OPINION AND ORDER

STANCEU, Judge.

Before the court is the redetermination made by the International Trade Administration, United States Department of Commerce (“Commerce” or the “Department”) in response to the court’s remand order in Jinan Yipin Corp. v. United States, 31 CIT -, 526 F.Supp.2d 1347 (2007) (“Jinan Yipin I ”). Final Results of Redetermination Pursuant to Ct. Remand (“Remand Redetermination”). Because the Remand Redetermination does not comply in all respects with the court’s order in Jinan Yipin I, the court issues a second remand order with specific instructions.

I. Background

The background of this litigation is discussed in the court’s opinion in Jinan Yipin I. See 31 CIT at-, 526 F.Supp.2d. at 1349-51. Additional background is presented below to address events that have occurred since Jinan Yipin I was decided.

Jinan Yipin I ruled on plaintiffs’ several challenges to the final results that Commerce issued in the eighth administrative review (“Final Results”) of the antidumping duty order on fresh garlic (“subject merchandise”) imported from the People’s Republic of China (“China” or the “PRC”). Jinan Yipin I, 31 CIT-, 526 F.Supp.2d 1347; see Fresh Garlic From the People’s *1185 Republic of China: Final Results of Anti-dumping Duty Admin. Review and New Shipper Reviews, 69 Fed.Reg. 33,626 (Jun. 16, 2004) (“Final Results ”). In the Final Results, Commerce assigned plaintiff Jinan Yipin Corporation, Ltd. (“Jinan Yipin”), a Chinese producer and exporter of the subject merchandise, a weighted average antidumping duty margin of 115.81%. Final Results, 69 Fed.Reg. at 33,627, 33,629; see Summons 1. Commerce assigned plaintiff Shandong Heze International Trade and Developing Company (“Shandong”), a Chinese exporter of the subject merchandise, a margin of 43.30% in the Final Results. Final Results, 69 Fed.Reg. at 33,-627, 33,629.

Based on the conclusions the court reached on the various issues plaintiffs raised in contesting the Final Results, Jinan Yipin I ordered Commerce to redetermine the weighted average antidumping duty margins assigned to each of the plaintiffs. Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1382-83. The Remand Re-determination assigns a 9.70% margin to Jinan Yipin. Remand Redetermination 26. Because Commerce, in the Remand Redetermination, did not make any changes in response to the court’s conclusions on the issues on which Shandong contested the Final Results, the Remand Redetermination assigns to Shandong an unchanged margin of 43.30%. Id. at 26-27.

II. Discussion

The court will sustain the Department’s determination upon remand if it complies with the court’s remand order, is supported by substantial evidence on the record, and is otherwise in accordance with law. See 19 U.S.C. § 1516a(b)(l)(B)(i) (2000). Substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Consol. Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 83 L.Ed. 126 (1938). In considering whether substantial record evidence supports the Department’s findings, the court considers “the record as a whole, including evidence that supports as well as evidence that ‘fairly detracts from the substantiality of the evidence.’ ” Huaiyin Foreign Trade Corp. v. United States, 322 F.3d 1369, 1374 (Fed.Cir.2003) (quoting Atl. Sugar, Ltd. v. United States, 744 F.2d 1556, 1562 (Fed.Cir.1984)).

A, The Court Affirms the Department’s Resolution in the Remand Redetermination of the “Affiliation” and “Indirect Selling Expense” Issues Affecting Jinan Yipin

In Jinan Yipin I, the court concluded that Commerce acted unlawfully in assigning a 376.67% antidumping duty rate to the sales of subject merchandise that Jinan Yipin’s U.S. sales affiliate, American Yipin, made to “Houston Seafood,” one of its customers in the United States. Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1351. Commerce based this rate on the “facts otherwise available” and “adverse inferences” provisions of 19 U.S.C. § 1677e. See 19 U.S.C. § 1677e(a)-(b) (2000). Commerce found that Jinan Yipin failed to cooperate to the best of its ability in providing, in response to the Department’s questionnaires, information pertaining to Jinan Yipin’s affiliates during the period of review. See Jinan Yipin I, 31 CIT at -, 526 F.Supp.2d at 1353-54. The court concluded that the Final Results, which presented vague and inconsistent conclusions on the issue of whether an affiliation existed between Jinan Yipin, through American Yipin, and Houston Seafood, could not be sustained upon a finding that Jinan Yipin withheld information on the affiliation issue or substantially impeded the Department’s access to information needed to resolve that issue. Id. at-, -, 526 F.Supp.2d at 1354, 1358-61.

*1186 After reopening the record, Commerce found on remand that Jinan Yipin and Houston Seafood were not affiliated at any point from November 1, 2001 to October 31, 2002 (the “period of review”), which is the time period covered by the administrative review at issue in this proceeding. Remand Redetermination 2, 4. Based on this new finding, the Remand Redetermination treats all sales between American Yipin and Houston Seafood during the period of review as unaffiliated sales and does not apply to the sales to Houston Seafood a rate based on facts otherwise available or adverse inferences. Id. at 4-5. Jinan Yipin makes no objection to the Department’s resolution of this issue. See Jinan Yipin’s Comments Regarding the Dep’t’s Remand Redetermination (“Jinan Yipin’s Comments”). The court affirms the Department’s conclusion, in the Remand Redetermination, that American Yipin and Houston Seafood were not affiliated for the period of review and its decision not to apply facts otherwise available and adverse inferences to American Yipin’s sales of Jinan Yipin’s merchandise to Houston Seafood during the period of review. Remand Redetermination 4-5.

