. at 26-38. SolarWorld challenges four aspects of Commerce’s final determination. See So-larWorld Americas Inc.’s Mem. Supp. Rule 56.2 Mot. J. Agency R. Revised Confidential Version, Apr. 15, 2016, ECF No. 61 (“SolarWorld Br.”). First, SolarWorld challenges Commerce’s selection of surrogate value .import data for two inputs (steel frames and semi-finished polysilicon ingots and blocks) .used in the production of subject merchandise. See id. at 14-24. Second, SolarWorld questions Commerce’s decision to include a line item identified as “other income” in its surrogate financial ratio calculation. See id. at 42-44. Third,' Solar-World’challenges Commerce’s inclusion of sales made by mandatory respondent Wuxi Suntech Power Co., Ltd. (“Suntech”) to an affiliated company in the United States as constructed : export price (“CEP”) sales when calculating Suntech’s dumping margin. See id. at 29-33. In relation to this challenge, SolarWorld also claims that Commerce should have determined the date of sale for Suntech’s reported sales to its affiliate based on the date the contract whs initially executed. See id. at 33-39. Fourth, SolarWorld contests Commerce’s use of factors of production (“FOP”) usage data from Suntech and certain tollers. See id. at 39-42. For the reasons that follow, the court sustains Commerce’s application- of the China-wide AFA rate of 249.96 percent to ERA Solar as supported by substantial evidence and in accordance with law, Commerce’s surrogate value data selections to value solar frames and semi-finished silicon ingots and blocks, Commerce’s inclusion of the “other income” line item in its surrogate financial profit calculation, the inclusion of sales to Suntech as CEP sales, and Commerce’s acceptance of FOP usage data from Suntech and certain tollers as facts available. However, the court remands for further explanation or reconsideration consistent with this opinion Commerce’s determination to use the date of shipment rather than the date of contract for Suntech’s sales to its affiliate. BACKGROUND On February 3, 2014, Commerce initiated the first administrative review of the ADD order on crystalline silicon photovoltaic cells, whether or not assembled into modules, from China. See Crystalline Silicon Photovoltáic Cells, Whether or Not Assembled Into Modules, From the [PRC], 79 Fed. Reg. 6,147 (Dep’t Commerce Feb. 3, 2014) (initiation of antidumping and countervailing duty administrative reviews and request for revocation in part). Commerce- selected Yingli Energy (China) Company Limited (“Yingli”) and Suntech as mandatory respondents in this review. See Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, From the [PRC], 80 Fed. Reg. 1,021 (Dep’t Commerce Jan. 8, 2015) (preliminary results of ADD administrative review and preliminary determination of no shipments; 2012-2013) (“Prelim. Results”). To calculate a surrogate value (“SV”) for Yingli’s aluminum frames, Commerce preliminarily used import data corresponding to Thai Harmonized Tariff System (“HTS”) heading 7604. Decision Memorandum for the Preliminary Results of the 2012-2013 [ADD] Administrative Review of Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, From the [PRC] at 1, PD 513, bar code 3250265-01 (Dec. 31, 2014) (“Prelim. Decision Memo”). Commerce used the world price for polysilicon to calculate a SV for semi-finished ingots and blocks purchased by Yingli in the production of subject merchandise. Id. at 2-3. Commerce also preliminarily excluded a line item labeled “other income” from the surrogate profit ratio calculation. See [ADD] Administrative Review of Crystalline Silicon Photovoltaic Cells from the [PRC]: Factor Valuation Memo, A-570-979, 9, Attach. V, PD 530, bar code 3252163-01 (Dec. 31, 2014). Commerce preliminarily assigned Yingli Energy (China) Company Limited a rate of 1.82 percent. See Prelim. Results, 80 Fed. Reg. at 1,023. Commerce preliminarily assigned Suntech the China-wide rate of 238.56 percent, id. at 1,023 n.11, because Commerce determined Suntech and entities with which it was collapsed had not demonstrated an absence of de facto government control over export activities. Prelim. Decision Memo at 14. Commerce preliminarily determined that ERA Solar did not establish eligibility for a separate rate because ERA Solar did not submit a separate rate application. See Prelim. Decision Memo at 15. Further, because certain entities did not respond to Commerce’s request for quantity and value (“Q & V”) information, Commerce determined that the China-wide entity failed to cooperate to the best of its ability. Prelim. Results, 80 Fed. Reg. at 1,022, 1,023 n.11. Accordingly, Commerce preliminarily assigned ERA Solar the AFA rate of 238.56 percent assigned to the China-wide entity, id., which is a dumping margin from the petition. See Prelim. Decision Memo at 36. In its final results, Commerce continued to use import data under Thai HTS heading 7604 to value Yingli’s aluminum frames inputs. Final Decision Memo at 76, 81-84. Commerce also continued to use the world price for polysilicon to obtain a SV for semi-finished ingots and blocks. See id. at 76. Commerce determined that the “other income” line item that had been preliminarily excluded from its surrogate financial profit ratio calculation should be included in surrogate financial profit for the final results. See id. at 60. To calculate the amount of certain FOPs consumed in the production of subject merchandise, Commerce used FOP data on the record from Suntech and from certain unaffiliated tol-lers that supplied FOP usage data as a substitute for missing FOP usage data of other unaffiliated tollers who had not provided usage information. See id. at 40-41. Commerce included sales reported by Sun-tech to its affiliate as CEP sales, because Commerce found that the reported sales price from Suntech to its affiliate was the same price as the price the affiliate sold the merchandise to the first unaffiliated party. See id. at 91-94. To determine the date of sale of Suntech’s U.S. sales in order to ensure a temporal nexus between home market sales and U.S. sales for Commerce’s margin calculation, Commerce used the date of shipment because Commerce determined that the prices in Suntech’s contracts for sale of subject merchandise can change after the original contract is executed. Id. at 93. These changes resulted in Commerce assigning a 33.08 percent margin to mandatory respondent Suntech and a de minimis margin to Yingli. Final Results, 80 Fed. Reg. at 41,001. Commerce also continued to find that ERA Solar was ineligible for a separate rate. Final Decision Memo at 11-12. Commerce assigned the China-wide entity, including ERA Solar, a rate of 238.95 percent. Final Results, 80 Fed. Reg. at 41,002, 41,002 n.49. 