*1317Plaintiff Habas Sinai ve Tibbi Gazlar Istihsal Endustrisi A.S. ("Habas") and Consolidated Plaintiff Icdas Celik Enerji Tersane ve Ulasim Sanayi A.S. ("Icdas") (together, "Plaintiffs") challenge the U.S. Department of Commerce's ("Commerce" or the "agency") final affirmative determination in the sales at less than fair value investigation of steel concrete reinforcing bar ("rebar") from the Republic of Turkey ("Turkey").1 See Steel Concrete Reinforcing Bar From the Republic of Turkey, 82 Fed. Reg. 23,192 (Dep't Commerce May 22, 2017) (final determination of sales at less than fair value) ("Final Determination"), ECF No. 17-5, as amended by Steel Concrete Reinforcing Bar From the Republic of Turkey and Japan, 82 Fed. Reg. 32,532 (Dep't Commerce July 14, 2017) (am. final affirmative antidumping duty determination for the Republic of Turkey and antidumping duty orders) ("Am. Final Determination "), ECF No. 17-7, and accompanying Issues and Decision Mem., A-489-829 (May 15, 2017) ("I & D Mem."), ECF No. 17-6.
Plaintiffs challenge several aspects of the Final Determination.2 See Confidential Pl.'s Rule 56.2 Mot. for J. on the Agency R. and Mem. in Supp. of Mot. of Pl. Habas Sinai ve Tibbi Gazlar Istihsal Endustrisi A.S., for J. on the Agency R. Pursuant to Rule 56.2 ("Habas's Mem."), ECF No. 22; Confidential Mot. for J. on the Agency R., ECF No. 24, and Confidential Pl. Icdas Celik Enerji Tersane ve Ulasim Sanayi A.S.'s Mem. of Law in Supp. of Mot. for J. on the Agency R. Pursuant to Rule 56.2 ("Icdas's Mem."), ECF No. 29. Habas and Icdas challenge Commerce's calculation of their respective duty drawback adjustments and refusal to use a quarterly cost-averaging methodology in the determination of normal value. See Habas's Mem. at 4-25; Icdas's Mem. at 9-31. Habas challenges Commerce's selection of the invoice date as the date of sale for its U.S. sales and rejection of its zero-interest short-term loans to calculate imputed credit expenses. Habas's Mem. at 25-39. Icdas challenges Commerce's use of partial adverse facts available in relation to certain sales for which it could not provide manufacturer codes. Icdas's Mem. at 31-36.3 Defendant United States ("Defendant" or the "Government") and Defendant-Intervenor Rebar Trade Action Coalition ("Defendant-Intervenor" or "RTAC") urge the court to sustain Commerce's Final Determination *1318in full. See Confidential Def.'s Resp. to Pls.' Mots. for J. Upon the Agency R. ("Def.'s Resp."), ECF No. 37; Confidential Resp. Br. of Def.-Int. Rebar Trade Action Coalition ("Def.-Int.'s Resp."), ECF No. 40.
For the reasons discussed herein, the court remands Commerce's calculation of Plaintiffs' duty drawback adjustment and application of partial adverse facts available to Icdas. The court sustains the Final Determination in all other respects.
BACKGROUND
On October 18, 2016, Commerce initiated this antidumping duty investigation of rebar from Turkey in response to a petition filed by RTAC and the domestic rebar producers that constitute RTAC's individual members. See Steel Concrete Reinforcing Bar From Japan, Taiwan and the Republic of Turkey, 81 Fed. Reg. 71,697 (Dep't Commerce Oct. 18, 2016) (initiation of less-than-fair-value investigations), PR 28, PJA Tab 1. Commerce selected Habas and Icdas as mandatory respondents in the investigation. I & D Mem. at 1. The period of investigation ("POI") ran from July 1, 2015 to June 30, 2016. Final Determination, 82 Fed. Reg. at 23,192.
On March 7, 2017, Commerce issued its preliminary determination. See Steel Concrete Reinforcing Bar From the Republic of Turkey, 82 Fed. Reg. 12,791 (Dep't Commerce Mar. 7, 2017)("Prelim. Determination"), and accompanying Prelim. Decision Mem., A-489-829 (Feb. 28, 2017) ("Prelim. Mem."), PR 161, PJA Tab 30. Commerce preliminarily calculated a weighted-average dumping margin of 5.29 percent for Habas and 7.07 percent for Icdas. Prelim. Determination, 82 Fed. Reg. at 12,792.
On May 22, 2017, Commerce issued the Final Determination. 82 Fed. Reg. at 23,192. Commerce issued an amended final determination on July 14, 2017. See Am. Final Determination, 82 Fed. Reg. at 32,532. Therein, Commerce calculated a weighted-average dumping margin of 5.39 percent for Habas and 9.06 percent for Icdas. Id., 82 Fed. Reg. at 32,533.
On July 31, 2017, Habas timely commenced this action. See Summons, ECF No. 1. On August 11, 2017, Icdas timely commenced a separate action also challenging the Final Determination. See Summons, ECF No. 1 (Court No. 17-00218). On October 5, 2017, the court consolidated the two actions under lead Court No. 17-00204. See Order (Oct. 5, 2017), ECF No. 15. The court heard oral argument on November 29, 2018. See Docket Entry, ECF No. 58.4
JURISDICTION AND STANDARD OF REVIEW
The court has jurisdiction pursuant to § 516A(a)(2)(B)(i) of the Tariff Act of 1930, as amended, 19 U.S.C. § 1516a(a)(2)(B)(i) (2012),5 and 28 U.S.C. § 1581(c) (2012). The court will uphold an agency determination that is supported by substantial evidence and otherwise in accordance with law.
*131919 U.S.C. § 1516a(b)(1)(B)(i). "Substantial evidence is 'such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.' " Huaiyin Foreign Trade Corp. (30) v. United States, 322 F.3d 1369, 1374 (Fed. Cir. 2003) (quoting Consol. Edison Co. v. NLRB., 305 U.S. 197, 229, 59 S.Ct. 206, 83 L.Ed. 126 (1938) ). Substantial evidence "requires more than a mere scintilla," but "less than the weight of the evidence." Nucor Corp. v. United States, 34 CIT 70, 72, 675 F.Supp.2d 1340, 1345 (2010). The court may not "reweigh the evidence or ... reconsider questions of fact anew." Downhole Pipe & Equip., L.P. v. United States, 776 F.3d 1369, 1377 (Fed. Cir. 2015).
DISCUSSION
I. Duty Drawback
A. Background
To determine whether the subject merchandise is being sold at less than fair value, Commerce compares the export price ("EP") or constructed export price ("CEP")6 of the subject merchandise to its normal value ("NV"). See generally 19 U.S.C. § 1673 et seq. Generally, an antidumping duty is the amount by which the normal value of a product-generally, its price in the exporting country-exceeds export price, as adjusted. See id. § 1673. One of the adjustments Commerce makes to export price pursuant to 19 U.S.C. § 1677a(c) is known as the "duty drawback adjustment." Specifically, Commerce will increase export price by "the amount of any import duties imposed by the country of exportation which have been rebated, or which have not been collected, by reason of the exportation of the subject merchandise to the United States." Id. § 1677a(c)(1)(B).
