FMC Corp. v. United States

*241 OPINION I. Introduction BARZILAY, Judge: Plaintiff’s Motion for Judgment upon an Agency Record under USCIT Rule 56.2 challenges two elements of the Department of Commerce’s final results of the third administrative review in Persulfates From the People’s Republic of China: Final Results of Antidumping Duty Administrative Review (“Final Results”), 66 Fed. Reg. 42,628 (Aug. 14, 2001), with regard to persulfates exported from the People’s Republic of China (“PRC”) by the Respondent Shanghai Ai Jian Import and Export Corporation (“Ai Jian”). Plaintiff FMC Corporation (“FMC”) is a domestic manufacturer of the subject merchandise. FMC contends that the decision of the Department of Commerce (“Department” or “Commerce”) was wrong to base the selling, general and administrative (“SG&A”) expense ratio on the financial statements of the surrogate company because it “contraven [ed] the Department’s duty, when conducting an anti-dumping investigation, to calculate margins as accurately as possible.” Pl.’s Mem. in Supp. of Mot. for J. on the Agency R. (“Pl.’s Br.”) at 4 (citing Rhone-Poulenc, Inc. v. United States, 899 F. 2d 1185, 1191 (Fed. Cir. 1990)). Plaintiff claims the calculations were not accurate due to the dissimilar cost structure between the company’s subject and non-subject merchandise, which resulted in a severe understating of SG&A expenses. Id. at 4-5. Plaintiff also argues that Commerce’s decision to rely on Respondent’s market economy ocean freight rates was “unsupported by substantial evidence and is otherwise contrary to law” because there were indications the invoices were fictitious. Pl.’s Br. at 7. For the reasons detailed below, the court denies Plaintiff’s motion and upholds Commerce’s determination. II. Background Commerce, in the 1997 final determination of an antidumping investigation, found that persulfates imported from China were being sold in the United States at less-than-fair-market-value. See Notice of Final Determination of Sales at Less Than Fair Value: Persulfates From the People’s Republic of China (“Final Determination”, 62 Fed. Reg. 27,222 (May 19, 1997). No other company with publically available financial statements in an economy comparable to the PRC produced persulfates at the time of the initial investigation. Id. at 27,228. FMC is the sole producer in the United States of persulfates. Pl.’s Br. at 2. Commerce initiated its original investigation of persulfates in 1996, in response to an antidumping petition filed by FMC. See Initiation of Antidumping Duty Investigation: Persulfates From the People’s Republic of China, 61 Fed. Reg. 40,817 (Aug. 6, 1996). The investigation set an antidumping duty rate of 34.41 per *242 cent for Ai Jian. See Notice of Amended Antidumping Duty Order: Persulfates From the People’s Republic of China, 62 Fed. Reg. 39,212 (July 22, 1997). There have been three subsequent reviews. It is the third review that is the subject of this case. See Final Results, 66 Fed. Reg. at 42,628. The third review set a duty rate of 0.04 percent, which is de minimis. See id. at 42,629. Over the course of the three reviews the dumping margin has fallen with each review. The first review set a rate of 5.54 percent. See Persulfates From the People’s Republic of China: Amended Final Results of Antidumping Duty Administrative Review (“First Amended Administrative Review”), 65 Fed. Reg. 1,356 (Jan. 10, 2000). The second review set a rate of 2.62 percent. See Persulfates from the People’s Republic of China: Final Results of Antidumping Duty Administrative Review and Partial Rescission of Administrative Review (“Second Administrative Review”), 65 Fed. Reg. 46,691, 46,692 (July 31, 2000). During the initial investigation, Commerce had used the financial statements of National Peroxide (“NPL”), an Indian company that produces similar, but not identical merchandise to calculate surrogate values. See Final Determination, 62 Fed. Reg. at 27,229. During the first annual review, Commerce found that another Indian company had begun producing persulfates. Citing a traditional practice of using surrogate companies that manufacture the subject merchandise, Commerce switched and began using the financial statements from Calibre Chemicals Pvt. Ltd. (“Calibre”). See Persulfates from the People’s Republic of China: Final Results of Antidumping Duty Administrative Review (“First Administrative Review”), 64 Fed. Reg. 69,494, 69,499-500 (Dec. 13, 1999). Selling, general and administrative expenses are one element that Commerce uses to establish a total cost of goods sold in a non-market economy. To establish a market-economy value for those elements, Commerce will use the financial data of a surrogate producer. The first step is for Commerce to establish an SG&A ratio, which is derived “by dividing the company’s general expenses by its total costs of sales.” Issues and Decision Memorandum for the Antidumping Duty Administrative Review of Persulfates from the People’s Republic of China for the Period July 1, 1999 through June 30, 2000; Final Results (“Issues and