Jinan Yipin I concluded that Commerce also erred in calculating Jinan Yipin’s indirect selling expenses for purposes of determining constructed export price. Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1366. Commerce found at verification that American Yipin shared certain employees with Bayou Dock, a company located near the business location of American Yipin that did not sell the subject merchandise. See id. at -, 526 F.Supp.2d at 1363. Commerce concluded that American Yipin did not properly record salary expenses for these shared employees for its first three months of operation in its Westwego, Louisiana location. See id. In the Final Results, Commerce, invoking its authority to use facts otherwise available and adverse inferences under 19 U.S.C. § 1677e, added all twelve months of Bayou Dock’s salary and benefit expenses to Jinan Yipin’s reported indirect selling expense. Id. at-, 526 F.Supp.2d at 1364. The court concluded in Jinan Yipin I that substantial record evidence supported a finding that some of American Yipin’s indirect selling expenses were under-reported or irregularly reported but also concluded that the Final Results did not contain a satisfactory explanation of why adding Bayou Dock’s salary and expenses for all twelve months of the period of review was a reasonable exercise of the authority provided by § 1677e. Id. at -, 526 F.Supp.2d at 1365-66.

On remand, Commerce applied to American Yipin’s indirect selling expenses the equivalent of three full months of salary and benefit expenses for the shared employees. Remand Redetermination 7. According to Commerce, this added amount “continues to represent adverse facts available because it includes the full salaries of the shared employees which were incurred while performing work for both American Yipin and Bayou Dock.” Id. In its comments on the Remand Redetermination, Jinan Yipin does not object to this method of recalculating its indirect selling expenses. See Jinan Yipin’s Comments. The court affirms the Remand Redetermination with respect to the recalculation of the indirect selling expenses.

B. The Department’s Surrogate Values for Garlic Seed, Water, and Cardboard Cartons Are Not in Accordance with Law

The statute, 19 U.S.C. § 1677b(c) (2000), establishes requirements that Commerce must satisfy in determining surrogate values but also “accords Commerce wide discretion” in selecting the “best available information” for determining those values. Nation Ford Chem. Co. v. United States, 166 F.3d 1373, 1377 (Fed. *1187 Cir.1999); see Shakeproof Assembly Components, Div. of Ill. Tool Works, Inc. v. United States, 268 F.3d 1376, 1381-83 (Fed.Cir.2001). Informing the Department’s exercise of discretion to select the “best available information” pursuant to 19 U.S.C. § 1677b(c)(l) is the broader purpose of the antidumping law “to calculate dumping margins as accurately as possible.” Parkdale Int’l v. United States, 475 F.3d 1375, 1380 (Fed.Cir.2007); see Lasko Metal Prods., Inc. v. United States, 43 F.3d 1442, 1443 (Fed.Cir.1994); Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed.Cir.1990).

In Jinan Yipin I, the court remanded to Commerce the determinations of surrogate values for garlic seed, water, and cardboard cartons. Jinan Yipin I, 31 CIT at -, 526 F.Supp.2d at 1383. In the Remand Redetermination, Commerce provided additional explanation but did not change its selection of any of these surrogate values. See Remand Redetermination 7-27. Because the Remand Redetermination, in setting forth the additional explanation, did not cure the deficiencies that affected the determinations of surrogate values for garlic seed, water, and cardboard cartons, the court directs that Commerce reconsider each of these surrogate values and redetermine them as necessary to address the shortcomings in the Remand Redetermination that are identified in this Opinion and Order.

1. The Court Rejects the Department’s Determination of a Surrogate Value for Garlic Seed

In the Final Results, Commerce valued respondents’ garlic seed at 50 Indian rupees ($1.03 USD) per kilogram, the price set forth in several “News Letters” of the National Horticultural Research and Development Foundation (“NHRDF”) for two Indian varieties of garlic seed, “Agrifound Parvati” and ‘Yamuna Safed-3,” that were developed and sold by NHRDF in India. Jinan Yipin I, 31 CIT at -, 526 F.Supp.2d at 1368 (citing Factors Valuations for the Prelim. Results of the Admin. Review and New Shipper Reviews 2-3. (Dec. 1, 2003) (Admin.R.Doc. No. 170)). Commerce considered these varietals of garlic seed to be comparable to plaintiffs’ garlic in bulb size and number of cloves per bulb. Id. Commerce chose the NHRDF value over prices for garlic obtained from Indian import statistics, which prices Commerce had used to value garlic seed in the previous (seventh) administrative review. Id. at-, 526 F.Supp.2d at 1369. Commerce stated, in support of its choice, its findings that the NHRDF prices were “ ‘the most product-specific information on the record’ ” and that the Indian import data are “ ‘considerably less product-specific because we cannot ascertain the quality or nature (ie., bulbs, loose cloves, etc.) of the garlic products entered under the applicable [Harmonized Tariff Schedule] category.’ ” Id. at -, 526 F.Supp.2d at 1370 (quoting Issues and Decision Mem. for the Admin. Review and New Shipper Reviews of the Antidumping Duty Order on Fresh Garlic from the People’s Republic of China 11 (June 7, 2004) (Admin.R.Doc. No. 254)).