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. Commerce’s Application of AFA to Goal Zero Goal Zero challenges Commerce’s application of the non-market economy (“NME”) presumption to ERA Solar, Goal Zero’s exporter, as unsupported by substantial evidence. See Goal Zero Br. 10-19. Further, Goal Zero argues that the application of a China-wide rate based on AFA is contrary to law and unsupported by substantial evidence. See id. at 19-25. Lastly, Goal Zero challenges the AFA rate assigned to the China-wide entity as contrary to law and unsupported by substantial evidence because it is uncorroborated. See id. at 26-38. The court addresses each challenge in turn. A. Application of the NME Presumption Goal Zero argues that Commerce’s application of a rebuttable presumption that all respondents are government controlled is contrary to law because it is inconsistent with Commerce’s findings in countervailing duty (“CVD”) proceedings that market-oriented reforms render many companies’ export activities independent of the Chinese government. See Goal Zero Br. 10-15. Defendant responds that the court should decline to decide this issue because Goal Zero failed to exhaust this issue at the administrative level. Def.’s Mem. Opp’n Pis.’ Rule 56.2 Mots. J. Upon Agency R. Confidential Version 9-10, Dec. 14, 2016, ECF No. 80 (“Def.’s Resp. Br.”). Defendant argues that, even if Goal Zero exhausted its administrative remedies, Commerce’s application is consistent with its longstanding practice and supported by substantial evidence. Id. at 10-17. The court first addresses Defendant’s exhaustion argument. Thereafter, the court proceeds to address the legal bases for Commerce’s application of the NME presumption to ERA Solar. 1. Exhaustion of Administrative Remedies Defendant argues that Goal Zero failed to exhaust its administrative remedies. Def.’s Resp. Br. 9-10. Goal Zero responds that its challenge should be reviewed by the court because other parties raised and extensively briefed the challenge to Commerce’s NME presumption at the administrative level. Reply Br. Consolidated PI. Goal Zero, LLC Confidential Version 2-7, Feb. 16, 2017, ECF No. 100 (“Goal Zero Reply Br.”). The court reaches Goal Zero’s challenge to the NME presumption because Commerce had ample opportunity to address the arguments supporting this challenge. Congress has directed that the Court “shall, where appropriate, require the exhaustion of administrative remedies.” 28 U.S.C. § 2637(d). Although the statute requires that parties exhaust their remedies before the-pertinent administrative agencies, it also allows the court some degree of discretion to consider circumstances where a strict application of the rule may be inappropriate. See 28 U.S.C. § 2637(d); see also Corus Staal BV v. United States, 502 F.3d 1370, 1379 (Fed. Cir. 2007) (stating that the Court should ordinarily insist that parties exhaust their remedies before the pertinent agency, but the statutory injunction is not absolute). The overarching purpose of the exhaustion doctrine is to “allow[ ] the agency to apply its expertise, rectify administrative mistakes, and compile a record adequate for judicial review—advancing the twin purposes of protecting administrative agency authority and promoting judicial efficiency.” Carpenter Tech. Corp. v. United States, 30 CIT 1373, 1374-75, 452 F.Supp.2d 1344, 1346 (2006) (citing Woodford v. Ngo, 548 U.S. 81, 88-90, 126 S.Ct. 2378, 165 L.Ed.2d 368 (2006)). Here, Commerce had sufficient opportunity to address the legal viability of the NME presumption and its applicability in this proceeding in its final determination because Suntech’s brief before the agency raised the same issues. See Case Brief of Wuxi Suntech Power Co., Ltd. at 6-12, CD 577, bar code 3275378-01 (May 8, 2015). Indeed, Commerce squarely addressed these arguments, which are materially identical to those raised by Goal Zero here, in its final determination. See Final Decision Memo at 98-101. Commerce had ample opportunity to apply its expertise, address any errors made at the administrative level, and develop a record upon which to ensure adequate judicial review. 2. Commerce’s Application of the NME Presumption to ERA Solar Goal Zero contends that Commerce’s application of the NME presumption is unsupported by substantial evidence because the presumption is outdated and inconsistent with Commerce’s recent findings that the Chinese economy is not composed of a single economic entity. See Goal Zero Br. 10-15. Defendant responds that the NME presumption is in accordance with law and that Commerce’s findings that the NME presumption should be applied to ERA Solar are not inconsistent with Commerce’s findings that CVD law is applicable to China. Def.’s Resp. Br. 17-22. The court agrees with the Defendant. A NME country is defined in the statute as “any foreign country that [Commerce] determines does not operate on market principles of cost or pricing structures, so that sales of merchandise in such country do not reflect the fair value of the merchandise.” 19 U.S.C. § 1677(18)(A). The statute provides a non-exhaustive list of factors for Commerce to take into account in determining whether a country is a NME country. See 19 U.S.C. § 1677(18)(B)(i)-(vi). Once Commerce determines that a country is a nonmarket economy, that determination shall remain in effect until revoked. See 19 U.S.C. § 1677(18)(C). In cases involving imports from a NME country, Commerce’s regulations permit it to calculate a single dumping margin applicable to all exporters and producers that are not individually examined. See 19 C.F.R. § 351.107(d) (2014). Nothing in the statute requires that Commerce determine that exporters operating within a NME country may or may not operate freely of government control. Commerce acts within its discretion to develop a reasonable methodology for determining whether an exporter operates independently of government control in a NME country. Therefore, it is not unreasonable for Commerce to presume government control of an exporter in a NME country. Commerce’s practice requires that a party’s separate rate status be established in each segment of the proceeding in which the party is involved because a company’s corporate structure, ownership, or relationship with the government may change from one segment of a proceeding to the next. Final Decision Memo at 13 (citing Wooden Bedroom Furniture from the [PRC]: Issues and Decision Memorandum for the Final Results of Review, A-570-890, 7 (Jun. 5, 2013), available at http://ia.ita.doc.gov/frn/summary/prc/ 2013-13987-l.pdf (last visited Jun. 23, 2017)). Here, Commerce justified the continued viability of the NME presumption on the grounds that it reasonably determines that governments may “use a variety of legal and administrative levers to exert influence and control (both direct and indirect) over the assembly of economic actors across the economy” even where the economy is not comprised entirely of the government. See Final Decision Memo at 100. Goal Zero points to no reason why it is unreasonable to conclude that an entity could be controlled by the government where the government does not comprise the entire economy. Commerce reasonably explains why its presumption that an exporter in a NME country is benefiting from government control is compatible with its finding that the Chinese economy has undergone reforms to eliminate the government as a single central authority comprising the entire economy. Goal Zero argues that Commerce’s NME presumption of state control is unreasonable in light of Commerce’s findings that changes in the Chinese economy have left Chinese respondents with “ ‘the discretion to change [their] export and/or production decisions.’ ” See Goal Zero Br. 11-14 (quoting Issues and Decision Memorandum for the Antidumping Investigation of Certain New Pneumatic Off-the-Road Tires from the [PRC], A-570-912, 10 (July 7, 2008), available at http://ia.ita.doc.gov/ frn/summary/prc/E8-16156-l.pdf (last visited Jun. 23, 2017)). Further, Goal Zero cites Commerce’s recognition that more companies’ export activities in China are independent from the government in comparison to the early to mid-1990s, Id. at 11-12. However, Commerce explains that the NME presumption does not stem from a determination that individual companies operating within a NME country are government entities," but rather that the NME-government can exert influence over economic actors in such an economy. See Final Decision Memo at 100. Commerce’s practice allows a respondent to prove its independence from ; the government by means of a separate rate application, and it is reasonable to place the burden on a respondent to prove its independence from government control even where economic and legal reforms have progressed to a point where some firms operate freely of government control. B. Assignment of China-Wide Rate to ERA Solar . Goal Zero claims that Commerce’s assignment of the China-wide rate to ERA Solar on the basis that it is government-controlled is unsupported by substantial evidence. Goal Zero Br. 16-19. Defendant counters that Commerce’s determination to include ERA Solar ’within the China-wide entity is supported by substantial evidence because ERA Solar failed to put forth any evidence supporting its eligibility for a separate rate. Def.’s Resp. Br. 17-22. The court agrees with Defendant. Com-merce’s assignment of the China-wide rate to ERA Solar is supported by substantial evidence. As already discussed, Commerce’s regulations permit it to calculate a single dumping margin applicable to all exporters and producers that are not individually examined in cases involving a NME country. See 19 C.F.R. § 351.107(d). Here, Goal Zero does not dispute Commerce’s finding that ERA Solar did not meet its burden to demonstrate independence from state control because it did not submit a separate rate application in this administrative review. See Final Decision Memo at 11; Goal Zero Br. 18. Because ERA Solar failed to submit any separate rate documentation, Commerce determined that ERA Solar did not qualify for a separate rate. Final Decision Memo at 11. Commerce would not excuse ERA Solar’s failure to submit separate rate information because Commerce reasonably determined that ERA Solar had notice that it was subject to the review. H. at 12. Goal Zero argues that it is unreasonable to presume that ERA Solar is government-controlled for this review where Commerce preliminarily recognized ERA Solar’s entitlement to a separate rate in the second administrative review of this same order and in the companion CVD administrative review. See Goal Zero Br. 18-19. Commerce explained that the agency cannot rely on a party’s separate rate eligibility in one segment of a proceeding as evidence of eligibility in a prior or subsequent segment. Final Decision Memo at 13 (“[A] party’s separate rate status must be established in each segment of the proceeding in which the party is involved because a company’s corporate structure, ownership, or relationship with the government can change from one segment of a proceeding to the next.”). Although Goal Zero claims that no such changes to ERA Solar’s corporate structure actually occurred from the investigation to the second period of review, see Goal Zero Br. 19, it is reasonably discernible that Commerce relies upon the record of each proceeding because it is not burdensome for a company, which is in the best position to produce such information, to provide such information in each segment. See Final Decision Memo at 13. Based upon the record here, Commerce’s determination to deny separate rate status based upon ERA Solar’s failure to submit a separate rate certification is supported by substantial evidence. C. Application of AFA to the China-Wide Entity Goal Zero argues that, in applying an adverse inference, Commerce unreasonably imputed non-cooperation to all members of the NME entity when most of the respondents fully cooperated. Goal Zero Br. 20-25. Defendant responds that Commerce reasonably applied AFA to determine the China-wide rate because Commerce determined that the China-wide entity failed to cooperate by acting to the best of its ability, because PRC exporters or producers that are part of the China-wide entity failed to respond to quantity and volume (“Q & V”) questionnaires. Def.’s Resp. Br. 24-26. Commerce’s application of an AFA rate to the China-wide entity is in accordance with law and supported by substantial evidence. Commerce shall use “facts otherwise available in reaching the applicable determination” when a respondent: (1) withholds information that has been requested by Commerce; (2) fails to provide such information by Commerce’s deadlines for submission of the information or in the form and manner requested; (3) significantly impedes an antidumping proceeding; or (4) provides information that cannot be verified. 19 U.S.C. § 1677e(a)(2). Commerce may apply an adverse inference in choosing among facts available if the respondent’s failure to act to the best of its ability to comply with Commerce’s request for information caused the deficiency. 19 U.S.C. § 1677e(b); see also Nippon Steel Corp. v. United States, 337 F.3d 1373, 1381 (Fed. Gir. 2003). Here, Commerce justified its determination to apply AFA to the China-wide entity because Chinese exporters of subject merchandise during the POR that are part of the China-wide entity did not respond to Commerce’s Q & V questionnaires. See Final Decision Memo at 15; Prelim. Decision Memo at 17. Goal Zero offers no record evidence to detract from Commerce’s finding that these companies withheld requested information and, therefore, failed to act to the best of their ability. Therefore, Commerce’s determination that the China-wide entity failed to cooperate by acting to the best of its ability is in accordance with law and supported by substantial evidence. Goal Zero further argues that Commerce lacked substantial evidence to determine that the China-wide entity failed to cooperate to the best of its ability because only two companies for which Commerce completed the review did not respond to its Q & V questionnaires. Goal Zero Br. 22-23. However, the statute only requires that Commerce assess the extent of the China-wide entity’s abilities, efforts, and cooperation. See Nippon Steel, 337 F.3d at 1382. Although Commerce could conceivably have taken the extent to which individual members of the China-wide entity cooperated with its requests into account, nothing in the statute requires it to make such an individualized inquiry of re1 spohdents that are part of the China-wide entity. Commerce found some respondents that Were sent Q & V questionnaires failed to provide information that they reasonably could, have been expected to maintain. See Final Decision • Memo at 15. Therefore, Commerce’s determination that the China-wide entity as a whole failed to cooperate to the best of its ability is supported by substantial evidence. D. Selection and Corroboration of the China-Wide Rate Goal Zero argues that Commerce’s selection of the 249.96 percent rate as the AFA rate is unsupported by substantial evidence because-it is not sufficiently corroborated, as required by the statute. Goal Zero Br. 26-38. Defendant responds that Commerce selected the highest rate in the history of the proceeding and corroborated its rate by comparing the petition margins to the margins calculated for the individually' examined respondents to determine their probative value.