Polyethylene Retail Carrier Bag Committee v. United States

*1419 OPINION BARZILAY, Judge: In this consolidated action, the plaintiffs and defendant-intervenors filed USCIT R. 56.2 Motion for Judgment upon an Agency Record, challenging certain aspects of the final determination of the U.S. Department of Commerce (“Commerce”) in the antidumping investigation Final Determination of Sales at Less Than Fair Value: Polyethylene Retail Carrier Bags from the People’s Republic of China, 69 Fed. Reg. 34,125 (June 18, 2004) (RR. 505) {“Final Determination”), amended, 69 Fed. Reg. 42,419 (July 15, 2004) (P.R. 530) {“Amended Final Determination”). Plaintiffs Polyethylene Carrier Bag Committee, and its individual members, Vanguard Plastics, Inc., Hilex Poly Co., LLC, and Superbag Corp. (collectively “PRCB Committee Plaintiffs”) are domestic manufacturers. Plaintiffs Glopack, Inc. (“Glopack”), Elkay Plastics Co., CPI Packaging, Inc., and PDI Saneck International are importers of polyethylene retail carrier bags from China to the United States; Plaintiffs Sea Lake Polyethylene Enterprise, Ltd. (“Sea Lake”) and Rally Plastics Co. (“Rally”) are Chinese producers and exporters to the United States of polyethylene retail carrier bags, (collectively “Glopack Plaintiffs”). Before the court are four issues raised by the PRCB Committee Plaintiffs and two issues raised by the Glopack Plaintiffs. For the reasons outlined below, the court AFFIRMS Commerce’s determination regarding five challenges and remands the case on Commerce’s calculations of Hang Lung’s electricity usage as a factor of production. I. Background This action challenges Commerce’s determination of sales at less than fair value in the underlying investigation on polyethylene retail carrier bags (“PRCBs”) from the People’s Republic of China (“PRC”), covering the period of review from October 1, 2002 to March 31, 2003 (“POR”). Specifically, the subject merchandise included t-shirt sacks, merchandise bags, grocery bags, and checkout bags. See Final Determination, 69 Fed. Reg. at 34,125. In the Final Determination, Commerce determined that PRCBs from PRC were or likely were sold in the United States at less than fair value as provided in section 735 of the Tariff Act of 1930. See Final Determination, 69 Fed. Reg. at 34,125. See Issues and Decision Mem. for the Administrative Review of Certain Stainless Steel Bar from India (July 5, 2002) {“Issues and Decision Mem.”) (containing explanations for Commerce’s determinations). The PRCB Committee Plaintiffs contest the following aspects of Commerce’s antidumping duty determination: 1) Commerce’s selection of facts otherwise available with respect to Hang Lung Plastic Manufactury Ltd. (“Hang Lung”), a Chinese manufacturer and exporter to the United States of the PRCBs and a mandatory respondent in the underlying investigation and defendant-intervenor in this case; 2) Commerce’s acceptance of certain prices for polyethyl *1420 ene resin reported by Nantong Huasheng Plastic Products Co. (“Nantong”), a PRCBs exporter and mandatory respondent in the underlying investigation and defendant-intervenor in this case; 3) Commerce’s decision to accept Nantong’s reported factors of production data; and 4) Commerce’s selection of a surrogate value for cardboard inserts consumed in the production of PRCBs. In the amended preliminary determination, the Department calculated a dumping margin of 0.12 percent for Hang Lung. See Notice of Amended Preliminary Determination of Sales at Less Than Fair Value: Polyethylene Retail Carrier Bags from the PRC, 69 Fed. Reg. 7908, 7909 (Feb. 20, 2004) (“Amended Preliminary Determination”). The final margin assigned to Hang Lung was 0.24 percent. See Final Determination, 69 Fed. Reg. at 42420. Since Hang Lung’s margin was less than the 2.0 percent de minimis threshold, Hang Lung’s customs entries of the subject merchandise were excluded from anti-dumping duties. Commerce’s preliminary determination for Nantong was based on neutral “facts otherwise available” pursuant to 19 U.S.C. § 1677e(a) because the respondent did not report its factors of production information on a product-specific basis. See Preliminary Determination, 69 Fed. Reg. at 3549. Nantong’s preliminarily determined margin was 18.43 %. Preliminary Determination, 69 Fed. Reg. at 3549. Commerce ultimately verified Nantong’s questionnaire responses and did not apply facts otherwise available in calculating the final margin. Issues and Decision Mem., at 77-80. Nantong’s final margin was 0.01 %. See Final Determination, 69 Fed. Reg. at 42420. The Glopack Plaintiffs challenge the following aspects of the final determination: 1) Commerce’s use of average unit values calculated from Indian basket category import statistics for certain black and color printing ink types to value Sea Lake and Rally’s reported color-specific flexographic and gravure printing inks used in the manufacture of the subject merchandise and 2) Commerce’s use of surrogate values rather than the prices reported paid for certain raw material inputs purchased from a Hong Kong trading company. Glopack Pis Br. at 2-3. Plaintiffs Glopack, Inc., Sea Lake, and Rally participated in the investigation and submitted detailed responses to the Department’s information requests. Plaintiff Sea Lake submitted responses covering two production plants: Shanghai Glopack Packing Ltd. (“Shanghai Glopack”) 1 and Sea Lake. Commerce calculated a margin of 19.79 % for Shanghai Glopack and Sea Lake. Final Determination, 69 Fed. Reg. at 42420; Antidumping Duty Order: Polyethylene Retail *1421 Carrier Bags from the People’s Republic of China, 69 Fed. Reg. 48,201. Rally’s final margin was determined to be 23.85 %. Amended Final Determination, 69 Fed. Reg. at 42420. In the preliminary determination, the Department calculated a dumping margin of 18.56 % for Rally. Amended Preliminary Determination, 69 Fed. Reg. at 7909. The six issues contested in this action concern different sets of facts Commerce relied on in the administrative record. The relevant facts will be set forth separately in the discussion section for each separate challenge. II. Jurisdiction and Standard of Review The court exercises its jurisdiction over this case pursuant to 28 U.S.C. § 1581(c) (2004). The court “must sustain ‘any determination, finding or conclusion found’ by Commerce unless it is ‘unsupported by substantial evidence on the record, or otherwise not in accordance with the law.’ ” Fujitsu Gen. Ltd. v. United States, 88 F.3d 1034, 1038 (Fed. Cir. 1996) (quoting 19 U.S.C. § 1516a(b)(l)(B)). “Substantial evidence is more than a mere scintilla. It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Consol. Edison Co. of New York v. NLRB, 305 U.S. 197, 229 (1938). “[T]he possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Matsushita Elec. Indus. Co. v. United States, 750 F.2d 927, 933 (Fed. Cir. 1984) (quoting Consolo v. Fed. Mar. Comm’n, 383 U.S. 607, 619-20 (1966)). When the court applies this standard, it affirms the agency’s factual determinations “so long as they are reasonable and supported by the record as a whole, even if there is some evidence that detracts from the agency’s conclusions.” Olympia Indus., Inc. v. United States, 22 CIT 387, 389, 7 F. Supp. 2d 997, 1000 (1998) (citing Atlantic Sugar, Ltd. v. United States, 744 F.2d 1556, 1563 (Fed. Cir. 1984)); see Granges Metallverken AB v. United States, 13 CIT 471, 474, 716 F. Supp. 17, 21 (1989) (The court may not reweigh the evidence or substitute its own judgment for that of the agency.). The court reviews Commerce’s surrogate value determinations for reasonableness. Coalition for the Preservation of Am. Brake Drum & Rotor Aftermarket Mfrs. v. United States, 23 CIT 88, 44 F. Supp. 2d 229, 258 (1999) (“Commerce need not prove that its methodology was the only way or even the best way to calculate surrogate values for factors of production as long as it was a reasonable way.”). III. Discussion 1. Commerce’s Application of Adverse Facts Available by Allocating Hang Lung’s Total Electricity Usage Rate The factors of production methodology used in nonmarket economy (“NME”) cases requires Commerce to “determine the normal value of the subject merchandise on the basis of the value of the *1422 factors of production utilized in producing the merchandise . . . based on the best available information regarding the values of such factors in a market economy country or countries considered to be appropriate.” 19 U.S.C. § 1677b(c). Factors of production include “(A) hours of labor required, (B) quantities of raw materials employed, (C) amounts of energy and other utilities consumed, and (D) representative capital cost.” 19 U.S.C. § 1677b(c). Respondents in such cases have to report on a model-specific basis the precise quantity of each factor required for production of one unit of merchandise. When Commerce “determines that it is unable to verify the respondent’s submission, it may substitute for the information submitted by the respondents, facts otherwise available.” Chia Far Industrial Factory Co. v. United States, 28 CIT_, 343 F. Supp. 2d 1344, 1363 (2004) (“Chia Far”) (citing 19 U.S.C. § 1677e(a)(2)(A), (D)). If Commerce finds that “the failure to fully respond is the result of the respondent’s lack of cooperation in . . . failing to put forth its maximum efforts to investigate and obtain the requested information from its records [,]” the agency is authorized to adopt an adverse inference when selecting facts otherwise available, pursuant to 19 U.S.C. § 1677e(b) 2 . Nippon Steel Corp. v. United States, 337 F.3d 1373, 1383-84 (Fed. Cir. 2003) (“Nippon Steel”); F.LLI De Cecco Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir. 2000). In its Final Determination, Commerce applied an adverse inference to determine the value of the electricity usage related to the production of PRCBs. Commerce explained that it “was able to verify the usage amounts that were listed on model-specific usage worksheets for the individual models that [it] examined, but the per-unit amounts [it] verified did not appear in Hang Lung’s FOP database for most of the models examined.” Issues and Decision Mem. at 64. Finding that Hang Lung did not act to the best of its ability in reporting usage rates, Commerce decided to use adverse inferences in restating these usage rates. Issues and Decision Mem. at 64. However, Commerce concluded that because it was “able to verify the total electricity used by Hang Lung during the POI for subject mer *1423 chandise[,]” it “allocated Hang Lung’s total electricity usage that [it] verified to its reported U.S. sales.” Issues and Decision Mem. at 64. Commerce further explained that it “allocated electricity based on the total amount of extruded resin and concentrate reported by Hang Lung in its United Stats factors-of-production database. In addition, [it] only allocated printing electricity to printed bags and [it] only allocated handwork electricity to handworked bags.” Analysis for the Final Determination of PRCBs from PRC: Hang Lung Manufactory Ltd., June 9, 2004, at 2. During the oral argument and in its motion for leave to clarify its calculation, Commerce, however, evidenced an inconsistent