Decision Memo”) at 13. The cost of sales includes labor, materials, factory overhead, and energy costs. See Preliminary Results Factors Valuation Mem., Attachment 11: SG&A Expenses and Profit (Apr. 2, 2001). This ratio for SG&A is then applied to the cost of manufacturing the subject merchandise to determine an SG&A expense amount for the subject merchandise. According to Commerce,“general expenses are so indirectly related to a particular production process that the most reasonable allocation basis is the company’s total cost of manufacturing.” Issues and Decision Memo at 13. *243 Plaintiff, in the administrative proceedings below, challenged the use of Calibre’s financial data for surrogate values. Id. at 11. Calibre produces multiple product lines. One is subject merchandise, and the others are non-subject merchandise. FMC argued that the high cost of raw materials to produce the non-subject merchandise resulted in an overstatement of the SG&A ratio for non-subject merchandise, and an understatement of those costs for persulfates. Id. Commerce rejected this argument. Id. at 13. However, Commerce did accept Plaintiff’s argument that Calibre was not an accurate source for financial information related to factory overhead (“FOH”) costs, and agreed to reallocate Calibre’s FOH expenses to correct inaccuracies. Persulfates from the People’s Republic of China: Preliminary Results of Antidumping Duty Administrative Review, 66 Fed. Reg. 18,439, 18,443 (Apr. 9, 2001). Plaintiff now appeals Commerce’s use of Calibre’s SG&A expenses, which Plaintiff claims are distorted in a manner similar to the FOH expenses. Pl.’s Br. at 4. After the third review preliminary results, Plaintiff raised objections to Ai Jian’s submission of freight invoices. See Issues and Decision Memo at 2. FMC argued that the freight invoices were probably fraudulent because two of the four submitted had dates that differed by eleven months from the actual invoice payment date. Pl.’s Br. at 18. In addition, the mistake was made by two different shipping companies and the date was the same on both misdated invoices. See id. In response to this discrepancy, Commerce requested an explanation and supporting documentation from Respondent Ai Jian. Issues and Decisions Memo at 2. Ai Jian claimed that the dates were incorrect due to a clerical error by the shipping companies. Id. at 2. To support its claim of market economy freight costs, Ai Jian provided an internal accounting voucher, an operating sub-ledger, a bank deposit sub-ledger, and a bank statement. Id. at 5. It also provided a foreign exchange voucher, which was reconciled to the company’s internal documents. Id. In the final results, Commerce rejected Plaintiff’s argument and found that Ai Jian paid market economy prices for ocean freight expenses. Id. Plaintiff appeals this decision as well. This court has jurisdiction over this matter under 28 U.S.C. § 1581(c). III. Standard of Review The court must evaluate whether Commerce’s decisions are supported by substantial evidence on the record or are otherwise in accordance with law. See 19 U.S.C. § 1516a(b)(l)(B). Substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Consolidated Edison Co. of New York v. NLRB, 305 U.S. 197, 229 (1938) (citations omitted); Matsushita Elec. Indus. Co., Ltd. v. United States, 750 F.2d 927, 933 (Fed. Cir. 1984). This Court noted, “[i]n applying this standard, the court affirms [the agency’s] factual determinations so long as they *244 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)). The court may not reweigh the evidence or substitute its own judgment for that of the agency. See Granges Metallverken AB v. United States, 13 CIT 471, 474, 716 F. Supp. 17, 21 (1989). Substantial evidence is “something less than the weight of the evidence, and the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Id., 716 F. Supp. at 21 (citations omitted). Additionally, absent a showing to the contrary, the agency is presumed to have considered all of the evidence in the record. Nat’l Ass’n of Mirror Mfrs. v. United States, 12 CIT 771, 779, 696 F. Supp. 642, 648 (1988) (citations omitted). To determine if the agency’s interpretation of the statute is in accordance with law “we must first carefully investigate the matter to determine whether Congress’s purpose and intent on the question at issue is judicially ascertainable.” Timex V.I. v. United States, 157 F.3d 879, 881 (Fed. Cir. 1998) (citing Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842 — 43 (1984)). The expressed will or intent of Congress on a specific issue is dispositive. See Japan Whaling Association v. American Cetacean Society, 478 U.S. 221, 233-237 (1986) (citing Chevron, 467 U.S. at 843.). If the court determines that the statute is silent or ambiguous, the question to be asked is whether the agency’s construction of the statute is permissible. See Chevron, 467 U.S. at 843. This deference is due “when it appears that Congress delegated authority to the agency generally to make rules carrying the force of law, and that the agency interpretation claiming