This statutory duty drawback adjustment is intended to prevent the dumping margin from being increased by import taxes that are imposed on raw materials used to produce subject merchandise, but which are rebated or exempted from payment when the subject merchandise is exported to the United States. See Saha Thai Steel Pipe (Public) Co. Ltd. v. United States, 635 F.3d 1335, 1338 (Fed. Cir. 2011) ; Wheatland Tube Co. v. United States, 30 CIT 42, 60, 414 F.Supp.2d 1271, 1286 (2006), rev'd on other grounds, 495 F.3d 1355 (Fed. Cir. 2007). The adjustment accounts for the fact that producers are subject to the import duty when merchandise is sold in the home market, "which increases home market sales prices and thereby increases [normal value]." Saha Thai, 635 F.3d at 1338. The statute increases constructed export price "to the level it likely would be absent the duty drawback" to prevent the absence of import duties from generating or increasing any dumping margin. Id.
Commerce has developed a two-prong test to determine whether a respondent is entitled to a duty drawback adjustment: first, "that the exemption [from import duties] is linked to the exportation of subject merchandise"; and second, "that there [were] sufficient imports of the raw material to account for the duty drawback on the export of subject merchandise." I & D Mem. at 12; see also Saha Thai, 635 F.3d at 1340 (affirming the lawfulness of Commerce's two-prong test).
Commerce determined that Plaintiffs had demonstrated their entitlement to the duty drawback adjustment. I & D Mem. at 12.7 At issue, however, is Commerce's method of calculating the adjustment.
*1320Until recently, Commerce calculated the duty drawback adjustment to U.S. price (referred to as the sales-side adjustment) by dividing rebated or exempted duties by total exports and adding the resultant per unit duty burden to the export price. See Rebar Trade Action Coalition v. United States ("RTAC I "), Slip Op. 15-130, 2015 WL 7573326, at *4 (CIT Nov. 23, 2015) (granting Commerce's request for a voluntary remand to reconsider the sales-side adjustment methodology as set forth in the Issues and Decision Mem. for the Final Negative Determination in the Less than Fair Value Investigation of Steel Concrete Reinforcing Bar from Turkey, A-489-818 (Sept. 8, 2014) ("Rebar from Turkey Mem.") ).
When producers participate in a duty exemption program, Commerce also makes a corresponding upward adjustment to the cost of production ("COP") and constructed value ("CV") (referred to as the cost-side adjustment)8 to account for the cost of the unpaid import duties for which the producer remains liable until the merchandise containing the dutiable input(s) is exported and the exemption program requirements are satisfied. See Saha Thai, 635 F.3d at 1341-44. In affirming Commerce's inclusion of implied duty costs in its calculations, the Saha Thai court reasoned that the purpose of the statutory increase to export price "is to account for the fact that the import duty costs are reflected in ... home market sales prices[ ] but not ... sales prices in the United States[ ]." Id. at 1342. Thus, "[i]t would be illogical to increase EP to account for import duties that are purportedly reflected in NV, while simultaneously calculating NV based on a COP and CV that do not reflect those import duties." Id. Accordingly, "[u]nder the 'matching principle,' EP, COP, and CV should be increased together, or not at all."9 Id. at 1342-43.10
In 2016, on remand pursuant to RTAC I , Commerce modified its sales-side adjustment by allocating exempted duties over total production rather than exports. See Rebar Trade Action Coalition v. United States ("RTAC II "), Slip Op. 16-88, 2016 WL 5122639, at *3 (CIT Sept. 21, 2016) ; Final Results of Redetermination Pursuant to Court Remand, Consol. Ct. No. 14-00268 (Apr. 7, 2016), available at http://ia.ita.doc.gov/remands/15-130.pdf (last visited Dec. 19, 2018) ("Rebar from Turkey Remand Mem."). Commerce developed this methodology in response to arguments by domestic producers regarding alleged distortions in the margin calculations that may arise when the respondent uses fungible inputs both from foreign sources, which incur import duties, and domestic sources, which do not. See RTAC II, 2016 WL 5122639, at *3-4. Commerce claimed that adhering to its prior methodology generated "distortions" in the margin calculations because the larger denominator *1321on the cost-side resulted in a smaller adjustment to normal value than U.S. price. Id. at *3 (citing Rebar from Turkey Remand Mem. at 16). Thus, according to Commerce, equalizing the denominators used in each adjustment "ensure[d] that the amount added to both sides of the comparison of EP or CEP with NV is equitable, i.e., duty neutral[,] meeting the purpose of the adjustment as expressed in Saha Thai ." Id. at *4, (citing Rebar from Turkey Remand Mem. at 18).
In subsequent administrative proceedings involving respondents that source inputs from foreign and domestic suppliers, including Plaintiffs here,11 Commerce has applied its modified sales-side adjustment. See I & D Mem. at 12; Final Results of Redetermination Pursuant to Remand, Consol. Ct. No. 16-00218 (July 7, 2018) at 11-12, ECF No. 106; Issues and Decision Mem. for the Final Results of Antidumping Duty Admin. Review: Welded Carbon Steel Standard Pipe and Tube Products from Turkey; 2014-2015, A-489-501 (Dec. 12, 2016) at 5-6, available at https://enforcement.trade.gov/frn/summary/turkey/2016-30541-1.pdf (last visited Dec. 19, 2018); Issues and Decision Mem. for the Final Determination of the Antidumping Duty Investigation of Certain Corrosion-Resistant Steel Products from India, A-533-863 (May 24, 2016) at 7-11, available at https://enforcement.trade.gov/frn/summary/india/2016-12986-1.pdf (last visited Dec. 19, 2018). In doing so here, Commerce reiterated the need for an "equitable, i.e., duty neutral" comparison of export price with normal value to maintain consistency with "the purpose of the adjustment as affirmed in Saha Thai ." I & D Mem. at 12 (citing Saha Thai, 635 F.3d at 1344 ).12
B. Parties' Contentions
Plaintiffs contend that Commerce's modified sales-side adjustment is unlawful because it ignores the statutory linkage between foregone duties and exported subject merchandise and reduces the full upward adjustment to which they are entitled. Habas's Mem. at 9-15; Icdas's Mem. at 9-17; see also Icdas's Mem. at 17-18 (asserting that Commerce's methodology impermissibly attributes duty drawback to domestic sales, which do not qualify for drawback under the Turkish duty drawback scheme). Plaintiffs further contend that Commerce's reliance on Saha Thai to support the modified sales-side adjustment as ensuring a "duty neutral" approach is misplaced. Habas's Mem. at 16; Icdas's Mem. at 21-22. Icdas further contends that Commerce's methodology requires a rulemaking procedure pursuant to the Administrative Procedure Act ("APA"), 5 U.S.C. § 553. Icdas's Mem. at 24-25.
The Government contends that Commerce's calculation of the duty drawback adjustment represents a permissible construction of the statute, which is silent on the issue of allocation. Def.'s Resp. at 15. According to the Government, "[h]ad Congress intended to limit Commerce's discretion in performing the EP/CEP duty drawback calculation, ... the statute would state that for each unit of subject merchandise exported, the EP/CEP shall be increased by the amount of duty rebated or not collected on that unit." Id. at 16.
*1322While recognizing that Saha Thai "does not address allocation," the Government contends that the U.S. Court of Appeals for the Federal Circuit ("Federal Circuit") "endorsed the concept of a 'matching principle,' which would ensure [duty] neutrality by requiring equal adjustments to both the NV and EP/CEP sides of the equation." Id. at 17 (citing Saha Thai, 635 F.3d at 1338, 1342-43 ). The Government further contends that Commerce need not conduct a rule-making procedure pursuant to the APA when it changes it practice. See Def.'s Resp. at 22-23.