- Def.’s Resp. Br. 29. Commerce’s selection of an AFA rate is in accordance with its practice and it is adequately corroborated. Therefore, Commerce’s selection of an AFA rate for the China-wide entity is supported by substantial evidence and in accordance with law. Commerce may rely on information from several sources in selecting an AFA rate, including: (1) the petition; (2) the final determination in the investigation; (3) any previous administrative review; or (4) any other information placed on the record. 19 U.S.C. § 1677e(b); see 19 C.F.R. § 351.308(c) (mirroring the statute’s directives regarding the sources Commerce may rely upon when selecting an AFA rate). When Commerce uses secondary information, it must, to the extent practicable, corroborate the rate with independent sources that are reasonably at its disposal. 19 U.S.C. § 1677e(c). Commerce’s regulation refers to “secondary information” as information derived from: (1) the petition; (2) a final determination in a CVD investigation or an ADD investigation; (3) any previous administrative review, new shipper review, expedited antidumping review, section 753 review, or section 762 review, 19 C.F.R. § 351.308(c)(l)(i)~(iii), and defines the corroboration requirement as providing that Commerce “examine whether secondary information to be used has probative value.” 19 C.F.R. § 351.308(d). However, Commerce’s regulation explicitly states that the fact that information may not be practicably corroborated “will not prevent [Commerce] from applying an adverse inference as appropriate and using the secondary information in question.” Id. Here, Commerce referenced its finding in the investigation that the price and normal value used to derive the highest margin contained in the petition were within the range of the U.S. prices and normal values for the respondents in the investigation. Final Decision Memo at 16 (citing Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, From the [PRC], 77 Fed. Reg. 31,309, 31,318 (Dep’t Commerce May 25, 2012) (preliminary determination of sales at less than fair value, postponement of final determination and affirmative preliminary determination of critical circumstances); Crystalline Silicon Photovoltaic Cells, Whether or Not Assembled Into Modules, From the [PRC], 77 Fed. Reg. 63,791, 63,795 (Dep’t Commerce Oct. 17, 2012) (final determination of sales at less than fair value, and affirmative final determination of critical circumstances, in part)). Commerce also compared the 249.96 percent rate applied to the China-wide entity here to the margins calculated for individually examined respondents, and determined that the margins fall within the range of Suntech’s calculated weighted average dumping margins from this review. Final Decision Memo at 17. Commerce noted that several margins calculated for Sun-tech’s sales were “significantly above” the AFA rate applied here and that many more were at a similar level. Id. Goal Zero points to no data on the record that undermines Commerce’s conclusions that the margins calculated for some of Suntech’s sales are at a similar level or above the selected AFA rate. Therefore, Commerce corroborated the China-wide rate to the extent practicable by finding that the China-wide rate was within the range of the margins applicable to respondents from both the investigation and this review, which were calculated from sources allowed by.regulation to corroborate secondary information. See 19 C.F.R. § 351.308(c)(l)(i)-(iii). Goal Zero argues that Commerce lacked substantial evidence to conclude that the 249.96 percent rate in the investigation is corroborated because the AFA rate is disproportionately high relative to the calculated rates for any party whose rates were individually calculated. See Goal Zero Br. 31-32. However, Commerce enjoys broad discretion to develop a methodology for calculating an AFA rate so long as the rate is corroborated. See 19 U.S.C. §§ 1677e(b),(c). Goal Zero points to no information on the record calling into doubt Commerce’s corroboration of the rate as applied to the China-wide entity. Moreover, none of the authority relied upon by Goal Zero requires that the AFA rate selected for the China-wide entity reflect commercial reality of individually investigated respondents. Lastly, Goal Zero argues that the petition rate used by Commerce to corroborate the rate applied to the China-wide entity is unreliable because it is based on different methodologies than those used by Commerce in the investigation and administrative review, including different primary surrogate countries and values for certain key inputs. See Goal Zero Br. BB-SS. However, Goal Zero failed to exhaust its administrative remedies because it did not raise these arguments before Commerce. See Case Brief of Goal Zero, LLC, PD 568, bar code 3265921-01 (Mar. 23, 2015) (“Goal Zero Adm. Case Br.”). None of the arguments raised by Goal Zero before the agency implicate the methodological differences between the rates calculated in the petition and prior segments of the proceeding and the calculated rates in this administrative review. See id. at 5-10. The purpose of the exhaustion requirement would not be served by reviewing the reliability of the rates Commerce used to corroborate based upon methodological distinctions without Commerce having the opportunity to make a determination, finding, or conclusion on those challenges. II. Surrogate Value Data SolarWorld challenges Commerce’s selection of import data under Thai HTS heading 7604 to value Yingli’s aluminum frames inputs, see SolarWorld Br. 14-21, and Commerce’s use of world polysilicon prices to value semi-finished polysilicon ingots and blocks purchased by Yingli, see iff at 21-25, as unsupported by substantial evidence. SolarWorld also challenges Commerce’s determination to offset selling, general, and administrative (“SG & A”) expenses by income labeled in its surrogate profit ratio calculation. See iff at 42-44. The court first reviews Commerce’s selection of SV data for aluminum frames and next reviews Commerce’s selection to value semi-finished polysilicon ingots and blocks. Thereafter, the court turns to Commerce’s