Baoding Mantong Fine Chemistry Co. v. United States

222 F. Supp. 3d 1231, 2017 CIT 44, 39 I.T.R.D. (BNA) 1248, 2017 Ct. Intl. Trade LEXIS 44, 2017 WL 1403280

• Court: United States Court of International Trade
• Decided: April 19, 2017
• Docket: Slip Op. 17-44; Court 12-00362
• Status: Published

Opinion

OPINION AND ORDER

Timothy C. Stanceu, Chief Judge

Stanceu, Chief Judge: In this litigation, plaintiff Baoding Mantong Fine Chemistry Co., Ltd. (“Baoding Mantong” or “Baod-ing”) contested the final determination (“Final Results”) that the International Trade Administration of the U.S. Department of Commerce (“Commerce” or the “Department”) issued to conclude an administrative review of an antidumping duty order (the “Order”) on glycine from the People’s Republic of China (“China” or the “PRC”). Glycine from the People’s Republic of China: Final Results of Antidumping Duty Administrative Review, 77 Fed. Reg. 64,100 (Int’l Trade Admin. Oct. 18, 2012) (“Final Results”). In the review, Commerce assigned to Baoding Mantong, a Chinese producer and exporter of glycine, a weighted-average dumping margin of 463.79%. Id. at 64,101.

Before the court is the Department’s decision on remand (“Remand Redetermination”) issued in response to this court’s opinion and order in Baoding Mantong Fine Chemistry Co. v. United States, 39 CIT -, 113 F.Supp.3d 1332 (2015) (“Baoding Mantong”). The Remand Redetermination calculated a new weighted-average dumping margin of 64.97% for Baoding Mantong. The court concludes that the Remand Redetermination is in some respects unsupported by the record evidence and orders that it be reconsidered.

I, Background

The court’s prior opinion and order presents background information on this case, which is summarized briefly and supplemented herein with developments since the issuance of that opinion and order. Baoding Mantong, 39 CIT at -, 113 F.Supp.3d at 1334-36.

A. The Administrative Review Proceeding before Commerce

Commerce issued the antidumping duty order on glycine from China (the “Order”) in 1995. 1 Antidumping Duty Order: Glycine From the People’s Republic of China, 60 Fed. Reg. 16,116 (Int’l Trade Admin. Mar. 29, 1995). On April 27, 2011, Commerce initiated the administrative review at issue, for which the period of review (“POR”) was March 1, 2010 through February 28, 2011, and in which Baoding Mantong was the sole respondent. Initiation of Antidumping and Countervailing Duty Administrative Reviews, 76 Fed. Reg. 23,545 (Int’l Trade Admin. Apr. 27, 2011). Defendant-intervenor GEO Specialty Chemicals, Inc. (“GEO”), a domestic producer of glycine and petitioner in the anti-dumping duty investigation, also participated in the review.

On April 11, 2012, Commerce published the preliminary results of the review (“Preliminary Results”), determining a preliminary margin of zero for Baoding Mantong. Glycine From the People’s Republic of China: Preliminary Results of Antidumping Duty Administrative Review and Partial Rescission of Antidumping Duty Administrative Review, 77 Fed. Reg. 21,738, 21,743 (Int’l Trade Admin. Apr. 11, 2012). Subsequently, in response to GEO’s allegation that Commerce made a currency conversion error in the Preliminary Results, Commerce released “Revised Preliminary Results of review to all interested parties on June 27, 2012.” Final Results, 77 Fed. Reg. at 64,100; see Revisions to Certain Surrogate Valuations and the Preliminary Margin-Calculation Program for Baoding Fine Chemistry, Co., Ltd., Mem. from Edythe Artman to File 2 (Int’l Trade Admin. June 27, 2012), (Ad min.R.Doc. No. 84) ECF No. 50-5. 2 Commerce explained that “correction of this error has a significant impact on Baoding Mantong’s dumping margin,” which increased from zero to 457.76%. Issues and Decision Mem. for the Final Results in the Administrative Review of Glycine from the People’s Republic of China, A-570-886 ARP 10-11, at 29 (Int’l Trade Admin. Oct. 9, 2012), (Admin.R.Doc. No. 127), available at http://enforcement.trade. gov/frn/summary/PRC/2012-25595-1.pdf (last visited Mar. 1, 2017) (“Final I & D Mem.”).

Commerce issued the Final Results on October 18, 2012, in which it made a minor change to the analysis in the Revised Preliminary Results and calculated a final margin of 453.79% for Baoding Mantong. 3 Final Results, 77 Fed. Reg. at 64,101.

B. Baoding’s Initiation of this Action and Filing of a Rule 56.2 Motion

Baoding Mantong initiated this action by filing a summons on November 16, 2012 and a complaint on December 7, 2012. Summons, ECF No. 1; Compl., ECF No. 7. Baoding Mantong subsequently moved for judgment on the agency record under USCIT Rule 56.2. Mot. for J. on the Agency R. and Mem. in Supp. (July 22, 2013), ECF No. 30 (“Baoding Mantong’s Br.”).

In its Rule 56.2 motion, Baoding Man-tong contested various individual surrogate values Commerce used, under the procedure of section 773(c)(1) of the Tariff Act of 1930, 19 U.S.C. § 1677b(c)(1), to determine the normal value of Baoding Mantong’s subject merchandise in calculating the 453.79% weighted-average dumping margin. It also claimed generally that the 453.79% margin “defies commercial and economic reality.” Baoding Mantong, 39 CIT at -, 113 F.Supp.3d at 1336. “Regarding the enormity of the margin assigned to it, Baoding argues that if the 453.79% margin ‘reflected commercial reality, Baoding would have suffered huge operating losses during the period of review.’ ” Id., 39 CIT at -, 113 F.Supp.3d at 1338 (citing Baoding Mantong’s Br. 13) (footnote omitted). Baoding Mantong maintained that “the administrative record establishes that Baoding operated profitably during both the period of review and the prior year” and “could not be in busi ness if Commerce’s results were a reasonable reflection of commercial reality.” Baoding Mantong’s Br. 13. Baoding Mantong pointed out, further, that “the 458.79 percent margin is approximately three times higher than the 151.89 percent China-wide margin, the margin applied as total adverse facts available for parties who refuse to participate or otherwise impede Commerce’s investigations.” Id. at 11.

Baoding Mantong challenged the individual surrogate values Commerce determined for four of its factors of production: chlorine, ammonia, formaldehyde, and steam coal. It also claimed that Commerce erred in its choice of financial information for use in calculating surrogate financial ratios for Baoding Mantong’s factory overhead, its selling, general and administrative (“SG & A”) expenses, and its surrogate profit. In the Final Results, Commerce used financial statements from three producers of pharmaceutical products in Indonesia, which was the Department’s chosen surrogate country, see 19 U.S.C. § 1677b(c)(1)-(4), to calculate the three financial ratios. Baoding Mantong argued that these Indonesian companies are dissimilar to it because they are “conglomerate high-tech pharmaceutical companies that employ thousands of workers and have huge R & D budgets, elaborate distribution systems and advertising campaigns for their branded and proprietary products.” Baoding Mantong’s Br. 39. Baoding Mantong stated that, in contrast, during the POR Baoding Mantong produced only glycine as a commodity product and had minimal selling expenses. Id. at 38.

Following oral argument, defendant filed a motion for a partial voluntary remand to allow Commerce to reconsider the selection of the financial statements for use in calculating the surrogate financial ratios. Def.’s Mot. for Voluntary Remand (Aug. 6, 2014), ECF No. 64. GEO and Baoding Mantong opposed this motion. Def.-Int.’s Opposition to Def.’s Mot. for Voluntary Remand (Aug. 25, 2014), ECF No. 65; Pl.’s Opposition to Def.’s Mot. for Voluntary Remand (Aug. 25, 2014), ECF No. 66. The court granted defendant’s motion for leave to reply to Baoding Mantong’s opposition to its motion. Baoding Mantong, 39 CIT at -, 113 F.Supp.3d at 1342.