RTAC adopts Defendant's arguments, see Def.-Int.'s Resp. at 2, and further contends that Commerce's methodology properly accounts for distortions that may arise when a respondent uses a mix of domestic and imported inputs or "otherwise manages its imports and exports such as to effectively pay no import duties regardless of the market for which its goods are destined," Def.-Int.'s Resp. at 6; see also id. at 6-10.
C. Commerce's Methodology is Remanded
The Government relies on the purported statutory silence regarding the way Commerce must calculate the duty drawback adjustment to support Commerce's allocation of exempted duties over total production. The court's review of Commerce's interpretation and implementation of a statutory scheme is guided by Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). See Apex Frozen Foods Private Ltd. v. United States, 862 F.3d 1322, 1329 (Fed. Cir. 2017). First, the court must determine "whether Congress has directly spoken to the precise question at issue." Id. (quoting Chevron, 467 U.S. at 842, 104 S.Ct. 2778 ). If Congress's intent is clear, "that is the end of the matter," and the court "must give effect to the unambiguously expressed intent of Congress." Id. (quoting Chevron, 467 U.S. at 842-43, 104 S.Ct. 2778 ). Only "if the statute is silent or ambiguous," must the court determine whether the agency's action "is based on a permissible construction of the statute." Id. (quoting Chevron, 467 U.S. at 843, 104 S.Ct. 2778 ).
Several members of the court, including the undersigned, have previously held that Commerce's allocation of foregone duties over total production is inconsistent with the clear statutory linkage between those duties and exported merchandise. See Eregli Demir ve Celik Fabrikalari T.A.S v. United States ("Erdemir II "), 42 CIT ----, Slip Op. 18-180 at 14-15, 357 F.Supp.3d 1325, 2018 WL 7050292 (CIT Dec. 27, 2018) ; Tosçelik Profil ve Sac Endustrisi A.S. v. United States, 42 CIT ----, ----, 321 F.Supp.3d 1270, 1275-78 (2018) ; Uttam Galva Steels Ltd. v. United States, 42 CIT ----, ----, 311 F.Supp.3d 1345, 1355 (2018) ; RTAC II, 2016 WL 5122639 at *4. Commerce offers nothing new meriting a different outcome here.13
Section 1677a(c)(1)(B) requires Commerce to increase EP/CEP by "the amount of any import duties imposed by the country of exportation which have been rebated, or which have not been collected, by reason of the exportation of the subject merchandise to the United States."
*132319 U.S.C. § 1677a(c)(1)(B) (emphasis added). Congress, thus, clearly intended the adjustment to capture the amount of duties Plaintiffs would have paid on their export sales but for the exportation of that merchandise. Allocating Plaintiffs' exempted duties over total production is contrary to the plain language of section 1677a(c)(1)(B)"because it attributes some of the drawback to domestic sales, which do not earn drawback, and fails to adjust export price by the amount of the import duties exempted by reason of exportation." Erdemir II, Slip Op. 18-180 at 14-15, 357 F.Supp.3d at 1333 ; see also Tosçelik, 321 F.Supp.3d at 1278. Thus, instead of calculating the amount of the adjustment based on duties foregone solely in relation to the exported merchandise eligible for drawback, as the statute requires, Commerce has calculated a lesser amount that is based on the distribution of some of the exempted duties to domestic sales, which is contrary to the statute's plain language. See Erdemir II, Slip Op. 18-180 at 14-15, 357 F.Supp.3d at 1333-34.
The Government's appeal to agency discretion pursuant to Chevron prong two also fails. See id. at 15; Def.'s Resp. at 15. Even if the statute was ambiguous for lack of an explicit methodology, Commerce must "exercise [ ] its gap-filling authority" in a "reasonable" manner. See Apex Frozen Foods, 862 F.3d at 1330 (citing Chevron, 467 U.S. at 843-44, 104 S.Ct. 2778 ). Commerce's exercise of any discretionary authority it has in this regard was unreasonable because it substantively departed from the guidance Congress did provide by decoupling the amount of the adjustment from duties forgiven solely on exported merchandise. See Ningbo Dafa Chem. Fiber Co., Ltd. v. United States, 580 F.3d 1247, 1253 (Fed. Cir. 2009) (an agency's statutory interpretation is unreasonable when it is "manifestly contrary" to the statutory terms) (citation omitted).14
Commerce's-and, by extension, the Government's-reliance on Saha Thai is also misplaced. In Saha Thai, the Federal Circuit approved Commerce's decision to utilize the cost-side adjustment in conjunction with its original sales-side adjustment to ensure that normal value and U.S. price are compared on a mutually-duty-inclusive basis. See 635 F.3d at 1342 (finding that Commerce "reasonably decided" to accompany an increase to EP with a "corresponding increase to COP and CV" because "[i]t would be illogical to increase EP to account for import duties that are purportedly reflected in NV, while simultaneously calculating NV based on a COP and CV that do not reflect those import duties"); see also id. at 1342-43 ("Under the 'matching principle,' EP, COP, and CV should be increased together, or not at all."). The Federal Circuit never stated or otherwise inferred that the adjustments to EP/CEP and normal value must be "equal," Def.'s Resp. at 17, in order to render the comparison between U.S. price and normal value "duty neutral," I & D Mem. at 12. An interpretation of the Federal *1324Circuit's discussion of duty inclusivity to espouse such a position, which would neutralize the duty drawback adjustment, goes further than the opinion supports and is inconsistent with the purpose of the statute. Accordingly, this issue is remanded to the agency to revise its calculation of the duty drawback adjustment using exports as the denominator rather than total production.15
II. Quarterly Cost Averaging Methodology
A. Background
Commerce calculates the normal value of the subject merchandise based on home market sales that are made "in the ordinary course of trade." 19 U.S.C. § 1677b(a)(1)(B)(i). Commerce, therefore, disregards sales at prices that are less than the cost of production, id. § 1677b(b)(1), because those sales are not made within the ordinary course of trade, id. § 1677(15)(A). The cost of production "equal[s] of the sum of ... the cost of materials and of fabrication or other processing of any kind employed in producing the foreign like product, during a period which would ordinarily permit the production of that foreign like product in the ordinary course of business." Id. § 1677b(b)(3)(A) (emphasis added).