adjustments to its surrogate profit ratio calculation. A. Aluminum Frames SolarWorld argues that Commerce unreasonably selected import data for Thai HTS subheading 7604.29, which includes “Aluminum bars, rods and profiles: Other” because the import data under this category is not specific to aluminum solar frames. SolarWorld Br. 14-21. Specifically, SolarWorld argues that the aluminum solar frames purchased by Ying-li “have been further manufactured and processed into a good” that can no longer be considered a profile and has assumed the character of a different heading. Iff at 19. Defendant responds that Commerce reasonably concluded that import data under HTS heading is the best available information to value the frames purchased because it is most specific to the aluminum solar frames purchased by Yingli, which Yingli describes as “non-hollow, aluminum profiles.” Def.’s Resp. Br. 36-40. For the reasons that follow, Commerce’s selection of import data under HTS subheading 7604.29 is supported by substantial evidence. In cases involving imports from NME countries, Commerce obtains a normal value by adding the value of the FOPs used to produce the subject merchandise and “an amount for general expenses and profit plus the cost of containers, coverings, and other expenses.” 19 U.S.C. § 1677b(c)(l). Commerce values the FOPs “based on the best available information regarding the values of such factors in a market economy country or countries.” Id. Commerce’s methodology for selecting the best available information evaluates data sources based upon their: (1) specificity to the input; (2) tax and import duty exclusivity; (3) contemporaneity with the period of review; (4) representativeness of a broad market average; and (5) public availability. See Import Admin., U.S. Dep’t Commerce, Non-Market Economy Surrogate Country Selection Process, Policy Bulletin 04.1 (2004), available at http:// enforcement.trade.gov/policy/bull04-l.html (last visited Jun. 23, 2017) (“Policy Bulletin 04.1”); Prelim. Decision Memo at 28. Here, Commerce concluded that import data under HTS subheading 7604.29, covering “aluminum alloy bars, rods and profiles, other, other than hollow profiles, other” is the best available information to value aluminum frames purchased by Yingli because the data is more specific to Yingli’s solar frames than import data under HTS subheading 7616.99, covering “articles of aluminum nesoi.” Final Decision Memo at 81-82. Commerce concluded that HTS heading 7604 is more specific than HTS subheading 7616.99 because HTS 7616.99 is “an ‘other’ category that includes products dissimilar to aluminum frames,” id. at 81, including “ ‘nails, tacks, staples, screws, bolts, nuts, screw hooks, rivets, cotters, cotter pins, washers, knitting needles, bodkins, crochet hooks, embroidery stilettos, safety pins, other pins and chains, and cloth, grill and netting of aluminum wire.’” Id. at 84. Therefore, Commerce reasonably determined that import data under HTS heading 7604 is more specific because SolarWorld points to no evidence indicating that Yingli’s aluminum frames are more similar to the products in HTS heading 7616 than those in HTS heading 7604. SolarWorld’s claim that HTS heading 7604 only includes items that are unfinished and Yingli’s frames have become finished products is unavailing. First, the notes to Chapter 76 state that profiles include products that “ ‘have been subsequently worked after production ... provided that they have not thereby assumed the character of articles or products of the other headings.’ ” See Final Decision Memo at 82-83 (emphasis in original, quoting Jiangsu Jiasheng Photovoltaic Technology Co. v. United States, 38 CIT -, -, 28 F.Supp.3d 1317, 1337 (2014)). Further, it is reasonably discernible from Commerce’s reference to the International Trade Commission’s definition of aluminum profiles as “cast sintered, and worked after production,” Final Decision Memo at 82, that the work performed on the profiles highlighted by SolarWorld is not sufficient to make Yingli’s profiles more similar to the finished products enumerated in a different subheading. SolarWorld points to no evidence detracting from Commerce’s determination. SolarWorld also argues that Commerce unreasonably concluded that Yingli’s aluminum frames were “profiles” because profiles must have a uniform cross section. SolarWorld Br. 19-20 (citing the Notes to Chapter 7 of the HTS). SolarWorld suggests that Yingli’s profiles are not uniform along their entire length. W. at 20 (citing Yingli’s Response to the Department’s Fourth Section D Supplemental Questionnaire at Ex. D-30, CD 428-430, bar codes 3232605-01-03 (Oct. 1, 2014) (‘Yingli 4th Suppl. Sec. D. Resp.”)). However, Commerce’s task is not to classify the solar frame inputs used by Yingli for customs purposes, but to- select the best available data to value the FOPs in question. Solar-World offers no evidence that calls into question Commerce’s conclusion that the processing performed by Yingli leaves its solar frames more similar to the unfinished items included in HTS heading 7604 than the dissimilar finished items included in HTS subheading 7616.99. Therefore, Commerce’s determination is supported by substantial evidence. B. Semi-Finished Ingots and Blocks SolarWorld challenges Commerce’s' determination to value Yingli’s semi-finished.ingots and blocks input with the world market price for : polysilicon, claiming that price does not ^reflect, the substantial additional processing that po-lysilicon, must undergo tó become an ingot or block. See SolarWorld Br. 23-25. Defendant responds that Commerce’s selection is reasonable because there was no surrogate value data for ingots and blocks, and because additional processing was captured elsewhere in Commerce’s- calculations. See Defi’s Resp. Br. 40-41. .For the reasons that follow, Commerce’s determination is supported by substantial evidence. ”, ■; As already discussed, Commerce selects the best available information to value FOPs by evaluating data' Sources based upon their: (1) 'specificity-to the input; - (2) tax and import duty exclusivity; (3) contemporaneity with the period of review; (4) representativeness of a broad market average; and (5) public availability. See Policy Bulletin 04.1; Prelim. Decision Memo at 28. Here, Commerce determined that, because no party submitted a SV for ingots and blocks, the world market price was the best information available to value ingots and blocks purchased by Yingli. Final Decision Memo at 76. Commerce noted that the inputs in question are primarily composed of polysilicon. Id Further, because Yingli self-produces most of its ingots and blocks, Commerce could account elsewhere in its calculations for the additional processing costs to turn polysilicon feedstock into ingots and blocks. Id. Therefore, Commerce determined that no reliable record evidence suggests that the world market price for polysilicon would result in unrepresentative pricing for ingots and blocks. Id. SolarWorld contends that Commerce’s use of a price for polysilicon fails to capture all the costs associated with