Opinion
Wallach, Judge:Plaintiff Rhone-Poulenc moved pursuant to USCIT Rule 56.2, for judgment on the agency record challenging the Department of Commerce’s (“Commerce”) Final Determination of Sales at Less Than Fair Value: Coumarin From The People’s Republic of China, 59 Fed. Reg. 66895 (Dep’t Comm. 1994)(the “Final Determination”). For the reasons set forth below, the motion is granted in part and denied in part.
I
Background
On December 30, 1993, Rhone Poulenc filed a petition with Commerce and the International Trade Commission (“ITC”) alleging that imports of coumarin from the People’s Republic of China (the “PRC”) were being sold in the United States at less than fair value, and that an American industry was materially injured by these sales. On January 19, 1994, Commerce commenced an antidumping duty investigation.
Rhone-Poulenc is a domestic producer of coumarin, an aroma chemical used in the manufacture of perfumes, among other things. Coumarin has the chemical formula C sub9 H sub6 O sub2, and also is known by the following names: 2H-l-benzopyran-2-one; 1,2-benzopyrone; cis-o-coumaric acid lactone; coumarinic anhydride; 2-Oxo-l,2-benzopyran; 5,6-benzo-alpha pyrone; ortho-hydroxyc innamic acid lactone; cis-ortho-courmaric (sic) acid anhydride; and tonka bean camphor. Commerce’s investigation included all forms of this chemical.
Because the PRC has a non-market economy (Final Determination, 59 Fed. Reg. at 66896), Commerce determined the foreign market value (“FMV”) of the subject merchandise using the factors of production (“FOP”) method. See 19 U.S.C. §1677b(c); Final Determination, 59 Fed. Reg. at 66896-97. Under this method, Commerce determined the amount of labor, raw materials, energy and other manufacturing inputs used to make the subject coumarin in the PRC. It then determined the value of these inputs using surrogate prices from India, which Com*574merce determined has a market economy at a level of development similar to China. Id. Pursuant to the statute, Commerce made adjustments to ensure the comparability of the United States and foreign prices that it used to calculate the anti-dumping margins. Id.
Amongthese adjustments was an offset for the production of by-products in the coumarin manufacturing process. Id. at 66897. Commerce reduced the cost of materials included in FOP by the value of by-products, in order to determine the cost of manufacturing.
Based on the foregoing analysis, Commerce found that coumarin from the PRC had been sold in the United States at less than fair value. It promulgated preliminary dumping margins for two exporters separately from the rest of the PRC coumarin industry. For Jiangsu Native Produce Import/Export Corporation (“JNPIEC”), Commerce set a margin of 15.04%; for Tianjin Native Produce Import/Export Corporation (“TNPIEC”), a margin of 50.35%; and for all other exporters, a margin of 160.80%.
JNPIEC and TNPIEC received separate dumping margins because information that they supplied convinced Commerce that they operate independently of government control, and negotiate at arm’s length with coumarin manufacturers in China. Id. at 66895-96. Other exporters failed to make this showing, and therefore were collectively subject to a higher antidumping margin based on best information available (“BIA”). The separate margins for JNPIEC and TNPIEC were based on the FOP of the manufacturers of the coumarin that they exported, Changzhou Perfumery and Tianjin Perfumery, respectively. There was no evidence that JNPIEC or TNPIEC exported coumarin produced by any other manufacturer.
Rhone-Poulenc brought this action timely pursuant to 19 U.S.C. § 1516a and 28 U.S.C. § 1581 (1988).
Rhone-Poulenc raises four issues:
A. Did Commerce err in determining the value of by-products of coumarin production, and in deducting it from the cost of manufacturing coumarin?
B. Did Commerce err in valuing FOP for Jiangsu Native’s and Tianjin Native’s coumarin suppliers separately from the rest of the Chinese coumarin manufacturing industry?
C. Did Commerce err in failing to limit the application of the lower dumping margins for Jiangsu Native and Tianjin Native to imports of coumarin made by the producers on whose costs the margins were based?
D. Did Commerce err in failing to include resellers’ selling, general and administrative costs and profit in its calculation of the constructed value of coumarin?