The statute does not define the "period" to be used or the method by which Commerce must calculate the costs of production. SeAH Steel Corp. v. United States, 34 CIT 605, 617, 704 F.Supp.2d 1353, 1363 (2010). Commerce's usual methodology is to rely on "an annual weight-average cost" for the period of investigation. I & D Mem. at 15. Commerce may depart from its usual methodology and rely on quarterly cost-averages when "significant cost changes are evident [and] ... sales can be accurately linked with the concurrent quarterly costs." Pastificio Lucio Garofalo, S.p.A. v. United States, 35 CIT ----, ----, 783 F.Supp.2d 1230, 1235-36 (2011), aff'd 469 F. App'x 901 (Fed. Cir. 2012) ; see also I & D Mem. at 15. The significance of any cost changes must be demonstrated before Commerce analyzes the linkage between costs and sales. I & D Mem. at 15. A significant cost change "is defined as a greater than 25 percent change in [cost of manufacturing] between the high and low quarters during the POI ...." Id. at 15.16
In the underlying proceeding, Plaintiffs urged Commerce to conduct its quarterly cost test using changes in the cost of primary inputs or total direct raw material costs ("DIRMAT") instead of the total cost of manufacturing ("TCOM"). See, e.g., Habas § DQR, Ex. D-3.B at 4-7, CR 69-75, PR 91, CJA Tab 11, PJA Tab 11; Icdas § DQR at D-9, Ex. D-2. Plaintiffs reasoned that its primary input and DIRMAT costs fluctuated by more than 25 percent throughout the POI, Habas § DQR, Ex. D-3.B at 1; Icdas § DQR at D-9, Ex. D-2, and, in Habas's case, its prices closely followed changing costs, Verification of the Sales Resp. of Habas (Apr. 12, 2017) ("Habas Sales Verification Report") at 5, CR 458, PR 210, CJA Tab 44, PJA Tab 44 (explaining that Habas sets its prices daily based on the daily market price for steel scrap). Habas further argued that Commerce's *1325addition of a POI-average transformation cost (consisting of labor and overhead) to quarterly DIRMAT costs "smooths out any quarterly fluctuations" and biases the test against using quarterly costs, particularly when price closely follows cost. See I & D Mem. at 13 (summarizing Habas's argument).17
Commerce denied Plaintiffs' request on the basis that neither respondent's quarterly cost of manufacturing fluctuated by more than 25 percent during the POI. Prelim. Mem. at 14; I & D Mem. at 14-16. Noting that its 25 percent threshold is derived from generally accepted international accounting standards, Commerce explained that input cost changes "were not significant enough to impact the reported [T]COM," and its usual methodology "accounts for both the significant changes in the cost of inputs and their impact on the cost of manufacturing." I & D Mem. at 15. Commerce expressed its preference for conducting the quarterly cost test using changes in total cost of manufacturing because the measure "accounts for all production costs" that "impact pricing." Id. at 16 ("[U]sing [T]COM is more meaningful as it is the total cost of manufacturing that prices must be set to recover, not just material costs."). According to Commerce, because "material costs as a percentage of [T]COM may vary significantly from product to product, using [T]COM as the denominator in our significant cost change test results in a more consistent test." Id. Thus, Commerce "disagree[d] with Habas that a bias here creates mismatches between sales and costs when price follows cost closely. To the contrary, by keeping the test linked to [T]COM we prevent mismatches." Id.
B. Parties' Contentions
Plaintiffs contend that Commerce abused its discretion by refusing to conduct its quarterly cost test using changes in raw material costs. Habas's Mem. at 17-24; Reply Br. of PI. Habas ("Habas's Reply") at 9-13, ECF No. 45; Icdas's Mem. at 30-31; Reply Br. of Pl. Icdas to Def. and Def.-Ints.' Resp. to Pl.'s Rule 56.2 Mot. for J. on the Agency R. ("Icdas's Reply") at 10-11, ECF No. 46.18 Habas contends that Commerce's preference for using total cost of manufacturing on the basis that total production costs impact pricing ignores record evidence that "Habas sets its prices daily based on the daily change in scrap price." Habas's Mem. at 19. Habas reiterates that Commerce's addition of POI-average transformation costs (labor and overhead) to quarterly DIRMAT costs "intrinsically dilutes any [T]COM fluctuations," particularly for products with a low DIRMAT to TCOM ratio. Id. at 20 (asserting that the resulting bias leads to "disparate treatment" of respondents that are similarly situated with respect to significant input cost changes); see also id. at 21 (noting that Commerce's test is more likely *1326to deny the use of quarterly costs to respondents with products reflecting a low DIRMAT to TCOM ratio). Icdas contends that Commerce's 25 percent threshold is "too rigid." Icdas's Mem. at 28. In sum, according to Plaintiffs, Commerce's methodology results in the exclusion of more sales from the calculation of normal value as below the cost of production and a distorted dumping margin. See Habas's Mem. at 23-24; Habas's Reply at 11-12; Icdas's Mem. at 28.
The Government contends that the court should defer to Commerce's methodology because it accounts for significant input cost changes, which may be moderated by "countervailing trends in other types of costs," ensures consistent and predictable policy, and adheres to international accounting standards. Def.'s Resp. at 27-28. RTAC contends that Commerce's test does not result in disparate treatment of similarly situated respondents but instead reflects "each producer's actual overall cost experience." Def.-Int.'s Resp. at 11. RTAC further contends that Plaintiffs' DIRMAT-based test could result in Commerce's use of quarterly costs even when "an input represented a small fraction of overall manufacturing costs."Id. at 12.
C. Commerce's Determination is Sustained
The absence of a statutory definition of the period or methodology to be used when calculating the cost of production for the sales below cost test provides Commerce with broad discretion in this regard. See SeAH Steel Corp., 34 CIT at 617, 704 F.Supp.2d at 1363. In assessing the agency's methodology, the court "ask[s] whether Commerce's exercise of its gap-filling authority and its explanation are reasonable." Apex Frozen Foods, 862 F.3d at 1330 (citing Chevron, 467 U.S. at 843-44, 104 S.Ct. 2778 ).
From the outset, Plaintiffs do not deny that their cost data fail to meet the agency's test based on the total cost of manufacturing. Instead, they suggest that Commerce abused its discretion by using that test when a different test would have supported a different result. In order to challenge Commerce's methodology for conducting its cost test, Plaintiffs must show that the agency's methodology was unreasonable. See Whirlpool Corp. v. United States, Slip Op. 13-155, 2013 WL 6980820, at *11 (CIT Dec. 26, 2013) (citation omitted). Plaintiffs' challenge fails in that regard. The fact that Plaintiffs' alternative methodology would have achieved a different result is insufficient to suggest that Commerce's methodology, consistently applied for at least a decade and rooted in International Financial Reporting Standards, is unreasonable. See I & D Mem. at 15 and nn.57-58. As Commerce explained, its test examines, in the first instance, the total cost of manufacture because "it accounts for all production costs, the total of which impact pricing." Id. at 16. The fact that Plaintiffs, in this case, may alter their pricing based on price changes for direct materials does not change the reasonableness of Commerce's total cost of manufacturing approach.
Habas's argument that Commerce's "methodology results in disparate treatment of respondents that are in the same position" with respect to changing input costs based solely on differing DIRMAT to TCOM ratios is misplaced. See Habas's Mem. at 20; Habas's Reply at 12. The existence of different DIRMAT to TCOM ratios means, quite simply, that respondents are not similarly situated. The higher the DIRMAT to TCOM ratio the greater the likelihood that changes in DIRMAT will be reflected by changes in TCOM. As Commerce explained, Plaintiffs' changing raw material costs "were just not significant enough to impact the reported [T]COM." I & D Mem. at 15. This is not arbitrary, but instead reflects "each producer's *1327actual overall cost experience." Def.-Int.'s Resp. at 11.
Icdas's argument that Commerce's 25 percent threshold is "too rigid" also fails. See Icdas's Mem. at 28. Commerce pointed to generally accepted International Financial Reporting Standards to support its use of the 25 percent threshold and explained that it "is high enough to ensure that [it] do[es] not move away from [the agency's] normal practice without good cause and forgo the benefits of using an annual average cost, but allows for a change in methodology when significantly changing input costs are clearly affecting [its] annual average cost calculations." I & D Mem. at 15 (citation omitted). Determining the most appropriate threshold for departing from the agency's usual methodology is the "type of line-drawing exercise" properly left to the agency's discretion. Baoding Yude Chemical Industry Co., Ltd. v. United States, 25 CIT 1118, 1126, 170 F.Supp.2d 1335, 1343 (2001). Without more, the court sees no reason to disturb the agency's exercise of that discretion. Accordingly, Commerce's determination on this issue is sustained.