C. The Court’s Opinion and Order

In Baoding Mantong, the court directed Commerce “to reconsider all aspects” of its determination of the 453.79% margin. Id. The court stated that it “will not assume that a remand confined to the question of the financial ratios could suffice for correction of the serious, fundamental deficiencies affecting the Final Results.” Id. The court noted that Commerce had not relied upon facts otherwise available or an adverse inference in calculating the 453.79% margin. Id., 39 CIT at -, 113 F.Supp.3d at 1338 n.7. The court concluded that “Commerce failed to fulfill its obligation to determine the most accurate margin possible when it assigned Baoding a weighted average dumping margin of 453.79%, which on the record of this case was not realistic in any commercial or economic sense and punitive in its effect.” Id., 39 CIT at -, 113 F.Supp.3d at 1334. The court directed Commerce to “determine a new margin for Baoding that is ... grounded in the commercial and economic reality surrounding the production and sale of Baoding’s subject merchandise, and that is fair, equitable, and not so large as to be punitive.” Id. The court also stated that:

If, in the process of determining a new margin for Baoding, Commerce concludes that the record information is insufficient to allow it to determine a margin that satisfies these fundamental requirements, it either must reopen the record ... or it must follow the statutory directive to determine a margin according to the method of 19 U.S.C. § 1677b(c)(2) ....

Id., 39 CIT at -, 113 F.Supp.3d at 1341.

D. The Department’s Remand Redetermination

In response to the court’s opinion and order in Baoding Mantong, Commerce issued the Remand Redetermination on March 29, 2016. Final Results of Redeterm. Pursuant to Court Remand (March 30, 2016), ECF No. 73 (“Remand Redeterm.”). On remand, Commerce “recalculated certain aspects of Baoding Mantong’s dumping margin,” basing the financial ratios for factory overhead, SG & A expenses, and profit “upon the financial information for an Indonesian producer of urea, rather than the financial information of three Indonesian pharmaceutical companies, in determining various aspects of Baoding Mantong’s normal value pursuant to section 773(c)(1) of the [Tariff] Act [of 1930 (19 U.S.C. § 1677b(c)(1)) ].” Id. at 5. Commerce stated that “respectfully, and under protest, we have also reconsidered the remaining aspects of Baoding Man-tong’s normal value calculation.” Id. Commerce added that it “has not reconsidered the remaining aspects of Baoding Man-tong’s dumping margin, such as export price or constructed export price.” Id.

Although discussing the surrogate values it determined in the Final Results for Baoding Mantong’s use of the production inputs of liquid chlorine, ammonia, formaldehyde, and steam coal, Commerce made no changes to these surrogate values as determined in the Final Results. See id. at 12-20. The change in the financial ratios resulting from use of a new source of financial information, i.e., the financial statements of the Indonesian urea fertilizer producer PT Pupuk Kujang (“Pupuk”), reduced the weighted-average dumping margin from 453.79% to 64.97%. Id. at 36.

In comments to the court, GEO opposed the Remand Redetermination in general and objected specifically to the change Commerce made that departed from the Final Results. Def.-Intervenor’s Comments on Remand Redeterm. (Apr. 29, 2016), ECF No. 75 (“GEO’s Comments”). In its comments, Baoding Mantong supported the Remand Redetermination in part and opposed it in part. Pi’s Comments on Remand Redeterm. (Apr. 29, 2016), ECF No. 77 (“Baoding Mantong’s Comments”). Defendant filed a reply to the comments of Baoding Mantong and GEO on June 1, 2016. Def.’s Reply to Comments on Remand Redeterm. (June 1, 2016), ECF No. 81 (“Defi’s Reply”).

II. Discussion

A. Jurisdiction and Standard of Review

The court exercises jurisdiction of this action under section 201 of the Customs Courts Act of 1980, 28 U.S.C. § 1581(c), pursuant to which the court reviews actions commenced under section 516A of the Tariff Act of 1930 (the “Tariff Act”), as amended, 19 U.S.C. § 1516a. 4 In reviewing an agency determination, the court “shall hold unlawful any determination, finding, or conclusion found ... to be unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i).

B. Remaining Issues in this Litigation

In its comments, GEO relies on Nan Ya Plastics Corp. v. United States, 810 F.3d 1333 (Fed. Cir. 2016), which the Court of Appeals for the Federal Circuit (“Court of Appeals”) decided after the court issued the opinion and order in Baoding Mantong. GEO’s Comments 4-6. GEO argues that “Commerce should have rejected the Court’s Opinion and Order in its entirety since the Federal Circuit’s 2016 Nan Ya decision is binding precedent that invalidates the basis of the Court’s 2015 Opinion and Order.” Id. at 4. According to GEO,

Because the Nan Ya decision invalidates the basis of the Opinion and Order, Commerce in its Remand Results should have relied on the Nan Ya decision to either (1) uphold without caveat its entire Final Results or (2) acquiesce to the protest—including the surrogate financial ratios in the Final Results.

Id. at 6. GEO adds that “[w]hat Commerce should not do—but did do—is effectively overturn the Court’s denial of Commerce’s voluntary remand motion.” Id. GEO argues, in the alternative, that it was unlawful for Commerce to replace the financial information of the three Indonesian pharmaceutical manufacturers with the financial information of the Indonesian urea fertilizer producer Pupuk. GEO maintains that Commerce improperly applied “the three-part comparable merchandise test, endorsed by this Court,” under which Commerce is to examine “the physical characteristics, end uses and production processes of a product produced in a surrogate country to determine whether it is comparable.” Id. at 6-7.

In its comments to the court on the Remand Redetermination, Baoding Man-tong supports the Department’s new financial ratios resulting from use of the financial information of the Indonesian urea fertilizer producer in place of the financial information pertaining to the three Indonesian pharmaceutical producers. Baoding Mantong’s Comments 2. Baoding Mantong opposes the Department’s decision not to revise the surrogate values for liquid chlorine, ammonia, formaldehyde, and steam coal. Id. at 2-7.

C. The Decision of the Court of Appeals in Nan Ya Plastics Corp. Does Not Invalidate the Court’s Previous Opinion and Order

GEO is incorrect in its view that Nan Ya Plastics Corp. invalidates the basis of the court’s order in Baoding Mantong. GEO further errs in concluding from its interpretation of the decision of the Court of Appeals that Commerce either should have upheld the Final Results or designated as entirely under protest its response to the court’s order in Baoding Mantong.

In arguing that Baoding Mantong is now invalidated, GEO relies upon certain language in the opinion of the Court of Appeals that mentions the court’s opinion in Baoding Mantong and other opinions of this Court as illustrations of recent instances in which the Court of International Trade has relied upon the terms “commercial reality” and “accurate” as used in past decisions of the Court of Appeals. Nan Ya Plastics Corp., 810 F.3d at 1341-42. Taking the opportunity to clarify these terms, the Court of Appeals stated that “[w]e clarify that ‘commercial reality' and ‘accurate’ represent reliable guideposts for Commerce’s determinations.” 810 F.3d at 1343. The appellate court added, however, that “[tjhose terms must be considered against what the antidumping statutory scheme demands.” Id. (citing Chevron, U.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984)). Further, the Court of Appeals opined that “[w]hen Congress directs the agency to measure pricing behavior and otherwise execute its duties in a particular manner, Commerce need not examine the economic or commercial reality of the parties specifically, or of the industry more generally, in some broader sense.” Id. at 1344 (citing United States v. Eurodif S.A., 555 U.S. 305, 317-18, 129 S.Ct. 878, 172 L.Ed.2d 679 (2009)). The Court of Appeals continued,

Our ease law and the statute thus teach that a Commerce determination (1) is “accurate” if it is correct as a mathematical and factual matter, thus supported by substantial evidence; and (2) reflects “commercial reality” if it is consistent with the method provided in the statute, thus in accordance with law.

Id. (citation omitted). For two reasons, Nan Ya Plastics Corp. is not properly interpreted to invalidate the court’s opinion and order.

First, Nan Ya Plastics Corp. and this case involve different issues and different facts. Nan Ya Plastics Corp. upheld an antidumping duty rate of 74.34% as facts otherwise available, and an adverse inference, determined according to 19 U.S.C. § 1677e(b) for a respondent that refused to participate in an administrative review of an antidumping duty order. Id. at 1339, 1350. This case involves the final results of an administrative review of an antidumping duty order in which Commerce assigned an individual weighted-average dumping margin of 453.79% to a cooperative respondent without resorting to an adverse inference. Due to these differences, Nan Ya Plastics Corp. does not establish a precedent under which the order remanding the Final Results for reconsideration and redetermination must be invalidated.

Second, even if Nan Ya Plastics Corp. were considered to be a holding controlling the outcome of this case (which it is not), the guidance the Comí; of Appeals provided in its opinion would not support the notion that the court’s order remanding the Final Results is invalid. Under that guidance, a margin is accurate if it is supported by substantial evidence and reflects commercial reality if it is consistent with the statutory method. Nan Ya Plastics Corp., 810 F.3d at 1344. The Department’s determination of a 453.79% margin for Baoding Mantong in the Final Results did not satisfy this standard.