position, claiming that “Commerce actually allocated the total electricity used in Hang Lung’s production of all plastic bags, regardless of destination, to its United States sales.” Def. Mot. Leave Clarify Commerce’s Electricity Calculations for Hang Lung, at 2. In support, Commerce references Hang Lung Verification Exhibit 11 that showed: 1) the amounts of electricity used by Hang Lung for each month of the period of investigation for each department involved in its bag-producing activities (extrusion, printing, gusseting, cutting, and handwork), calculating the total electricity used for each department during the period of investigation by adding these monthly amounts, and 2) Hang Lung’s reported electricity usage for each bag-producing and non-bag producing activity in December 2002. Verification Hang Lung’s U.S. Sales and Factors-of-Production Data, Mar. 11, 2004, Ex. 11. Commerce explained that Exhibit 11 demonstrates that Hang Lung calculated total electricity used to produce all merchandise by subtracting “overhead electricity” from the expenses it incurred in the monthly electricity bills. Def. Mot. Leave to Clarify Commerce’s Electricity Calculations for Hang Lung, at 2. These numbers, Commerce now argues, are not broken down by country of destination. While this reading of Exhibit llsupports Commerce’s explanation of its final calculations of the electricity usage in Issues and Decision Memorandum, at 64, it is inconsistent with Commerce’s explanation in Analysis of the Final Determination for Hang Lung, where Commerce stated that it allocated the total electricity “based on the total amount of extruded resin and concentrate reported by Hang Lung in its United States factors-of-production database.” See Analysis Final Determination PRCBs from PRC: Hang Lung Manufactory Ltd., June 9, 2004, at 2. On appeal, the PRCB Committee Plaintiffs do not contest the application of adverse facts available, but challenge Commerce’s calculations as based on neutral, non-adverse facts. Plaintiffs maintain that Commerce recalculated Hang Lung’s electricity consumption based on “a precise, multi-step allocation methodology and then restricted the amount to be allocated to the total verified electricity usage associated with production of subject merchandise during the POI.” PRCB Committee Br. at 15. Plaintiffs suggest that Commerce did not adhere to its decision to use adverse facts, instead calculat *1424 ing the electricity usage in a neutral way. PRCB Committee Br. at 16. Thus, Plaintiffs claim that the methodology adopted by the Department in the final results is contrary to law and must be reversed. See 19 U.S.C. 1516a(b)(l)(B)(i). They demand that the court remand this issue, instructing the Department to select as adverse facts available, the highest reported and verified electricity usage rate on the record. PRCB Committee Br. at 20. In response and opposition to Commerce’s motion for leave to clarify its calculation, the PRCB Committee Plaintiffs argued that the government attempted to “completely change its position” by claiming that Commerce allocated the electricity used by Hang Lung to produce plastic bags, regardless of the type of bag or destination of shipment. PRCB Committee Resp. in Opp’n Def. Mot. Leave Clarify, at 3. The court must evaluate whether Commerce’s selection of partial adverse facts available was supported by substantial evidence in the administrative record. See Fujitsu, 88 F.3d at 1038. In this case, Commerce claims that it decided to take the total electricity usage calculated for each department involved in the productions of subject merchandise, less the overhead electricity usage, and to allocate it to the United States sales. Commerce’s final determination specifically cited to 19 U.S.C. § 1677m(e), which provides: [Commerce] shall not decline to consider information that is submitted by an interested party and is necessary to the determination but does not meet all the applicable requirements established by the administering authority or the Commission, if— (1) the information is submitted by the deadline established for its submission, (2) the information can be verified, (3) the information is not so incomplete that it cannot serve as a reliable basis for reaching the applicable determination, (4) the interested party has demonstrated that it acted to the best of its ability in providing the information and meeting the requirements established by the administering authority or the Commission with respect to the information, and (5) the information can be used without undue difficulties. Commerce explained that it verified Hang Lung’s reported total electricity usage rate, and nothing in the record indicates that this information was unusable or incomplete. In applying facts adverse to a party’s interest, Commerce’s goal is to “encourage compliance while determining current margins as accurately as possible.” Nat’l Steel Corp. v. United States, 20 CIT 100, 103, 913 F. Supp. 593, 596 (1996). Simultaneously, Commerce must select non-aberrant facts rationally related to what they are used to calculate. See id. In this case, the court finds that using the total electricity consumed during the period of review in producing all bags, regardless of destination, was neither unreasonable nor aberrant. However, Commerce took incon *1425 sistent positions in explaining how it allocated that verified value. In the Hang Lung Analysis for the Final Determination, Commerce stated that it “allocated that value based on the total amount of extruded resin and concentrate reported by Hang Lung in its factors of production database.” Analysis for the Final Determination of PRCBs from PRC: Hang Lung Manufactory Ltd., June 9, 2004, at 2. In its motion to clarify, the government insisted that it allocated the total electricity usage to Hang Lung’s reported United States sales, not conditioned on any other factor. The record does not furnish any more information for the court to ascertain how Commerce allocated the value. Lacking this information, the court cannot determine whether Commerce’s methodology in this instance was contrary to law or unsupported by substantial evidence in the record. Therefore, the court remands'this issue, instructing the government to explain its calculation. Commerce must address the seeming inconsistency between the Hang Lung Analysis for the Final Determination and the information in Commerce’s motion to clarify and to reconcile this information. 2. Use of the Price Nantong Paid to Market Economy Suppliers for Certain Raw Material Inputs During the investigation, Nantong reported that it bought certain raw material inputs from market economy suppliers. Nantong'Ques-tionnaire Response, Oct. 1, 2003. Upon verification of the reported information, Commerce examined the invoices for each raw material input, including invoices for inputs from market economy suppliers. See Nantong Verification Report, April 15, 2004, at 8. The record shows that Nantong reported low prices for polyethylene resin purchased from a supplier located in Hong Kong. Plaintiffs point out that Nantong’s reported prices for resin were [a certain range] 3 percent below contemporaneous price indices published by commodities exchanges and the verified prices reported by all other respondents participating in the investigation. In this case, Nantong was required to report any relationship with its supplier of polyethylene resin, [ ]. Nantong disclosed its relationship with [the supplier] during verification, claiming that it was able to negotiate the low resin prices because of a “long-established relationship” with [that supplier], and because its arrangements with [the supplier] were subject to certain minimum purchase requirements. Nantong Verification Report, Apr. 15, 2004, at 8. In addition, Nantong reported the following facts regarding its relationship with [the supplier] : 1) In the nine months preceding the POI, Nantong sold PRCBs to [the supplier] as a downstream product valued at more than $ [ ]. *1426 2) Nantong and [the supplier] had a joint customer - [ ]. 3) The [ ] dollars in sales to [the supplier] in the months immediately preceding the POI involved merchandise identical to the merchandise sold to [the joint customer], and all the merchandise sold to [the supplier] was resold to [the joint customer]. 4) Nantong granted preferential prices to [the supplier]. Specifically, the prices charged to [the supplier] were up to [ ] percent less than the prices charged to [the joint customer] for identical merchandise. Commerce was satisfied with Nantong’s reported prices and found no discrepancies between the information reported by Nantong in its questionnaire and the results of the administrative verification. See Nantong Verification Report, Apr. 15, 2004, at 8. Commerce concluded that it could use the prices paid by Nantong to its Hong Kong supplier instead of surrogate values, because Nantong purchased the inputs in arm’s length transactions from a market economy supplier. Plaintiffs challenge Commerce’s use of these reported prices in the final margin calculations, which resulted in a de minimus margin for Nantong. The PRCB Committee argued on the administrative level that prices reported by Nantong for key material inputs such as polyethylene resin 4 (“PE resin”) were unacceptable because they were significantly lower than prices reflected in published price indi-ces or prices reported by all other respondents. Plaintiffs further argued that the Department was obligated by law to investigate whether the relationship between Nantong and its supplier distorted prices paid by Nantong to its supplier. Plaintiffs presented evidence that the reported prices were [a certain range] percent lower than contemporaneous prices published by the London Metals Exchange or Independent Commodity Information Service - London Oil Report (“ICIS-LOR”) and that prices reported by other respondents were tightly clustered (within 5 and 6 percent of the mean, respectively, for both HDPE and LLDPE), as would be expected form a commodity like PE Resin. Plaintiffs argue that because PE Resin is a commodity, and Commerce has recognized that commodity purchasers base their purchasing decisions primarily on price and availability, there exists little price variability in a given market at a given time. For example, in one case Commerce stated that products traded as commodities are “price sensitive and sales are thus often made or lost based on relatively small differences in price.” Notice of Final Determination of Sales at Less Than Fair Value: Stainless Steel Plate in Coils from Taiwan, 64 Fed. Reg. 15,493, 15,504 (Dep’t Commerce, Mar. 31, 1999), app’d on other grounds, Allegheny Ludlum Corp. v. *1427 United States, 24 CIT 1424, 215 F. Supp. 2d 1322 (2000). In contrast, significant differences appear between the prices reported by Nantong and other respondents, who reported PE Resin purchases from market economy suppliers. Plaintiffs also claim that Nantong failed to disclose facts surrounding its relationship with its market economy supplier of polyethylene bags. Finally, Plaintiffs maintain that Commerce’s failure to thoroughly investigate