II
Standard of Review
When reviewing Commerce’s Final Determination in an antidump-ing investigation, the court will “hold unlawful any determination, find*575ing or conclusion found * * * to be unsupported by substantial evidence on the record, or otherwise not in accordance with law * * 19 U.S.C. § 1516a(b)(1)(B) (1988). Substantial evidence on the record means “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Pierce v. Underwood, 487 U.S. 552, 565, 108 S.Ct. 2541, 2550 (1988) (citations omitted). Under this standard, the Court neither weighs the evidence nor substitutes its own judgment for that of the agency. Negev Phosphates, Ltd. v. U.S. Dep’t of Commerce, 12 CIT 1074, 1076-77, 699 F.Supp. 938, 942 (1988). Nonetheless, the court’s deference cannot allow the agency to deviate from the clearly expressed intent of Congress. Bd. of Governors of the Fed. Reserve Sys. v. Dimension Fin. Corp., 474 U.S. 361, 368, 106 S.Ct. 681, 686 (1986). The court will find a determination unlawful where Commerce has failed to carry out its duties properly, relied on inadequate facts or reasoning, or failed to provide an adequate basis for its conclusions. See Budd Co. Ry. Div. v. United States, 1 CIT 67, 70-76, 507 F. Supp. 997, 1000-04 (1980); Indus. Fasteners Group v. United States, 2 CIT 181, 190, 525 F.Supp. 885, 892-93 (1981).
Ill
Discussion
A
Commerce Erred In Valuing By-Products and In Adjusting Cost of Manufacturing to Reflect Their Value
As a general matter, Commerce will determine whether by-products of the manufacture of subject merchandise have a market value. Final Determination, 59 Fed. Reg. at 66900-01. If they do, Commerce will subtract that amount from the cost of manufacturing of the subject merchandise. Here, Rhone-Poulenc argues that Commerce over-stated the value of the coumarin manufacturing by-products produced by JNPIEC’s and TNPIEC’s suppliers, i.e., Changzhou Perfumery (“Changzhou”) and Tianjin Perfiimery (“Tianjin”), respectively. The alleged over-statement arose from Commerce’s failure to account for the reduction in the by-products’ value caused by their contamination with impurities.
During the investigation, Rhone-Poulenc informed the government that by-product valuation would be an issue in determining the couma-rin producers’ costs of production. “The Department also needs information regarding the quality and composition of the by-products in order to attribute any value to them * * *. Information on impurities is essential because their presence can wipe out the value of a chemical. [Respondents] reported consumption of phenol, which contaminates by-products and renders them nearly worthless if not removed * * *.” Letters from Baker & Botts, L.L.P to Hon. Ronald H. Brown, May 10, 1994, A.R. Fiche 44 at 62 & 78.
As a consequence of Rhone-Poulenc’s letters, Commerce asked Changzhou and Tianjin, in deficiency questionnaires, to describe the *576quality and quantity of their by-products. In the questionnaire to Changzhou, Commerce directed: “Explain in more detail Changzhou’s disposal of waste and by-products * * *. For each by-product, identify any distinguishing product characteristics (e.g., grade, form, etc.) that may be necessary in determining the appropriate surrogate value.” I.T.A. Changzhou Section D Deficiency Questionnaire, May 19, 1994, A.R. Fiche 15 at 75. In the questionnaire to Tianjin, Commerce directed: “ Explain in more detail Tianjin’s disposal of waste and by-products * * * In addition, for each by-product, report any distinguishing product characteristics (i.e., grade, quantity of impurities) that are necessary in determining the appropriate surrogate values.” (Emphasis added) I.T.A., Tianjin Perfumery Section D Deficiency Questionnaire, May 19, 1994, A.R. Fiche 15 at 82. It is not clear why Commerce noted impurities specifically in the latter questionnaire and not in the former.
Changzhou’s response to its questionnaire was deficient. In describing the characteristics of its by-products, it said only: “The HCL acid produced as a result of the salicylaldehyde production is at a [ ] concentration level. The second by-product, hypochlorite, is produced at [ ] concentration level (of HCL).” A.R. Fiche 50 at 75. It made no mention of acetic acid, another by-product of Changzhou’s process and none of any impurities. Tianjin’s response to its questionnaire similarly was deficient. It described the disposition of by-products, but neither their composition nor any impurities. A.R. Fiche 53 at 20.
After reviewing the deficiency questionnaire responses, Rhone-Pou-lenc sent additional letters to Commerce arguing that the government “should not make adjustments to FMV for alleged by-products” for either Changzhou or Tianjin. Letter from Baker & Botts, L.L.R to Hon. Ronald H. Brown, July 18, 1994, A.R. Fiche 52 at 1. The primary reason stated was the two entities’ failure to provide sufficient information regarding the composition and disposition of their by-products. The lack of information rendered “the Department * * * unable to make any reasonably accurate offset to * * * constructed value as a result of any net sales revenue that may have been derived from sales of by-products. ” Id.