III. Habas's U.S. Date of Sale
A. Background
The antidumping duty statute does not provide a methodology for determining the "time of sale" for purposes of Commerce's comparison between export price and normal value when determining whether goods are being sold at less than fair value. See 19 U.S.C. § 1677b(a)(1)(A) (noting that normal value is to be determined "at a time reasonably corresponding to the time of the sale used to determine the export price or constructed export price"). However, the Statement of Administrative Action ("SAA") accompanying the Uruguay Round Agreements Act defines "date of sale" for the purposes of currency conversion as the "date when the material terms of sale are established." Uruguay Round Agreements Act, Statement of Administrative Action, H R. Doc. No. 103-316, vol. 1, at 810 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4153.19
Consistent with the SAA, Commerce's regulations prescribe that "[i]n identifying the date of sale of the subject merchandise ..., the [agency] normally will use the date of invoice" unless it "is satisfied that a different date better reflects the date on which the exporter or producer establishes the material terms of sale." 19 C.F.R. § 351.401(i) ; see also Antidumping Duties; Countervailing Duties, 62 Fed. Reg. 27,296, 27,349 (Dep't Commerce May 19, 1997) (final rule) ("Preamble") ("If [Commerce] is presented with satisfactory evidence that the material terms of sale are finally established on a date other than the date of invoice, [Commerce] will use that alternative date as the date of sale"). In other words, Commerce's date of sale regulation establishes a "rebuttable presumption" in favor of the invoice date unless the proponent of a different date produces satisfactory evidence that the material terms of sale were established on that alternate date. Colakoglu Metalurji A.S. v. United States, 29 CIT 1238, 1240, 394 F.Supp.2d 1379, 1380 (2005).20
*1328In its initial questionnaire responses, Habas reported the invoice date as the date of sale. See Sec. A Questionnaire Resp. of Habas (Dec. 19, 2016) ("Habas § AQR") at 15-16, CR 25-25, PR 66-68, CJA Tab 5, PJA Tab 5; Sec. C Questionnaire Resp. of Habas (Jan. 17, 2017) ("Habas § CQR") at 15, CR 69-75, PR 91, CJA Tab 11, PJA Tab 11. Habas further informed Commerce that, "[f]or U.S. sales, the parties may amend orders and letters of credit to change price, quantity, product mix, or delivery shipment date," and "there may be multiple such amendments for a given order." Habas § AQR at 18. Commerce issued a supplemental questionnaire requesting Habas to report all U.S. sales that were invoiced or negotiated to agreement during the period of investigation, observing that Habas's U.S. sales process "indicate[d] that the essential terms of sale are reached upon conclusion of the purchase order and/or contract negotiations." Suppl. Sec. A Questionnaire Resp. of Habas (Jan. 25, 2016) ("Habas Suppl. § AQR") at 2, CR 144-46, PR 103, CJA Tab 14, PJA Tab 14. In response to Commerce's request for examples of sales for which material contract terms changed between the contract date and invoice date, Habas stated that it was unable to locate any sales "where the shipment was not within the terms and tolerances of the contract." Id.
Commerce preliminary determined to use the invoice date as the U.S. date of sale. See I & D Mem. at 19. At verification, Commerce afforded Habas latitude to "[d]emonstrate that the date of invoice/date of shipment is the appropriate [U.S.] date of sale." Verification Outline (Feb. 24, 2017) at 7, CR 339, PR 155, CJA Tab 26, PJA Tab 26. Commerce examined four U.S. sales during Habas's sales verification. Habas Case Br. at 22.
Habas subsequently urged Commerce to select the contract date for its U.S. date of sale. Id. at 20-24. Habas pointed to evidence demonstrating that all four verified sales shipped in accordance with contractual quantity tolerances, prices, and delivery dates. See id. at 22-23.
For the Final Determination, Commerce continued to use the invoice date as the U.S. date of sale. I & D Mem. at 18. In so doing, Commerce relied on Habas's initial questionnaire responses, Habas's failure to clarify those responses, and the possibility of amendments to material terms. Id. at 18-19. Commerce further found that although changes to the material terms of the verified sales were within contractual tolerances, those sales represented "only [a] few out of the numerous U.S. sales," and, thus, "an insufficient basis on which to ... change the date of sale determination." Id. at 19.
B. Parties' Contentions
Habas contends that Commerce's selection of the invoice date lacks substantial evidence because the record does not contain evidence of changes to the material terms of Habas's U.S. sales between the contract date and the invoice date. Habas's Mem. at 30-32. The Government contends that Commerce properly relied on the invoice date because the possibility of contract amendments establishes that material terms of sale were not final until invoicing. Def.'s Resp. at 29, 32. RTAC contends that Habas failed to timely notify Commerce of its request to use the contract date or build the factual record supporting that request. Def.-Int.'s Resp. at 13-14.
C. Commerce's Determination is Sustained
Commerce's determination must be assessed in view of the allocation of the *1329burden on Habas to overcome the presumptive use of invoice date. That is, the precise question before the court is whether substantial evidence supports Commerce's determination that Habas failed to present "satisfactory evidence that the material terms of sale are finally established on [the contract] date." Preamble, 62 Fed. Reg. at 27,349. A decision regarding the sufficiency of the evidence presented to the agency "lies primarily within Commerce's discretion." Koyo Seiko Co. v. United States, 551 F.3d 1286, 1292 (Fed. Cir. 2008). Upon review of the record as a whole, Commerce's decision is supported by substantial evidence.
By waiting until the factual record had closed and verification completed to argue for the use of contract date, Habas constrained Commerce's ability to fully test Habas's claim that material terms were set on that date. Although the record contained Habas's assertion that it was unable to identify sales for which material terms had changed before invoicing, Habas Suppl. § AQR at 2, Habas did not provide Commerce with supporting documentation or clarify its prior statement regarding the potential for changes to material terms of sale, see id.; Habas § AQR at 17-18. Habas also did not indicate any changes in its position regarding the date of sale in its supplemental questionnaire response or at verification, which would have alerted Commerce to the potential need for supporting documentation. See Habas Suppl. § AQR at 2; Verification Outline at 1-2 (noting that Commerce "reserve[s] the right to request any additional information or materials necessary for a complete verification," and Commerce may accept new information at verification when "the need for that information was not evident previously"). In view of the foregoing, Commerce reasonably declined to find that the evidence gathered at verification provided satisfactory evidence supporting Habas's last-minute request for the use of contract date as the date of sale for all U.S. sales. See I & D Mem. at 19.