Nantong’s reported prices was contrary to law and its acceptance of those prices unsupported by substantial evidence. Commerce claims that it adequately investigated how Nantong negotiated the price for resin. Commerce accepted Nantong’s explanation that it relied on a website for market prices, which it used to negotiate the price, per metric ton, that it would pay its market economy supplier. Nantong Verification Report, Apr. 15, 2004, CR 176. Commerce also concluded that the prices that Nantong paid arose through market-driven, arm’s length negotiations. Commerce maintains that there is nothing unusual in the discounted prices that Nantong’s market economy supplier provided to Nantong, a long-standing customer. The government argues that a longstanding relationship between Nantong and its supplier and Nantong’s sales of its products back to the supplier do not indicate the type of an affiliation between Nantong and its supplier that would make transactions between the two distorted. Importantly, there is evidence in the record that Nantong and [the supplier] competed against each other for customers in the United States. See Nantong Verification Report, Apr. 15, 2004, at 9. During an investigation, Commerce aims to determine whether a relationship exists between respondents and their suppliers that would distort the reported prices. Commerce has the authority to value raw material inputs used to determine normal value in NME cases using the actual market prices paid by respondents (“ME inputs”) instead of surrogate values. Commerce’s relevant regulation states that “where a factor is purchased from a market economy supplier and paid for in a market economy currency, the Secretary normally will use the price paid to the market economy supplier.” 19 C.F.R. § 351.408(c)(1). Before Commerce can use reported ME input prices, the record must show that such prices are “market determined.” Shakeproof Assembly Components v. United States, 268 F.3d 1376, 1382 (Fed. Cir. 2001). The Preamble to the Department’s regulation notes that although the Federal Circuit upheld Commerce’s practice of using prices paid to market economy suppliers, it “do[es] not view this decision as permitting us to use distorted (i.e., non-arm’s length) prices.” Antidumping Duties; Countervailing Duties: Final Rule, 62 Fed. Reg. 27,296, 27,366 (Dep’t Commerce, May 19, 1997). In non-market economy cases, the statute authorizes Commerce to disregard prices paid to affiliated suppliers if the price “does not fairly reflect” market prices, 19 U.S.C. § 1677b(f)(2), or if *1428 the suppliers sell major raw material inputs at “less than the cost of production.” 19 U.S.C. § 1677b(f)(3). In this case, the record evidence supports Commerce’s position because, even though Nantong reported prices that were lower than the prices reported by other respondents in the investigation, Nantong provided sufficient explanation on how it negotiated lower prices in the normal course of its business. Commerce verified that the transactions between Nantong and its supplier were at arm’s length. Plaintiffs’ comparison of Nantong’s prices to other respondents’ reported prices and to contemporaneous prices published by the London Metals Exchange or ICIS-LOR does not evidence market price distortion. Commerce did investigate the validity of the market prices that Nantong reported and determined that they were the “best available information” for valuing market economy inputs. See Lasko Metal Prods. v. United States, 43 F.3d 1442, 1446 (Fed. Cir. 1994) (finding that the best available information on what the supplies used by the Chinese manufacturers would cost in a market economy country was the price charged for those supplies on the international market). Commerce is not required to scrutinize the reported prices other than satisfactorily verify them. The record supports Commerce’s conclusion that the prices accurately reflected market prices in accordance with 19 U.S.C. § 1677b(c)(l) and 19 C.F.R. § 351.408(c)(1). See Luoyang Bearing Factory v. United States, 26 CIT 1156, 1187, 240 F. Supp. 2d 1268, 1298 (2002) (refusing to apply plaintiff’s mode of examination that required determining “whether the price paid by a PRC bearing manufacturer to a market-economy supplier was market-driven or representative of market-prices.”). 3. Commerce’s Decision to Accept Nantong’s Reported Factors of Production A. Commerce’s Acceptance of Nantong’s Allocation Methodology for HDPE and LLDPE Resin In its Preliminary Determination, Commerce applied “facts otherwise available” under 19 U.S.C. § 1677e(a) to value all of Nantong’s reported sales. Preliminary Determination, 69 Fed. Reg. at 3548-49. Nantong explained that because of its usual business practices, it could not allocate its use of different inputs to the production of its different products. Id. (citing Nantong Letter to Commerce, Jan. 12, 2004, at 2). Commerce preliminarily determined that Nantong’s data, as provided, distorted the amount of raw material inputs consumed in production and that Commerce would have to use facts otherwise available to value the inputs. Preliminary Determination, 69 Fed. Reg. at 3529. Specifically, Nantong reported the same factor usage rate for all five inputs reported for each of the 94 individual products exported to the United States during the POI. Then Nantong submitted a re *1429 vised factors database, which included four sets of material input factors and usage rates for 95 individual products. Nantong reported four different costs of production and normal values for 95 unique bag types. In the preliminary determination, Commerce declined to use any of Nantong’s factors data and based its entire margin on neutral facts available, because it found the factor information distorting. Preliminary Determination, 69 Fed. Reg. at 3549. In its final determination, the Department reversed itself and used Nantong’s data to calculate Nantong’s de minimis final dumping margin. Issues and Decision Mem., Comment 23; Amended Final Determination, 69 Fed. Reg. at 42420. Commerce explained that it verified Nantong’s assertion that its business practices prevented it from reporting an allocation of its factors of production on a product-specific basis. See Nantong Verification Report, at 12 (Def. Tab 16). Commerce verified Nantong’s reported factors of production of resin, ink, and scrap. Nantong advised the Department that the bag production involved mixing resin with scrap and pigment in accordance with a specified recipe stated on “production order slips.” Nantong Verification Report, at 3. The verifiers examined the production order slips, which included production codes for each model ordered, the number of cartons and pieces per carton for each model ordered, resin percentage instructions for each model, and total raw material inputs required to produce the order (whether new or recycled). Nantong Verification Report, at 6. Nantong used only two types of resin: HDPE and LLDPE. Commerce found that Nantong’s allocation methodology for these raw material inputs per kilogram of finished product, based on the total consumption of raw materials in the POI and the total production of finished goods in that period was satisfactory. Nantong explained to the verifiers that the company did not “follow the production order consumption ratios exactly because it need[ed] to take into account recycle scrap in its mixture of resin . . . which can vary between 10 to 20 percent and can go as high as 50 percent.” Nantong Verification Report, at 6-7. “Nantong officials explained that they are unable to provide the Department with more specific information because they do not keep track of that type of information in the accounting system.” Nantong Verification Report, at 12. Nantong also explained that it could provide the actual amount of scrap in inventory by using its end-of-month scrap inventory ledger, but it could not tell how much scrap was consumed for each production run. Nantong Verification Report, at 12. Commerce found no discrepancies resulting from this generalized methodology. It supported its decision by concluding that there was “no evidence that Nantong did not act to the best of its ability in providing the necessary information to calculate a dumping margin.” Issues and Decision Mem., at 78. As a consequence of using Nantong’s reported data, Nantong received a de minimus margin, and was, pursuant to 19 C.F.R. § 351.204(e), excluded from the order. See Amended Final, 69 Fed. Reg. at 42,420 (Tab 3). *1430 Plaintiffs claim that Nantong’s reported factors were not accurate, alleging that Nantong falsely stated that it did not keep business records that would allow a more detailed allocation of product-specific costs. Nantong’s Resp. at 13. Plaintiffs claim that Commerce’s reversal of its approach in the final determination was not warranted by any enhanced data accuracy. They maintain that the varied amount of scrap reported on the order slips does not make the slips unreliable because it is ordinary for PRCB producers to recycle scrap. Plaintiffs argue that Commerce should have used the data based on the “specific recipes” stated in the order slips. Instead, Nantong based its reported factors of production on average resin consumption ratios reported to Chinese customs officials for purposes of claiming import tax exemptions on imported PE Resins consumed in production of exported products: 75 % HDPE and 25 % LLDPE; 5 % HDPE and 95 % LLDPE. Nantong Verification Report, at 7, 12. These ratios are supported by a letter dated August 5, 2001, more than a full year before the October 1, 2002, beginning of the POI. Nantong Verification Report, Ex. 16. Plaintiffs argue that Commerce failed to discuss or acknowledge the basis of Nantong’s reporting in the published notice of final determination. Issues and Decision Mem,., Comment 23. Plaintiffs’ central argument is that Nantong maintained records in the normal course of business that would have allowed preparation of considerably more accurate factors of production data. Because Commerce accepted less accurate factors of production, its final determination was not supported by substantial evidence and was contrary to law. 19 U.S.C. § 1516a(b)(l)(B). To show that Nantong produced imprecise factors of production, Plaintiffs cite the data reported by other respondents in the investigation. While Nantong reported only five raw material input factors for its production of t-shirt bags, other respondents reported no fewer than 15 and as many as 29 raw material inputs for exactly the same type product. Additionally, other respondents reported a broad range of normal values among the various sizes, colors, and printings of t-shirt bags sold in the United States. Nantong reported 4 normal values for the 95 products it exported. PRCB Committee Br. at 38-39. Plaintiffs argue that because Nantong had more accurate data, such as production order slips, Nantong’s reporting of factors of production data presents a concrete example of a respondent “failing to put forth its maximum efforts to investigate and obtain the requested information from its records.” Nippon Steel, 337 F.3d at 1384. Plaintiffs