deference was promulgated in the exercise of that authority.” United States v. Mead Corp., 533 U.S. 218, 226-27 (2001). This deference is not limited to notice and comment rulemaking but is also given to those “statutory determinations that are articulated in any ‘relatively formal administrative procedure.’” Pesquera Mares Australes Ltda., v. United States, 266 F.3d 1372, 1381 (Fed. Cir. 2001). Therefore, statutory interpretations articulated by Commerce during antidumping proceedings are entitled to judicial deference under Chevron. Id. at 1382. IV. Discussion Establishing antidumping duty rates begins by comparing the normal value of the product in its home market with the export price of the goods sold in the United States. See 19 U.S.C. § 1677b(a) (1999). In the case of a non-market economy like the PRC, if Commerce cannot rely on the information available from the respondent, it will determine normal value “on the basis of the value of the factors of *245 production . . . based on the best information available regarding the values of such factors in a market economy country.” § 1677b(c)(l). In this case, Commerce established India as the appropriate surrogate country. Within that surrogate country, Commerce will then choose among various sources to value specific elements of normal value. See e.g., Final Determination, 62 Fed. Reg. at 27,229. A. Commerce’s Use of Calibre’s SG&A Expenses Is Permissible. Commerce is charged under the antidumping laws with establishing dumping margins that are as accurate as possible. See Rhone-Poulenc, 899 F.2d at 1191. Consistent with this charge, Commerce has developed various methodologies for assessing the likely market value of goods that are produced in non-market economies. See, e.g., First Administrative Review, 64 Fed. Reg. at 69,499. As part of this practice, Commerce has developed a preference for selecting surrogate value sources that rely on financial statements of companies that produce the subject merchandise of an investigation. Id. at 69,500. However, if the surrogate data is distorted or otherwise unreliable, Commerce will not rely on that data. Id. During the initial period of investigation Commerce did not have access to the financial statements of a foreign persulfates manufacturer existing in an economy comparable to the PRC. See Final Determination, 62 Fed. Reg. at 27,229. Commerce, unable to rely on Ai Jian’s non-market economy data, and unable to use a producer of the subject merchandise, used the financials of NPL, an Indian based company that produces similar merchandise. First Administrative Review at 69,500. After the initial investigation, Calibre, also an Indian company, began producing persulfates. Id. Commerce, citing its preference to use companies that produce the subject merchandise as surrogates, began to use Calibre’s financial statements to calculate the SG&A expenses for a persulfate producer in calculating Ai Jian’s normal value and export price. FMC, during the administrative process and before this court, argues that Calibre’s financial statements are not representative and, therefore, aye not reliable as a surrogate value for SG&A expenses. Pl.’s Br. at 9. FMC points out that 90 percent of Calibre’s raw material costs are exclusively related to the material used in production of non-subject merchandise. Id. Plaintiff claims that this results means: Calibre’s cost of goods sold will greatly exaggerate the amount of SG&A expenses attributable to the company’s production of non-subject merchandise, i.e., potassium brómate and potassium iodate, causing the SG&A expenses attributable to the company’s production of persulfates to be severely understated. Id. In this case, Commerce did acknowledge that for the purposes of calculating the FOH costs, the use of Calibre’s financial statements *246 produced results that required adjustment because of the disparity in the cost structure differences between subject and non-subject merchandise. In the preliminary results, Commerce explained how those adjustments were made: During the first administrative review, we determined that it was appropriate to allocate Calibre’s overhead expenses between its product lines because of the differing cost structures between Calibre’s production of subject and non-subject merchandise. . . . For purposes of this analysis, we considered Cali-bre’s raw material groups “sulfates” and “acids” to be related to its production of potassium persulfates, and the raw material groups “halogens” and “alkalis” to be related to its production of potassium brómate and potassium iodate. Because of the differing cost structures attributable to subject and non-subject merchandise, we allocated Calibre’s factory overhead expenses between subject and non-subject merchandise on the basis of raw material consumption. Prelim. Results Factors Valuation Mem. at 7-8. Plaintiff claims that like the FOH calculation, the SG&A ratio calculation is not reliable without adjustment. Pl.’s Br. at 11. However, unlike in the FOH numbers, there is no “cost driver” which can be relied on to correct