Although the opinion and order in Baod-ing Mantong cited appellate decisions discussing the concepts of accuracy and commercial reality, the problem posed by the enormity of the margin in the Final Results, and by the inconsistency with the record facts, was more serious than those terms might be understood to imply. As the court pointed out, a margin of 453.79% that is not an adverse inference but instead is an actual, calculated weighted-average dumping margin would have to have been based on normal value that is between five and six times the U.S. price, i.e., a U.S. price reflecting goods sold at a huge loss. Baoding Mantong, 39 CIT at -, 113 F.Supp.3d at 1339. Yet, Baoding Mantong identified record evidence supporting a finding that its export sales incurred no such loss during the POR. Id. Apart from the question of what might be considered a result grounded in accuracy or commercial reality, the margin Commerce assigned in the Final Results, considered according to record evidence, signified commercial impossibility. Id., 39 CIT at -, 113 F.Supp.3d at 1340 (“Commerce has assigned a margin that, on the record facts, has not been demonstrated to be anything other than commercially impossible.”) Nan Ya Plastics Corp. did not hold that Commerce is free to assign to a cooperative respondent a weighted-average dumping margin that is shown by record evidence—in particular, the evidence that the merchandise in question was not sold at a loss during the POR—to be so enormously high as to be punitive.

GEO’s summary conclusion that in the Remand Redetermination Commerce “should have relied on the Nan Ya decision to either ... uphold without caveat its entire Final Results” or “acquiesce to the Court and reconsider its entire Final Results under protest,” GEO’s Comments 6, is similarly misguided. Just as nothing decided by Nan Ya Plastics Corp. requires the court to vacate Baoding Mantong, nothing in that appellate decision could justify the Department’s disregarding of the court’s order, whether by reverting to the Final Results or by otherwise signaling to the court that it does not consider itself bound to reconsider the determination contested in this litigation. Baoding Mantong has challenged the 453.79% margin in general terms as commercially impossible and specifically has challenged the surrogate values for four factors of production and the source of information underlying the Department’s financial ratios. The premise underlying GEO’s argument is that the holding of the Court of Appeals in Nan Ya Plastics Corp. somehow invalidates or defeats all of Baoding Mantong’s claims in this litigation. This premise is unsound, if for no other reason than that claims analogous to Baoding Mantong’s were not before the Court of Appeals.

D. Commerce Permissibly Used the Financial Data of the Indonesian Producer of Urea Fertilizer, Pupuk, to Calculate the Financial Ratios

According to section 773(c)(1) of the Tariff Act, Commerce, as a general matter, determines the normal value of subject merchandise from a nonmarket economy (“NME”) country “on the basis of the value of the factors of production utilized in producing the merchandise and to which shall be added an amount for general expenses and profit plus the cost of containers, coverings, and other expenses.” 5 19 U.S.C. § 1677b(c)(1). Commerce typically calculates surrogate values for factory overhead, for SG & A expenses, and for profit by applying surrogate financial ratios derived from the financial statements of one or more producers of comparable merchandise in the primary surrogate country. Commerce stated in the Remand Redetermination that “[ajceording to 19 CFR 351.408(c)(4), we will normally use non-proprietary information from producers of identical or comparable merchandise in the surrogate country as the basis for our calculation of the surrogate ratios.” Remand Redeterm,. 8 (citing 19 C.F.R. § 351.408(c)(4)). The regulation does not provide guidance as to the meaning of “comparable merchandise.” 6 Commerce explained that “when selecting financial statements for the purpose of calculating surrogate financial ratios, the Department’s policy is to use data from market-economy surrogate companies based on the specificity, contemporaneity, and quality of the data.” Remand Redeterm. 8-9 (footnote omitted). Commerce explained, further, that “the Department has ... developed a three-part test for identifying comparable merchandise which examines, where appropriate, the physical characteristics, end uses, and production process.” Id. at 9 (footnote omitted). Commerce added that it also “examines how similar a proposed surrogate producer’s production experience is to the NME producer’s production experience.” Id. (footnote omitted).

In the Preliminary and Final Results, Commerce “relied on the financial information of three Indonesian pharmaceutical companies,” PT Darya-Varia Laboratoria Tbk (“Darya-Varia”), PT Pyridam Farma Tbk (“Pyridam”), and PT Kalbe Farma Tbk (“Kalbe”), whose financial statements GEO had placed on the record, “as there was no information on the record for companies in Indonesia that produced glycine.” 7 Final I & D Mem. 26. Commerce declined to use financial statements submitted by Baoding Mantong, which were those of the Indonesian producer of urea fertilizer and of Indian producers of chlorine, ammonia, and formaldehyde. 8 Id. at 26-27. Upon reconsidering the question for the Remand Redetermination, Commerce found “that the financial statements of the Indonesian urea fertilizer producer, PT Pupuk Kujang (Pupuk),” which Baoding had placed on the record, “constitute^] the best available information for determining overhead, SG & A, and profit for Baoding Mantong.” Remand Redeterm. 8.

With respect to the first criterion of the three-part test, physical characteristics, Commerce acknowledged and maintained its earlier finding, as made in the Final Results, that “the pharmaceutical companies produce amino acids (used in pharmaceutical products), and that glycine is also an amino acid.” Id. at 10 (footnote omitted). Commerce further stated in the Remand Redetermination that “[w]e also continue to recognize that urea fertilizer is a raw-material input which is used in other chemicals, and thus not akin to glycine.” Id. (footnote omitted). Commerce added that “[h]owever, upon further examination of the record, we find that the pharmaceutical companies produce high-value patented and branded medical products which, by their nature, are much more complex products than glycine, which, if used, is merely an additive to some of these products.” Id. at 10 n.37 (citing GEO Comments on Selection of Surrogate Country for Valuing Factors of Production and Surrogate Value Data for Valuing Baoding’s Factors of Production at Exhs. 6-7 (Nov. 1, 2011), (Second Admin.R.Doc. No. 16) (referencing Darya-Varia’s and Kalbe’s high-value brands)).

GEO argues that Commerce “improperly applied” the three-part test. GEO’s Comments 7. With respect to physical characteristics, GEO argues that Commerce “did not provide any record evidence about glycine itself to support its claims that the products of Darya-Varia, Pyridam, and Kalbe are ‘ultimately more complex than glycine’ and that glycine is ‘merely an additive.’ ” Id. at 8 (emphasis in original). According to GEO, Commerce ignored record evidence that “glycine itself is complex and is sold primarily in the United States as USP-grade glycine, which is often used as a high-value patented and branded pharmaceutical end product.” Id. GEO asserts that “[t]he record shows that the vast majority of glycine sold in the United States is USP-grade glycine, an FDA-regulated product used in patented, biotech and commercially branded products.” Id. (citation omitted). GEO cites evidence it placed on the record, including a list of “commercially branded” glycine-based medical products and a finding by the U.S. International Trade Commission that USP-grade glycine represents “the vast majority of glycine sold in the United States.” 9 Id. at 8-9 (citing GEO Rebuttal to Baoding’s July 16, 2012 Arguments and Surrogate Value Information 4 n.4-5 (July 23, 2012), (Second Admin.R.Doc. No. 122)). GEO also argues that in comparing physical characteristics, Commerce ignored record evidence that Pupuk produced ammonia, a raw material input of glycine, in addition to urea and that urea is also a raw material. GEO’s Comments 8.

The court is not persuaded by the arguments GEO directs against the Department’s finding as to physical characteristics. Evidence that most glycine sold in the United States is USP-grade glycine and is sometimes a pharmaceutical end product, or the basis of one, does not undermine the validity of the Department’s considering of the glycine at issue, which is the glycine produced and sold by Baoding Mantong, to be physically different in a significant respect than the pharmaceutical products of Darya-Varia, Pyridam, and Kalbe. The record contained substantial evidence supporting the finding by Commerce that the glycine produced and sold by Baoding Mantong during the POR differed from the high-value, patented and branded medical products of Darya-Varia, Pyridam, and Kalbe because it was a type of product sold “in bulk quantities to customers for use in the production of retail products.” Remand Redeterm. 10 (citing, inter alia, Baoding Mantong Section C and D Response D-16 (Aug. 2, 2011), (Admin.R.Doc. No. 41)). In concluding that glycine, “if used, is merely an additive to some of these [high-value, patented and branded medical] products,” Remand Redeterm. 10, Commerce did not expressly take the position that glycine produced by companies other than Baoding Mantong could not be sold as branded pharmaceutical end products. Thus, Commerce sought to compare the physical characteristics of the products of the Indonesian companies to the specific glycine produced by Baoding Mantong, the NME producer for which Commerce was determining overhead, SG & A expenses, and profit.