claim that Commerce’s decision to disregard the existence of more accurate data resulted in a calculation unsupported by substantial evidence in the record. Nantong claimed that its resin allocation methodology “in actuality increases its factors-of-production cost .’’Nantong Verification Report, Apr. 15, 2004, at 7. It explained that HDPE resin costs more than LLDPE, so that by over-reporting its HDPE consumption, *1431 Nantong used a conservative methodology for reporting resin consumption. Nantong Verification Report April 15, 2004, at 7. In addition, Nantong argues that the HDPE/LLDPE percentages listed in the production order slips exist only in narrow bands, differing no more than five percentage points. Nantong’s Br. at 12. Thus, Nantong calculates that using Nantong’s market-economy purchase prices, the difference in constructed value using the 75 % HDPE / 25 % LLDPE ratio versus a 70 % HDPE / 30 % LLDPE ratio would yield only a minimal per unit change of [ ]. Nantong’s Br. at 12. Nantong’s calculation of [ ] reflects the most extreme range of difference in the production order slip ratios. Nantong presents other calculations to illustrate that because it always reported the highest percentage for the more expensive HDPE resin in its factors, Plaintiffs’ proposed allocation methodology based on the order slips could only reduce Nantong’s constructed value. Nantong also claims that Plaintiffs have not offered evidence demonstrating that its production order slips would yield a more accurate calculation of the constructed value. At verification, Nantong’s officials said that they could not provide “the amount of actual resin, pigment, or ink consumption by day, production order or model because the mixing workshop does not record amounts inputted during the mixing process.” Nantong Verification Report, at 12. In addition, Nantong did not follow its recipes reflected on the production order slips: “[BJecause the recipes ... do not vary substantially, the employee who mixes the resin will regularly mix multiple production orders at the same time.” Nantong Verification Report, at 3. The production order slips provide only a guide to the mixing of HDPE and LLDPE resins, supporting Commerce’s finding that the company did not maintain records of the actual amount of resin consumed in the production process by model, production run, or other basis. See Nantong Verification Report, at 6. Regarding the use of scrap resin, Nantong explains that although the production order slips have specific percentages, in reality, Nantong does not follow the production order consumption ratios exactly because it needs to take into account recycle scrap in its mixture of resin. . . . [T]he production order slip . . . does not take scrap into account and therefore the percentages reported on the slip do not reflect the actual percentages produced. Nantong Verification Report, at 6. Nantong explains that even if the production order slips were followed with respect to virgin resin consumption, the high scrap percentage would reduce the accuracy of the production order slips. Plaintiffs concede that the presence of scrap renders any allocation methodology less accurate. See PRCB Committee Br., at 40. *1432 Commerce maintains that it had discretion to accept Nantong’s factors of production responses. Section 1677e of the antidumping statute provides that Commerce apply “facts otherwise available” on the record if, among other things, necessary information is not available on the record. 19 U.S.C. § 1677e(a). Furthermore, section 1677m(e) provides that Commerce shall not decline to consider information that is submitted by an interested party and is necessary to the determination but does not meet all the applicable requirements established by the administrative authority . . . if— (1) the information is submitted by the deadline established for its submission, (2) the information can be verified, (3) the information is not so incomplete that it cannot serve as a reliable basis for reaching the applicable determination, (4) the interested party has demonstrated that it acted to the best of its ability in providing the information and meeting the requirements established by the administering authority or the Commission with respect to the information, and (5) the information can be used without undue difficulties. 19 U.S.C. § 1677m(e). Commerce is charged with “determining current margins as accurately as possible.” Rhone-Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed. Cir. 1990); Shakeproof, 268 F.3d at 1382. When factors of production are identified, the statute directs the Department to use the “best available information” to value each one. 19 U.S.C § 1677b(c)(4). “In determining the valuation of the factors of production, the critical question is whether the methodology used by Commerce is based on the best available information and establishes an-tidumping margins as accurately as possible.” Shakeproof, 268 F.3d at 1382. Plaintiffs cite to other cases where the Department requested product-specific cost data, requiring reasonable cost allocations among various products under investigation if the respondent’s normal accounting records did not contain product-specific cost information. See, e.g., Certain Cold-Rolled Flat-Rolled Carbon Quality Steel Products from Turkey, 65 Fed. Reg. 15123 (Dep’t Commerce, Mar. 21, 2000) (noting that frequently respondents’ normal cost accounting systems do not differentiate among products or provide product-specific costs and that Commerce’s “consistent practice” is to require reasonable allocation methodologies to achieve product-specific costs.”). In reviewing whether Commerce’s decision is supported by substantial evidence on the record, the court “tak[es] into account the *1433 entire record, including whatever fairly detracts from the substanti-ality of the evidence.” Atlantic Sugar, Ltd. v. United States, 744 F.2d 1556, 1562 (Fed. Cir. 1984). The court “will find a determination unlawful where Commerce . . . relied on inadequate facts or reasoning, or failed to provide an adequate basis for its conclusions.” Rhone-Poulenc, Inc., 20 CIT 573, 575, 927 F. Supp. 451, 454 (1996) (citations omitted). In this case, Commerce accepted the average non-model-specific ratios provided by Nantong. Commerce specifically verified Nantong’s reported factors of production of resin, ink, and scrap and found Nantong’s methodology of allocating raw material inputs per kilogram of finished product, based upon the total consumption of raw materials in the POI and the total production of finished goods in that period; reliable. It also found “no evidence that Nantong did not act to the best of its ability in providing the necessary information to calculate a dumping margin.” Issues and Decision Mem., at 78. Commerce did not find the production order slips more accurate or reliable because verification revealed that the slips were not strictly followed in production. Commerce found that Nantong did not keep any production or accounting records that tracked costs on a model-specific basis. See Nantong Verification Report, at 12; Issues and Decision Mem., at 77-80. As a general rule, Commerce has the discretion and “authority to determine the extent of investigation and information it needs.” PPG Indus., Inc. v. United States, 978 F.2d 1232, 1238 (Fed. Cir. 1992). In this case, Commerce’s conclusion was supported by substantial evidence in the record, and the court will not re-weigh the evidence. B. Nantong’s Allocation of Ink Consumption Plaintiffs argue that Commerce also improperly accepted Nantong’s ink consumption data. The record shows that Nantong produced [ ] styles of bags that contained printing, that the print images varied in total print area, and that Nantong reported using red, blue, green, yellow, and black ink. Commerce accepted Nantong’s two consumption ratios for ink; one of the values was zero. The Department examined the size of bags and the number of colors used in printing and found that these factors “are not necessarily an accurate indicator of ink consumption.” Issues and Decision Mem., at 80. Commerce also found no “correlation between bag size, the number of printed sizes, and ink consumption.” Id. It concluded “that the size of the bag and the number of colors are not necessarily an accurate indicator of ink consumption.” Final Determination Mem., Comment 23. Plaintiffs argue that Commerce had model-specific image design and color information for each product and that Nantong could have developed a model-specific allocation methodology for black and color inks. For instance, Nantong could have reported more precise factors of production based on information maintained in the normal course *1434 of business, such, as the production order slips. Plaintiffs argue that Nantong failed “to put forth its maximum efforts to investigate and obtain the requested information from its records,” and Commerce is thus authorized to adopt adverse inferences when selecting facts available, pursuant to 19 U.S.C. § 1677e(b). The Department’s task is to “articulate a satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made.’ ” Motor Vehicle Mrfs. Ass’n of United States v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168 (1962)). In this case, the Department found no correlation among size, the number of printed sides, and the number of colors, and concluded that there was no basis to allocate ink consumption to various bag types. Lacking a basis for allocation, Nantong could only assign a single ink consumption amount to printed bags. Although there were variations among Nantong’s printed bags, Commerce found Nantong’s allocation methodology reasonable given the information that the company kept in its normal course of business. Plaintiffs have not demonstrated a more discernable pattern for a different allocation. Commerce’s conclusion cannot be disturbed unless unsupported by substantial evidence or contrary to law. Fujitsu, 88 F.3d at 1038. 