the distortions. For that reason FMC asked that Commerce return to using the financial statements of another Indian company, NPL, but Commerce rejected that request. Issues and Decision Memo at 11. FMC points to at least one other occasion where Commerce has refused to use the financial statements of a'company that produced the subject merchandise for surrogate purposes, because “the product mix” made the financials unreliable. Pl.’s Br. at 12. In Certain Non-Frozen Apple Juice Concentrate from the People’s Republic of China, Commerce determined, according to Plaintiff, that “the nature of the operations of the company, unrelated to production of the subject merchandise, made it unsuitable as a source of surrogate value. . . .” Id. at 12 (citing the Issues and Decision Memo for Notice of Final Determination of Sales at Less Than Fair Value: Certain Non-Frozen Apple Juice Concentrate from the People’s Republic of China, 65 Fed. Reg. 19,873 (Apr. 13, 2000)). FMC contends that because Commerce decided that NPL’s financial data were sufficiently accurate for surrogate purposes in the initial investigation, for Commerce to now use Calibre’s data, which includes non-correctable distortions, is unsupported by substantial evidence and otherwise not in accordance with law. See id. at 15. Defendant replies that using Calibre’s financial information is consistent with past practice and the statute, and that there is ample justification for the policy. The statute requires that when products are imported into the United States from a non-market economy *247 Commerce will determine the normal value of the “subject merchandise on the basis of the value of the factors of production utilized in producing the merchandise and to which shall be added an amount for general expenses and profit plus the cost of containers, coverings, and other expenses.” 19 U.S.C. § 1677b(c)(l). The statute does not specify how to calculate the individual factors of production. Through practice Commerce has enunciated a policy of calculating the general and administrative expenses “based upon the company-wide G&A costs incurred by the producing company allocated over the producing company’s company-wide cost of sales.” Def’s Br. at 19. Commerce uses a company-wide calculation for general expenses because they do not relate to a specific product line. Id. (citing, inter alia, Large Newspaper Printing Presses and Components Thereof, Whether Assembled or Unassembled, From Japan: Final Results of Antidumping Duty Administrative Review, 66 Fed. Reg. 11,555 (Feb. 26, 2001)). 1. Calculating FOH and SG&A Costs Differently Is Permissible. Commerce grants that it does not treat the calculations for FOH and SG&A costs in the same manner. As Commerce explained in its Issues and Decision Memo, FOH costs are traditionally calculated to factor in specific product costs for subject merchandise. SG&A costs, in contrast, are not traditionally based on product specific costs: Unlike factory overhead costs, SG&A expenses are not considered to be directly related to the production of merchandise. In fact, in most cases, general expenses are so indirectly related to a particular production process that the most reasonable allocation basis is the company’s total cost of manufacturing. Thus, while it may be appropriate to allocate the factory overhead costs between subject and non-subject merchandise on a basis other than cost, we find no basis to allocate SG&A expenses to specific product lines using any other method. Issues and Decision Memo at 13-14 (citing First Administrative Review, 64 Fed. Reg. at 69,499-500). FMC’s argument that the acknowledged distortions in the FOH costs, due to the disparity in production costs for subject and non-subject merchandise, require discounting the value of the SG&A expenses is not supportable. FOH costs are directly tied to material costs. However, raw material costs do not generally directly impact the SG&A ratio, which is calculated on a company-wide basis. This Court has previously upheld Commerce’s policy to base SG&A costs on a company-wide determination when a company produces subject and non-subject merchandise. See U.S. Steel Group A Unit of USX Corp. v. United States, 22 CIT 104, 106, 998 F. Supp. 1151, 1154 (1998) (citing Rautaruukki Oy v. United States, 19 CIT 438, 444 (1995)). That distinction is rooted in a consistent practice that this *248 Court has found is proper under the statute. Therefore, it is not contrary to law for Commerce to employ different methodologies for calculating FOH and SG&A costs. 