Despite Rhone-Poulenc’s urgings, Commerce made adjustments for coumarin by-products to Changzhou’s and Tianjin’s costs of manufacturing. It made an allowance for varying concentrations of the by-product chemicals, but none for impurities. Moreover, there is no evidence in the record that Commerce attempted any investigation into the purity of coumarin by-products generally, or the effect of impurities on their value other than the questionnaires cited above. Nor did Commerce make an attempt to gather “best information available”, pursuant to 19 U.S.C. § 1677e (1988)1, in the absence of complete responses to its questionnaires.
*577In at least one other antidumping duty investigation, Commerce recognized the importance of grade and impurities in determining the value of a product. Notice of Final Determination of Sales at Less Than Fair Value: Ferrovanadium and Nitrided Vanadium From the Russian Federation, 60 Fed. Reg. 27957, 27961-62 (Dep’t Comm., 1995). There, Commerce valued vanadium slag, an intermediate product of ferrova-nadium, the final product, which came from Russia and South Africa. After Commerce accounted for the different concentrations of vanadium pentoxide in the Russian slag and the South African slag, the respondents commented that the Russian slag should have been further reduced in value because it contained impurities:
Comment 7: Valuation of Vanadium Slag
* * * Simply adjusting the value for vanadium pentoxide content (“straight-line proportionality” method) is not sufficient, respondents claim, because the additional impurities contained in the Russian slag add to the cost of extracting vanadium pentoxide from the raw material. They argue that this renders the Russian slag less valuable than the prime grade South African Highveld slag, even after adjusting for the different concentration levels of vanadium pentoxide
% * * * 4s *
Based on the submitted information, verification findings, and the Department’s own research, we agree with the respondents that the South African vanadium slag value should be adjusted to reflect the lower purity of the Russian vanadium slag.
Id. at 27961. The foregoing passage demonstrates that Commerce recognizes the difference between concentration and impurities in determining the value of an industrial substance.
The issue before the court is whether, under these circumstances, Commerce’s decision to allow an adjustment for coumarin by-products without accounting for the downward effect of impurities on the byproducts’ value is supported by substantial evidence on the record. The court finds that even with the deference owed to Commerce’s decision, it must vacate and the final determination is remanded insofar as it applies an offset to cost of manufacture for by-products without accounting for the impurities they are alleged to contain.
First, it is impossible for this court to perform fully the required review of Commerce’s action without a complete statement of the reasoning behind it. Comment 8 of the Final Determination states that Rhone-Poulenc argued “that no by-product offsets should be made to FMV in this investigation because * * * the respondents failed to provide the Department with sufficient information regarding the grade, quality, purity, and after-separation costs of the by-products. ” Final Determination, 59 Fed. Reg. at 66900-01. Commerce responded, in pertinent part, that “the cost to manufacture coumarin should be offset by the value of by-product recovered * * * adjusted for concentration lev*578els.” Id. at 66901. The response makes no mention of purity, and offers no justification for ignoring it in valuing the respondents’ by-products.
Commerce’s failure to provide the basis for its decision not to adjust by-product values for impurities “precludes the court from fulfilling its statutory obligation on review.” Indus. Fasteners Group, 2 CIT at 190, 525 F.Supp. at 893. The court cannot infer Commerce’s position on this issue, nor may it substitute its own judgment for that of the agency. Negev Phosphates, Ltd, 12 CIT at 1076-77, 699 F.Supp. at 942.
Furthermore, Commerce’s failure to perform an independent investigation of the facts related to this issue falls short of its statutory duty to investigate antidumping cases and assign fair antidumping margins. See Bomont Indus. v. United States, 13 CIT 546, 552, 718 F. Supp. 958, 964 (1989), citing Budd Co., 1 CIT at 72, 507 F. Supp. at 1001; see also Freeport Minerals Co. v. United States, 4 Fed. Cir. (T) 16, 776 F.2d 1029 (1985) (ITA abused its discretion by revoking antidumping duty order without first obtaining certain sales data); Timken Co. v. United States, 10 CIT 86, 91-92, 630 F. Supp. 1327, 1332-33 (1986) (ITA abused its discretion in failing to obtain information relevant to the revocation of an antidumping duty order). “Antidumping proceedings are investigatory, not adjudicatory.” Timken, 10 C.I.T. at 92, 630 F. Supp at 1333.