4. Commerce’s Application of Adverse Facts Available to Hang Lung In the final determination, Commerce applied a surrogate value for cardboard inserts using the weighted-average unit value of cardboard inserts imported into India during the POI. Issues and Decision Mem., at 97, 100-01. When providing Commerce with information regarding the cardboard inserts, Glopack explained that “certain companies use untreated cardboard” inserts while others “use treated cardboard” ones. See Surrogate Value Submission, Nov. 20, 2003, at 6. The investigation respondents did not specify which type they used in their production. Commerce therefore relied on a combination of HTS subheading 4810.29.00 (treated cardboard) and 4805.80.09 and 4805.70.09 (untreated cardboard) to value all inserts. The PRCB Committee argues that Commerce failed to analyze and explain its conclusion that HTS 4810.29.00 is the correct tariff classification for valuing coated cardboard inserts and maintains that HTS 4810.39.09, which includes lower-quality treated inserts, might be more appropriate. PRCB Committee Br. at 47. Specifically, Plaintiffs argue that the Department failed to make a reasonable connection between the facts on the record and its conclusion, and, therefore, its selection of the surrogate value for cardboard inserts was contrary to law and unsupported by substantial evidence. PRCB Committee Br. at 47. During the investigation, Commerce instructed participating respondents to “describe each type and grade of material used in the *1435 production process.” Rally replied that the cardboard inserts were “low grade, recycled cardboard inserts.” See, e.g., Rally’s Section D Response, Oct. 2, 2003, at 6. Later, Glopack, Rally, and Hang Lung provided different information, stating that “the treated cardboard inserts . . . [are] higher quality cardboard that can be used for graphic purpose, but [they are] used by the respondents for inserts.” Surrogate Value Submission, Nov. 20, 2003, at 6. The respondents further provided certain Indian import data showing that “[t]he treated cardboard inserts used by respondents are classified under US HTS [sic] item 4810.29.10.00.” Surrogate Value Submission, Nov. 20, 2003, at 6. The PRCB Committee Plaintiffs requested that Commerce use the value for HTS subheading 4810.39.09, rather than HTS subheading 4810.29.00, as the proper surrogate value for treated cardboard inserts, claiming that HTS subheading 4810.29.00 included higher quality inserts than those included under HTS subheading 4810.39.09, and that inconsistent statements made Glopack’s assertion that it used higher quality inserts incredible. In its final determination, Commerce did not address this inconsistency in the respondents’ responses. However, Commerce explained that “[b]ecause none of the respondents specified what type of cardboard insert (treated or untreated) it used in the production of subject merchandise, [it] applied our methodology . . . valuing] cardboard inserts using the weighted-average of the surrogate values for treated and untreated cardboard inserts.” Issues and Decision Mem., at 100. As a result, for most respondents, Commerce used “the weighted average of the values ... for HTS subheadings 4810.29.00, 4805.70.09, and 4805.80.09.” Issues and Decision Mem., at 100. Commerce did not explain, however, why the selection of subheading 4810.29.00 over subheading 4810.39.09 was more appropriate, stating that petitioners “have not demonstrated that the use of HTS subheading 4810.29.00 is inappropriate or that the use of HTS subheading 4810.39.09 is more appropriate.”Issues and Decision Mem., at 100. The PRCB Committee argues that the Department failed to address the official descriptions of the competing tariff headings. Heading 4810 is defined as: “Paper and paperboard, coated on one or both sides with kaolin (China clay) or other inorganic substances.” The heading is further divided into the following relevant divisions: 1) subheadings 4810.21 and 4810.29 encompasses “Paper and paperboard of a kind used for writing, printing or other graphic purposes, of which more than 10 percent by weight of the total fiber content consists of fibers obtained by a mechanical process” and 2) subheadings 4810.31, 4810.32 and 4810.39 cover “Kraft paper and paperboard, other than that of a kind used for writing, printing, or other graphic purposes.” See Surrogate Value Submission, Mar. 22, 2004 (citing to Chapter 48 of the Harmonized Tariff Schedule of the United States, which is harmonized to the six-digit level with the Indian HTS.). *1436 In rebuttal, the government argues that the PRCB Committee Plaintiffs did not demonstrate how subheading 4810.39.09 is more appropriate for valuing treated cardboard inserts. .The Department explains that the respondents did not provide specific detail regarding whether the cardboard inserts they used were “treated” or “unreated,” let alone specific types of inserts they used. Commerce therefore selected a weighted average of the HTS categories for both treated and untreated cardboard inserts as the “best available information” for cardboard inserts. See 19 U.S.C. § 1677b(c)(l). However, as accurately pointed out by the PRCB Committee Plaintiffs, the record shows that the description of the treated cardboard supplied by the respondents during the investigation changed, and Commerce did not specifically address that change in its final analysis. While Commerce has the authority to use “best information available” when it finds petitioners’ submissions incomplete or inconsistent, see 19 U.S.C. § 1677e, it also must, “to the extent practicable, provide [petitioners] with an opportunity to remedy and explain the deficiency.” 19 U.S.C. § 1677m(d). Plaintiffs do not claim that they lacked an opportunity to show Commerce that the use of subheading 4810.39.09 was more appropriate. In addition, although the detailed submission by Glopack, Rally, and Hang Lung, which stated that the treated cardboard inserts respondents used were classified under subheading 4810.29.10.00 differed from Rally’s earlier response, it was not unreasonable for Commerce to rely on a more detailed explanation, especially since Commerce had to use “the best information available.” See Surrogate Value Submission, Nov. 20, 2003, at 6. Commerce chose one among several HTS categories to value treated cardboard inserts, and the respondents’ submission supported that choice. Thus, the PRCB Committee Plaintiffs did not show that record evidence did not support Commerce’s methodology. 5. Selection of the Surrogate Value for Black and Color Inks In response to Commerce’s request that the respondents provide publicly available information for valuing their factors of production, Glopack offered a list of average prices for flexographic and gravure inks, in black and other colors, from Hindustan Inks and Resins, Ltd. (“Hindustan”), an individual Indian manufacturer. Glopack Surrogate Value Submission 1 (Nov. 20, 2003). Glopack argued that these prices should be used as surrogate values for ink because they included only those inks used in printing on polyethylene retail carrier bags, namely flexographic and gravure inks, and because they were color-specific. Id. at 3-5. Commerce noted that the proposed prices were not accompanied by source documentation. In the preliminary determination, Commerce concluded that India represented the appropriate surrogate for the PRC and relied on publicly available Indian import statistics for valuing black and color inks. See Preliminary Determination, 69 Fed. Reg. at 3549. In the Factor Valuation Memorandum, Commerce explained that it used *1437 cumulative Indian import statistics as surrogate values for each material input. Commerce also explained why it did not use the surrogate value data submitted by Glopack to value the ink. Commerce explained that the Indian import statistics were “reliable” because they were “based on the sum of all imports into India during the POI.” Commerce did not accept the Hindustan data, finding that it was derived from only one producer and varied greatly from the import statistics. Factor Valuation Mem., at 3. Following the preliminary determination, Glopack submitted U.S. import statistics of gravure and flexographic printing inks, price lists for United States sales from one of Glopack’s own importers for gravure and flexographic printing inks used to print plastic bags, and a price list from a Malaysian company for sales to Vietnam of gravure printing inks used to print plastic bags. Glopack Submission, Mar. 22, 2004, at 3-5. Another respondent in the investigation provided worldwide average data from the United Nations. Using these various data, Glopack contended that the Indian import statistics that Commerce used were less accurate than the Hindustan data. In the Final Determination, Commerce continued to value ink inputs using the publicly available import statistics that it used in the preliminary determination. It determined that the Indian import statistics presented the best available surrogate value because they were 1) sufficiently product-specific, 2) country-wide, 3) tax and export exclusive, and 4) contemporaneous with the period of investigation. Commerce did not find the Hindustan data more accurate or representative than the Indian import statistics. Id. The basket category tariff provisions Commerce used were based on the Indian HTS items 3215.11.90 (“Other black printing ink”) and 3215.19.90 (“Other printing ink & printing colors”). These broad basket provisions include a large number of products Plaintiffs did not use to produce the subject merchandise. In addition to the flexographic and gravure printing inks, they include: 1) all types of printing inks other than newspaper inks, rotary inks, and screen and lithographic printing inks; and 2) gravure and flexographic printing inks used in applications other than printing on polyethylene bags, including more costly applications of printing on paper, coated paper, and cardboard; 3) printing inks of higher quality than necessary for printing on polyethylene bags; 4) printing inks of different concentrations, such as printing ink jellies and paste, necessarily more expensive on a per-unit basis; and 5) for color inks a single combined value for all distinct colors, regardless of the relative value of each individual color. 1. Glopack’s Arguments Glopack argues that the Department’s selection of surrogate values for the color-specific factors of production for flexographic and gravure printing inks used by Plaintiffs Sea Lake and Rally to *1438 manufacture the subject merchandise did not meet the statutory test for two reasons: (1) the surrogate prices selected by the Department were derived from basket category tariff provisions not specific to the ink type used to produce the subject merchandise and 2) the Department rejected alternative surrogate values Plaintiffs submitted that consisted of the average unit values based on actual Indian sales of color-specific flexographic and gravure printing inks used to manufacture polyethylene bags. Regarding the specificity of the data Commerce used, Glopack argues that the basket tariff provisions were overly broad in several respects. First, as a residual basket category, the data for printing types included all types of specialty and computer printing inks (in addition to the gravure and flexographic) valued substantially higher than gravure and flexographic printing inks used to print polyethylene bags. Plaintiff argues that the import statistics also used both liquid printing inks of normal concentrations used by Plaintiffs to print plastic bags and more highly concentrated inks in jelly and paste forms used in other applications. The record shows that ink’s per unit value necessarily increases as the ink concentration of the product increases, thus making the data used inaccurate. Finally, the basket import statistics for color ink fail to account for the different colors of the printing inks. The cost of red, violet, and pink tone printing inks substantially exceeds those of blue, yellow and green tone. Glopack pinpoints a significant discrepancy between the alternative surrogate values: the Hindustan data showed black ink valued at $1.96/kg and the most expensive color ink at $ 4.27 per kg, whereas the basket categories had values of $7.63/kg for black ink and $12.47/kg