2. Significant Raw Materials Variations. While Commerce is able to justify its different calculation methodologies for FOH and SG&A costs, Commerce must also address Plaintiff’s claim that the SG&A ratio is distorted on its own terms, not just in relation to the FOH costs. In the administrative proceeding below, FMC challenged the use of Calibre as a surrogate for calculating SG&A costs despite the fact that Calibre produces the subject merchandise. FMC claimed that the “characteristics of the goods manufactured by Calibre” and Cali-bre’s “product mix” made its financial information “unreliable.” Perkins Coie Case Brief on behalf of Petitioner FMC (“Pet. ’s Br”) (May 9, 2001) at 23, in Def.’s Proprietary App. Tab C, at 69. FMC pointed out that two other products Calibre produced, potassium brómate and potassium iodate, accounted for “90 percent of Calibre’s 1999-2000 raw material costs” and for “over 68 percent of Calibre’s cost of sales for the 2000 fiscal year.” Pet.’s Br. at 24. To explain the “characteristics of the goods manufactured” and “product mix” of Calibre, which FMC claims make the financials unreliable, FMC points to Commerce’s own findings in the third review which found: . . . Calibre consumed a relatively low quantity of raw materials related to non-subject merchandise (i.e., potassium brómate and potassium iodate) which incurred high raw material costs. On the other hand, the majority of Calibre’s consumption of raw materials is attributable to subject merchandise (i.e., potassium persulfates) with relatively low costs. Prelim. Results Factors Valuation Mem. at 7-8. Commerce rejected FMC’s challenge to Calibre’s financials. As a preliminary matter, Commerce pointed to its “long-standing practice in market-economy cases with respect to allocating general expenses to individual products is to calculate a rate by dividing the company’s general expenses by its total cost of sales.” Issues and Decision Memo at 13. Commerce justifies using company-wide expenses because “general expenses are costs that relate to the company’s overall operations.” Id. Commerce noted that the “company-wide” calculation was developed for investigations involving market economies, and the investigation of Ai Jian involves a non-market economy. However, “the issue at hand involves deriving an SG&A ratio using the financial data of a market-economy company.” Id. Commerce also states that “in most cases, general expenses are so indirectly related to a particular production process that the most reasonable allocation basis is the company’s total cost of manufacturing.” Id. Commerce’s position can be summarized as follows: it has a “preference” *249 for using producers of the subject merchandise, and it has a “longstanding practice” of using company-wide costs for general expenses, which is based on a general observation that “in most cases” SG&A expenses are only “indirectly related” to a particular product line, and, therefore, the “most reasonable allocation basis is the company’s total cost of manufacturing.” Commerce concludes its explanation of its general practice by observing that it “find[s] no basis to allocate SG&A expenses to specific product lines using any other method.” Id. at 14. FMC does not challenge Commerce’s interpretation of the anti-dumping statute to permit these general principles and practices. FMC is pointing to specific evidence in the record that indicates that the general practice should not be applied in this circumstance. FMC argues that there is a “basis” for allocating SG&A costs along product lines, because the material costs of non-subject merchandise are so large in comparison to subject merchandise that they cannot be considered merely “indirectly related” to a specific product line. See Pl.’sBr. at 11. The task before this court is to determine if there is “substantial evidence” to justify Commerce’s decision to use the financials of Cali-bre over NPL. Responding to FMC’s contention that Calibre’s data is unreliable, Commerce states that “petitioner has presented no new evidence.” Issues and Decisions Memo at 14. Commerce explains further in a footnote: The petitioner’s argument relies solely on the fact that Cali-bre produces more non-subject than subject merchandise. We find that this fact does not lead to the automatic conclusion that it is distortive to allocate SG&A expenses on the basis of cost of sales. On the contrary, we find that the two other products produced by Calibre are comparable to persulfates. Issues and Decision Memo at 14 n.7 (citations omitted). Commerce found the various product lines to be comparable because they are used for similar purposes. See id. (citing Aug. 7, 2001, memorandum from Shawn Thompson to the file entitled “Source Data Used to Identify the Uses of Potassium Persulfates, Potassium Iodate and Potassium Brómate in the 1999-2000 Antidumping Duty Administrative Review of Persulfates from the [PRC]”). Commerce also distinguishes two instances when it did not rely on the financial statements of a surrogate producer of the subject merchandise. In Certain Helical Spring Lock Washers From the People’s Republic of China: Final Results of Antidumping Duty Administrative Review {“Lock Washers”), Commerce rejected using data from a company that produced both subject and non-subject merchandise because the company produced only a “minuscule” amount of the subject merchandise. See Def.’s Br. at 22 n.l (citing 61 Fed. Reg. 41,994 (Aug. 13, 1996)). Defendant points out that, unlike in Lock *250 Washers, Calibre produces the subject merchandise in commercial quantities. See id. In