In Bomont, Commerce’s negative determination in an antidumping duty investigation was vacated and remanded for Commerce’s failure to investigate alleged transshipments of goods. Bomont Indus., 13 CIT at 552, 718 F.Supp. at 963-64. The court held that “the provisions of the Trade Agreements Act of 1979 clearly express congressional intent that the administrative agency will obtain information through its own investigative efforts.” Id. at 552, 718 F.Supp. at 964. Bomont cited Budd Co. as authority for this proposition, taking a holding in an ITC case and applying it to Commerce. Id. In Budd, the ITC made a finding of “no injury” to the domestic railcar industry from imports of finished railcars or railcar parts. 1 CIT at 69, 507 F.Supp. at 999. The ITC was held to have failed to make a diligent investigation of allegations regarding the importation of railcar parts. Id., passim. This failure violated the requirement implicit in the antidumping law, that the ITC “take an active role in obtaining information prior to its determination,” id. at 71, 507 F.Supp. at 1001, and that “there has been a lack of diligence by the Commission in seeking information available to it in the absence of such evidence having been presented by any of the parties to the investigative proceedings.” Id. at 74, 507 F. Supp. at 1003.
Commerce’s primary argument in defense of its actions is that Rhone Poulenc failed to provide information on the valuation in India of low-grade or impure chemicals. Rather, Commerce relied on the only information that it received from Rhone-Poulenc, viz., published lists of prices for chemicals in India that make no mention of impurities. The antidumping law, 19 U.S.C. § 1677e© (1988), however, required Commerce to use the “best information otherwise available” in the absence of information provided by the parties to the investigation. And as the *579Budd case established, “best information otherwise available” means that, in this context, “all information that is $accessible or may be obtained,’ from whatever its source may be, must be reasonably sought by” Commerce. 1 CIT at 75, 507 F. Supp. at 1003.
The statute does not say “best information readily available.” BIA is neither the absolute best information on an issue, nor merely information already in Commerce’s possession. Rather, BIA implies that Commerce has a duty to reasonably avail itself of its own powers to obtain relevant information in the absence of cooperation by a respondent. Commerce may not rest and require a petitioner to perform its investigation for it. This case does not present, however, the issue of how far Commerce must go to gather BIA, because here Commerce made no effort at all to do so. It is this failure that mandates remand.
For the foregoing reasons, the Court remands this matter to Commerce in order for it to perform a thorough investigation of the proper valuation of the respondents’ by-products, or to obtain and apply BIA regarding coumarin by-product amounts and surrogate country pricing. In the alternative, Commerce may set forth its valid reasons, if any, for ignoring effect of impurities on the value of Respondents’ by-products. Moreover, Commerce is to promulgate its reasons or results and rationale as to the proper valuation of these substances and the valuation’s effect on the respondents’ costs of manufacturing and dumping margins in the Federal Register within 90 days of the entry of this Opinion and Order.
B
Commerce Did Not Err By Not Using BIA to Calculate FMV for TNPIEC and JNPIEC
Rhone-Poulenc next asserts that Commerce erred in not applying BIA to derive the FMV for JNPIEC’s and TNPIEC’s coumarin exports. This argument focuses on PRC coumarin producers, not exporters. The PRC government did not cooperate with Commerce’s investigation in that it failed to provide information about the PRC’s coumarin manufacturing industry. Moreover, unlike JNPIEC and TNPIEC, there is no evidence in the record that coumarin manufacturers are independent of PRC government control. Therefore, Rhone-Poulenc concludes, Commerce should have applied the BIA-based “all others” antidumping margin to the coumarin that JNPIEC and TNPIEC exported to the United States. In contrast, Commerce’s approach was to ascertain which specific producers made the coumarin exported by JNPIEC and TNPIEC, and apply their actual factors of production data to determine FMV and the appropriate antidumping margins.
In examining this argument, the court is mindful of the deference that must be accorded Commerce’s determination. See, 19 U.S.C. § 1516a(b)(1)(B). Commerce’s method for determining JNPIEC’s and TNPIEC’s anti-dumping margins can be invalidated only if it is “not in accordance law”. 19 U.S.C. § 1516a(b).
*58019 U.S.C. § 1677b(c)(l) mandates a method for determining foreign market value of goods from non-market economy countries:
If—
(A) the merchandise under investigation is exported from a nonmarket economy country, and2
(B) (Commerce) finds that available information does not permit the foreign market value of the merchandise to be determined under subsection (a) of this section,
(Commerce) shall determine the foreign market value of the 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, as required by subsection (e) of this section. Except as provided in paragraph (2), the valuation of the factors of production shall be based on the best available information regarding the values of such factors in a market economy country or countries considered to be appropriate by (Commerce).
19 U.S.C. § 1677b© (1988). Subsection (a), referred to above, instructs Commerce to determine FMV based on actual market sales, which clearly is not possible in a state-controlled economy such as the PRC’s.