for all color ink. In addition, Glopack asked Commerce to compare the Hindustan data with U.S. import statistics for calendar years 2000, 2001, 2002, and 2003 (HTSUS 3215.11.00.20, black flexographic; 3215.11.00.20, black gravure; 3215.19.00.20, color flexographic; and 3215.19.00.30, color gravure). Pub. Doc. # 424 (Tab 7). Plaintiffs argue that, unlike the Indian tariff provisions, these U.S. tariff provisions are not basket categories, but are limited to gravure and flexographic printing inks, the types of ink specifically used in production of the subject merchandise. Specifically, Glopack argues that the U.S. import statistics for calendar years 2002 and 2003 that it provided in its surrogate value submission, showed U.S. import prices substantially closer to the Hindustan prices than to the basket category import prices. For example, Plaintiffs calculated that the calendar year 2003 combined average unit value of U.S. import statistics for black flexographic and black gravure inks came to $3.74/kg. The Hindustan data provided $1.96/kg for black flexo-graphic and gravure inks and the Department’s surrogate value for black gravure and flexographic ink was $7.63/kg. Glopack also submitted a signed price list from a Malaysian exporter of gravure inks to Vietnamese producers of polyethylene bags. *1439 The price list included the sale of white, red, and blue gravure inks ranging from $2.00/kg to $2.25/kg. Glopack calculated that the average Hindustan price reported for red ink was $2.47/kg, for blue ink $2.88/kg and for white ink $2.10/kg. In comparison, the average unit value for color inks for the basket category Indian tariff provision was $ 10.22 higher than the highest priced color ink offered by the Malaysian producer. Consequently, Glopack maintains that the Malaysian price list corroborates the color-specific average sales prices reported by Hindustan and confirms aberration of the basket category import statistics and does not reflect commercial prices of flexographic and gravure printing inks. Such data also shows that Commerce used distorted data that includes all other types, qualities, and concentrations of printing inks in addition to flexographic and gravure printing inks. Glopack argues that Commerce erroneously cited to the U.N. data to support the reliability of the basket category import statistics because Commerce’s analysis focused on the U.N. data for India derived from official Indian statistics and was “comparable [to the official Indian import statistics] . . . with regard to both black and color ink.” Issues and Decision Mem., at 48. While engaging in this circular reasoning, Commerce completely ignored the reason why the respondent submitted the data - to highlight the great disparity between the average unit Indian values and the weighted-average global unit average import price of black and color flexographic inks. In this case, the color-specific average Indian prices for flexographic and gravure printing inks reported by Hindustan constituted the most specific surrogate value information for flexographic and gravure printing inks because the data provided surrogate values on a color-specific basis and included the types of ink used in production of the subject merchandise. Issues and Decision Mem., Comment 9, at 46. Glopack argues that the administrative record indicated that a significant price differential existed between different colored inks. In fact, the record shows color inks more expensive than black ink, and certain color inks are significantly more expensive than others. Average ink prices ranged from 94.35 rupees per kg for black ink to 205.12 rupees per kg for purple ink. Likewise, a significant price differential among color inks was evident in the two U.S. ink price lists and the Malaysian price list the respondents submitted. In the final determination, Commerce concluded that while “Hindustan’s pricing data is more specific to black and color inks, the data is less preferable in terms of the other factors we considered because the data is not contemporaneous, the pricing data is based on an experience by a single Indian producer of ink, and, therefore, not completely representative of the cost of this input, and the pricing data has little or no supporting documentation.” Issues and Decision Mem., at 46-47. The import data Commerce used is more contemporaneous than the Hindustan data. The POI in this case spans from *1440 October 1, 2002 to March 31, 2003. The Hindustan data covers the six-month period immediately after the POI, April 1, 2003 to September 30, 2003. Id. In response, Plaintiffs maintain that the specificity of the Hindustan data takes precedence over other factors, such as contemporaneity. In addition, Glopack argues that the Hindustan data is reasonably contemporaneous and covers the six-month period immediately after the POI. Plaintiffs make one appealing argument. They claim that indexing the reported prices to the period of review - an adjustment routinely made in surrogate value calculations - can remedy any concerns about the contemporaneity of the data. Glopack argues that even so, to calculate certain surrogate value factors, Commerce applied inflation adjustments by using the Indian wholesale price index (“WPI”) data reported in the International Financial Statistics published by the International Monetary Fund. Glopack Br. at 24. In addition, the Hindustan data post-dates the period of review, and therefore, any WPI adjustment necessary to account for inflation would reduce, not increase, the prices Hindustan reported. Glopack Br. at 24. The unadjusted Hindustan data possibly overstates the relevant ink prices. Glopack argues that when weighing the contemporaneity factor against the specificity of import statistics, the balance should tip toward the specificity factor because Commerce’s ability to index the data can mend the modest shortcomings in the Hindustan data’s contemporaneity. The Department therefore should have chosen the Hindustan data as superior in product-specificity. Based on this reasoning, Glopack claims that the Department’s choice of data is not supported by substantial evidence on the record. Glopack also argues that Commerce incorrectly characterized the color-specific Hindustan data as a series of prices quotes. Commerce explained that it considered the Hindustan data deficient because it did not reflect “numerous transactions between many buyers and sellers because the experience of a single producer is less representative of the cost of an input in a surrogate country.” Issues and Decision Mem., at 46. Glopack points out that the Hindustan data concerned average unit prices of sales based on actual sales transactions of flexographic and gravure printing inks for the printing on plastic bags in India from April 2003 to September 2004. The record shows that the Hindustan data, was derived from actual sales transactions widely applicable throughout India. According to a Hindustan official, Hindustan.sold those products throughout India, and its sales accounted for approximately 30% of the Indian market. The Hindustan data was derived as follows: the company reviewed all sales of flexographic and gravure printing inks made to home market customers who purchased ink for the purpose of printing on polyethylene bags during the April 2003 - September 2003 period; then on a color-by-color basis, the company aggregated the total sales quantity of each ink color (in kilograms) and the total sales value of each ink color (in rupees); then the total sales value of *1441 each color ink was divided by the total sales quantity of that color ink. The total average sales price formed a weighted average rupee per kilogram rate derived from the total sales value of all color inks (including black) divided by the sales quantity of all color inks (including black). Glopack also notes that nothing in the record addresses the size of the Indian market for printing inks. There was no basis for Commerce to conclude that the small quantity of ink imports constituted a reliable domestic price for printing inks in India. Plaintiffs request that the court take judicial notice of new information relating to the size of the Indian market for printing inks, obtained from the website of the All India Printing Ink Manufacturers Association, Ltd., reporting that the annual domestic Indian market for flexographic and gravure printing inks is approximately 32,100 metric tons. Based on this new information, Glopack argues that even if all imports in the basket category Commerce used were flexographic and gravure printing inks, these imports constitute only 3.64 % of the Indian market for gravure and printing inks. Meanwhile, Glopack states that Hindustan’s sales accounted for 30% of the Indian market. Consequently, the Department’s determination that the basket category import statistics better represented the Indian prices than the color-specific average unit prices derived from actual sales transactions of flexographic and gravure printing inks in India, as reported by the largest Indian printing inks seller, is not supported by substantial evidence in the administrative record. In addition, Glopack argues that the import statistics used by Commerce are unreliable because they are inconsistent over time. Specifically, “when viewed over time, the basket category import statistics, being of relatively small volume and covering numerous types of printing inks of varying concentrations and quality, are highly volatile and are not representative of domestic Indian ink pricing.” Glopack Br., at 29. In support, Glopack presents new information, not considered on the administrative level, of quarterly import statistics for basket tariff provisions for each quarter from January 1999 to June 2004 obtained from the official Indian import statistics and reported by Global Trade Atlas. Having calculated quarterly average unit values for each provision, Glopack argues that its analysis exposes the unrepresentative nature of the basket category import statistics of the Indian pricing for gravure and flexographic printing inks and demonstrates that the import statistics for any particular period are unrepresentative of import pricing over time. As a general matter, Glopack maintains that due to the volatility of import statistics over time, they are too inaccurate to provide a reliable basis for the calculation of surrogate values in this case. Finally, Glopack points out that the petitioners on the administrative level declined to furnish the prices that they paid for the flexographic and gravure printing inks that they used in printing on *1442 the polyethylene bags in their U.S. or foreign facilities. The absence of such readily available information further confirms that the average unit values of the basket category import statistics greatly overstated the actual commercial prices of flexographic and gravure printing inks used to produce the subject merchandise. 