Further, GEO objects that “Commerce in its Remand Results attempted to claim without any supporting record evidence that the admittedly upstream product of Pupuk, urea, should now be considered to have physical characteristics more comparable to glycine than a downstream, amine-based pharmaceutical product.” GEO’s Comments 10. According to GEO, “[t]he record evidence shows, if anything, that glycine is a complex, downstream product that is more than an additive; however, Commerce ignored such evidence because it would have invalidated Commerce’s physical characteristics finding.” Id. GEO adds that “[i]f it had considered that record evidence, Commerce would have found that the products of Darya-Varia, Pyridam and Kalbe have physical characteristics much more similar to glycine than the upstream products of Pupuk.” Id. GEO would have the court conclude that Commerce, in order to use the financial data of Pupuk, was required to find that glycine as general matter was, in every respect, more similar in physical characteristics to urea fertilizer than it was to the pharmaceutical products of Darya-Varia, Pyridam, and Kalbe. The Department’s three-part test does not require that generalized finding, which is one Commerce did not make. Rather, Commerce acknowledged that “urea fertilizer is a raw-material input which is used in other chemicals, and thus not akin to glycine.” Remand Redeterm. 10. Nor did Commerce find that glycine is not complex; instead, its comparison to the branded pharmaceuticals was in relative terms. Commerce applied its “physical characteristics” criterion more narrowly than GEO assumes. Commerce reached a finding that the products of Darya-Varia, Pyridam, and Kalbe, as high-value, patented and branded medical products, were dissimilar in that respect to, and more complex than, the bulk glycine of Baoding Mantong, which according to record evidence was used as an ingredient in more advanced products. Commerce reached this finding based on substantial evidence and reasonably applied its own “physical characteristics” criterion in the context of the inquiry it was making. GEO is mistaken both in concluding that Commerce reached a broader finding and in insisting that it was required to do so to support its ultimate decision to use the Pupuk financial statements.

As to the second criterion, end uses of the products, Commerce relied on essentially the same finding that it made as to physical characteristics, but it did so without the reference to relative complexity. Commerce found that “the pharmaceutical companies are engaged in the development, production and sale of retail products (i.e., products that are packaged for consumer use) whereas Baoding Mantong sells glycine in bulk quantities to customers for use in the production of retail products.” Remand Redeterm. 10 (footnote omitted). Commerce added that “[i]n this respect, urea is also most commonly sold in bulk quantities to customers for use in fertilizer applications.” Id.

GEO raises, in effect, the same objection to the Department’s application of its end uses criterion as it raised with respect to physical characteristics. GEO argues as follows:

Like Commerce’s physical characteristics determination, Commerce’s finding on end uses in its Remand Results—that the end use of Pupuk’s urea product is more like the end use of glycine because both are used in the production of retail products while the products of Darya-Varia, Pyridam and Kalbe are retail products themselves—relied on unsubstantiated claims that contradicted record evidence and ignored record evidence: Commerce ignored information about the end uses of glycine from the very companies that use glycine as an end product and ignored information that Pupuk makes ammonia also, which is a raw material input for glycine itself. Instead, Commerce referenced Baod-ing’s unsupported views of the end uses of glycine and accepted without proof Baoding’s empty mantra that glycine is a commodity product sold in “bulk quantities.”

GEO’s Comments 11 (citations omitted). GEO identifies the information “Commerce ignored” as the evidence “that the ammonia product of Pupuk is an early raw material input of glycine itself’ and evidence that “glycine is sold primarily in the United States as USP-grade glycine, an FDA-regulated product which is often used as a high-value patented and branded pharmaceutical end product.” Id. at 12 (citations omitted).

GEO’s argument misses the point that the “ignored” evidence to which it directs the court’s attention does not establish that Commerce lacked substantial record evidence with which to find that the glycine Baoding Mantong actually produced was a bulk product used in manufacturing retail products. This was not an “unsubstantiated claim” or an “empty mantra.” It was a valid finding supported by the evidence that Baoding Mantong provided and that Commerce cited in the Remand Rede-termination. GEO does not point to record evidence showing that Baoding Mantong misrepresented to Commerce the facts about its own products.

The Department’s analysis under its “production processes” criterion is based on findings similar to those made under its first two criteria. Commerce found that “the production processes for the pharmaceutical products are much more complex than that of glycine,” that “[t]he production process for glycine consists of chemical reactions between a few inputs, as demonstrated by the low number of factors of production reported by Baoding Mantong,” and that “urea is similar to glycine in that the production process relies upon the chemical reactions between a few inputs.” Remand Redeterm. 10-11 (footnotes omitted). Commerce found that, in contrast, “pharmaceutical products can involve the production and combination of several ingredients, including glycine, for one product and, in the case of Darya-Varia, Pyridam, and Kalbe, involves packaging processes that are not necessary for Baoding Mantong’s sales of glycine.” Id. at 11 (footnote omitted).

GEO objects that Commerce omitted from its “production processes” analysis “any consideration of what Baoding perceived to be Pupuk’s primary product, ammonia, which is used in the early stages of the glycine production process.” GEO’s Comments 13 (citations omitted). According to GEO, the ammonia production process is not “most like” the glycine production process, which, “especially for the glycine primarily sold, to the United States, USP-grade glycinef,] involves many more downstream stages than the production process of one of its initial raw material inputs.” Id. (citations omitted). GEO concludes that “Commerce’s failure to consider Pupuk’s ammonia product is a fatal flaw in its analysis.” Id.

GEO’s objection does not convince the court that the Department’s analysis of the production processes criterion is fatally flawed. As defendant points out in response to GEO’s argument, there is record evidence that Pupuk described itself as being in the urea fertilizer business and that “[t]he record also demonstrates that, for the relevant period of 2010-2011, Pupuk produced more urea than ammonia.” Def.’s Reply 9 (citing Baoding Submission of Surrogate Value Information and Comment (July 16, 2012), (Second Admin.R.Doc. No. 99) (“Baoding’s Surrogate Value Submission”)). Moreover, the fact that Pupuk produced ammonia as well as urea does not refute or otherwise invali date the finding, which was supported by substantial evidence, that the bulk glycine produced by Baoding Mantong was less complex in terms of production and packaging processes than the pharmaceuticals of Darya-Varia, Pyridam, and Kalbe.

As to production experience, Commerce concluded from its findings as to production processes that Baoding Mantong’s production experience was “most similar to that of the urea fertilizer, rather than the pharmaceutical companies” and added a finding that the pharmaceutical producers had relatively high research and development (“R & D”) and selling costs compared with Baoding Mantong. Remand Redeterm. 11 (footnotes omitted) (“We also have no evidence that Baoding Mantong incurred high R & D expenses and high selling costs”). Commerce found that “[i]n contrast, glycine, like urea fertilizer, is a low-value commodity product, in which the raw materials account for a larger share of the production costs.” Id. (citation omitted).

GEO raises objections parallel to those it raised in attacking the Department’s analysis under the other criteria, arguing that Commerce relied on Baoding Man-tong’s “unsubstantiated claims” and “conveniently ignored Pupuk’s ammonia production experience.” GEO’s Comments 14. The latter, GEO argues, “cannot be logically described as a production experience similar to the production experience of the downstream product, glycine, using that product as a key input.” Id. GEO adds that “[t]he entire glycine production experience involves many more downstream stages than the production experience of one of its initial raw material inputs, ammonia.” Id. (citations omitted).

The argument GEO directs to the production experience criterion does not convince the court that Commerce relied on invalid findings and unsubstantiated claims or ignored contrary evidence. The record contained substantial evidence allowing Commerce to conclude that the production experience of Baoding Mantong differed from that of the three Indonesian pharmaceutical producers in certain respects that Commerce reasonably considered to be significant to its analysis, in particular as to R & D and sales expenses. The Department’s analysis under this criterion, and its analyses under the previous ones, are not invalidated simply because Baoding Mantong’s production experience and production processes, and the end uses and physical characteristics of its subject merchandise, are not analogous in all respects to those of Pupuk. Commerce analyzed the competing information sources under each of its criteria, reached findings supported by substantial evidence, and adequately explained its reasoning. While GEO emphasizes record evidence of the ways in which Baoding Mantong differs from Pu-puk and is similar to the three pharmaceutical producers, it has not made the case that the Department’s findings and ultimate conclusion to use the Pupuk statements were unsupported by substantial record evidence. Accepting GEO’s argument that Commerce improperly applied its criteria and reached an invalid result would require the court to ignore the substantial evidence supporting the Department’s choice of the Pupuk statements and in so doing reweigh the evidence, substituting its own judgment for that of the agency. The court, therefore, will not order Commerce to reconsider or alter its decision to use those statements in calculating the financial ratios.