In Jinan Yipin I, the court concluded that the findings of fact regarding the product specificity of the import data were not supported by substantial evidence on the record. Id. at -, 526 F.Supp.2d at 1370. The court observed that the record contained evidence that Chinese garlic imports constitute the overwhelming majority of all imported garlic in India, that Chinese garlic is imported in the form of whole bulbs, not loose cloves, and that these imports are comparable to the subject merchandise with respect to bulb diameter and number of cloves per bulb, both of which criteria Commerce considered important to comparability. Id. This evidence, which Com *1188 merce did not reference or discuss in the Final Results, was contained in a June 2003 report, Market Research on Fresh Whole Garlic in India, placed on the record by petitioners. See Letter from Collier Shannon Scott, PLLC to Sec’y of Commerce (June 30, 2003) (setting forth petitioners’ surrogate value submission), Ex. 7 at 27, 29 (“Market Research Report”) (Case No. 03-00636 Admin. R. Doc. No. 100). Commerce relied on the same report for record evidence in support of its finding that certain bulb varietals listed in the NHRDF price list, Agrifound Parvati and Yamuna Safed-3, are comparable to the subject merchandise with respect to bulb diameter and the number of cloves per bulb. Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1370. The court instructed Commerce to reconsider its selection of the NHRDF data over the import data based on the record evidence indicating that the import data consisted largely of garlic that was highly similar to the subject merchandise. Id. at -, 526 F.Supp.2d at 1370-72.

On remand, Commerce considered three sets of data on the record: the NHRDF data it had relied upon in the Final Results, the Indian import statistics data from Monthly Statistics of the Foreign Trade of India (“MSFTI import data”), and the import, export, and domestic price data in the Market Research Report. Remand Redetermination 8; Letter from Grunfeld, Desiderio, Lebowitz, Silverman & Klestadt LLP to Sec’y of Commerce (June 30, 2003) (Case No. 03-00636 Admin. R. Doc. No. 101) (setting forth Jinan Yipin’s surrogate value submission). Commerce again concluded that the NHRDF data, which set a 50-rupee surrogate value for garlic seed, were the best available information. Commerce reiterated its preference for “review period-wide price averages, prices specific to the input in question, prices that are net of taxes and import duties, prices that are contemporaneous with the period of investigation or review, and publicly available data.” Remand Redetermination 8-9 (internal quotation marks and citation omitted). Commerce then made separate findings upon which it found the other two data sources to be unusable. Id. at 8-13. Jinan Yipin contests the Department’s garlic seed re-determination, reiterating the court’s conclusions in Jinan Yipin I that the MSFTI import data for China are product-specific and that “a reasonable mind could not conclude that an Indian garlic producer would incur the significantly higher cost for NHRDF garlic as opposed to using the less costly garlic imports.” Jinan Yipin’s Comments 4 (citing Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d. at 1372).

Commerce concluded that the MSFTI import data, like the NHRDF data, are “period-wide price averages, prices that are net of taxes and import duties, contemporaneous, and publicly available” and, in a departure from the Final Results, found that the MSFTI data pertaining to imports from China were product-specific. Remand Redetermination 9. Commerce, however, rejected the MSFTI import data on grounds that it had not relied upon in the Final Results, specifically, that 94% of Indian garlic imports consist of Chinese imports, that China has a non-market economy (“NME”), and that prices in a non-market economy are unreliable and invalid. Id. at 9-10. Citing a “practice” of excluding imports from NME countries, Commerce stated that “[ijmports from NME countries are excluded for the same reason that we do not rely on an NME producer’s cost to calculate normal value in an anti-dumping duty calculation” and, in giving the reason, stated that “the presence of government control on various aspects of NMEs renders production costs, prices, and price comparisons unreliable and invalid.” Id. at 10 (citing Natural Bristle Paint Brushes and Brush Heads From the *1189 People’s Republic of China: Final Results of Antidumping Duty Admin. Review, 65 Fed.Reg. 45,758 (July 25, 2000)).

Commerce rejected the MSFTI import data for China for the sole reason that these data pertain to products exported from an NME country, concluding that the “Chinese garlic values within the Indian import data are necessarily eliminated from the value under consideration.” Remand Redetermination 10. However, a blanket policy of refusing to use import data pertaining to products exported from an NME country is inconsistent with the statutory obligation to value factors of production according to the best available information on the individual record in the specific investigation or review. See 19 U.S.C. § 1677b(c)(l) (requiring that “the valuation of the factors of production shall be based on the best available information regarding the values of such factors in a market economy country” (emphasis added)). Commerce reasonably may infer, based on its findings regarding the presence of government control on various aspects of NMEs, that import data on goods from an NME country are inferior to import data for goods from a market economy country. Nevertheless, a fair comparison of the various competing sets of data on the record in a hypothetical investigation or review in which all available information is affected by at least one flaw could lead to a finding that data on imports from an NME country are the best available information with which to value a factor of production. Moreover, the analogy Commerce draws between an NME producer’s cost as an element of normal value, and prices in a market economy country of products imported from an NME country, is not a meaningful one. The costs of production in an NME country will depend on costs and prices prevailing in that country. Value information for Chinese-origin imports in India’s MSFTI import data, by definition, are based on purchases made by importers in India. In this case, Commerce foreclosed the opportunity to make a qualitative comparison between the NHRDF data and the MSFTI import data for China, in addition to the other competing data on the record. Because the Remand Redetermination fails to make this comparison and state findings based on an actual comparison of the data sets, the court concludes that the Department’s decision to reject the MSFTI import data for China is unsupported by findings of fact grounded in record evidence and also is unsupported by adequate reasoning.