the ingots and blocks. No party submitted SV data for ingots and blocks and additional manufacturing is required to transform polysilicon into ingots and blocks. Id. Yet, Commerce concluded that additional manufacturing costs were already accounted for because “Yingli self-produces most of its ingots and blocks.” Id. Commerce found further adjustments unnecessary. Id. However, SolarWorld claims that, because evidence on the record indicated that Yingli purchased a significant portion of the blocks consumed in the production of subject merchandise, Commerce erroneously concluded that the additional processing costs required to manufacture ingots and blocks were accounted for elsewhere in the calculation. SolarWorld Br. 24 (citing Yingli’s Response to the Department’s Section D Questionnaire, Including Related Appendices at Exh. D-5, CD 173-176, bar codes 3201697-01-04 (May 13, 2014)). However, the data relied upon by Solar-World does not undermine Commerce’s conclusion that the agency accounted for processing costs required to manufacture the ingots and blocks for most merchandise. Yingli highlights record evidence demonstrating that the total purchases of ingots and blocks relative to the volume of ingots and blocks consumed during the period of review (“POR”) was not significant. See Oral Arg. 01:21:22-01:24:36, May 5, 2017, ECF No. 129 (comparing Yingli 4th Suppl. Sec. D. Resp. at Revised Exh. D-7 (outlining consolidated purchases by Yingli entities of monocrystalline and polysilicon blocks and ingots during the POR) with Yingli 4th Suppl. Sec. D. Resp. at Exh. D-20 (outlining total consolidated quantities of ingots and blocks consumed by Yingli entities during the POR)). It is reasonably discernible that Commerce considered the imperfect nature of the surrogate value, noting that no party had submitted SV for ingots and blocks that were purchased, and reasoned that it had chosen “the best SV information on the record to value Yingli’s ingots and blocks” because it had accounted for most of the additional processing costs. See Final Decision Memo at 76. Although SolarWorld claims that the record contained information to allow Commerce to add processing costs onto the value of polysilicon to build up a price for these inputs, see SolarWorld Br. 23-24, the availability of another methodology does not make Commerce’s determination unreasonable given the record evidence. Lastly, SolarWorld argues that Yingli’s market economy purchases demonstrate that the SV chosen by Commerce is aberrational. See SolarWorld Br. 25. Commerce cited its practice, which is not to use a respondent’s market economy purchase prices as benchmarks because those prices are specific to one respondent and not necessarily representative of industry-wide prices available to other producers. Final Decision Memo at 76. It is reasonably discernible from Commerce’s statement that most of the ingots and blocks used by Yingli in production are self-produced that Commerce considered the small scale of Yingli’s market economy purchases to be a weak indicator of industry-wide pricing. See id. SolarWorld points to no evidence that Yingli’s market economy purchases were more representative in this case to justify a deviation from Commerce’s cited practice, which is not to use a respondent’s market economy purchase prices as a benchmark to determine if a surrogate value is aberrational. See id. (citing Issues and Decision Memorandum for the 2003-2004 [ADD] Administrative Review: Certain Cased Pencils from the [PRC], A-570-827, 8 (July 6, 2006), available at http://ia. ita.doc.gov/frn/summary/prc/E6-10568-l. pdf (last visited Jun. 23, 2017) (stating that pricing data can be selectively obtained and does not serve as a basis for evaluating whether other pricing from other data sources is unreliable)). C. Calculation of Surrogate Financial Profit Ratio to Include “Other Income” SolarWorld challenges as unsupported by substantial evidence Commerce’s deduction of a line item labeled “Other Income” in the financial statements of PT Len Industri (Persero) (“PT Len”), an Indonesian producer of electrical equipment including solar modules, whose financial statements Commerce selected to calculate respondents’ surrogate financial profit ratio. See SolarWorld Br. 43-44; see also Prelim. Decision Memo at 33. As already discussed, Commerce obtains a normal value in NME cases by adding the value of the FOPs used to produce the subject merchandise and “an amount for general expenses and profit plus the cost of containers, coverings, and other expenses.” 19 U.S.C. § 1677b(c)(l). Neither the statute nor Commerce’s regulations further define how Commerce is to calculate its surrogate financial profit ratio. Nonetheless, as a matter of practice, Commerce typically offsets a company’s SG & A expenses by any income related to the general operations of the company for the current period, when calculating the surrogate profit ratio. See Final Decision Memo at 60; 1,1,1,2-Tetrafluoroethane from the [PRC]: Issues and Decision Memorandum for the Final Determination of Sales at Less Than Fair Value [ADD] Investigation, A-570-998, 46-47 (Oct. 14, 2014), available at http://ia.ita.doc.gov/frn/ summary/prc/2014-24903-l,pdf (last visited Jun. 23, 2017) (“1,1,1,2 Tetra fi*om PRC I & D”). Here, Commerce offset the SG & A expenses in the surrogate profit ratio calculation with the “other income” line item on PT Len’s financial statements. See Final Decision Memo at 60. Commerce explained that the “other income” line item consists of “foreign exchange,” “interest from bank,” and “interest income,” which are' among the types of expenses that it hormálly allows to offset SG & A expenses. Id. (citing SolarWorld Submission of Factual information at Ex. 24 at 624, CD 491-612, bar code 3241320-01-22 (Nov. 10, 2014) (“SolarWorld Factual Info.”)). Commerce further explained its determination by referencing its practice in'NME cases of assuming that unassigned income relates to a company’s general operations, absent evidence indicating that such income relates to a specific expense. Id (citing 1,1,1,2 Tetra from PRC I & D at 46-47). SolarWorld claims that no record evidence indicates that the sums deducted were overhead expenses normally deducted by Commerce. See SolarWorld Br. 43. Yet, SolarWorld does not challenge the reasonableness of Commerce’s practice, nor does it point to record evidence indicating that the “other income” line item relates to a specific expense unrelated to PT Len’s general operations. Here, not only does Commerce point to record information indicating that the “other” income line item was grouped with line items that are attributable to general overhead, but Commerce also points to record evidence indicating that PT Len actually treated the “other income” line item as income attributable to the general operations of the company. Final Decision Memo at 60. Therefore, Commerce’s determination is supported by substantial evidence. III. Suntech’s CEP Sales The court first addresses SolarWorld’s challenge to Commerce’s inclusion of Sun-tech’s U.S. sales to an affiliated party as CEP sales in its calculation of Suntech’s dumping margins. See SolarWorld Br. 25-33. The court next addresses SolarWorld’s alternative challenge, which argues that Commerce inappropriately determined the date of sale for Suntech’s reported CEP sales. See id. at 29-32. A. Treatment of Suntech’s Sales to U.S. Affiliate as CEP Sales SolarWorld argues that Commerce lacked substantial evidence to include sales to an affiliated party in Sun-tech’s margin calculation. SolarWorld Br. 25-33. Defendant responds that Commerce properly calculated CEP based on the price at which subject merchandise was first sold in the United States by Suntech’s affiliated seller to an unaffiliated customer. Def.’s Resp. Br. 42-44. For the reasons that follow, Commerce’s inclusion of Sun-tech’s sales as CEP sales is in accordance with law and supported by substantial evidence. Commerce determines the dumping margin by calculating the amount by which the normal value exceeds the export price or CEP of the subject merchandise. 