E. Surrogate Values for Liquid Chlorine, Ammonia, Formaldehyde, and Steam Coal

In its comments on the Remand Rede-termination, Baoding Mantong concurred in the Department’s decision to change the source of information for the financial ratios and added that it “strongly disagrees with the remainder of Commerce’s remand results.” Baoding Mantong’s Comments 2. However, the comment submission does not object specifically to any aspect of the Remand Redetermination other than the surrogate values for liquid chlorine, ammonia, formaldehyde, and steam coal, which Commerce in the Remand Redetermination left unchanged from the Final Results. The court, therefore, construes the comments as a waiver of any challenge to the decisions Commerce reached in the Remand Redetermination other than the decision to maintain those four surrogate values as determined in the Final Results.

The Tariff Act directs Commerce to value the factors of production “based on the best available information regarding the values of such factors in a market economy country or countries” that Commerce considers “appropriate.” 19 U.S.C. § 1677b(c). The Tariff Act further directs that Commerce, when valuing factors of production, “utilize, to the extent possible, the prices or costs of factors of production in one or more market economy countries that are—(A) at a level of economic development comparable to that of the nonmarket economy country, and (B) significant producers of comparable merchandise.” Id. § 1677b(c)(4).

Commerce identified six countries as at a level of economic development comparable to that of China and as significant producers of merchandise comparable to glycine based on adjusted quantities and values for each of the factors of production: the Philippines, Indonesia, Ukraine, Thailand, Colombia, and South Africa. 10 Letter from Angelica L. Mendoza, Surrogate Country List 1-2 (Sept. 8, 2011), (Second Admin.R.Doc. No. 5). It was from these six countries that Commerce selected Indonesia as its primary surrogate country. Factors of Production Valuation for the Preliminary Results 1 (Mar. 30, 2012), (Second Admin.R.Doc. No. 58) (“Preliminary FOP Mem.”). A Department regulation provides, in pertinent part, that “the Secretary normally will value all factors in a single surrogate country.” 19 C.F.R. § 351.408(c)(2).

During the review, Baoding opposed the selection of Indonesia as the primary surrogate country and placed on the record sources of possible surrogate data pertaining to India. See Baoding, Surrogate Country Comments and the Submission of Proposed Surrogate Values 7 (Nov. 1, 2011), (Second Admin.R.Doc. No. 12) (“Baoding’s Surrogate Country Comments”). In its Rule 56.2 motion, Baoding Mantong stated that “[i]n the current period of review, Commerce changed the primary surrogate country from India (the primary surrogate country since the initial investigation in 1994) to Indonesia” but added that “Baoding does not challenge Commerce’s decision to change the primary surrogate country from India to Indonesia.” Baoding Mantong’s Br. 14. Commerce decided to retain Indonesia as its primary surrogate country in the Remand Redetermination, and again Baoding Mantong raises no objection to that decision.

During the review, GEO placed on the record, for use as surrogate values, import data for Indonesia published in the Global Trade Atlas (“GTA”). See GEO Specialty Chemicals’ Pre-Preliminary Results Comments 3-4 (Mar. 6, 2012), (Second Admin.R.Doc. No. 55). Commerce chose to use the Indonesian GTA import data to value Baoding’s raw 'material inputs, including the inputs of liquid chlorine, ammonia, formaldehyde, and steam coal. See Preliminary FOP Mem. 3, Attachment 1; Final I & D Mem. 3-18. As it had in the Final Results, Commerce mentioned in the Remand Redetermination its practice of choosing surrogate data that are product-specific, representative of a broad market average, publicly available, contemporaneous with the period of review, and exclusive of taxes and duties. Remand Redeterm. 12 (addressing surrogate value for liquid chlorine).

1. Surrogate Value for Liquid Chlorine

In the Final Results and again in the Remand Redetermination, Commerce chose, as the best available information for use in valuing Baoding Mantong’s liquid chlorine production input, import data for Indonesian harmonized tariff schedule (“HTS”) subheading 2801.10 (“Chlorine”), obtained from the GTA. From these import data, Commerce determined a surrogate value of $0.56 per kilogram. Baoding Mantong comments that “Commerce’s chosen surrogate value for liquid chlorine was more than five times greater than Baoding’s alternative surrogate value,” Baoding Mantong’s Comments 3, which was $0.11 per kilogram, a value Baoding Mantong obtained by calculating an average unit value from the sales of two producers of chlorine in India during the period of review. See Baoding Mantong’s Br. 23.

In its comments on the Remand Rede-termination, Baoding argues that the Indonesian GTA data Commerce used to calculate the surrogate value for liquid chlorine (and for the other three challenged surrogate values) were aberrational and inferior when compared to other data on the record, incorporating by reference the objections it raised in its Rule 56.2 brief and its reply brief. Baoding Mantong’s Comments 6-7 (citing Baoding Mantong’s Br. 11-34 and Pl.’s Reply Br. 3-12 (Mar. 10, 2014), ECF No. 51). Baoding Mantong submits that the Department’s chlorine surrogate value was “commercially and statistically insignificant.” Baoding Mantong’s Br. 20. Baoding Mantong pointed to record data showing that the two Indian chlorine producers it identified for its alternate surrogate value sold a combined volume of 86,497 metric tons of chlorine during the POR.

In the Remand Redetermination, Commerce stated that “[w]e have reevaluated the Indonesian GTA data for liquid chlorine by comparing it to that of the other five countries and continue to find that Indonesia’s average unit value (AUV) of 0.56 USD/kilogram falls within a range from 0.25 USD/kilogram to 12.54 USD/kilogram from the six countries,” referring to the six countries it considered economically comparable to China and significant producers of glycine, i.e., the Philippines, Indonesia, Ukraine, Thailand, Colombia, and South Africa. Remand Redeterm. 13. Commerce noted that the volume of the Indonesian imports exceeded 2000 metric tons and was “the highest volume of all potential surrogate countries,” which, Commerce added, “supports the conclusion that liquid chlorine was imported into Indonesia in commercial quantities during the period of review.” Id. at 13-14 (footnote omitted). Baoding Mantong opposed use of the chlorine import data from any of the individual six countries, maintaining that “[t]he import volumes of the other five potential surrogate countries were also too small upon which to base a reliable surrogate value.” Baoding Mantong’s Br. 23. In so doing, Baoding Mantong based its argument entirely on a contention that Commerce should have chosen the average unit value Baoding Mantong calculated from the two Indian chlorine producers instead of the value determined from the Indonesian import data.

The court is not convinced by Baoding Mantong’s argument. The import quantity upon which the $0.56 per kilogram surrogate value was based was relatively small in relation to the sales of the two Indian producers, but Baoding Mantong does not offer a standard, or record evidence, demonstrating that this quantity, 2,110 metric tons, was too commercially “insignificant” a quantity to serve as a surrogate value. Indonesia had the highest volume of any of the six countries shown in the GTA data for the six countries Commerce found to be economically comparable. Baoding’s Surrogate Value Submission at Attachment 2. Therefore, even were the court to presume, arguendo, that Commerce permissibly could have chosen the Indian price data over the Indonesian import data based on the much larger quantity from which they were derived, it also would conclude that the record data considered as a whole are insufficient to compel that choice. Commerce reasonably gave weight to the fact that the Indian data were from a country that it did not find economically comparable to China, a finding Baoding Mantong does not contest in commenting on the Remand Redetermination. See 19 U.S.C. § 1677b(c)(4) (requiring that Commerce, when valuing factors of production, “utilize, to the extent possible, the prices or costs of factors of production in one or more market economy countries that are ... at a level of economic development comparable to that of the nonmarket economy country”).