The MSFTI import data include data on imports from countries other than China, including developed countries, at prices either similar to, or substantially lower than, the prices for imports of Chinese garlic. See Remand Redetermination, Attach. 1. Concerning these data, Commerce states in the Remand Redetermination that it found “no information on the record which speaks to the quality, size, or number of cloves in garlic from the remaining countries in the MSFTI data.” Id. at 10. Jinan Yipin objects to the Department’s decision to disregard, on the ground of product specificity, the non-China portion of the MSFTI import data and, specifically, the data for Hong Kong and Malaysia, arguing that Commerce erred in dismissing the finding in the Market Research Report that “Hong Kong and Malaysia are ‘Chinese origin garlic.’ ” Jinan Yipin’s Comments 3. The court construes Jinan Yipin’s argument to be that the imports from Hong Kong and Malaysia, being Chinese-origin garlic, were highly similar in physical characteristics to the subject merchandise and that the value data for these imports are reliable because they represent transactions between exporters in Hong Kong and Malaysia and importers in India. The flaw in Jinan Yipin’s argument is that the Department is not required, on *1190 the basis of the record evidence, to make a finding that these imports actually were of Chinese origin. The state of the record does not make such a finding unavoidable. The Market Research Report indicates only that there is belief or speculation in the garlic industry that the garlic imports from Hong Kong and Malaysia are of Chinese origin. Market Research Report 27 (stating that “[i]mports from Hong Kong SAR (believed to be essentially garlic of Chinese origin) have also increased in the last few years” and that “it is speculated by the trade that a significant part of these imports [from Hong Kong SAR and Malaysia] may possibly be of garlic of Chinese origin”). The report makes no definitive conclusions with respect to the origin of garlic imports from countries other than China. Id. at 27-30. Jinan Yipin cites no other record evidence to support its argument. Commerce, accordingly, was justified in determining that the record lacked sufficient information on the physical characteristics of the garlic represented by the MSFTI import data pertaining to countries other than China.

Commerce declined to use the garlic price data in the Market Research Report, stating that the report is not publicly available and does not provide source information for the import, export, and domestic pricing data summarized therein. Remand Redetermination 9. Under 19 U.S.C. § 1677b(c), Commerce has broad discretion in the valuation of a factor of production, Nation Ford, 166 F.3d at 1377, so long as the Department’s determination is reasonable and supported by substantial record evidence. The public availability of data is a factor, among others, that the Department reasonably may consider. See Allied Pacific Food (Dalian) Co. v. United States, 30 CIT 736, 762, 435 F.Supp.2d 1295, 1317 (2006) (“In applying the preference for publicly available data, Commerce must balance the interests of transparency and verifiability that are served by public availability with other considerations, including the desirability of data that are as specific as possible to the raw material being valued.”). In this case, however, the court concludes that the Department’s finding that the information on garlic prices in the Market Research Report is not publicly available is not supported by substantial evidence on the record considered as a whole. Commerce stated that the “data are not publicly available and the source data for the values in the report are not included on the record for evaluation.” Remand Redetermination 12. However, Commerce itself requested that petitioners, who submitted the Market Research Report during the review, submit the publicly available source information for the garlic prices in the Market Research Report. Letter from Office Dir., AD/CVD Enforcement S, Dep’t of Commerce, to Collier Shannon Scott, PLLC (Aug. 1, 2003) (Case No. 03-00636 Admin. R. Doc. No. 123) (requesting the “considerable number of sources of information that are publicly available” from pages one and two of the report). Petitioners responded by providing 167 pages of prices for garlic. Letter from Collier Shannon Scott, PLLC to Sec’y of Commerce (Aug. 8, 2003) (Admin.R.Doc. No. 127) (“Publicly Available Sources ”). The court, upon review of the petitioners’ submission, observes that import, export, and domestic price data for garlic were included in petitioners’ submission. The Remand Redetermination does not contain a satisfactory explanation of why the supplemental submission by petitioners does not constitute publicly available information.

Commerce also found that the source data for the Market Research Report were not specifically identified throughout the report. The report states that the “study [summarized therein] was based primarily on secondary data from various sources as well as visits to the main garlic growing *1191 regions across the country.” Market Research Report 1. The report then lists numerous sources ranging from the NHRDF to “[discussions with people involved in the Indian garlic industry.” Id. at 1-2. But many of the graphs and tables in the report identify an organization that provided the information and may indicate some specific source information, e.g., Exhibit 3.1 citing “NHRDF Annual Reports ” or Exhibits 6.2 and 6.4 citing “DGCIS, Kolkata.” Id. at 5, 23-25, 28-29. The titles of several exhibits from the petitioners’ submission correlate to sources indicated in the tables. Publicly Available Sources 4, 155-56, 161. The Remand Redetermination does not discuss the record information supporting a finding that the data on values of garlic in India were obtained from specific public sources. Nor does the Remand Redetermination mention that Commerce asked for and received source data for the data in question. In this circumstance, the failure to address the considerable amount of evidence contrary to the Department’s finding that the data were not from publicly available sources casts doubt upon the validity of the finding itself. See Universal Camera Corp. v. NLRB, 340 U.S. 474, 488, 71 S.Ct. 456, 95 L.Ed. 456 (1951) (“The substantiality of evidence must take into account whatever in the record fairly detracts from its weight.”); see also Cleo Inc. v. United States, 501 F.3d 1291, 1296 (Fed.Cir.2007) (“Although a reviewing court must take into account contradictory evidence or any evidence in the record that undermines the agency’s finding, the substantial evidence test does not require that there be an absence of evidence detracting from the agency’s conclusion.”) (citing Universal Camera, 340 U.S. at 487-88, 71 S.Ct. 456).