19 U.S.C. § 1677(35)(A). CEP is the price, subject to certain adjustments, “at which the subject merchandise is first sold (or agreed to be sold) in the United States before or after the date of importation by or for the account of the producer or exporter of such merchandise or by a seller affiliated with the producer or exporter” to an unaffiliated purchaser. 19 U.S.C. § 1677a(b). Here, Commerce classified sales made by Suntech’s affiliated seller after the date of importation to the first unaffiliated customer as CEP sales. Final Decision Memo at 92-93. Suntech reported the price and quantity of modules sold to an unaffiliated purchaser and Commerce could verify those terms against documentation provided by Suntech. Id. Commerce also determined that it could make the required adjustments under the statute to CEP. Id. SolarWorld claims that the unaffiliated party paid the affiliated party for the installation of a completed solar farm (a “turnkey solar project”). SolarWorld Br. 25-30. SolarWorld points to evidence indicating that the terms for this turnkey solar project were fixed terms. Id. at 31. There was no indication that the cost of solar modules was negotiated or priced separately. See id. Therefore, SolarWorld argues that the record evidence relied upon by Commerce cannot support its determination that the sales prices for modules to Suntech’s affiliate were identical to the sales price for modules to the unaffiliated customer. See id. However, Commerce found that the gross unit price for the modules was verified in parts of the contract between Suntech and the unafSliated purchaser. See Comments in the Issues and Decision Memorandum Containing Business Proprietary Information at 12, CD 583, bar code 3289927-01 (July 7, 2015). SolarWorld points to no record evidence detracting from Commerce’s determination that the price and quantity reflected in the invoice to the unafSliated purchaser are identical to the invoice issued by Suntech’s affiliate to the purchaser. Therefore, Commerce’s inclusion of module sales by Suntech’s affiliate to its unaffiliated customer as CEP sales is supported by substantial evidence and in accordance with law. B. Determination of Date of Sale for Suntech’s Reported Sales SolarWorld also challenges as unsupported by substantial evidence Commerce’s use of the date of shipment rather than the date of contract for Suntech’s reported CEP sales of solar modules to its unaffiliated purchaser. See SolarWorld Br. 33-39. Defendant responds that Commerce supported its determination with record evidence showing that the material terms of the contracts were subject to change up to the date of shipment, which was also consistent with usual solar industry contract pricing practices, including those of Suntech. Def.’s Resp. Br. 45-47. The court agrees with SolarWorld that Commerce’s determination is not supported by substantial evidence because the record evidence relied upon by Commerce reflects a transaction for the supply of solar modules, not a contract for the provision of a turnkey solar project. On remand, Commerce must explain what record evidence supports a determination that the key terms of the sales between Suntech’s affiliated seller and the purchaser were subject to change after the contract date or must reconsider its determination. As already discussed, the statute provides that CEP is the price at which merchandise is “first sold (or agreed to be sold) before or after the date of importation by or for the account of the producer or exporter of such merchandise or by a seller affiliated with the producer or exporter, to a purchaser not affiliated with the producer or exporter.” 19 U.S.C. § 1677a(b). The statute does not define when a product is deemed sold, nor does it define how Commerce determines date of sale for purposes of determining what sales should be included in calculating CEP for the POR. See id. Commerce’s regulations provide that Commerce “normally will use the date of invoice, as recorded in the exporter or producer’s records kept in the ordinary course of business” as the date of sale. 19 C.F.R. § 351.401(i). However, Commerce has discretion to use an alternative date if it determines that “a different date better reflects the date on which the exporter or producer establishes the material terms of sale.” Id. Here, Commerce determined that a date other than the date of invoice reflects the date that the material terms of sale were established. Final Decision Memo at 93. Commerce justified this determination by referencing contracts on the record for the sale of solar modules that are subject to change, id. (citing Section A Questionnaire Response—Wuxi Suntech Power Co., Ltd. at Ex. 16, CD 120-122, bar code 3196404r-01-03 (Apr. 18, 2014)), and referencing its findings in past proceedings regarding contract pricing practices within the solar industry that are subject to change. Id. (explaining that “the Department’s knowledge of solar industry contract pricing practices, including those of Wuxi Suntech, indicates that the material terms of its contracts are subject to change.”). Commerce relies upon language in specific contracts of sale for solar modules in separate transactions, unrelated to the sales it included as CEP sales, to support its finding that Suntech America, Inc.’s contracts with purchasers are generally subject to change after they are executed. However, SolarWorld contends that the specific contracts between Suntech and its unaffiliated purchaser that were included as CEP sales include payment for services other than the sale of solar modules. See SolarWorld Br. 36 (citing Supplemental Section C Questionnaire Response—Wuxi Suntech Power Co., Ltd. at Ex. 3, CD 214-218, bar codes 3209510-01-02, 3209387-01-03 (Jun. 16, 2014) (“Suntech Suppl. Sec. C Questionnaire Resp.”)). Commerce does not explain why it is reasonable to conclude that specific language in a contract that reflects only the sale of solar modules lends support for an interpretation of a contract for services other than the sale of solar modules. Further, SolarWorld points to specific