Apple Juice Concentrate, cited by Plaintiff in its brief, Commerce rejected the use of a producer because the primary business of the proposed surrogate company was not manufacturing. See Comment 8 to Issues and Decision Mem. for the Investigation of Certain Non-Frozen Apple Juice Concentrate from the People’s Republic of China, 65 Fed. Reg. 19,873 (Apr. 6, 2000) (finding that the total revenues of the proposed surrogate company were primarily derived from service-oriented rather than manufacturing operations). If this were all that were in the record to provide a foundation for Commerce’s decision it would not be sufficient. Contrary to what Commerce claims, FMC did not rely “solely” on the relative size of Calibre’s product lines. FMC rested its challenge on the fact, verified by Commerce in its FOH analysis, that non-subject merchandise represents a disproportionate amount of the costs of raw materials which has significant impact on the SG&A ratio. That is a more substantial challenge than the fact that Calibre merely makes more of one product than another. In addition, the disparity between the product lines that Plaintiff claims creates a problem is not that they have different ultimate uses, but that raw materials for non-subject merchandise are significantly more expensive than for subject merchandise. See Pl.’s Br. at 11; Pet’s Br. at 23. Commerce misrepresents FMC’s argument and dismisses it for irrelevant reasons. Standing alone, a decision is not based on substantial evidence if it relies on general preferences which have been called into doubt by evidence on the record. See Rhodia, Inc. v. United States, 25 CIT _, __, 185 F. Supp. 2d 1343, 1348 (2001) (stating that Commerce must articulate a rational connection between the facts found and the choice made). However, Commerce cites to more evidence when explaining its decision to use Calibre instead of NPL as a surrogate. The relief FMC seeks here is a direction to Commerce to use the NPL financials instead of the Calibre financials. Pl.’s Br. at 11. FMC argues Calibre’s financials are not reliable for a specific reason, but it does not believe they can be adjusted to reflect an accurate SG&A ratio. This is because there is no “cost driver” which could permit allocation of material costs among the product lines. See id. This puts Commerce in the position of weighing two flawed sets of financial data for calculating a surrogate value for SG&A costs. Commerce explicitly undertook this task and weighed several pieces of relevant evidence. See Issues and Decision Memo at 14. First, Commerce restates its preference for “selecting surrogate value sources that are producers of identical merchandise, provided that the surrogate data is not distorted or otherwise unreliable.” Id. Second, Commerce casts doubt on the reliability of NPL’s financial data for the period of review because NPL “incurred certain expenses related to property development.” Id. at 14 n.8 (citations *251 omitted). Third, Commerce valued FOH and profit based on Calibre’s financials because to value SG&A based on NPL’s financials would “result in [Commerce’s] applying a profit ratio that would bear no relationship to the SG&A ratio.” Id. at 14 (footnote omitted). Fourth, mixing the two financial statements raises the “potential for double counting or under-counting of expenses because different companies may classify expenses differently.” Id. at 15. Plaintiff does not challenge any of these findings. Commerce does acknowledge that in at least one case it has used multiple sources. See id. at 15 (citing Notice of Final Determination of Sales at Less Than Fair Value: Beryllium Metal and High Beryllium Alloys From the Republic of Kazakstan, 62 Fed. Reg. 2,648 (Jan. 17, 1997)). However, in that case, a lack of detailed information required using more than one source. In this case, Commerce found Calibre’s financial data are detailed enough to use across the board. Id. Commerce has wide discretion in choosing among various surrogate sources. “When Commerce is faced with the decision to choose between two reasonable alternatives and one alternative is favored over the other in their eyes, then they have the discretion to choose accordingly.” Technoimportexport, UCF America Inc. v. United States, 16 CIT, 13, 18, 783 F. Supp. 1401, 1406 (1992). Commerce has provided a thorough and complete explanation for why it chose Calibre over NPL. Despite the flaws Plaintiff has identified in Commerce’s analysis of SG&A costs, it does not seek to have those costs adjusted. Rather, Plaintiff seeks to replace one set of flawed data with another set of flawed data. Plaintiff has raised a question about one element of Calibre’s financials. However, it has not countered the various problems Commerce raised with regard to using NPL’s financials. Commerce did not choose Calibre as a surrogate because it was a perfect match. It chose Calibre because it was a better option than NPL. Plaintiff’s argument does not overcome the substantial evidence supporting Commerce’s ultimate conclusion. Faced with a difficult