Commerce, although rejecting the Market Research Report on the ground of public availability, discussed in the Remand Redetermination the possible use, as best available information, of the import data, export data, and grade-specific, domestic price data set forth in the report. Remand Redetermination 11. With respect to the import data, the Department’s analysis suffers the same flaws as its analysis of the Chinese MSFTI data. See id. (confining its analysis of the data in the Market Research Report to the sentence “[sjimilar to our discussion of the Indian import data above, most of the imports included in the import data in the Market Research Report are from the PRC, and for that reason, the import data in the report are not usable.”). Here again, Commerce relied on an overly broad practice without making findings specific to this review and without comparing the import data to the other data sets on the record. With regard to the export statistics, Commerce relied on a similar practice, stating that “India is a country with generally available export subsidies and it is the Department’s longstanding practice to eliminate exports from India from consideration in our surrogate value calculations.” Id. Commerce did not find, based on record evidence, that India subsidized the production of garlic for export or that any other identified export subsidy affected garlic exported from India. The unsupported finding that generally available export subsidies possibly affected garlic production in India does not justify the Department’s decision to reject the export data on garlic without comparing the data set to other available data sets.

Commerce also concluded in the Remand Redetermination that the domestic garlic prices in the Market Research Report are less product specific than the NHRDF data because the largest garlic bulb size identified in the price data is 40 mm in diameter while the Market Research Report states that Chinese garlic is 50-65 mm in diameter. Id. at 12; Market Research Report 29. Based on the record evidence pertaining to bulb size, the De *1192 partment’s finding that the NHRDF data are more product specific than the data on domestic garlic prices in the Market Research Report is supported by substantial record evidence.

Overall, the court concludes that the Department’s rejections of the two alternate data sources rely on overly broad practices, lack key findings and sound reasoning, and rely on certain findings that are unsupported by substantial record evidence. With respect to the MSFTI import data in particular, Commerce erred in failing to consider the quality of the MSFTI import data for China as compared to the quality of the NHRDF data after accounting for the flaw identified in the MSFTI import data for China, specifically that it is import data for goods from a nonmarket economy, and the possible flaw in the NHRDF data, specifically that Commerce has failed to find that a garlic producer is likely to spend substantially more to obtain clonal varietals from NHRDF. Commerce similarly erred in failing to make adequate findings, based on substantial evidence, to support its determination to disregard summarily the import and export data from the Market Research Report. Nor did Commerce undertake a qualitative comparison of the NHRDF data with the import data or the export data in the Market Research Report to produce a finding or findings that the court could affirm. Finally, the Department’s finding as to the public availability of the source information for the Market Research Report is not supported by substantial evidence.

The shortcomings affecting the Department’s rejection of the two alternate data sets preclude the court from affirming the Department’s decision, as set forth in the Remand Redetermination, that the NHRDF data represented the best available information with which to value the garlic seed input. On remand, Commerce must reconsider its decision to use the NHRDF data and base a new determination on a fair comparison of the three data sets. In so concluding, the court does not hold that a decision by Commerce on remand to use the NHRDF data in a determination of the surrogate value for garlic seed necessarily would be rejected as contrary to law. The record contains various evidence, in particular, the evidence in the Market Research Report, that may bear on the question of what garlic seed typically would be used by a garlic cultivator in India. However, it is for Commerce, not the court, to weigh this and other record evidence possibly relevant to a surrogate value for garlic seed and to reach a decision on this issue that may be sustained based on valid findings and adequate reasoning. Should Commerce determine that the record is inadequate to enable it to make a proper selection, it may reopen the record to obtain additional evidence.

2. The Court Rejects the Department’s Determination of a Surrogate Value for Irrigation Water

In determining a surrogate value for the irrigation water that Jinan Yipin used in its garlic cultivation, Commerce, in the Final Results, used averages of municipal water rates in India obtained from a reference issued by the Asian Development Bank. Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1372. In Jinan Yipin I, the court concluded that Commerce neither set forth sufficient findings of fact supported by substantial record evidence to support its determination nor provided a rational explanation for its choice. Id. at -, 526 F.Supp.2d at 1375. The Remand Redetermination makes no change in the valuation of irrigation water. See Remand Redetermination 13-19. The court concludes that Commerce has failed to comply with the remand ordered by Jinan Yipin I and again has determined a surrogate value that is not in accordance with law.

*1193 The Remand Redetermination states that Commerce, as it typically does, treated water as a material input, rather than an overhead expense, because of the significant quantity of water used in garlic cultivation and the incorporation of water into the finished product. Id. at 15-17. It also states that “[i]n selecting a surrogate value for water, we attempt to approximate the price that would be paid for the input in a market-economy country” and that “the fact that Jinan Yipin incurred no cost for the water does not negate the requirement that the Department value water as an input.” Id. at 17.