Favorable to Baoding Mantong’s argument is the fact that the Indonesian value of $0.56 per kilogram was considerably higher than the AUV for Ukraine, which was $0.25 per kilogram based on a comparable volume (2,000 metric tons), and for the Philippines ($0.31 per kilogram, based on a volume of 1,007 metric tons). Baoding’s Surrogate Value Submission at Attachment 2. The AUV for Colombia was comparable at $0.49 per kilogram, but it was based on a volume of only 42 metric tons. Id. The AUVs for South Africa ($12.54 per kilogram) and Thailand ($10.37 per kilogram) were much higher, but the quantities involved (0.8 metric tons and 0.6 metric tons, respectively) were very small; therefore, the data for these two countries can be described as aberrational. See id. Still, while the AUV for the imports in Indonesia was the highest for the countries with the largest, non-insignificant volumes, the court cannot conclude that Commerce was required to find on this record that the data for Indonesia (an AUV of $0.56 per kilogram on a quantity of 2,110 metric tons) were aberrational. It is possible that a wider set of data could have shown the Indonesian surrogate value to be aberrationally high; data used for these comparison purposes need not have been confined to the aforementioned six countries. Baoding Mantong did not submit such a wider set of data for the record during the review, leaving Commerce to consider the question of whether the surrogate price was aberrational, and to make its ultimate decision, from a limited record. For these reasons, the court sustains the Department’s surrogate value of $0.56 per kilogram for valuation of liquid chlorine.

2. Surrogate Value for Ammonia

Baoding Mantong included in its Rule 56.2 brief, and incorporated by reference in its comments on the Remand Re-determination, a claim that Commerce, based on a mistake by Baoding Mantong’s counsel, incorrectly found that Baoding Mantong’s ammonia production input was aqueous ammonia of the formula NH40H rather than anhydrous ammonia of the formula NH3. Baoding Mantong’s Br. 23-26. Baoding Mantong argues that substantial evidence does not support the Department’s finding that the input was aqueous ammonia and that Commerce should have valued the input as anhydrous ammonia instead. Id. Rejecting this argument during the review, Commerce used GTA value data for Indonesian HTS subheading 2814.20 to calculate a surrogate value of $4.06 per kilogram. Remand Redeterm. 17. Baoding Mantong argues for a surrogate value it calculates as $0.43 per kilogram, determined according to GTA value data for anhydrous ammonia, Indonesian HTS subheading 2811.41. Baoding Mantong’s Br. 23.

Commerce considered, and rejected, Baoding’s argument on the identity of the input in the Final Issues and Decision Memorandum. Final I & D Mem. 11-12. Commerce found that “[t]he record shows that Baoding Mantong used ammonia with the formula NH40H in its production of glycine for this review and, accordingly, we have based the valuation of the input on the available GTA Indonesian data for aqueous ammonia under HTS subheading 2814.20.” Id. at 12. In the Remand Redetermination, Commerce stated that “[t]he record indicates that in Baoding Mantong’s Section D response [i.e., the response to Section D of the Department’s questionnaire], it provided the formula for the ammonia it uses as NH4OH, which is the formula for aqueous ammonia.” Remand Redeterm. 16 (citing Baoding Mantong Section C and D Response D-5 (Aug. 2, 2011), (Admin.R.Doc. No. 41)). Commerce added that “Baoding Mantong offered no correction to this information at the start of verification, and the Department made no findings at verification to contradict Baoding Mantong’s reporting of aqueous ammonia as the liquid ammonia it used.” Id. at 14.

In its brief in support of its Rule 56.2 motion, Baoding Mantong stated as follows:

Counsel admits that in a fit of panic and confusion, he purposefully, but incorrectly, changed the provided chemical formula for the liquid ammonia inputs reported in Exhibit D5 of the Section D Response from NH3 (for anhydrous ammonia) to NH4OH (aqueous ammonia), mistakenly believing that “liquid ammonia” corresponded with “aqueous ammonia” (anhydrous ammonia in a water solution).

Baoding Mantong’s Br. 23-24. The brief continues, “[c]ounsel did not recognize this error until after the release of the revised preliminary results on July 27, 2013,” that “[c]ounsel requested that Commerce correct the error and value liquid ammonia inputs using the HTS item corresponding to anhydrous ammonia,” and that Commerce refused to do so. 11 Id. at 24.

As Commerce concluded, Exhibit D-5 of Baoding Mantong’s questionnaire response, filed on August 2, 2011, contains evidence that Baoding Mantong used aqueous ammonia rather than anhydrous ammonia. Specifically, Commerce found that in that exhibit Baoding Mantong listed the formula for its ammonia input as NH40H, which is the formula for aqueous ammonia, and in the same exhibit proposed valuing ammonia accordingly based on Indonesian HTS subheading 2814.20. Final I & D Mem. 11. Commerce found, further, that in November of 2011, Baoding Mantong again proposed valuing its ammonia input based on subheading 2814.20 in a surrogate value submission. Id. (citing Baoding’s Surrogate Country Comments at Attachment 3). Exhibit D-5 and the November 1, 2011 submission are probative evidence in support of the Department’s finding that aqueous ammonia was used, but this evidence is not conclusive due to the presence of contrary evidence on the record. Baoding Mantong argued that Exhibit D-5 also identified the input as liquid ammonia, and not as aqueous ammonia, “which by definition is ammonia in a water solution,” adding that the exhibit “reported a purity level of the liquid ammonia as greater than 99.8% and a molecular weight of 35.04, values that are consistent with anhydrous ammonia.” Baoding Mantong’s Br. 25. It also points out that Exhibits D-1 and D-2 to the Section D questionnaire response contains a “diagram of the production flowchart and the technical description of the glycine production process” that “refers to liquid ammonia, not aqueous ammonia.” 12 Id.

The molecular weight of 35.04 reported in Exhibit D-5 is contrary to Baoding Mantong’s argument, as this is given in technical sources as the molecular weight of aqueous ammonia, not anhydrous ammonia (17.03). See Table of Molecular Weights: A Companion Volume to the Merck Index, Ninth Edition 1 (Martha Windholz et al. eds., 1st ed. 1978); Dale L. Perry, Handbook of Inorganic Compounds 27 (2nd ed. 2016). In other words, Exhibit D-5 is internally inconsistent as to the reporting of the identity of the material. Nevertheless, Commerce, despite this internal inconsistency, treated Exhibit D-5 as a principal reason for its conclusion, along with the November 1, 2011 submission. In addition to the ambiguity in Exhibit D-5, the record contained other evidence supporting Baoding Mantong’s argument, detracting from the Department’s finding that aqueous ammonia was used. The other exhibits Baoding Mantong identified (D-1 and D-2) discuss liquid ammonia, not aqueous ammonia. The evidence detracting from the Department’s finding includes the Department’s own discussion in the Final Issues and Decision Memorandum of the input as it related to prior reviews:

In the previous reviews, we determined that liquid ammonia should be valued as anhydrous ammonia but only after finding that it was supported by the record; specifically, we found the respondent (which, in the case of the 2006/2007 and 2007/2008 reviews, was Baoding Mantong) in each review had identified the molecular formula of the ammonia used in their production of glycine as NH3—the formula for anhydrous ammonia.

Final I & D Mem. 11 (citations omitted). In summary, the court cannot conclude from the Department’s discussion and the record information that Commerce correctly made its decision based on substantial record evidence. The Department’s statement in the Remand Redetermination that “Baoding Mantong offered no correction to this information [reporting of the input as NH40H] at the start of verification, and the Department made no findings at verification to contradict Baoding Mantong’s reporting of aqueous ammonia as the liquid ammonia it used,” Remand Redeterm. 14, appears to treat Exhibit D-5 as unequivocally reporting the input as aqueous ammonia, which is not the case. Moreover, Baoding Mantong did not actually use the term “aqueous ammonia” in reporting its input in Exhibit D-5; instead it gave the formula as NH40H (incorrectly, according to counsel for Baoding Mantong) and proposed a surrogate value for aqueous ammonia accordingly. The court does not conclude that the finding that the production input was aqueous ammonia necessarily was incorrect as a factual matter, but in light of the deficiencies in the Department’s explanation, the court directs Commerce to review the relevant record evidence and reach a'well-reasoned and adequately explained finding as to what the input actually was.