Habas's dilatory argument for the use of contract date distinguishes this case from those upon which Habas seeks to rely. Habas's Mem. at 26 (citing Habas Sinai ve Tibbi Gazlar Istihsal Endustrisi A.S. v. United States, 33 CIT 695, 737-38, 625 F.Supp.2d 1339, 1374-75, (2009) ; Nakornthai Strip Mill Public Co. Ltd. v. United States, 33 CIT 326, 336-38, 614 F.Supp.2d 1323, 1333-34 (2009) ). Although Habas cites to the court's post-remand opinions, a careful review of the preceding opinions reveals that each plaintiff advocated for the use of contract date early in the proceeding. See Habas Sinai ve Tibbi Gazlar Istihsal Endustrisi A.S. v. United States, 31 CIT 1793, 1795, 2007 WL 3378201 (2007) ; Nakornthai Strip Mill Public Co. Ltd. v. United States, 32 CIT 553, 556, 558 F.Supp.2d 1319, 1322 (2008). The Habas Sinai court therefore found that Commerce was within its discretion to rely on a single sample sale to select the contract date when, unlike here, it had the opportunity to seek additional documentation if necessary. See 33 CIT at 738, 625 F.Supp.2d at 1375 (noting that "Commerce found no indicia [in the sample sale] which prompted the agency to require ... further documentation"). The Nakornthai court looked for evidence of significant changes to material contract terms when assessing whether Commerce properly denied the plaintiff's prompt request for contract date, because, when faced with such a request, Commerce typically disregards insignificant changes. See 33 CIT at 336, 614 F.Supp.2d at 1333. Here, however, Commerce's determination did not turn solely upon the degree to which Habas's sales reflected any changes to material terms, but on Habas's request for the use of invoice date throughout the fact gathering stage of the proceeding and concomitant representations regarding the potential for *1330multiple contract amendments. I & D Mem. at 18-19.
In sum, Habas had the burden of "remov[ing] any doubt about when material terms are firmly and finally set." Toscelik Profil ve Sac Endustrisi A.S. v. United States, 41 CIT ----, ----, 256 F.Supp.3d 1260, 1263 (2017). Commerce's determination that Habas's supplemental questionnaire response and the evidence gathered at verified failed to fulfill that burden is supported by substantial evidence. Accordingly, Commerce's date of sale determination is sustained.
IV. Habas's Zero-Interest Short Term Loans
A. Background
Commerce may adjust normal value to account for "the amount of any difference (or lack thereof) between the export price or constructed export price and [normal value]" that Commerce determines is "wholly or partly due to ... differences in the circumstances of sale." 19 U.S.C. § 1677b(a)(6)(C)(iii). One such adjustment pertains to credit expenses. Hornos Electricos de Venezuela v. United States, 27 CIT 1522, 1538, 285 F.Supp.2d 1353, 1368 (2003) ; see generally Import Admin. Policy Bulletin 98.2: Imputed Credit Expenses and Interest Rates (Feb. 23, 1998) ("Policy Bulletin 98.2") at 3, available at https://enforcement.trade.gov/policy/bull98-2.htm (last visited Dec. 18, 2018). "[T]o account for differences in credit terms," Commerce "imputes a U.S. credit expense and a foreign credit expense on each sale." Policy Bulletin 98.2 at 1. Commerce "measures the credit expense on a sale by the amount of interest that the sale revenue would have earned between date of shipment and date of payment." Id. Credit expenses "must be imputed on the basis of usual and reasonable commercial behavior." Id. (quoting LMI-La Metalli Industriale, S.p.A. v. United States, 912 F.2d 455, 460-61 (Fed. Cir. 1990) ).21
For home market sales transactions, Habas reported that "[it] did not incur an imputed credit expense during the POI as its only short-term [Turkish Lira] loans in the period were at zero-interest." Sec. B Questionnaire Resp. of Habas (Jan. 17, 2017) ("Habas § BQR") at 31, CR 69, PR 91, Suppl. CJA Tab 3, Suppl. PJA Tab 3. Elsewhere, Habas noted that its reporting of zero-interest short-term loans was consistent with Commerce's prior use of such loans to impute credit expenses. Suppl. Secs. A-D Questionnaire Resp. of Habas (Feb. 13, 2017) at 6 & n.1, CR 249-53, PR 145, CJA Tab 23, PJA Tab 23 (citations omitted).
The domestic producers challenged Habas's short-term interest rate, pointing to Commerce's practice of allowing negative credit expenses when "the date of payment occurs before the date of shipment" because "the seller receives the benefit of the time value of money, resulting in revenue or income." Pre-Prelim. Comments Regarding Habas (Feb. 17, 2017) ("Pet'r's Comments") at 16 & n.52, CR 337, PR 153, CJA Tab 25, PJA Tab 25 (citing Bottle-Grade Polyethylene Terephthalate (PET) Resin From Indonesia, 70 Fed. Reg. 13,456 (Dep't Commerce Mar. 21, 2005) (notice of final determination of sales at less than fair value) ).22 Arguing that Habas's *1331short-term interest rate "does not reflect commercial reality in Turkey," the domestic producers sought to persuade Commerce to instead use "[p]ublicly available information from the Central Bank of the Republic of Turkey [ ("CBRT") ]," which showed that short-term lending rates in Turkey during the period of investigation ranged from 9.0 percent to 10.75 percent, with an average rate of 10.42 percent (the "CBRT rate"). Id. at 16-17 & n.54 (citing id., Ex. 2).23
Agreeing with the domestic producers, Commerce preliminarily recalculated Habas's home market credit expenses using the CBRT rate on the basis that the rate conforms with commercial reality. Prelim. Determination Margin Calculation for Habas (Feb. 28, 2017) at 1-2, CR 342, PR 165, CJA Tab 33, PJA Tab 33. Commerce applied the interest rate to Habas's reported "customer-specific average age of receivables." Id. at 2 & n.4 (citing Habas § BQR at 19-21).24
Commerce affirmed its imputed credit expense calculation for the Final Determination. See I & D Mem. at 22-23. Pointing to evidence of prepayment,25 Commerce explained that Habas benefitted from that prepayment "for longer periods than what it paid (zero-interest) on its [short-term] borrowings." Final Determination Margin Calculation for Habas (May 15, 2017) ("Habas Final Calc. Mem.") at 4-5, CR 467, PR 223, CJA Tab 50, PJA Tab 50.26 Commerce determined that the CBRT rate afforded a "more comparable" measure of the time value of Habas's prepaid sales and was consistent with the non-zero-interest rate reported in Habas's financial statement in connection with short-term trade receivables27 and Icdas's reported rate. Id. at 5; I & D Mem. at 23.
B. Parties' Contentions
Habas contends that Commerce's determination is unsupported by substantial evidence and not in accordance with law. Habas's Mem. at 37-39. Habas assails Commerce's reliance on the CBRT data supplied by Haver Analytics because it is not "a primary source" and does not represent commercial rates. Id. at 37-38. Habas *1332further asserts that neither Icdas's reported rates nor its earned interest rate on trade receivables are relevant to assessing the commerciality of its short-term interest rate. Id. at 39; Habas's Reply at 18. Habas also contends that Commerce has previously included zero-interest loans in its credit expense calculations. See Habas's Mem. at 38 (citations omitted); Habas's Reply at 19-20 (citing Issues and Decision Mem. for the Final Aff. Determination in the Less-Than-Fair-Value Investigation of Welded Line Pipe from the Republic of Turkey, A-489-822 (Oct. 5, 2015) ("Line Pipe from Turkey Mem.") at Comment 13, 80 ITADOC 61632 ; Issues and Decision Mem. for the Final Results of the Antidumping Duty Admin. Review: Certain Welded Carbon Steel Pipe and Tube from Turkey, A-489-501 (Dec. 2, 2011) ("Steel Pipe from Turkey Mem.") at Comment 10, 76 ITADOC 76939 ).