The reasons Commerce offers for retaining its earlier method of valuing irrigation water do not confront the fundamental errors identified in Jinan Yipin I. The court in Jinan Yipin I rejected the Department’s implied assumption that Indian garlic producers typically irrigate their crop using water supplied by municipal utilities, at costs associated with such utilities. Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1375. The Department’s claim in the Remand Redetermination that “we attempt to approximate the price that would be paid for the input in a market-economy country,” Remand Redetermination at 17 (emphasis added), is difficult to reconcile with what Commerce actually did on remand, which was to value water according to a method that is not linked by any factual finding to irrigation methods typically employed by garlic producers in India. See Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1375. As did the Final Results, the Remand Redetermination fails to make a finding that Indian garlic producers typically use water from a municipal water utility to irrigate their crops. Absent such a finding, it is unreasonable to consider irrigation water to be comparable to municipal water with respect to acquisition cost. Rather than make a finding, Commerce merely stands behind the lack of record evidence, stating that “there is no evidence on the record that demonstrates that garlic growers in market-economy countries incur no cost for water, nor that they incur only the cost of the energy to pump their water from a local water source.” Remand Redetermination 17. Observing that the municipal water rates are the only value for water on the record of the review, Commerce concludes that “valuing water with the municipal water rates is the best approximation of the water expense.” Id. at 18-19.

It is the Department’s role to conduct the review and determine such facts as are required for the calculation of surrogate values according to the “best available information” in the surrogate country. See 19 U.S.C. § 1677b(c)(l). However, Commerce offers no factual findings to support its conclusion that the municipal water rates are the' “best approximation of the water expense,” nor is an adequate reason stated in the Remand Redetermination as to why the municipal water rates are better information for this purpose than the cost of the energy used to pump Jinan Yipin’s irrigation water. Even though it makes no finding on how a garlic grower in India acquires irrigation water, Commerce paradoxically gives as its reason that “[i]f we were to only value the energy that Jinan Yipin used to pump the water from a water source, we would not be capturing the expense as experienced by a garlic grower in a market-economy country.” Remand Redetermination 19. The Department’s choice of the municipal utility water rates, absent any finding on how a garlic grower in the market economy country it chose as a surrogate obtains irrigation water, is entirely arbitrary and speculative and, therefore, not in accordance with the statutory requirement to use the best available information.

The Department’s insistence on using the municipal water rates is the more ob *1194 jectionable because Commerce had the opportunity to reopen the record to obtain additional values for irrigation water but did not do so. See Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1376; Remand Redetermination 14. In this regard, the administrative record reveals that Jinan Yipin attempted, in its reply to the Department’s draft remand results, to place on the record new information regarding agricultural water use in India and that Commerce rejected this information as an untimely submission of new information. Letter from Dir., AD/CVD Operations, Office 8, Dep’t of Commerce, to Grunfeld, Desiderio, Lebowitz, Silverman, & Klestadt LLP (Feb. 25, 2008) (Remand Admin. R. Doc. No. 7); Jinan Yipin’s Comments 6-7. Having refused to reopen the record to admit evidence relevant to a plausible valuation of the water Jinan Yipin pumped from a river in China, and having rebuffed Jinan Yipin’s attempt to submit additional evidence, Commerce cannot be sustained in a decision that is based on the lack of record information.

In Jinan Yipin I, the court also determined that the Department’s conclusion that double counting of water did not occur as a result of its calculation of a surrogate value for water was unsupported by factual findings. Jinan Yipin I, 31 CIT at -, 526 F.Supp.2d at 1375. In response, Commerce again relies on the absence of a separate line item for water expense on the financial statements of Parry Agro, an Indian tea producer, as evidence that double counting of water did not occur with respect to the calculation of the normal value of Jinan Yipin’s garlic. Remand Re-determination 18. Commerce reasons that irrigation water is likely excluded from the “indigenous” and “imported” category of “stores and spare parts consumed” on the Parry Agro statements. Id. As Commerce also states, “[tjhere is no record evidence that would demonstrate that the cultivation of tea in India requires irrigation,” id., but aside from the speculative assumption concerning categories on the financial statements, the Remand Redetermination points to no actual record evidence establishing that irrigation is not required. Here again, Commerce relies on the absence of record evidence; it does so in this instance to support its conclusion that “double counting does not occur by valuing water as a direct material.” Id.

Even if double counting did not occur, the Department’s surrogate value for irrigation water could not be sustained on remand. Although Commerce has considerable discretion in the valuation of factors of production, Nation Ford, 166 F.3d at 1377, it must exercise that discretion reasonably, according to actual findings of fact, and must not act arbitrarily or speculatively. On the administrative record as it now stands, the Department’s choice of municipal utility rates does not satisfy the “best available information” requirement of the statute. See 19 U.S.C. § 1677b(c)(l). If Commerce, on the remand, ordered herein, determines that it must value Jinan Yipin’s irrigation water according to a cost specific to irrigation water, it must reopen the record and make findings of fact, supported by new evidence, linking the value it chooses with the irrigation practice of garlic producers in India or another surrogate country that it chooses. Should Commerce choose not to reopen the record, it must value the water according to another method that is supportable on the current record, e.g., a method that accounts for the record fact that Jinan Yipin incurred only a pumping cost in obtaining irrigation water.