Baoding Mantong also claimed that the surrogate value Commerce chose, $4.06 per kilogram, was not supported by substantial evidence due to the small volume of imports under the Indonesian tariff provision for aqueous ammonia. Baoding Mantong’s Br. 26. Baoding Mantong argued that “Indonesian imports of aqueous ammonia were less than 82 metric tons during the period of review and are simply too small to be the basis of an accurate and reliable surrogate value” and that “[i]n comparison, Baoding Mantong alone purchased over 660 metric tons of liquid ammonia during the period of review.” Id. (citations omitted). Referring to the other five countries Commerce identified, Baoding Mantong argued, further, that “[t]hree of the countries (Ukraine, Colombia and South Africa) had imports [of aqueous ammonia] even smaller than even Indonesia’s miniscule import volume,” adding that “Thailand’s import volume was only 306 MT, which is also significantly smaller than Baoding’s consumption.” Id. at 26-27. Baoding Man-tong argues that if the court concludes that Commerce permissibly treated the input as aqueous ammonia, then the input should be valued according to aqueous ammonia import data for the Philippines, which had a total import volume of 52,304 metric tons. Id. at 27. Baoding Mantong calculated a proposed surrogate value of $0.28 per kilogram from eligible Philippine imports. As an alternative, it proposed a surrogate value of $0.39 per kilogram based on ammonia sales as reported in the financial statement of the Indian fertilizer producer. Id.

In the Remand Redetermination, Commerce mentioned the quantity of 82 metric tons for Indonesian imports of aqueous ammonia but did not address the question of whether this quantity is commercially significant or why the Philippine data, which are based on a much larger quantity, would- not be superior in that respect. Remand Redeterm. 17. Instead of addressing the question of quantity, Commerce discussed the question of whether the value was aberrational, concluding that it was not because “Indonesia’s AUV is 4.06 USD/kilogram, which falls within the range of eeonomic[ally] comparable countries of 0.28 to 6.94 USD/kilogram.” Id. Commerce does not address the point Baoding Mantong raised concerning the relatively low quantities upon which all of the GTA data were based other than the data from the Philippines. Commerce concluded that “[t]he GTA data from Indonesia is representative of a broad market-average of liquid ammonia that is specific to this product HTS code,” id. at 16, but does not explain in the Remand Redetermination why the Philippine GTA data, which is based on 52,304 metric tons as compared to 82 metric tons for Indonesia, would not reflect a much broader market average. Although Commerce prefers using data from a single surrogate country, see 19 C.F.R. § 351.408(c)(2), choosing the Indonesian data over the Philippine data, which were based on a substantially larger quantity, raises a question as to whether the Indonesian data were the “best available information” as required by 19 U.S.C. § 1677b(c)(1).

In summary, the court finds inadequate the Department’s explanation for its finding that Baoding Mantong’s input was aqueous ammonia and its explanation for its choice of the Indonesian GTA import data over the Philippine GTA import data, even though the latter would appear to represent a much broader market average based on the relative quantities. Commerce, therefore, must reconsider its ammonia surrogate value, supporting its choice with findings grounded in record evidence and an adequate explanation.

3. Surrogate Value for Formaldehyde

In the Final Results and again in the Remand Redetermination, Commerce valued Baoding Mantong’s formaldehyde input using Indonesian GTA data for HTS subheading 2912.11 to arrive at a surrogate value of $0.49 per kilogram. Remand Redeterm. 17-18. Commerce found the quantity of imports in those data, 357,277 kilograms, or just over 357 metric tons, to be sufficient for this purpose, noting that this was the fourth largest quantity of the six countries, “surpassing South Africa and Colombia with 823 kilograms and three metric tons, respectively.” Id. at 18 (footnote omitted). Commerce considered its surrogate value not to be aberrational, falling with the range of the six countries, which was $0.27 to $23.54 per kilogram. Id.

In contesting the surrogate value for formaldehyde, Baoding Mantong argued that the quantity for the Indonesian imports was not based on a commercially and statistically significant quantity. Baoding Mantong’s Br. 28-29 (citing Shanghai Foreign Trade Enterprises Co. v. United States, 28 C.I.T. 480, 318 F.Supp.2d 1339 (2004)). Baoding Mantong argued, further, that although South Africa and Colombia had even smaller import volumes and Ukraine and Thailand had volumes that it characterized as “insignificant,” the volume of formaldehyde imports into the Philippines, 6,025 metric tons, “was 17 times larger than the Indonesian import volume” and “several times larger than all of the other five countries combined.” Id. at 29. Baoding Mantong submitted that the average unit value shown by the Philippine GTA import data, $0.27 per kilogram, should be used to value the formaldehyde factor of production. Id.

The court agrees with Baoding Man-tong’s point that the data on Colombia and South Africa reveal aberrantly low quantities. See Baoding’s Surrogate Value Submission at Attachment 3. The data for South Africa, which showed 828 kilograms (approximately 0.8 metric tons) of formaldehyde at $23.54 per kilogram, and the data for Colombia, which showed 3 metric tons at an AUV of $5.64 per kilogram, contain insignificant quantities when compared to the remainder of the record evidence. The AUVs for Colombia and South Africa are also aberrational when compared to the AUVs for the other four countries, which ranged from $0.27 (the Philippines) to $0.90 (Ukraine). Compared to these values under $1.00 per kilogram, the disproportionately high values of $5.64 per kilogram (Colombia) and $23.54 per kilogram (South Africa) are an additional reason why the data for South Africa and Colombia must be rejected as offering nothing meaningful to the analysis of what constitutes the best available information. This leaves only four countries (the Philippines, Thailand, Ukraine, and Indonesia) for which meaningful value data existed on the record. Therefore, the court views the Department’s conclusion that the chosen data for Indonesia are the fourth largest of the six quantities as supported by record evidence but also misleading: the Indone sian data showed the smallest quantity of the four data sets that actually merited consideration. While Commerce has discretion to give some weight to the effect of its regulation on the valuation of the factors of production, which provides that “[e]x-cept for labor ... the Secretary normally will value all factors in a single surrogate country,” 19 C.F.R. § 351.408(c)(2), the court questions whether the Indonesian data, with the lowest quantities of the four, are the best available information when the preference for a single surrogate country is the only factor in favor of using these data as opposed to other meaningful data on the record. Moreover, because the court holds below that Commerce erred in finding that the Indonesian import data for valuing steam coal were the best available information, Commerce will be required to use data other than Indonesian data when it responds to the court in its redetermination. Departure from the single surrogate country practice thus will be required in any event, which reduces, if not defeats, the relevance of the preference reflected in 19 C.F.R. § 351.408(c)(2).

For the reasons stated above, the court concludes that Commerce must reconsider its finding that the Indonesian import data were the best information with which to value formaldehyde. Although the court is not requiring that Commerce necessarily use the Philippine data (and Commerce could, for example, choose to reopen the record), Commerce must consider that alternative and explain why the Philippine data would have been, or would not have been, a better source of information than the Indonesian data for valuing formaldehyde. The explanation must consider the record data as a whole, including the data showing that the quantity for the Philippine data was substantially higher than those for the other countries (and between seven and eight times higher than the next largest quantity, which was the quantity for Thailand). The huge disparity between the Philippine quantity and the quantities for the other three countries that merited consideration must be considered in light of the Department’s stated preference for using data that represent a broad market average.

4. Surrogate Value for Steam Coal

Commerce determined a surrogate value of $0.66 per kilogram for steam coal, using the GTA import data for Indonesia under HTS subheading 2701.19 (“Coal, Other Than Anthracite or Bituminous ... ”). Commerce found these data, which were based on an import volume of 604 metric tons, to be “not aberrational” and therefore “usable and reliable.” Remand Redeterm. 20. The record does not contain substantial evidence to support the finding that the Indonesian GTA data were the best available information with which to value the steam coal input. The court reaches this conclusion because the value of $0.66 per kilogram is disproportionately high relative to the other value information on the record. The other GTA data on the record, which pertain to the Philippines, South Africa, Thailand, and Ukraine, demonstrate this point.

The average unit value shown by the data for Thailand and the Philippines was $0.05 per kilogram. See Baoding’s Surrogate Value Submission at Attachment 4. The AUV for Ukraine was $0.20. Id. The highest AUV among the four countries other than Indonesia was $0.21 per kilogram, for South Africa. Id.

The disparity in value alone is sufficient to call the Department’s analysis into question. The relatively small quantity on which the surrogate value was based corroborates this point. The import quantity for the Indonesian GTA import data, which was 603,684 kilograms, i.e., less than 604 metric tons, might or might not be considered commercially insignificant for a low value bulk product such as coal but in any event was substantially smaller than the quantities for the other data sources. While Commerce normally seeks a surrogate value reflecting a broad market average, the quantity of the Indonesian imports was far less than even Baoding Mantong’s own consumption of 1,037 metric tons. See Baoding Mantong’s Br. 30. Commerce itself acknowledged that “Indonesia has the lowest import volume (ie., 604 metric tons) and the highest unit value (ie., 0.66 USD/kilogram)” compared with the “other five [sic] economically comparable countries.” Remand Redeterm, at 19. By comparison, the two countries with an AUV of $0.05 per kilogram had far larger quantities, 954,648 metric tons for Thailand and 15,919,558 metric tons for the Philippines. Imports into Ukraine, which totaled 638,189 metric tons, had an average unit value of $0.20, and South African imports, the second smallest of the six countries, totaled 234,389 metric tons (with an average unit value of $0.21 per kilogram). See Baoding’s Surrogate Value Submission at Attachment 4. Commerce reasoned that “it appears that both Thailand and Philippines values are extremely low compared to the other three countries’ AUVs.” Remanid Redeterm. 20. This reasoning does not support the choice of the Indonesian GTA import data because the value obtained from those data ($0.66 per kilogram) was more than three times as high as the next highest value ($0.21 per kilogram, for South Africa).