The Government contends that substantial evidence supports Commerce's determination and Habas points to no evidence rebutting the accuracy of the CBRT rate. Def.'s Resp. at 35-36. RTAC contends that similarities between the CBRT rate, Icdas's rate, and the interest rate reported in Habas's financial statement "support[ ] the overall reasonableness of [Commerce's] decision" to reject Habas's reported rate. Def.-Int.'s Resp. at 16. RTAC further contends that Commerce "reasonably declined to treat [ ] advance prepayment as worthless." Id.
C. Commerce's Determination is Sustained
From the outset, Habas misstates Commerce's basis for rejecting its zero-interest short-term rates. Habas asserts that Commerce rejected its rates as "non-commercial" and, thus, seeks to persuade the court that its loans are indeed commercial. See Habas's Mem. at 38-39; Habas's Reply at 17. Commerce, however, never stated that Habas's loans were non-commercial; rather, Commerce found that Habas's short-term interest rate associated with those loans was not "reasonable or representative of usual commercial behavior" when considering the appropriate rate with which to impute revenue derived from prepayment. I & D Mem. at 23; see also Habas Final Calc. Mem. at 5. The issue confronting Commerce concerned the proper interest rate with which to calculate the benefit inuring to Habas from the advance payment, not the loss occasioned by delayed payment. See Habas Final Calc. Mem. at 4-5. Because longer lending periods are associated with higher interest rates, see id. at 4, Commerce determined that applying a zero-interest rate to Habas's negative receivables would not capture the benefit derived therefrom, id. at 5, and, thus, the rate was not "reasonable or representative of usual commercial behavior," I & D Mem. at 23.
Habas fails to persuade the court that Commerce should effectively treat prepayment as worthless. Habas's reliance on Commerce's prior inclusion of zero-interest loans in its credit expense calculations is misplaced because the cited determinations do not involve instances of prepayment. See Line Pipe from Turkey Mem. at 30-31; Steel Pipe from Turkey Mem. at 28-29. Indeed, Habas wholly fails to address that crucial distinction even though it underpinned Commerce's decision here. See Habas's Final Calc. Mem. at 3-5. Instead, Habas challenges Commerce's evidentiary foundation. Those challenges lack merit.
Habas argues that the source of the CBRT rate is not demonstrably reliable and is not a commercial rate. See, e.g., Habas's Mem. at 37-38. Habas points to no evidence, however, specifically undermining the source's reliability. While the CBRT rate may not represent a commercial rate, it is consistent with other rates on the record. See I & D Mem. at 23 & n.54 (citing Habas § AQR, Ex. A-11 at 32);
*1333Habas Final Calc. Mem. at 5. Moreover, Habas failed to offer Commerce any alternative rates other than its zero-interest short-term rate. See QVD Food Co. v. United States, 658 F.3d 1318, 1324 (Fed. Cir. 2011) (the burden of creating an adequate record before Commerce lies with interested parties).
Habas also argues that Icdas's "non-zero interest rates [do] not controvert the commercial nature of Habas's overnight loans." Habas's Reply at 17; see also Habas's Mem. at 39. As discussed above, Habas's argument is misdirected. Commerce did not base its decision on the commercially of Habas's loans (or lack thereof); rather, Commerce compared Habas's rate to other rates on the record and found that Habas's rate was not commercially reasonable given its negative receivables. I & D Mem. at 23; cf. Habas Final Calc. Mem. at 5.
Habas further argues that the interest rate in its financial statement "is simply a figure plugged by Habas's auditors" and "has no bearing on the actual rates at which Habas borrowed money in the POI." Habas's Reply at 18. Habas points to no evidence indicating that the interest rate in its financial statement does not, in fact, represent the interest earned on short-term trade receivables or provide a convincing reason as to why Commerce erred in considering this rate as supportive of its selection of the CBRT rate to determine Habas's home market credit expense.
In sum, Commerce reasonably determined that the consistency between the CBRT rate, Icdas's rates, and Habas's earned interest rate merited rejecting Habas's zero-interest short-term borrowing rate as a means of imputing its credit expenses. See l & D Mem. at 23. Accordingly, Commerce's use of the CBRT rate to calculate Habas's imputed credit expenses is sustained.
V. Use of Partial Adverse Facts Available
A. Background
When an interested party "withholds information" requested by Commerce, "significantly impedes a proceeding," "fails to provide [ ] information by the deadlines for submission of the information," or provides information that cannot be verified pursuant to 19 U.S.C. § 1677m(i), Commerce shall use the "facts otherwise available" in making its determination. 19 U.S.C. § 1677e(a)(2). Commerce's authority to use the facts otherwise available is subject to 19 U.S.C. § 1677m(c),28 (d),29 and (e).30
*1334Additionally, if Commerce determines that the party "has failed to cooperate by not acting to the best of its ability to comply with a request for information," it "may use an inference that is adverse to the interests of that party in selecting from among the facts otherwise available." Id., § 1677e(b)(1)(A).31 "Compliance with the 'best of its ability' standard is determined by assessing whether a respondent has put forth its maximum effort to provide Commerce with full and complete answers to all inquiries in an investigation." Nippon Steel Corp. v. United States, 337 F.3d 1373, 1382 (Fed. Cir. 2003).32 Before applying an adverse inference, Commerce must demonstrate "that the respondent['s] ... failure to fully respond is the result of the respondent's lack of cooperation in either: (a) failing to keep and maintain all required records, or (b) failing to put forth its maximum efforts to investigate and obtain the requested information from its records." Id. at 1382-83. "An adverse inference may not be drawn merely from a failure to respond." Id. at 1383. Rather, Commerce may apply an adverse inference when "it is reasonable for Commerce to expect that more forthcoming responses should have been made." Id.
In its section A questionnaire response, Icdas informed Commerce that it sold subject merchandise in the home market directly and through several affiliated resellers. Sec. A Questionnaire Resp. of Icdas (Dec. 19, 2016) ("Icdas § AQR") at A-12, CR 33, PR 69, CJA Tab 4, PJA Tab 4. In response to Commerce's request that Icdas report the manufacturer of merchandise sold directly and through its affiliates, Icdas stated, inter alia :
For sales made by affiliated resellers ... ICDAS has identified almost all the transactions whether the foreign like product sold is produced by ICDAS or other unaffiliated manufacturers. In a few cases Icdas was unable to verify that Icdas was the producer. For the transactions that are not identified, ICDAS leaves this field as blank. However, considering the magnitude of the ICDAS sales to its affiliated resellers, and the fact that ICDAS rarely purchases from other manufacturers, ICDAS assumes that these sales are also manufactured by ICDAS.
Sec. B Questionnaire Resp. of Icdas (Jan. 17, 2017) ("Icdas § BQR") at B-48, CR 124, PR 109, CJA Tab 10, PJA Tab 10.
Commerce issued a supplemental questionnaire requesting Icdas to remedy the deficient response by "provid[ing] the manufacture code for the ... home market transactions that reported no manufacturer code." Suppl. Questionnaire for Secs. B, C and D (Feb. 1, 2017) at 4, CR 182, PR 115, CJA Tab 15, PJA Tab 15. Commerce further requested Icdas to "explain the likelihood that you did not produce these products." Id. at 4. Icdas responded:
*1335Affiliated resellers normally purchase subject merchandise from Icdas. For the back-to-back sales[33 ], all the affiliated resellers are able to identify the manufacturer of subject merchandise. For the rest of the sales, even if the subject merchandise is kept in the inventories for an hour, affiliated resellers [do] not track the manufacturers. Therefore[,] neither Icdas nor the relevant affiliated resellers have this information.