8. The Court Rejects the Department’s Determination of a Surrogate Value for Cardboard Cartons

In determining a surrogate value for packing cartons, Commerce, in the Final Results, used select data from Indian *1195 Import Statistics to determine a surrogate value of 124.91 rupees per kilogram for the cardboard cartons in which Jinan Yipin packed its garlic. Jinan Yipin I, 31 CIT at-, 526 F.Supp.2d at 1376. In Jinan Yipin I, the court concluded that in terms of product specificity, Commerce had not demonstrated a reasonable correlation between the imported products represented by the import statistics on which Commerce relied and the cartons that Jinan Yipin actually used. Id. at -, 526 F.Supp.2d at 1377. The court concluded that Commerce failed to explain how it was reasonable, when valuing a single product, to rely on import statistics that ranged widely from 38.74 to 239.05 rupees per kilogram. Id. Having noted that numerous types of packing cartons fall under the same tariff category, the court observed that “it would be illogical to assume that the country of origin, rather than substantial variations in the types of boxes imported, produced the wide variation in listed values.” Id. at -, 526 F.Supp.2d at 1377-78, 1377 n. 16, 1378 n. 18. The court also concluded that Commerce did not explain adequately why it excluded data from six of the countries for which import statistics were listed. 1 Id. at -, 526 F.Supp.2d at 1377-78. Finally, the court dismissed the Department’s reasons for rejecting price-quote data submitted by Jinan Yipin. Id. at-, 526 F.Supp.2d at 1379. The court explained that the quotes “are vastly superior to the Indian import data in an important respect: they are specific to the factor being valued” and that “Commerce’s rationale concerning the date of the quotes, which were eight months after the close of the period of review, and its rationale concerning the temporal range are not convincing absent evidence of significant price fluctuation in a short time.” Id. The Remand Redetermination makes no change in the valuation of packing cartons.

Commerce, in the Remand Redetermination, did not change its determination to rely on only certain data in the Indian Import Statistics. Commerce explained its choice by reiterating the reasoning advanced in the redetermination of the garlic seed surrogate value, stating that “the Department has a longstanding practice of eliminating imports from NME countries and countries known to have widely available export subsidies.” Remand Re-determination 20. In calculating the surrogate value for cartons, Commerce therefore disregarded imports from China and imports included in the “Unspecified” category because imports from NME countries are, or may be, included. Id. Commerce “eliminated imports from Indonesia, South Korea, and Thailand because they are countries known to have widely available export subsidies.” Id. Finally, data for Finland were disregarded because these data did not include an import quantity from which Commerce could calculate a country-specific average unit value. Id.

Because an import quantity was essential to the Department’s analysis, the court concludes that Commerce did not err in excluding data on imports from Finland on the ground that these data were incomplete. The court also concludes that Commerce was justified in deleting the import data from China, although the court reaches this conclusion on reasoning different from that relied upon by Commerce. As the court stated above, it is reasonable for Commerce to infer that data on imports from an NME country are inferior to import data for goods from a market econo *1196 my country but not reasonable for Commerce to adopt a blanket policy under which data on imports in a market economy country that pertain to products from NME countries are always excluded. With respect to the exclusion of data on imports from Indonesia, South Korea, and Thailand, Commerce again relied on a general policy or practice. In so doing, Commerce failed to make a finding, supported by substantial evidence on the record of this administrative review, that export subsidy programs exist in these three countries that affected or likely affected exports of garlic. In the absence of such a finding and such evidence, the court is unable to sustain the exclusion of the import data from these three countries.

With regard to the Department’s reasoning and findings in the Remand Redetermination on the rejection of the price-quote data, Commerce did not address the issue of product specificity directly and instead focused on the aspects of the price-quote data that it considered inferior to the import data. Specifically, Commerce stated that the import statistics meet several of the Department’s criteria for selecting a surrogate value: the import statistics “are period-wide prices, net of taxes and import duties, contemporaneous, and publicly available.” Remand Redetermination 21. Commerce reiterated that the price-quote data do not meet those criteria because they are not contemporaneous with the period of review and they are not publicly available. Id. Commerce also stated that there is no evidence that the data were generated in the normal course of business, that “Jinan Yipin did not submit information to the Department regarding the parties that requested the prices, or whether or not an affiliation existed between the requester and the Indian companies,” id., and that there is no record evidence to indicate whether the price quotes represent a broad market average or even fall in the spectrum of packing-carton pricing. Id. at 21-22.

After examining the price quotes in light of the Department’s analysis, the court concludes that the Department’s new findings of shortcomings in the price-quote data are supported by substantial record evidence. Although record evidence does not support a finding that the import statistics are at all specific to the product, let alone “sufficiently specific to the product” as Commerce suggests, Remand Redetermination 23, the breadth of the Department’s discretion convinces the court that it is for Commerce to decide between two imperfect data sets, provided that decision is supported by valid findings and adequate reasoning. In addition, for the reasons stated above, the court concludes that the Department’s exclusion of certain import data was not based on findings supported by substantial record evidence.

C. Correction of Alleged Ministerial Error Discovered upon Redetermination

In its comments on the draft remand redetermination, Jinan Yipin requested that Commerce correct a clerical or ministerial error that it alleges to have affected the Department’s calculation of the surrogate financial ratios in this administrative review. 2 Jinan Yipin’s Comments 13-15; Remand Redetermination 35-36. Commerce refused to consider the issue, arguing that Jinan Yipin had an opportunity after the issuance of the preliminary results of the administrative review and after the issuance of the Final *1197 Results to identify the ministerial error and that Jinan Yipin’s attempt to raise the matter after a remand is untimely. Remand Redetermination 36-37 (asserting that “the Department does not have the administrative resources to continually reexamine the record of the review to test the authenticity and legitimacy of new ministerial error allegations”).