Subsection (e), referred to in the statute set forth above, instructs Commerce to use certain methods to determine constructed value:
(1) Determination — For the purposes of this subtitle, the constructed value of imported merchandise shall be the sum of —
(A) the cost of materials * * * and of fabrication or other processing of any kind employed in producing such or similar merchandise * * * ;
(B) an amount for general expenses and profit equal to that usually reflected in sales of merchandise of the same general class or kind as the merchandise under consideration which are made by producers in the country of exportation, in the usual commercial quantities and in the ordinary course of trade * * * ; and
(C) the cost of all containers and covering of whatever nature, and all other expenses incidental to placing the merchandise under consideration in condition, packed ready for shipment to the United States.
19 U.S.C. § 1677b(e) (1988). Commerce’s regulations and its Import Administration Anti-Dumping Manual reflect these statutory requirements. See 19 C.F.R. §§ 353.50, 353.52; Antidumping Manual, Import Admin., ch. 8, pp. 64-74 (Dep’t Comm., Int’l Trade Admin., 1994).
In 1991, Commerce recognized the relaxation of government control of the PRC’s economy, and began to allow PRC commercial entities to *581establish their independence from state control on a case-by-case basis. Notice of Final Determination of Sales at Less Than Fair Value: Sparklers from the People’s Republic of China, Comment 1 and DOC Position, 56 Fed. Reg. 20588, 20589 (Dep’t Comm. 1991); Notice of Final Determination of Sales at Less Than Fair Value: Silicon Carbide from the People’s Republic of China, 59 Fed. Reg. 22585, 22586-87 (Dep’t Comm. 1994) (embellishing Sparklers’ test for determining independence from state control) (Dep’t Comm. 1993). If Commerce is satisfied that an entity is indeed independent of state control, then Commerce will determine that entity’s anti-dumping margin separately from other, state-controlled enterprises. Id.
In practice, Commerce determines the separate rate for the exporter of merchandise. See, e.g., Sparklers, supra; Silicon Carbide, supra; Notice of Final Determination of Sales at Less Than Fair Value: Certain Cased Pencils From the People’s Republic of China, 59 Fed. Reg. 55625, 55626-28 (Dep’t Comm. 1994); Notice of Final Determination of Sales at Less Than Fair Value: Disposable Pocket Lighters from the People’s Republic of China, 60 Fed. Reg. 22359, 22359-61 (Dep’t Comm. 1995)3. It will identify or “pair” an exporter’s margin with a particular producer only when it finds that the antidumping margin for the pair is zero. The reason behind this “pairing” is that a zero-margin exporter is spared subsequent antidumping reviews, while other separate rate exporters are not. This latter group would be subject to higher antidumping duties on review if they purvey merchandise of a producer whose FMV is higher than their prior producer. See Final Determination at 66899; Pocket Lighters at 22363.
Nothing in Rhone-Poulenc’s briefs or oral argument convinces the court that Commerce’s methods, described above, are contrary to law. In fact, these methods are consistent with the letter and spirit of the anti-dumping law. Once it has determined that an exporter operates independently, Commerce must determine a margin that fairly reflects the true value of the imported goods using the statutory methods. Identifying, where possible, the actual FMV of those goods requires analysis of the factors of production that went into them. If Commerce can identify the facility that produced the goods, and the factors of production at that facility, then Commerce may determine FMV based on those specific factors.
Commerce did that here. It applied the Sparklers and Silicon Carbide tests to determine that JNPIEC and TNPIEC are independent of state control. It then determined that the two exporters’ coumarin was manufactured by Changzhou and Tianjian Perfumeries, respectively. Commerce ascertained the production factors that these manufacturers used and then determined the prices of those factors in India, an appropriate surrogate market country. This procedure was not contrary to law.
*582Despite Rhone-Poulenc’s arguments, Commerce was not bound to apply Best Information Available to JNPIEC’s and TNPIEC’s dumping margins. Commerce obtained accurate information about these exporters’ coumarin FMV’s, and therefore had no need to resort to BIA. Commerce was not required to “punish” these exporters for the PRC government’s failure to cooperate in the investigation as to other cou-marin exporters and producers. The statute indicates as much in plain language:
(B) Determinations to be made on best information available.— In making (its) determinations under this subtitle, (Commerce) shall, whenever a party or any other person refuses or is unable to produce information requested in a timely manner and in the form required, or otherwise significantly impedes an investigation, use the best information otherwise available.
19 U.S.C. § 1677e(c)(1988).4 JNPIEC and TNPIEC cooperated to Commerce’s satisfaction, provided the requisite information, timely and in proper form, and therefore did not come within the ambit of the BIA provision.