2. Government’s Arguments Commerce explained in its final determination that it chose the Indian import statistics because they were more “reliable,” as they were “based on the sum of all imports into India during the POI”. Factor Valuation Mem., at 3. Commerce found the Hindustan data unreliable because it was “not completely representative of the cost of this input” and “the experience of a single producer is less representative of the cost of an input in a surrogate country.” Issues and Decision Mem., at 46-47. Commerce argues that while more product-specific, the Hindustan data was not contemporaneous with the PIO and represented the experience of only a single Indian producer, and had little or no supporting documentation. Issues and Decision Mem., at 46-47. Therefore, Commerce’s determination that the Indian import statistics presented the best available information for use as a surrogate value for black and color ink is based on substantial evidence, consistent with the anti-dumping statute and Commerce’s practice, and is in accordance with law. In cases involving nonmarket economies such as the PRC, Commerce looks to surrogate value sources from market economy countries at the same level of economic development for the value of the factors of production. See 19 U.S.C. § 1677b(c)(l). Commerce needed to find surrogate sources for the values of black and color ink. Commerce solicited comments from all interested parties on possible sources for surrogate values. In response to Commerce’s solicitation, Glopack contended that Commerce should use data that it provided from a single Indian producer, Hindustan. Commerce considered the submissions by Glopack and other respondents and found that they did not support use of the Hindustan data over the Indian import statistics. Issues and Decision Mem., at 39-40, 46-49. Commerce found that while Indian prices were higher than the worldwide average, the statutory mandate required it to determine surrogate values based on data from a country at the same level of economic development as the PRC, despite an inconsistency with worldwide average import prices. Issues and Decision Mem., at 48. Commerce found it could not use the other data Glopack submitted because the data came from individual producers, was derived from importing countries not economically comparable to the PRC, and was not publicly available. Issues and Decision Mem., at 47. Commerce concluded that the United States import statistics confirmed that the Indian import statistics were not distorted for combining several colors within a single import category because *1443 the United States import prices for ink specific categories did not substantially differ from the import prices for basket categories that included several ink types. Issues and Decision Mem., at 47-48. Commerce criticized the Malaysian data for not being contemporaneous with the POI and for not following Commerce’s preference for publicly available data since it was based on a single producer’s experience. 3. Analysis The court decides whether Commerce’s choice of the surrogate value was supported by substantial evidence and is in accordance with law. “In determining the valuation of the factors of production, the critical question is whether the methodology used by Commerce is based on the best available information and establishes antidump-ing margins as accurately as possible.” Shakeproof, 268 F.3d at 1382. The statute requires that the Department’s “valuation of the factors of production shall be based on the best available information regarding the values of such factors.” 19 U.S.C. § 1677b(c)(l). “While § 1677b(c) provides guidelines to assist Commerce in this process, this section also accords Commerce wide discretion in the valuation of factors of production in the application of those guidelines.” Nation Ford Chem. Co. v. United States, 166 F.3d 1373, 1377 (Fed. Cir. 1999); see also Anshan Iron & Steel Co. v. United States, 2003 WL 22018898, at *3 (CIT, July 16, 2003); Timken Co. v. United States, 25 CIT 939, 166 F. Supp. 2d 608, 616 (2001). When assessing which particular surrogate represents the “best available information” for the factors of production reported to Commerce, the Department relies on surrogate values which are: 1) non-export average values, 2) most contemporaneous with the period of investigation, 3) product-specific, and 4) tax exclusive. Issues and Decision Mem., at 46. Commerce uses product-specific surrogate values, seeking surrogates most comparable in terms of design or materials to the actual input consumed by the Chinese respondents in production of the subject merchandise. See, e.g., Notice of Final Determination of Sales at Less Than Fair Value: Bicycles from China, 61 Fed. Reg. 19026, 19030 (Dep’t of Commerce, Apr. 30, 1996); Certain Helical Spring Lock Washers from China; Final Results of Anti-dumping Admin. Review, 61 Fed. Reg. 41994, 41996-7 (Dep’t of Commerce, Aug. 13, 1996) (it is Commerce’s practice to seek surrogate prices that most closely reflect the specific grade and physical characteristics of the input used). Plaintiffs cite to certain cases, involving activated carbon as a raw material input, where Commerce rejected the use of average unit values obtained from Indian import statistics as the appropriate surrogate because Indian import statistics broadly covered all grades and types of activated carbon. Commerce instead relied on an alternative surrogate value which more closely corresponded to the activated carbon incorporated into the subject merchandise. See, e.g., *1444 Notice of Final Determination of Sales at Less Than Fair Value and Negative Final Determination of Critical Circumstances: Certain Television Receivers from China, 69 Fed. Reg. 20594 (Dep’t of Commerce, Apr. 16, 2004); Notice of Final Determination of Sales at Less Than Fair Value'. Polyvinyl Alcohol from China, 68 Fed. Reg. 47538 (Dep’t of Commerce, Aug. 11, 2003). This Court has affirmed the Department’s selection of a surrogate value more specific than the average price in the Indian index numbers. See Kerr-McGee Chem. Corp. v. United States, 985 F. Supp. 1166, 1176 (1997) (affirming Commerce’s decision to use a surrogate value of the type of manganese with the ore content “ ‘more comparable to the ore used by the Chinese producers than the [surrogate price for] ore with the higher manganese content’ ” (internal citation omitted)). In previous cases, Commerce recognized that import statistics based on a basket tariff category are inappropriate if a more representative alternate surrogate is available. See, e.g., Notice of Preliminary Determination of Sales at Less Than Fair Value: Tetrahydrofurfuryl Alcohol from China, 69 Fed. Reg. 3887, 3892 (Dep’t of Commerce, Jan. 27, 2004); Freshwater Crawfish Tailmeat from China', Final Results of New Shipper Review, 64 Fed. Reg. 27961, 27962 (Dep’t of Commerce, May 24, 1999) (“[I]mport data from basket categories can be too broad to be reliable.”). This Court has ruled that Commerce can rely on Indian import statistics as the basis for a surrogate value only “after concluding that they [the import statistics] are based on commercially and statistically significant quantities.” Shanghai Foreign Trade Enter. Co. v. United States, 28 CIT_, 318 F. Supp. 2d 1339, 1352-53 (2004) (rejecting Commerce’s reliance on Indian import statistics for pig iron as surrogate value because import volume constituted only 1,132 metric tons of product, a quantity determined to be too small to reliably represent India market value). In this case, Commerce chose to use the Indian import statistics in accordance with its preference to use “countrywide data” when available. Issues and Decision Mem., Comment 2. Commerce considered the experience of one company, Hindustan, less representative of the cost of an input in an entire surrogate country. Issues and Decision Mem., at 46. The record shows that while the Hindustan data is more product-specific as it provides values for those input products valued in this case, it represents only 30% of the Indian sales of those products. Glopack moved to submit new information in support of its argument that the Indian import statistics account for even a smaller percentage of the sales of the relevant inks. However, the new information presented by Glopack relating to the size of the Indian printing inks market from the All India Printing Ink Manufacturers Association, Ltd.’s website was not supplied during the administrative review, and the court will not consider this evidence. See Atcor, Inc. v. United States, 11 CIT 148, 154, 658 F. Supp. 295, 300 (1987) (“In reviewing agency action, the Court must base its de- *1445 cisión upon the administrative record. New evidence may not be received.”); see also PPG Indus., Inc. v. United States, 5 CIT .282, 284 (1983) (“Thus, any data or memoranda not presented to, obtained by, considered or relied upon by [the agency] . . . [is] not part of the record.”). Commerce also found that the Hindustan data was not supported by sufficient documentation. Issues and Decision Mem., at 46-47. While Glopack presented several factors that detract from Commerce’s finding that the Indian import statistics were more accurate, those factors pertain only to the data’s product specificity. This court cannot substitute its own evidentiary evaluation for Commerce’s. Finally, Glopack has not shown that based on prior cases that Commerce acted contrary to the law when using the Indian imports statistics. 6. Use of Surrogate Values for Sea Lake’s and Glopack’s Purchases of Inputs from a Hong Kong Trading Company. Sea Lake provided Commerce detailed listings of its market economy purchases of raw material inputs used to produce the subject merchandise. The listings included purchases of raw material inputs from Hong Kong suppliers, including color concentrate, color ink, and cardboard inserts, bought with Hong Kong or U.S. dollars. In the preliminary determination, Commerce valued Sea Lake’s factors of production for these inputs using Sea Lake’s reported Hong Kong prices. During its verification of Sea Lake’s responses, Commerce found that a substantial volume of Sea Lake’s total market economy purchases of ink and color concentrate from Hong Kong suppliers was of Chinese origin. In the final determination, Commerce reversed its position and valued Sea Lake’s factors of production for inks, colorants, and cardboard inserts using surrogate values. Issues and Decision Mem., Comment 4. Commerce determined that its regulation requiring the valuation of reported factors of production with the market economy purchase prices of those inputs did not apply to inputs produced in China. Plaintiffs argue that the administrative record shows that Sea Lake purchased the Chinese-origin inputs in Hong Kong and had them shipped from Hong Kong to their factory in Shenzhen. Glopack argues that Sea Lake’s market economy purchases were not intra-China transfers transacted in Hong Kong dollars, but were transactions that left the NME stream of commerce, and were physically moved from China to Hong Kong and then re-exported back to China. Sea Lake owns a “processing” factory in Shenzhen, Sea Lake Shenzhen, which has a permit to import materials for processing and re-export. Issues and Decision Mem., at 4-5. Sea Lake Shenzhen is just a factory and is allowed to process imported materials only. The raw materials are purchased by the [sic] Sea Lake *1446 Hong Kong and the processed goods are sent to Sea Lake Hong Kong: The factory does not have any sales revenue and makes only processing fees to cover the labor wages and rent and utility expenses. Sea Lake Dec. 22, 2003 Supp. Response, at SA-3. Sea Lake provided Commerce with the information that its purchase of domestic Chinese raw material inputs for use in its production was limited. By operation of law, the factory is required to import most of its raw material inputs, in this case from Hong Kong, and then export the finished products back to Sea Lake in Hong Kong. Sea Lake Hong Kong, and not its Chinese factory, was responsible for the purchase of raw materials, including inks and color concentrate. Sea Lake Verification Report, at 4, Pub. Doc. # 447. In the pending appeal, Glopack argues that Commerce’s determination is contrary to law because the agency’s own regulation and longstanding administrative practice require the Department to use actual import prices to value reported factors of production if the inputs were purchased in a market economy country with market economy currency, without regard to the country of origin of the imported merchandise. Commerce’s regulation provides “where a factor is purchased from a market economy supplier and paid for in a market economy currency, the Secretary normally will use the price paid to the market economy supplier.” 