The Court of International Trade has recognized the Department’s general authority to correct ministerial errors in a remand proceeding. See Hyundai Elecs. Indus. Co. v. United States, 29 CIT 981, 993, 395 F.Supp.2d 1231, 1243 (2005). In this case, however, Commerce has refused to act in response to a clerical error that is alleged to have existed in the Final Results but was undiscovered until the remand proceeding. Remand Redetermination 36-37. In such a circumstance, the Court of International Trade previously has exercised its discretion to require Commerce to correct clerical errors raised upon remand so that the court would not be affirming knowingly a determination affected by such errors. Serampore Indus. Pvt. Ltd. v. U.S. Dep’t of Comm., 12 CIT 825, 834, 696 F.Supp. 665, 673 (1988) (ordering, as part of a second remand, that Commerce determine whether a clerical error existed and, if so, correct the error to effectuate the court’s interest in affirming a correct determination). In a more recent case, the court, in a similar circumstance, declined to order Commerce to consider a late-raised allegation of ministerial errors because of the length of time that had elapsed and the fact that the issue involved was not before Commerce on remand. Dorbest Ltd. v. United States, 32 CIT -, -, 547 F.Supp.2d 1321, 1348 (2008).

In this case, correction of a ministerial error affecting the surrogate financial ratios would not be the sole issue remanded to Commerce, nor would addressing the issue require Commerce to reconsider any factual findings or conclusions of law that previously were made. As such, correction of the error, if it is found to exist, is not likely to impose a significant burden on Commerce or delay these proceedings in any material way. Correction of any error that exists will further the principle of attaining the most accurate result under the antidumping laws. Therefore, the court will remand to Commerce the issue of the ministerial error claimed by Jinan Yipin and instruct Commerce to determine whether there is in fact a ministerial error, and if such an error is found, to make the necessary correction.

III. Conclusion

The court concludes that certain of the findings and determinations in the Remand Redetermination are unsupported by substantial evidence on the record and otherwise contrary to law. Accordingly, the court will remand the matter to Commerce for reconsideration and redetermination in accordance with this Opinion and Order. The court affirms the Department’s Remand Redetermination with respect to its finding that Jinan Yipin was not affiliated to Houston Seafood through American Yipin, its decision not to invoke facts otherwise available and adverse inferences, and its conclusion to include in American Yipin’s reported total indirect selling expenses three months of Bayou Dock’s total salary and benefits expenses.

The court concludes that the surrogate value determinations that Commerce made for Jinan Yipin’s and Shandong’s use of garlic seed and water, and for Jinan Yipin’s packing cartons, were not based on adequate reasoning and on findings supported by substantial record evidence. The court also remands the Remand Re-determination with the directive that Commerce determine whether the alleged error in the calculation of the surrogate fi *1198 nancial ratios exists and make any necessary correction.

ORDER

Based on the court’s conclusions and the foregoing discussion, the court remands the Department’s Remand Redetermination, and it is hereby

ORDERED that the Remand Redetermination is AFFIRMED IN PART and REMANDED IN PART; it is further

ORDERED that the Remand Redetermination is affirmed with respect to the Department’s finding that Jinan Yipin was not affiliated to Houston Seafood through American Yipin, the Department’s decision not to invoke facts otherwise available and adverse inferences, and the Department’s conclusion to include in American Yipin’s reported total indirect selling expenses three months of Bayou Dock’s total salary and benefits expenses; it is further

ORDERED that this matter is remanded for further administrative proceedings consistent with this Opinion and Order; it is further

ORDERED that Commerce shall redetermine the surrogate values of Jinan Yipin’s garlic seed, use of water or the energy cost of accessing water, and cardboard cartons according to this Opinion and Order and, in doing so, shall redetermine the weighted average percentage antidumping duty margin that it applied in the Final Results to Jinan Yipin’s merchandise for the period of review as required in this Opinion and Order; it is further

ORDERED that Commerce shall redetermine the surrogate values of Shandong’s garlic seed and use of water or the energy cost of accessing water according to this Opinion and Order and, in doing so, shall redetermine the weighted average percentage antidumping duty margin that it applied in the Final Results to Shandong’s merchandise for the period of review as required in this Opinion and Order; it is further

ORDERED that in redetermining surrogate values Commerce may reopen the record as necessary to obtain additional information; it is further

ORDERED that Commerce shall determine whether a ministerial error occurred in the calculation of surrogate financial ratios identified by Jinan Yipin and shall correct any such ministerial error that it finds to have occurred; it is further

ORDERED that Jinan Yipin’s complaint and motion for judgment upon the agency record be, and hereby are, deemed to be amended to contest the Final Results with respect to the claimed ministerial error; it is further

ORDERED that Commerce shall ensure that all redeterminations are based on sufficient findings of fact and that the findings of fact are supported by substantial record evidence and shall provide on remand the reasons supporting its various redeterminations; and it is further

ORDERED that Commerce shall have one hundred twenty (120) days from the date of this Order to complete and file its remand determination; plaintiffs shall have forty-five (45) days from that filing to file comments; and Commerce shall have thirty (30) days after plaintiffs’ comments are filed to file any reply.

. The court observed that the surrogate value for packing cartons would have been 109.88 rupees per kilogram if all the data were used. Jinan Yipin Corp. v. United States, 31 CIT -, 526 F.Supp.2d 1347, 1378 (2007).

. ''Clerical” or "ministerial” errors are “error[s] in addition, subtraction, or other arithmetic function, clerical error[s] resulting from inaccurate copying, duplication, or the like, and any other similar type of unintentional error which the Secretary considers ministerial.” 19 C.F.R. § 351.224(f) (2008).