Baoding Mantong argued that Commerce should have valued the steam coal input using data it placed on the record pertaining to a large Indian coal producer, Coal India Limited (“CIL”), which showed an average price of $0.03 per kilogram (based on the average prices of 19 of the company’s subsidiaries). Baoding Man-tong’s Br. 32. Baoding Mantong contends that this choice should be preferred because CIL’s coal was Grade B steam coal, which Baoding Mantong asserts is consistent with the steam coal it used to make the subject merchandise (useful heat value less than 6000 Kcal/Kg). Id. Baoding pointed out that Commerce used CIL price data to value its steam coal in two previous reviews due to the superior specificity of the CIL data to the input. Id. at 33.

The court does not hold that Commerce necessarily was required in this review to use CIL data, as it did in past reviews; those data might be advantageous in some respects but nevertheless are based on only one company and its subsidiaries. Rather, in reconsidering the matter, Commerce must ensure that its choice of what is the best available information for valuing the steam coal input is supported by substantial evidence, a standard the GTA import data for Indonesia do not meet.

III. Conclusion and Order

The court concludes that certain decisions by Commerce as set forth in the Final Results of Redeterm. Pursuant to Court Remand (Mar. 30, 2016), ECF No. 73 (“Remand Redetermination”) are not in accordance with law for the reasons set forth above. Specifically, Commerce must reconsider and redetermine, as appropriate, in conformance with this Opinion and Order, the surrogate values it applied to Baoding Mantong’s factors of production for ammonia, formaldehyde, and steam coal.

Therefore, upon consideration of all papers and proceedings had herein, and upon due deliberation, it is hereby

ORDERED that the Remand Redeter-mination be, and hereby is, remanded for reconsideration and redetermination in accordance with this Opinion and Order; it is further

ORDERED that Commerce shall issue, within ninety (90) days of the date of this Opinion and Order, a new determination upon remand (“Second Remand Redeter-mination”) that conforms to this Opinion and Order; it is further

ORDERED that Baoding Mantong and GEO may file comments on the Second Remand Redetermination within thirty (30) days from the date on which the Second Remand Redetermination is filed with the court; and it is further

ORDERED that defendant may file a response within fifteen (15) days from the date on which the last of such comments is filed with the court.

1 . As stated in the Order, glycine "is a free-flowing crystalline material, like salt or sugar” that "is produced at varying levels of purity and is used as a sweetener/taste enhancer, a buffering agent, reabsorbable amino acid, chemical intermediate, and a metal complexing agent.” Antidumping Duty Order: Glycine From the People’s Republic of China, 60 Fed. Reg. 16,116, 16,116 (Int'l Trade Admin. Mar. 29, 1995).

2 . In response to the preliminary results of the review, GEO’s case brief alleged that Commerce made a currency conversion error by extracting Global Trade Adas ("GTA”) import data, which Commerce used in surrogate value calculations, in Indian rupees rather than Indonesian rupiahs. See GEO Specialty Chem.’s Case Br. 13 (May 11, 2012), (Admin.R.Doc. No. 75). Commerce determined that the GTA data had been reported in U.S. dollars and, therefore, that it should not have performed a currency conversion. Revisions to Certain Surrogate Valuations and the Preliminary Margin-Calculation Program for Baoding Fine Chemistry, Co., Ltd., Mem. from Edythe Artman to File 2 (June 27, 2012), (Admin.R.Doc. No. 84) ECF No. 50-5; Issues and Decision Mem. for the Final Results in the Administrative Review of Glycine from the People’s Republic of China, A-570-836 ARP 10-11, at 2 (Int’l Trade Admin. Oct. 9, 2012), (Admin.R.Doc. No. 127), available at http:// enforcement.trade.gov/frn/summary/PRC/2012-25595-1.pdf (last visited Mar. 1, 2017) (“Final I & D Mem.").

3 . The minor change from the Revised Preliminary Results to the Final Results concerned Baoding Mantong’s international freight expenses on constructed export sales. Glycine from the People's Republic of China: Final Results of Antidumping Duty Administrative Review, 77 Fed. Reg. 64,100, 64,101 (Int’l Trade Admin. Oct. 18, 2012) (“Final Results"). After the Preliminary Results, Commerce asked Baoding Mantong to provide additional information concerning those expenses, and when the company did not respond within the provided deadline, Commerce applied surrogate freight expenses to construct export price sales for which freight services may have been provided by a non-market economy carrier. Id.

4 . Unless otherwise indicated, all citations to the United States Code herein are to the 2012 edition and all citations to the Code of Federal Regulations herein are to the 2015 edition.

5 . A "nonmarket economy country” is defined in 19 U.S.C. § 1677(18)(A) as "any foreign country that the administering authority 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.”

6 . The regulation provides in relevant part that:

(c) Valuation of Factors of Production. For purposes of valuing the factors of production, general expenses, profit, and the cost of containers, coverings, and other expenses (referred to collectively as "factors”) under section 773(c) of the Act the following rules will apply:

[[Image here]]

(4) Manufacturing overhead, general expenses, and profit. For manufacturing overhead, general expenses, and profit, the Secretary normally will use non-proprietary information gathered from producers of identical or comparable merchandise in the surrogate country.

19 C.F.R. § 351.408(c)(4). The regulatory history of the section does not expand upon the meaning of the term “comparable.” See Antidumping Duties; Countervailing Duties, 62 Fed. Reg. 27,296, 27,367-68 (Int’l Trade Admin. May 19, 1997).

7 . During the review, GEO described these companies as “in the pharmaceutical sector with products either identical or comparable to the subject merchandise.” GEO Comments on Selection of Surrogate Country for Valuing Factors of Production and Surrogate Value Data for Valuing Baoding's Factors of Production 6 (Nov. 1, 2011), (Second Admin.R.Doc. No. 16). GEO placed financial statements for five Indonesian pharmaceutical companies on the record, of which Commerce chose to use those of PT Darya-Varia Laboratoria Tbk, PT Pyridam Farma Tbk, and PT Kalbe Farma Tbk. See Factors of Production for the Preliminary Results 4-5 (March 30, 2012), (Second Admin.R.Doc. No. 58).

8 . Baoding Submission of Surrogate Value Information and Comment 4-5 (July 16, 2012), (Second Admin.R.Doc. No. 99).

9 . In a sunset review of the order on glycine from China, the U.S. International Trade Commission stated that:

USP-grade glycine is used in food, cosmetic or some medical uses, and accounts for approximately 89 percent of the U.S. glycine market. Pharmaceutical-grade glycine is a subset of the USP grade, but must meet additional specifications and testing. Pharmaceutical-grade glycine has the strictest purity standards and is used in intravenous injections.

GEO Rebuttal to Baoding's July 16, 2012 Arguments and Surrogate Value Information 4 n.4 (July 23, 2012), (Second Admin.R.Doc. No. 122) (citing Glycine from China at II—1, Inv. No. 731-TA-718, USITC Pub. 4255 (Aug. 2011)).

10 . When using import statistics to calculate surrogate values, Commerce typically disregards value data on imports that it has "reason to believe or suspect may be subsidized, such as those relating to imports from India, South Korea and Thailand, as well as countries deemed to be non-market economies.” Final I & D Mem. at 15 n.5. Here, Commerce adjusted the unit quantities and average unit values of the import data for the surrogate countries selected in this review to negate the influence of imports from nonmarket economy countries and countries that are suspected of subsidizing their exports.

11 . The actual date of release of the revised preliminary results of the review was June 27, 2012.

12 . Baoding Mantong also argued that ”[h]av-ing a different chemical composition, aqueous ammonia cannot be used to produce glycine.” Mot. for J. on the Agency R. and Mem. in Supp. 27 (July 22, 2013), ECF No. 30. However, Baoding Mantong did not cite record evidence to support this contention.