choice, Commerce made a reasonable decision. The “substantial evidence” standard of review is clear in granting deference to agency determinations. The court will not reweigh the evidence or substitute its own judgment for that of the agency. See Granges Metallverken AB, 716 F. Supp. at 21. Under this standard, “the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence." Id. (citations omitted). Commerce’s decision to use Calibre as a surrogate for calculating the SG&A ratio is sustained as based on substantial evidence and is in accordance with law. *252 B. Market Economy Ocean Freight Expenses. FMC challenges Commerce’s decision to value Ai Jian’s ocean freight expenses at market economy values. Pl.’s Br. at 15. Commerce will use the actual costs to value inputs in non-market economy cases when “a respondent purchased from a market economy supplier in a market economy currency.” Issues and Decision Memo at 5 (citing 19 C.F.R. § 351.408(c)(1)). Plaintiff points to a number of instances in the record that cast suspicion on the validity of the invoices Ai Jian submitted during the investigation. Id. FMC contends these discrepancies in the record demand that Commerce investigate further, to verify the legitimacy of the invoices submitted or to use “adverse facts available” under 19 C.F.R. § 351.308(a). Commerce issued its questionnaire for the period of review on August 22, 2000. Respondents filed a response on October 13, 2000. In that response Ai Jian stated that its market economy ocean freight expenses were paid in market economy currency and that the product was shipped with a market-economy freight company. Ai Jian Section C Response at 2 in Pl.’s Non-Confidential App Ex. 6. Commerce, following up on this claim, asked Ai Jian to submit additional documentation: For the invoices stated above in 5.a., provide a complete paper trail of documentation demonstrating that the international freight charges are set by market-economy carriers and paid for in a market economy currency. Such documentation should include but is not limited to: • contract(s) between Ai Jian and the shipping company; • invoices; • shipping documents such as bills of lading, airway bills, and delivery orders; and • records of payment such as canceled checks, letters of credit, debit/credit memos, wire transfers, promissory notes, bank deposit slips, and/or bank statements. Nov. 28, 2000 Commerce Letter to White & Case at 3-4 in Pl.’s Non-Confidential App. Ex. 7. In response to this request, Ai Jian submitted additional documentation. See Issues and Decision Memo at 5. Significantly, it also changed some elements of its original claim. Instead of claiming the goods were shipped by a single company, it submitted four invoices from two different companies. Id. at 3. Neither company matched the description of the company Ai Jian initially claimed carried the goods. See Pl.’s Br. at 17. Two of the invoices were dated eleven months after the payment dates that Ai Jian reported to Commerce. See Issues and Decision Memo at 3. Plaintiff claims that the two mistaken dates, and the switch in the company that Ai Jian claims carried the goods, indicates evidence that the invoices were fictitious. See Pl.’s Br. at 19. FMC argues that once such evidence raised a *253 “reasonable suspicion,” Commerce had a duty to further investigate the veracity of the invoices beyond the documents submitted. Pl.’s Br. b& 21 (citing U.S. Steel Group v. United States, 25 CIT_,_, 177 F. Supp. 2d 1325 (CIT 2001)). The government responds that Commerce “confirmed the validity of the documents by a thorough analysis of documentary and testimonial evidence surrounding the contested invoices.” Def’s Br. at 29. To establish the validity of the invoices Commerce evaluated the copies of the pages from [Ai Jian’s] expenses sub-ledger demonstrating that it recorded the freight expenses in question in its accounting system within a few days of making the payment to the freight supplier. Issues and Decision Memo at 5. Further, Ai Jian provided documents showing payment for these ocean freight transactions could be traced from the relevant foreign currency exchange vouchers issued by a commercial bank, to Ai Jian’s internal accounting vouchers and operating expense sub-ledger and bank deposit sub-ledger using the sales invoice numbers. Id. In addition, Commerce found that the “payment for ocean freight for these invoices can also be traced from Ai Jian’s bank deposit sub-ledger to Ai Jian’s bank statement, prepared by the Bank of China.” Id. Indeed, with the exception of the dates shown on documents generated by outside parties, all of the documents provided by Ai Jian have been internally consistent and are traceable to actual accounting records. Whatever the causes of the incorrect dates on the two invoices in question, the other information submitted by Ai Jian sufficiently demonstrates that the firm actually incurred and paid for the expenses. Id. at 5. It is clear from the record that, contrary to FMC’s assertion, Commerce did not rest its evaluation on the invoices themselves, but