Rhone-Poulenc argues at length about the policy and economic ramifications of the analysis that Commerce used. Those arguments do not address the real question here, which is not whether the court can devise a better analytical method, but rather whether Commerce’s actions are consistent with law and supported by evidence in the record. See 19 U.S.C. §1516a(b). The court finds that Commerce’s actions are consistent with law, and that they are amply supported by the record. Therefore, Commerce’s calculation of separate dumping rates for JNPIEC and TNPIEC based on FMV from their suppliers’ factories is not error.
C
Commerce Did Not Err In Failing to Limit TNPIEC’s and JNPIEC’S Separate Antidumping Rates to Coumarin Imports Manufactured by Tianjin and Changzhou Perfumeries
Rhone-Poulenc next asserts that Commerce erred in not limiting separate antidumping margins assigned to JNPIEC and TNPIEC to cou-marin manufactured by the two producers upon whom the margins were based. Rhone-Poulenc argues that Commerce should have ordered the application of the higher “all others” rate to coumarin exports by JNPIEC and TNPIEC if they change producers. Commerce responds that its actions were not contrary to law, and therefore not error. For the following reasons, the court agrees.
Commerce determines antidumping margins pursuant to 19 U.S.C. § 1673b(b) and (d)(2). Subsection 1673b(b) requires Commerce to
make a determination* * * of whether there is a reasonable basis to believe or suspect that the merchandise is being sold, or is likely to *583be sold, at less than fair value. If the determination of [Commerce] is affirmative, the determination shall include the estimated average amount by which the foreign market value exceeds the United States price.
19 U.S.C. § 1673b(b)(l)(A) (1988). Followiiig this determination, Commerce is directed to
order the posting of a cash deposit, bond, or other security, as it deems appropriate, for each entry of the merchandise concerned equal to the estimated average amount by which the foreign market value exceeds the United States price * * *.
19 U.S.C. § 1673b(d)(2) (1988).
Commerce complied with the foregoing requirements. After an investigation, it determined that there was reason to believe that JNPIEC and TNPIEC were selling coumarin from the PRC in the United States at less than FMV Commerce then determined the estimated average amount by which the actual United States price of coumarin sold by these exporters exceeded its FMV¡ calculated pursuant to the statute, in sales that came to light during the investigation. The coumarin that changed hands in these actual transactions was produced by the Changzhou and Tianjin perfumeries. There is nothing in the record to cause Commerce to believe that JNPIEC or TNPIEC had sold or would sell any other coumarin in the United States market. All Rhone-Poulenc offered was speculation regarding a potential occurrence unsupported by evidence of any actual transaction. There is nothing in the statute that requires Commerce to subject these companies to alternative anti-dumping margins based on a will-o'-the-wisp.
Moreover, the antidumping law does not require Commerce to investigate hypothetical transactions or speculate that an exporter may find suppliers of merchandise other than those it has used in the past. The statute’s administrative review procedures provides the method for accounting for any such changes, in order to assure that appropriate anti-dumping duties are assessed. See 19 U.S.C. § 1675 (1988). It is not for this Court to force its interpretation on Commerce where the statute is silent. See, e.g.,Daewoo Elecs. Co. v. The United States, 6 F.3d 1511 (Fed. Cir. 1993).
Rhone-Poulenc’s primary argument is that Commerce deviated from its prior practice in failing to limit the applicability of JNPIEC’s and TNPIEC’s separate antidumping rates. In a 1993 Federal Register Notice, Commerce adopted the position that Rhone-Poulenc asserts. Final Determination of Sales at Less Than Fair Value: Sulfur Dyes, Including Sulfur Vat Dyes, From the People’s Republic of China, 58 Fed. Reg. 7537, 7543 (Dep’t Comm. 1993). Sulfur Dyes is the sole example of “prior established practice” that Rhone Poulenc presented to the Court. Shortly thereafter, Commerce ceased limiting separate rates to specific producers except where the combination produced a “zero” antidump-ing margin. See Notice of Final Determination of Sales at Less Than Fair Value: Certain Cased Pencils from the People’s Republic of China, 59 *584Fed. Reg. 55625 (Dep’t Comm. 1994); Notice of Final Determination of Sales at Less Than Fair Value: Saccharin from the People’s Republic of China, 59 Fed. Reg. 58818 (Dep’t Comm. 1994); Notice of Final Determination of Sales at Less Than Fair Value: Certain Paper Clips From the People’s Republic of China, 59 Fed. Reg. 51168 (Dep’t Comm. 1994).