19 C.F.R. § 351.408(c)(1). The preamble to the regulation provides that “where the NME producer purchases inputs from a market economy producer and these inputs are paid for in a market economy currency, we would use the price paid by the NME producer to value that input.” Antidumping Duties; Countervailing Duties: Final Rule, 62 Fed. Reg. 27,296, 27,366 (Dep’t Commerce, May 19,1997). In its final decision, Commerce cited the preamble of 19 C.F.R. § 351.408(c)(1) and interpreted it as applying the regulation “to those inputs which were produced in a market economy country.” Issues and Decision Mem., at 26. Commerce concluded that, given the language in the preamble, the regulation did not require the use of the actual prices paid for inputs that were produced in a nonmarket economy. Issues and Decision Mem., at 25-26 (“[P]rices of products that originate in a NME country should not be used because of the inherent distortions involved in an economy that is not controlled by market forces.”). “ ‘[A]n agency’s interpretation of its own regulation[s] is entitled to deference’ when the language of the regulation is ambiguous or the regulation is silent about the issue at hand.” Timken Co., 25 CIT at 943 n.2, 166 F. Supp. 2d at 615 n.2 (citing Christensen v. Harris County, 529 U.S. 576, 588 (2000)). In this case, the regulation’s plain language provides that where the input is “purchased” in a market economy country with market economy currency from a market *1447 economy “supplier,” the purchase price is used to value the reported factor of production. See 19 C.F.R. § 351.408(c)(1). The term “supplier,” however, is open to interpretation because it arguably could either mean “vendor” or “producer.” See Def. Br. at 29. In past cases, Commerce has interpreted 19 C.F.R. § 351.408(c)(1) as not disqualifying transactions based on the goods’ country of origin. See Issues and Decision Mem. for the Antidumping Duty Investigation of Certain Color Television Receivers from the People’s Republic of China, Comment 8, at 39 (Dep’t Commerce, Apr. 16, 2004) (“We agree with the respondents that we should not reject prices of goods purchased in Hong Kong based on the country of origin of the goods.”); Issues and Decision Mem. for the Antidumping Duty Administrative and New Shipper Reviews on Certain Preserved Mushrooms from China, 88 ITADOC 31204, Comment 7 (Dep’t of Commerce, June 11, 2001) (stating that 19 C.F.R. § 351.408(c)(1) “does not require that the nonmarket economy respondent establish in which particular country the factor of production was produced, only that it was obtained from a market economy supplier.”). As a rule of thumb, agencies are required to interpret and apply regulations consistently from case to case. See Fort Stewart Sch. v. Fed. Labor Relations Auth., 495 U.S. 641, 654 (1990); Torrington Co. v. United States, 82 F.3d 1039, 1049 (Fed. Cir. 1996); China Steel Corp. v. United States 27 CIT_, 264 F. Supp. 2d 1339, 1354 (2003). Commerce may reach different determinations in separate administrative reviews, but it must either employ the same methodology or give reasons for changing its practice. Cinsa, S.A. de C.V. v. United States, 21 CIT 341, 349, 966 F. Supp. 1230, 1238 (1997) (involving challenge to Commerce’s method of calculation for cost of production and constructed value); Hussey Cooper, Ltd. v. United States, 17 CIT 993, 997, 834 F. Supp. 413, 418 (1993) (citations omitted) (“It is ‘a general rule that an agency must either conform itself to its prior decisions or explain the reasons for its departure.’ ”).When Commerce departs from its prior decision, it must provide a reasoned explanation for its departure in order for the court to judge the consistency of the administrative action. Hussey Cooper, Ltd., 17 CIT at 998, 834 F. Supp. at 419; see RHP Bearings Ltd. v. United States, 24 CIT 1218, 120 F. Supp. 2d 1116, 1124 (2000), aff’d in part and vacated in part, 288 F. 3d 1334, 1337 (Fed. Cir. 2002) (vacating the CIT decision sustaining Commerce’s calculation of profit component of constructed value and remanding case for further proceedings). In this case, Commerce explained why it reinterpreted its regulation: Unlike in [Color Television Receivers], in this case we have been presented with arguments as to why we should not use market-economy prices for inputs produced in a NME country. Based on our review of those comments, we have determined that prices of products that originate in a NME country should not *1448 be used because of the inherent distortions involved in an economy that is not controlled by market forces. [W]e have strong concerns that, were we to use the prices of inputs that were produced in a NME country, our methodology for valuing the factors of production would become easily open to manipulation. This is particularly worrisome where, as here, the inputs may never have left the stream of the NME commerce. It would not be difficult for a firm to open a paper company in Hong Kong (or other market-economy countries) and route “sales” through this company in order to take advantage of our market-economy-input methodology. For these reasons, our practice is not to use the prices of inputs that originated in a NME country even if the input is sourced from a market-economy supplier. Issues and Decision Mem., Comment 4, at 25-26. Commerce further distinguished this case from Certain Color Television Receivers from China, 69 Fed. Reg. 20594 and Certain Preserved Mushrooms from China, 66 Fed. Reg. 31204. In Color Televisions Receivers, the determination did not indicate whether the inputs purchased from the Hong Kong suppliers ever left China, and in Certain Preserved Mushroom from China, Commerce did not find evidence that the inputs were not produced in a market economy. Glopack argues that Commerce’s conclusion that the inputs never left the stream of NME commerce is not supported by the record, citing to Sea Lake’s response that Sea Lake, a Hong Kong company, purchased the Chinese-origin inputs in Hong Kong and shipped them from Hong Kong to its factory in Shenzhen. See Sea Lake Dec. 22, 2003 Supp. Response, at SA-3. Glopack argues that Sea Lake’s market economy purchases were not intra-China transfers transacted in Hong Kong dollars, but were transactions that left the NME stream of commerce and were physically moved from China to Hong Kong and re-exported back to China. Commerce argues that it used surrogate values rather than the actual prices paid by Sea Lake and Glopack for raw material inputs manufactured in the PRC but purchased from a Hong Kong trading company to avoid using distorted prices for factors of production. Commerce found that to avoid using prices influenced by the distortions inherent in the PRC’s nonmarket economy, it must disregard prices of inputs produced there, regardless of where the purchase took place. Commerce argues that this interpretation is consistent with its practice not to use prices distorted by nonmarket forces in its calculation. Where Commerce has reason to believe or suspect that actual prices are subsidized, the court will “look at the facts of [the] record to determine whether Commerce has sufficient reasons to suspect *1449 that actual prices are distorted such that the substitution of actual prices with surrogate values is warranted.” China Nat’l Mach. Imp. & Exp. Corp. v. United States, 27 CIT_, 293 F. Supp. 2d 1334, 1336 (2003), aff’d, 104 Fed. Appx. 183 (Fed. Cir. 2004). The court finds that Commerce provided sufficient explanation of why it applied its regulation differently in this case from some prior cases. Furthermore, Commerce’s decision was based on record evidence showing that Sea Lake purchased products produced in the PRC from its Hong Kong trading company, and that those goods may not have left the country on the way to Sea Lake. As a result, Commerce’s valuation of this factor is supported by substantial evidence and in accordance with the law. Conclusion In conclusion, Commerce’s determinations in this case are AFFIRMED with the exception of its calculation of Hang Lung’s electricity usage. This issue is remanded to the Department of Commerce for further proceedings consistent with this opinion. Footnotes [1] Shanghai Glopack, an exporter with no PRC ownership, was found to be affiliated with Sea Lake, a Hong Kong-based company with no PRC ownership. See Notice of Preliminary Determination of Sales at Less Than Fair Value and Postponement of Final Determination: PRCBs from PRC, 69 Fed. Reg. 3533, 3547. Because of these circumstances, Commerce did not engage in a separate-rate analysis for Glopack. See id. [2] Section 1677e(b) provides: If the administering authority or the Commission (as the case may be) finds that an interested party has failed to cooperate by not acting to the best of its ability to comply with a request for information from the administering authority or the Commission, the administering authority or the Commission (as the case may be), in reaching the applicable determination under this subtitle, may use an inference that is adverse to the interests of that party in selecting from among the facts otherwise available. Such adverse inference may include reliance on information derived from- (1) the petition, (2) a final determination in the investigation under this subtitle, (3) any previous review under section 1675 of this title or determination under section 1675b of this title, or (4) any other information placed on the record. 19 U.S.C. § 1677e(b). [3] Confidential business information has been redacted in this public version of the opinion. [4] Three types of PE Resin were involved in this investigation: high density polyethylene (“HDPE”), low density polyethylene (“LDPE”), and linear low density polyethylene (“LLDPE”).