Suppl. Secs. B-C and 2nd Suppl. Sec. D Questionnaire Resp. of Icdas (Feb. 13, 2017) ("Icdas Suppl. § B-C-D QR") at 13, CR 273, PR 146, CJA Tab 22, PJA Tab 22. Icdas provided Commerce with "percentages for each affiliated resellers' quantities sold to third parties by manufacturer," and "a summary of Icdas'[s] sales to its affiliates." Id. According to Icdas, that information indicated that its affiliates purchase "a very insignificant amount of resales ... from other manufacturers" and, thus, it believed that the transactions missing manufacturer codes most likely involved merchandise produced by Icdas. Id. According to Icdas, Commerce could therefore consider Icdas as the manufacturer for those transactions. Id.; see also Id., Ex. SB-26 (sales quantities of affiliated resellers by manufacturer).
For the Final Determination, Commerce concluded that "Icdas provided incomplete information with respect to the manufacturer of certain sales made by its home market affiliates," and, thus, determined to use facts available. I & D Mem. at 4-6. Commerce further found that an adverse inference was warranted when selecting from among the available facts. See Id. Commerce explained that the identity of the manufacturer of rebar sold by Icdas's affiliates "is critical to the [agency's] dumping analysis" and "is the type of information that a large steel manufacturer such as Icdas should be reasonably able to provide."Id. at 30. To support this finding, Commerce pointed to Icdas's mill test certificates that identify the manufacturer of the rebar and the waybills included in Icdas's home markets sales that "identify the Icdas mill where the rebar at issue was manufactured." See Id. at 6 & nn.18-19 (citations omitted); Id. at 30. According to Commerce, "had [Icdas] made the appropriate effort" to obtain the missing manufacturer codes "using the records over which it maintained control," it "would have been able to provide this information." Id. at 6. Commerce concluded that "Icdas did not act to the best of its ability" and applied partial adverse facts available to Icdas's downstream home market sales missing the manufacturer code. Id.; see also id. at 29-31. For those sales, Commerce "assigned the highest non-aberrational net price from Icdas'[s] downstream home market sales." Id. at 6, 31.
B. Parties' Contentions
Icdas contends that Commerce's use of partial adverse facts available lacks substantial evidence and is contrary to law. Icdas's Mem. at 31. Icdas asserts that it reported all the information it had regarding the manufacturer of its affiliates' resales but cannot "report information that it does not have." Id. at 36. Icdas further contends that use of adverse facts available is inappropriate when cooperative respondents are "unable to provide the information requested" and "offer[ ] a reasonable approximation or alternative." Icdas's Reply at 12 (citing, inter alia, *1336Husteel Co. v. United States, 39 CIT ----, ----, 98 F.Supp.3d 1315, 1361 (2015) ) (internal quotation marks omitted); see also Icdas's Mem. at 33.
The Government contends that Commerce correctly applied an adverse inference when selecting from among the facts available because Icdas's failure to report all the manufacturer codes reflected "inadequate recordkeeping." Def.'s Resp. at 37. The Government also contends that Commerce complied with predicate statutory requirements set forth in 19 U.S.C. § 1677m(d) and (e) before applying an adverse inference. Id. at 40-41. The Government further contends that Icdas's reliance on Husteel is misplaced because the identity of the manufacturer "cannot be derived from other data." Id. at 41.
RTAC contends that the requested information could have been obtained "through a review of mill test reports or other documentation associated with the sales," which "would [not] have required a particularly intense effort on Icdas's part." Def.-Int.'s Resp. at 17.
C. Commerce's Determination is Remanded
Commerce's use of adverse facts available is circumscribed by statutory procedural requirements and must be accompanied by specific factual findings. Here, Commerce's use of adverse facts available failed to comply with all statutory requirements and the agency's conclusion that Icdas failed to act to the best of its ability lacks substantial evidence.
As noted, section 1677m(c)(1) requires a respondent to promptly inform Commerce when it cannot comply with a request for information and suggest an alternative form for supplying that information. 19 U.S.C. § 1677m(c)(1). Icdas complied with that requirement when it explained that its affiliated resellers do not track the manufacturer of rebar sold in non-back-to-back sales and suggested, with supporting documentation, why Commerce could consider Icdas the manufacturer for those sales. Compare Maverick Tube Corp. v. United States, 857 F.3d 1353, 1360-61 (Fed. Cir. 2017) (respondent "never triggered" section 1677m(c)(1) when it "never claimed that it was unable to provide [the requested information]"), with World Finer Foods, Inc. v. United States, 24 CIT 541, 543-44, 2000 WL 897752 (2000) (producer triggered section 1677m(c)(1) when it informed Commerce that financial constraints prevented it from fully responding to questionnaire and offered to supply limited information). The statute, thus, required Commerce to consider Icdas's ability "to submit the information in the requested form and manner" and whether to "modify such requirements to the extent necessary to avoid imposing an unreasonable burden on that party." 19 U.S.C. § 1677m(c)(1) ; see also World Finer Foods, 24 CIT at 543-44 (Commerce erred in resorting to AFA without complying with section 1677m(c)(1) ); cf. Husteel, 98 F.Supp.3d at 1361 (Commerce properly declined to use AFA when the respondent "made an effort to provide its best estimate of the information Commerce had asked [it] to report").
Commerce did not comply with this requirement and instead resorted to use of AFA. Commerce's finding that Icdas could have undertaken additional efforts to obtain mill test certificates and waybills purportedly kept by its affiliates to identify the missing manufacturer codes, see I & D Mem. at 30-31, is contradicted by Icdas's statement that, for non-back-to-back sales, its affiliated resellers "[d]o not track the [identity of the] manufacturers" and, thus, do not "have this information," Icdas Suppl. § B-C-D QR at 13. Commerce's determination was, therefore, procedurally lacking, and its finding that Icdas failed to "act to the best of its ability" by failing to produce information at its disposal lacks *1337substantial evidence.34 See I & D Mem. at 6.
The Government's assertion that Commerce's determination is supported by evidence of Icdas's "inadequate recordkeeping" fails because Commerce did not base its use of AFA on such a finding. Here, Commerce's use of AFA is predicated on Icdas's purported failure to cooperate by not putting forth its maximum efforts to investigate and obtain the manufacturer codes from its records. I & D Mem. at 6, 30-31. The court may not sustain the agency's decision on a basis other than the one "articulated ... by the agency itself." Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168-69, 83 S.Ct. 239, 9 L.Ed.2d 207 (1962). For these reasons, Commerce's use of partial adverse facts available is remanded for reconsideration.
CONCLUSION AND ORDER
In accordance with the foregoing, it is hereby
ORDERED that Commerce's Final Determination is remanded for reconsideration regarding the agency's calculation of Plaintiffs' duty drawback adjustment, as set forth in Discussion Section I; it is further
ORDERED that Commerce's Final Determination is remanded with respect to the agency's use of partial adverse facts available to Icdas, as set forth in Discussion Section V; it is further
ORDERED that Commerce's Final Determination is sustained in all other respects; it is further
ORDERED that Commerce shall file its remand redetermination on or before 90 days from the restoration of appropriations; it is further
ORDERED that subsequent proceedings shall be governed by USCIT Rule 56.2(h) ; and it is further
ORDERED that any comments or responsive comments must not exceed 6,000 words.