Slip. Op. 99 -12 UNITED STATES COURT OF INTERNATIONAL TRADE BEFORE: RICHARD W. GOLDBERG, JUDGE MICRON TECHNOLOGY, INC., Plaintiff, v. THE UNITED STATES, Court No. 96-06-01529 Defendant, and LG SEMICON CO., LTD., and LG SEMICON AMERICA, INC., Defendant-Intervenors. [Final Results in the first administrative review of an antidumping duty order of the U. S. Department of Commerce is sustained in part, and remanded in part.] Dated: January 28, 1999 Hale & Dorr, LLP (Gilbert B. Kaplan, Michael D. Esch, Paul W. Jameson, Cris R. Revaz, and John M. Ryan), for plaintiff Micron Technology, Inc. Frank W. Hunger, Assistant Attorney General; David M. Cohen, Director, Commercial Litigation Branch, Civil Division, United States Department of Justice (Cynthia B. Schultz); Office of the Chief Counsel for Import Administration, United States Department of Commerce (Patrick V. Gallagher, Jr. of counsel, for defendant. Kaye, Scholer, Fierman, Hays & Handler, LLP (Michael P. House and Raymond Paretzky), for defendant-intervenors LG Semicon Co., and LG Semicon America, Inc. OPINION GOLDBERG, Judge: In this action, the Court reviews certain aspects of the Department of Commerce's (" Commerce") Notice of Final Results of Antidumping Administrative Review: Dynamic Random Access Memory Semiconductors of One Megabit or Above From the Republic of Korea, 61 Fed. Reg. 20216 (May 6, 1996) (" Final Results"). More specifically, plaintiff, Micron Technology, Inc. (" Micron"), petitioner in the underlying administrative review, contests five aspects of the Final Results. The Court exercises jurisdiction to review this motion for judgment on the agency record pursuant to 28 U. S. C. º 1581( c) (1994). The Court sustains the Final Results in part, and remands in part. I. BACKGROUND Micron, a U. S. manufacturer of dynamic random access memory semiconductors (" DRAMS"), filed a petition with Commerce in April 1992, alleging that Korean producers of DRAMS were selling subject merchandise in the United States at less than fair value. Following an antidumping investigation, Commerce published an antidumping order on DRAMS from Korea in May 1993. See 58 Fed. Reg. 27520 (May 10, 1993). In the first anniversary month of the order, three Korean respondents, including LG Semicon Co., Ltd. and LG Semicon America, Inc. (collectively "LG Semicon"), and Micron requested an administrative review of the DRAMS order. 1 On June 15, 1994, Commerce initiated a review of the three Korean manufacturers, covering the period October 29, 1992 through April 30, 1994. See Notice of Initiation of Antidumping Administrative Review, 59 Fed. Reg. 30770, 30771 (1994). In the Final Results of the review, Commerce assigned a dumping margin of 0.00% to LG Semicon. See 61 Fed. Reg. at 20222. Micron objects to five aspects of Commerce's Final Results as they pertain to LG Semicon. It asserts that Commerce erred (1) when it calculated LG Semicon's research and development (" R& D") costs; (2) in its treatment of LG Semicon's royalty payments; (3) when it decided to allocate certain indirect selling expenses reported by LG Semicon; (4) in its treatment of LG Semicon's reported loan fees; and (5) in its treatment of LG Semicon's U. S. trading company. Commerce agrees with Micron only insofar as it requests that the first issue, the calculation of R& D expenses, should be remanded for further review. Commerce opposes the remaining challenges to the Final Results. LG Semicon opposes all challenges. (1) The underlying administrative review was conducted prior to January 1, 1995. Consequently, the applicable law in this case is the antidumping statute as it existed prior to the amendments made by the Uruguay Round Agreements Act, Pub. L. No. 103-465, 108 Stat. 4809 (1994). See Torrington Co. v. United States, __ Fed. Cir. (T) __, __, 68 F. 3d 1347, 1352 (1995). II. STANDARD OF REVIEW Commerce's determination will be sustained if it is supported by substantial evidence on the record and is otherwise in accordance with law. See 19 U. S. C. º 1516a( b)( 1)( B) (1994). To determine whether Commerce's interpretation of the statute is in accordance with law, the court applies the two-prong test set forth in Chevron U. S. A., Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984). Chevron first directs the court "to determine whether Congress has directly spoken to the precise question at issue." Id. at 842-43 (internal quotations and citations