Commerce gave an adequate explanation of why it did not link .JNPIEC’s and TNPIEC’s rates to their suppliers in the Final Determination. Final Determination at 66899 and footnote 5. It explained the rationale behind its policy more fully in a later Federal Register Notice, published in May 1995:
Recent Department practice has been to assign rates only to exporters except in the case of producer/exporter combinations that have been found not to be dumping. (See e.g., Pencils, Saccharin, Couma-rin, and Final Antidumping Duty Determination: Certain Cased Pencils from the People’s Republic of China, 59 FR 55625, November 8,1994, where the Department assigned a zero rate to a producer/exporter for purposes of exclusion from the order, but the remaining rates were assigned to exporters only.) Where a producer/exporter combination is found not to be dumping, it is appropriate to publish a rate that applies to that producer/exporter combination because they are excluded from the order and, therefore, future administrative reviews. However, all other exporters remain subject to the order and administrative reviews. Hence* * * those exporters have no incentive to export the output of producers that might yield a high FMV unless they adjust their U.S. prices accordingly. If they fail to do so, an administrative review would result in an assessment of additional duties, with interest, and a higher cash deposit rate for future entries.
Notice of Final Determination of Sales at Less Than Fair Value: Disposable Pocket Lighters From the People’s Republic of China, 60 Fed. Reg. 22359, 22364 (Dep’t Comm. 1995)5.
Because the court finds the explanation given by Commerce in the Final Determination to be adequate, it need not reach the question of whether counsel’s offer of the explanation in Pocket Lighters is an impermissible post hoc rationalization. It would have been preferable, of course, to have had a fuller explanation of Commerce’s policy in the Final Determination. The court’s role here, however, is to determine the adequacy of the explanation, not to remand for a mere improvement of it.
Finally, Rhone-Poulenc asserts that Commerce’s action is inconsistent with Congressional policy. It suggests that Commerce’s approach fails to set accurate antidumping rates ab initio and that it would require an administrative review to achieve an accurate antidumping *585duty if TNPIEC and JNPIEC should sell the product of new suppliers of coumarin. This argument, however, is based on false premises.
As discussed above, with the exception of the allowance for by-products, Commerce calculated antidumping rates here consistently with statutory requirements. The rates accurately reflect the actual transactions that occurred in the period of investigation. Speculation on whether TNPIEC and JNPIEC might change their coumarin suppliers in the future does not change this conclusion.
D
Commerce Did Not Err in Its Calculation of General Expenses and Profit As Part of FMV
Rhone Poulenc’s final argument is that Commerce erred in determining the amount of general expenses and profit to be included in foreign market value. Rhone-Poulenc takes issue specifically with Commerce’s decision to include an amount for producer costs in FMV but not for reseller costs. Commerce’s determination was not contrary to law; therefore it need not be remanded.
Congress directed Commerce to treat cases involving goods from non-market economies differently from market economy cases. 19 U.S.C. § 1677b(c). The rationale behind this decision is that such countries “do{} not operate on market principles of cost or pricing structures, so that sales of merchandise in such country do not reflect the fair value of the merchandise.” 19 U.S.C. § 1677(18).
In the absence of useful pricing data from NME countries, Commerce constructs a hybrid price for subject goods, taking the amounts of inputs (“factors of production”) used by the NME manufacturer and valuing them with price data from a similar market economy. See 19 U.S.C. § 1677b(e)(l). The statute directs Commerce to add to this factors-of-production sum an additional amount for general expenses:
[Commerce] shall determine the foreign market value of the 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 * * * as required by subsection (e) of this section.
19 U.S.C. § 1677b(c)(1). Subsection (e) sets forth requirements for calculating general expenses and profit:
an amount for general expenses and profit equal to that usually reflected in sales of merchandise of the same general class or kind as the merchandise under consideration which are made by producers in the country of exportation, * * * except that —
(i) the amount for general expenses shall not be less than 10 percent of the cost [of materials and fabrication or other processing], and
(ii) the amount for profit shall not be less than 8 percent of the sum of such general expenses and cost * * *.
19 U.S.C. § 1677b(e)(1)(B) (emphasis added).
*586The foregoing subsection is the only statutory guidance on the calculation of general expenses and profit. In it, the plural “are made by producers” has only one plural antecedent, viz., “sales of merchandise.” Congress thus expressly directed Commerce to include in FMV general expense datá that relate only to producer sales. There is no mention of reseller sales. Similarly, profit is to be calculated with reference to producer sales. Therefore, it is reasonable to conclude that the profit cited is to be that of the producer and not a reseller. Finally, when read in conjunction with subsections 1677(18) and 1677b(c), supra, it is apparent that these- amounts are to be calculated based on surrogate country data.