language in the contract for the sales included as CEP sales to support its claim that the terms of the contract were fixed as of the date of contract. Id. at 37. Commerce failed to explain why its determination that these contracts are subject to change is reasonable in light of the affirmative language in the specific contract underlying the transaction included as CEP sales. On remand, Commerce must do so or reconsider its determination. IY. Use of FOP Data of Certain Sun-tech Tollers SolarWorld contends that Commerce’s acceptance of FOP usage data from Suntech and certain tollers in place of missing data from relevant toll processors was not supported by substantial evidence. SolarWorld Br. 39-42. Defendant responds that Commerce reasonably determined that the FOP usage data reported by Suntech and certain of its tollers could serve as a substitute for the missing FOP data and that it was appropriate to use facts otherwise available without an adverse inference because record evidence indicated that Suntech cooperated to the best of its ability. Def.’s Resp. Br. 49-50. For the reasons that follow, Commerce’s determination is supported by substantial evidence. Commerce shall determine the normal value of subject merchandise exported from a NME country on the basis of the value of FOPs, to which Commerce shall add an amount for general expenses and profit plus the cost of containers, coverings, and other expenses. 19 U.S.C. § 1677b(c)(l). The valuation of FOPs “shall be based on the best available information regarding the values of such factors in a market economy country or countries considered to be appropriate by [Commerce].” Id. Therefore, the statute gives Commerce discretion in assessing what the best available information for valuing FOPs and in calculating a normal value for subject merchandise in NME cases. See id. To make the applicable determination, Commerce shall generally apply facts otherwise available on the record in reaching an applicable determination if: (1) information necessary to Commerce’s administrative determination is not available on the record; (2) an interested party withholds information requested or fails to provide the information in a timely fashion or in the form and manner requested; (3) significantly impedes a proceeding; or (4) provides information but the information cannot be verified. See 19 U.S.C. § 1677e(a). The statute permits Commerce to use an inference that is adverse to the interests of a party in selecting from among the facts otherwise available if Commerce finds that an interested party has failed to cooperate by not acting to the best of its ability to comply with a request for information. 19 U.S.C. § 1677e(b). Here, Commerce acknowledged that Suntech was unable to provide the requested FOP consumption data for all of its tollers. Final Decision Memo at 40. Commerce accordingly determined that it was appropriate to apply facts available to calculate Suntech’s FOP usage. See id. at 40-41. Commerce justified its exercise of discretion to decline to apply an adverse inference to Suntech’s FOP consumption by recounting Suntech’s efforts to cooperate and act to the best of its ability to comply with Commerce’s request for such data. See id. at 41. Moreover, Commerce noted that its practice is to use FOP information from tollers as a substitute for missing FOP data where: (1) a respondent has a number of tollers; (2) a respondent identifies tollers in a timely manner; (3) a respondent documents its unsuccessful efforts to obtain FOPs from its tollers; and (4) non-reporting tollers account for only a small portion of FOPs and there is usable FOP information from other suppliers that could serve as a substitute for the missing FOPs. See Final Decision Memo at 40-41. Commerce found that Suntech had a large number of tollers and that Suntech identified those tollers in a timely manner. Id. at 40. Commerce also found that Suntech documented its unsuccessful efforts to obtain the requested data from its tollers. Id. at 41 (citing Memorandum re: Unreported Factors of Production, CD 585, bar code 329025-01 (July 7, 2015) (“Unreported FOP Memo”)). Commerce also justified its reliance on the tollers’ FOP data together with data provided by Suntech because it found that the non-reporting tollers accounted for a small portion of FOPs. Final Decision Memo at 41. Therefore, Commerce’s determination to follow its practice of using FOP information from Suntech, as well as cooperating tollers, as a substitute for missing FOP data from tollers used by Suntech is supported by substantial evidence. See id. at 40. Commerce also explained that it determined not to apply an adverse inference to Sun-tech’s FOP usage for non-reporting tollers because Commerce concluded that Suntech had acted to the best of its ability in complying with Commerce’s request for the information. See id. SolarWorld points to no evidence undermining Commerce’s conclusion. SolarWorld argues that Commerce’s use of Sunteeh’s own FOP data, for modules/laminates, as a substitute for missing unaffiliated tollers’ data is unreasonable because the unaffiliated tollers would likely have higher costs. SolarWorld Br. 40. So-larWorld argues that no record .eyidence indicates that a toll processor would, have similar per-unit costs to that of a large integrated producer like Suntech, Seq id. However, SolarWorld points to no evidence on the record that supports its suggestion that a relatively small producer of modules/laminates would have.higher per-unit costs than a large integrated producer like Suntech. Nor does SolarWorld support its assertion that Commerce’s acceptance of data for modules laminates would, have meaningfully affected Commerce’s margin calculation. See id. Without such affirmative evidence of a disparity between the missing costs, the court cannot say that Commerce’s determination is unsupported by substantial evidence. Finally, SolárWorld argues that Commerce’s decision to accept Suntech’s own FOP data for modules/laminates in place of its toll processors presents a significant risk for manipulation. SolárWorld Br. 41. However, Commerce notes that Suntech not only reported its own FOP data, but also reported data from cooperating unaffiliated tollers of modules. Unreported FOP Memo at 3. It is reasonably discernible that Commerce concluded, based on the absence of discrepancies in tolling data between the reporting tollers and the small proportion of module production for which it lacked data, that the data supplied could serve as a substitute for the missing data. Where SolárWorld cannot point to a discrepancy in the data actually produced which would render it unreasonable for Commerce to accept this data, the court cannot say that Commerce’s determination is unreasonable. CONCLUSION The court remands Commerce determination to use the date of shipment rather than the date of contract for Suntech’s reported CEP sales of solar modules to its unaffiliated purchaser for further consideration and explanation. The court sustains the Final Results in all other respects. In accordance with the foregoing, it