omitted). In doing so, the court must inquire "whether Congress's purpose and intent on the question at issue is judicially ascertainable." Timex V. I., Inc. v. United States, __ Fed. Cir. (T) __, , 157 F. 3d 879, 881 (1998) (citing Chevron, 467 U. S. at 842-43 & n. 9). Congress's purpose and intent must be divined using the traditional tools of statutory construction. Id. at 882 (citation omitted). Of course, the "first and foremost tool to be used is the statute's text," and "if the text answers the question, that is the end of the matter." Id. (citations and internal quotation omitted). In addition to the plain language of the statute, the other tools include, the statute's structure, canons of statutory interpretation, and legislative history. See id. (citing Dunn v. Commodity Futures Trading Comm'n, 117 S. Ct. 913, 916-20 (1997); Chevron, 467 U. S. at 859-63; Oshkosh Truck Corp. v. United States, 123 F. 3d 1477, 1481 (Fed. Cir. 1997)). If, using these tools, Congress's intent is unambiguous as to the issue at hand, then the court must give effect to the intent of Congress. On the other hand, if Congress's intent is "silent or ambiguous with respect to the specific issue, the question for the court is whether the agency's answer is based on a permissible construction of the statute." Chevron, 467 U. S. at 843 (footnote omitted). Thus, the second prong of the Chevron test directs the court to consider the reasonableness of an agency's interpretation. If asked to review Commerce's factual findings, the court will uphold the agency if its findings are supported by substantial evidence. "Substantial evidence is something more than a 'mere scintilla, ' and must be enough reasonably to support a conclusion." Ceramica Regiomenta, S. A. v. United States, 10 CIT 399, 405, 636 F. Supp. 961, 966 (1986) (citations omitted), aff'd, 5 Fed. Cir. (T) 77, 810 F. 2d 1137 (1987). In applying this standard, the court affirms Commerce'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. See Atlantic Sugar, Ltd. v. United States, 2 Fed. Cir. (T) 130, 744 F. 2d 1556, 1563 (1984). III. DISCUSSION A. Calculation of R& D Expenses Micron first argues that Commerce's calculation of LG Semicon's research and development (" R& D") expenses is incorrect. Micron points in particular to Commerce's statement that it "relied on LGS's accounting system to determine the total R& D figure applicable to the analysis: it amortized any R& D expenses that LGS amortized in its own books and records and it expensed any R& D expenses that LGS expensed." Final Results, 61 Fed. Reg. at 20219. Micron contends that Commerce acted contrary to this statement in the Final Results: it correctly included R& D expenses from LG Semicon's financial statements that were incurred in 1993, yet erroneously failed to include other R& D expenses that were also expensed in 1993, though incurred and amortized prior to 1993. Micron asserts that the nature of this error is clerical. Therefore, Micron requests that the Court remand to Commerce with instructions to include in its R& D calculation all costs expensed in 1993, regardless of whether the R& D expenses were actually incurred in 1993. Commerce agrees that a remand is appropriate on this issue. More specifically, Commerce requests that the Court remand "to reconsider its calculation of LG Semicon's R& D costs incurred in 1993 in light of this Court's remand to Commerce in the LTFV investigation, 2 ordering Commerce to amortize LG Semicon's R& D expenses, rather than expense them in its calculation." See Def. 's Br. In Opp. to Mot. for J. on the Agency R., at 2. (2) The parties appealed various aspects of Commerce's final determination in the less-than-fair-value investigation, including certain R& D issues. See Micron Technology, Inc. v. United States, 19 CIT 829, 893 F. Supp. 21 (1995) (" Micron I"). The Court first notes that a remand request by Commerce should not dictate the action subsequently taken by the court. See Gulf States Tube Div. of Quanex Corp. v. United States, 21 CIT __, __, 981 F. Supp. 630, 647 (1997). Nonetheless, the Court deems that a remand is appropriate in this instance, but for reasons other than those articulated by plaintiff and defendant. As noted above, Micron cites to Commerce's Final Results for the proposition that Commerce intended to "amortize[] any R& D expenses that LGS amortized in its own books and records and [] expense[] any R& D expenses that LGS expensed." Final Results, 61 Fed. Reg. at 20219. Micron's citation, however, obscures the context of Commerce's statement. As evidenced below, the statement excerpted by Micron falls, not within a discussion of how to calculate total R& D expenses, but rather how to calculate a specific type of R& D expense, purchased R& D. Comment 6: LGS asserts that the Department should accept amortization of purchased R& D amounts over the relevant contract period. LGS argues that the Department's decision in the preliminary determination to expense purchased R& D in the year incurred is inconsistent with the CIT decision in the less-than-fair-value investigation. See Micron I. LGS asserts that the Micron decision requires the Department to amortize R& D expenses over the life cycle of the product. The petitioner argues that LGS's own financial statements expensed purchased R& D in the year incurred. Therefore, all payments related to the purchased R& D should be acknowledged in the year in which they were incurred, since this is how the expenses were recorded in the company's books and records. DOC Position: We agree with petitioner that LGS's purchased R& D expenses should be acknowledged in the year in which they were incurred, since this is how the expenses were recorded in the company's books and records. See LGS COP/ CV Verification Report of July 26, 1995 at page 8. Moreover, the [Micron I] decision requires the Department to allow the allocation of R&D expenses over time, when the allocation is made in accordance with generally accepted accounting practices in effect in the home country, and when Commerce is satisfied that those principles reasonably reflect the costs associated with the production of the subject merchandise. In this case, although the Korean GAAP may allow LGS to amortize its purchased R&D over a given period, LGS did not do so. Rather, LGS expensed purchased R&D for its financial statements, and amortized it over a longer period for the antidumping response. In these calculations, the Department relied on LGS's accounting system to determine the total R& D figure applicable to the analysis: it amortized any R& D expenses that LGS amortized in its own books and records and it expensed any R& D expenses that LGS expensed. Final Results, 61 Fed. Reg. at 20219. Contrary to what Micron would have the Court believe, when placed in context it is not apparent whether Commerce intended the excerpted statement, i. e., the last sentence, to serve as a methodology for calculating all R& D expenses, or only purchased R& D expenses. That is, it can be inferred that the phrase "[ i] n these calculations," from the last sentence refers to the calculation of purchased R& D expenses, not total R& D expenses, thereby implying that the R& D expenses referred to later in the sentence are also purchased R& D expenses. (3) (3) If this is the case, the existing margin calculation, i. e., where purchased R& D costs are expensed in the year incurred, makes sense. In other words, when Commerce relied on LG Semicon's accounting system to calculate purchased R& D expenses, it effectively excluded the option of amortizing purchased R& D expenses because, as Commerce points out, LG Semicon's records acknowledged its purchased R& D expenses in the year in which they were incurred, not on an amortized basis. On the other hand, the Court agrees that the last sentence could also be construed as an attempt to establish Commerce's methodology for calculating all R& D expenses, not just purchased R& D expenses. Because the Court does not presume to opine on which view Commerce actually holds, it is appropriate to remand this issue so that Commerce may clarify the meaning of this last sentence from Comment 6. In doing so, Commerce should clarify the precise methodology it has used to calculate total R& D expenses. This is not a simple remand to correct for a clerical error; because Commerce failed to articulate clearly whether the methodologies used to calculate total R& D expenses as opposed to purchased R& D expenses are the same or different, it must do more than correct for a clerical error. The Court also cautions that Commerce should ensure that its clarified methodology is non-distortive and that it accurately and reasonably reflects costs. In particular, the Court notes that if Commerce continues to base its total R& D figure on those costs