requested additional information. The additional information, while not as complete as FMC would require, provided substantial evidence for Commerce to conclude that the market economy rates for shipping were actually paid by Ai Jian. The court agrees with Plaintiff that in cases where there is reasonable suspicion of wrongdoing Commerce must evaluate further. See U.S. Steel, 177 F. Supp. 2d at 1331 (citing Zenith Elec. Corp. v. United States, 15 CIT 394, 406-07, 770 F. Supp. 648, 659 (1991)). Here, Commerce did evaluate further and, based on substantial evidence, has verified that Ai Jian paid market economy rates for ocean freight. FMC argues that the two suspicious invoices, even if verified by Commerce, still demand that Commerce use total or at least partial *254 adverse facts available. See 19 C.F.R. § 351.308(a). 1 To support the use of total adverse facts Plaintiff cites Final Determination of Sales at Less Than Fair Value: Sulfanilic Acid from the Republic of Hungary (“Sulfanilic Acid”), 58 Fed. Reg. 8,256 (Feb. 12, 1993). In the Sulfanilic Acid case, Commerce found a document during an on-site inspection that indicated an attempt by the respondent to fabricate its submitted documents. Commerce decided that in order to maintain the integrity of the questionnaire process, it could not countenance such a blatant attempt at deception. Therefore, it substituted best information available for all the information submitted by the respondent. Id. at 8,257. In this case, however, Commerce merely had two invoices that were misdated. The improper date could have been the result of a clerical error or intentional deception. To determine if the dates were a minor and correctable problem, or significant and indicative of intentional misrepresentation, Commerce demanded and received from Ai Jian supporting documentation. Based on this documentation, Commerce determined that the invoices were valid and proper with the exception of the dates. In the Sulfanilic Acid case, there was substantial evidence that the documents were fraudulent. In this case, there was only a reasonable suspicion - disproved with substantial evidence gathered by Commerce - that the invoices might be fictitious. Plaintiff fails to grasp this distinction and risks making every clerical mistake discovered in an investigation grounds for the application of adverse facts available pursuant to 19 C.F.R. § 351.308(a). Plaintiff’s last objection to Commerce’s third review final results is that, even if the ocean freight documentation is authentic and reliable, there is still no evidence of actual payment to the carriers. Pl.’s Br. at 24. Plaintiff asks that in the absence of proof of payment Commerce be required to use partial adverse facts in granting Ai Jian an adjustment. Plaintiff cites to Sweaters Wholly or in Chief Weight of Man-Made Fiber from Taiwan: Final Results of Changed Circumstances Antidumping Duty Administrative Review, 58 Fed. Reg. 32,644 (June 11, 1993) (“Sweaters”). In that case, Commerce applied best information available to calculate -certain costs. It explained what kind of information it sought and what it deemed lacking in the respondent’s submissions: *255 We were unable in every instance to confirm that an actual payment had been made from Jia Farn to the yarn supplier. Although we saw evidence of payments coming out of Jia Farn’s bank account, we could not confirm to whom these payments were made. We did see receipts from yarn suppliers, but we could not confirm that the payment for the yarn was made from a Jia Farn bank account. We also encountered much difficulty in our attempts to match the amount of an invoice with the bank disbursements, because Jia Farn makes multiple payments for most invoices, and we were not able to reconcile multiple payments from the bank accounts to invoice amounts. Id. at 32,649. Unlike in the Sweaters case, in this case Commerce was able to trace the invoices to particular bank withdrawals, which were confirmed by Ai Jian’s bank statements. See Issues and Decision Memo at 5. As a result, Commerce found that the evidence “sufficiently demonstrates that the firm actually incurred and paid for the expenses.” Id. As discussed above, that finding was based on substantial evidence. V. Conclusion For the reasons detailed above, Plaintiff’s Motion for Judgment upon an Agency Record is denied, and the case is dismissed. Judgment will be entered accordingly. Footnotes [1] In the relevant portion the regulation reads: a) Introduction. The Secretary may make determinations on the basis of the facts available whenever necessary information is not available on the record, an interested party or any other person withholds or fails to provide information requested in a timely manner and in the form required or significantly impedes a proceeding, or the Secretary is unable to verify submitted information. If the Secretary 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,” the Secretary may use an inference that is adverse to the interests of that party in selecting from among the facts otherwise available.