Under the foregoing statutory regime, Commerce’s action cannot be found contrary to law. First, Commerce determined the cost of manufacturing using the factors of production of the PRC coumarin producers and economic data from India, an appropriate surrogate country. Calculation Memorandum, C.R. Doc. 59, p. 1. Then, Commerce consulted the Reserve Bank of India Bulletin (“RBI”), which contains data on income and expenditures of Indian companies engaged in the processing and manufacture of chemicals. Id. at p. 12, and Exh. 13 to Calculation Memorandum. Commerce selected the appropriate cost categories for inclusion in general expenses: managerial remuneration, “other expenses less debt”, bad debt, selling commissions, “other provisions”, non-operating deficit and interest. Calculation Memorandum at p. 13. Then, Commerce determined the percentage of Indian manufacturers’ cost of manufacturing that the general expense amount constituted. Id. This number was 17.87%, higher than the statutory minimum of 10%. Id. Therefore, Commerce used that percentage to determine the appropriate amount of general expenses to include in FMV
Similarly, Commerce attempted to calculate the profit rate by consulting the RBI. The Indian producers’ profit rate, based on RBI data, was 5.81%. Id. This number is less than the statutory minimum of 8%. Therefore, Commerce calculated profit for purposes of FMV at the rate of 8%. Id.
The foregoing methods of calculation are consistent with the law, and with Commerce’s regulations, 19 C.F.R. § 353.52(c). Rhone-Poulenc’s arguments to the contrary are unavailing. The foregoing statutory analysis answers Rhone Poulenc’s contention that reseller’s general expenses and profits should have been included here. It is a reasonable interpretation of the statute to conclude that they need not have been included.6 Nor has Rhone-Poulenc offered convincing evidence of a prior contrary practice. Commerce’s Antidumping Manual expressly *587provides for NME-related investigation methods distinct from those applicable in market economies. Antidumping Manual, Ch. 8, sec. XVI. The treatment of exports from market economies has no bearing here, whether contained in the Manual or in a prior Federal Register Notice, see Final Determination of Sales at Less Than Fair Market Value: Fresh and Chilled Atlantic Salmon From Norway, 56 Fed. Reg. 7661 (1991), given contrary statutory language.
IV
Conclusion
For the foregoing reasons, the court grants Plaintiffs Motion for Judgment on the Agency Record insofar as it challenges Commerce’s determination of the offset for by-products applicable to TNPIEC’s and JNPIEC’s cost of manufacturing and remands for further proceedings regarding same, consistent with this Opinion; and the Court denies Plaintiffs motion in all other respects.
Because Commerce’s investigation in this case predated January 1,1995, the 1994 amendments to the antidump-ing law do not apply. Pub. L. No. 103-465 sec. 291.
Paragraph (2) of 19 U.S.C. § 1677b© provides an exception to paragraph (l)’s method of determining FMV in a non-market economy applicable in cases where available information does not allow merchandise to be valued fully under paragraph (1). In such cases, Commerce is directed to base valuation on the price of “comparable” merchandise produced in one or more market economies. 19 U.S.C. § 1677b(c)(2). Paragraph (1) allows valuation here, so this provision does not apply.
Note that Commerce issued this final determination after the final determination in the instant case.
While this section of the Code was amended in 1994, the amendment does not affect this case. It applies only to those investigations initiated on or after January 1,1995 — well after the instant investigation was commenced. 19 U.S.C. § 1677e (1994); P.L. 103-465, sec. 291.
Commerce has sought public comment on whether to change its policy regarding the application of antidumping rate6 to exporters that are not also producers. Proposed Rule, 61 Fed. Reg. 7308, 7311 (Feb. 27, 1996). In the notice, Commerce suggested that one alternative is to “calculate a separate rate for each exporter/producer combination * * * “ Id. Commerce, however, ha6 neither changed its policy nor sought leave of the court to review its position in the context of this action. Therefore, Commerce’s potential rulemaking has no effect here.
Commerce, in its brief, suggested that it did not know whether resellers’ general expenses and profit are included in the RBI statistics that it used. Brief at 44-45. Therefore, Commerce concluded, adding an additional amount for resellers’ general expenses and profit could have resulted in double counting. In that argument Commerce seeks to have its cake and eat it too. Nonetheless, the court need not address this argument and the admissions that support it because the court does not find that Commerce was required to include reseller expenses and no coumarin exporter has claimed an overcharge for such expenses if they were included. Moreover, Rhone-Poulenc admitted that the RBI information did not include reseller general expenses and profit.