expensed in 1993, it should refrain from including in this figure those R& D costs expensed in 1993, yet incurred prior to 1993. Basing the total R& D figure on costs actually incurred and expensed in 1993 plus costs expensed in 1993, yet incurred prior to 1993 conflates the amortizing and expensing methodologies and is plainly distortive. It effectively results in double counting and, as such, should be rejected. See, e. g., Hussey Copper, Ltd. v. United States, 17 CIT 993, 999-1000, 834 F. Supp. 413, 420 (1993) (ordering remand to correct for possible double counting of credit expense). B. Treatment of Royalty Payments Micron contends that Commerce erred when it declined to make a circumstance of sale (" COS") adjustment to account for alleged differences in LG Semicon's royalty payments. Specifically, Micron argues that because different royalties were paid to two customers depending on whether the merchandise was sold in the United States or home market and because the royalties were paid on the basis of sales value, a COS adjustment to U. S. and home market prices should have been made to account for the alleged discrepancy. This claim is without merit. First, contrary to Micron's argument, the evidence of record plainly establishes that LG Semicon made royalty payments to one of the customers at the same rate in both the United States and Korea. More precisely, LG Semicon produced a royalty agreement for the customer, showing that the royalty rate was the same on both U. S. and Korean sales. See LG Semicon's Supplemental Sales Resp. (Oct. 19, 1994), C. R. Doc. 33, at App. SS-13 (providing a royalty agreement that defined "NET SALES BILLED" as DRAM sales "in the United States and Korea" and setting an identical fixed percentage for the royalty rate on the "NET SALES BILLED"). Thus, with respect to one of the two royalties at issue, the Court has reviewed the record evidence in detail and determined there is no factual basis for Micron's argument. (4) (4) Micron also argues that Commerce erred when it failed to verify the information contained in the royalty agreement. This argument does not withstand scrutiny. The terms of the royalty agreement provided to Commerce are plain, and there is no suggestion that the agreement itself is bogus. As such, the evidence presented to Commerce amounted to uncontroverted record evidence that the royalty rate for this customer was the same in the United States and Korea. Moreover, it is well established that Commerce has the discretion not to verify each piece of evidence made part of the record. See, e. g., Monsanto Co. v. United States, 12 CIT 937, 944, 698 F. Supp. 275, 281 (1988). Commerce certainly acted within its discretion when it decided that limited resources should not be allocated to verify the foundation of a royalty agreement. More generally, Commerce's decision to treat the royalty payments as a cost of manufacture, rather than as a selling expense that required a COS adjustment, was in accordance with law. As stated in the Final Results, "it has been [Commerce's] longstanding practice to treat royalty payments for production technology as [a] cost of manufacturing, even in circumstances where the royalty payments were based on sales revenue." 61 Fed. Reg. at 20218-19 (citing Extruded Rubber Thread from Malaysia; Final Determination of Sales at Less Than Fair Value, 57 Fed. Reg. 38465 (Aug. 25, 1992) (" Rubber Thread"); Certain Hot-Rolled Carbon Steel Flat Products from Canada; Final Determination of Sales at Less Than Fair Value, 58 Fed. Reg. 37099 (July 9, 1993) (" Canadian Steel")). In both Rubber Thread and Canadian Steel, Commerce expressly determined that a royalty fee paid for production technology should be treated as a cost of manufacturing, not as a selling expense. (5) See 57 Fed. Reg. at 38479-80; 58 Fed. Reg. at 37118. Similarly here, LG Semicon's royalty obligation was based on the purchase of production technology and, therefore, Commerce treated the expenses as a cost of manufacture. The Court finds Commerce's established practice reasonable, as it is based on sound logic: quite simply, a payment made for production technology more properly corresponds to the cost of manufacturing certain merchandise, than to expenses associated with the sale of the merchandise. Accordingly, Commerce's treatment of the royalty payments at issue was in accordance with law and supported by substantial evidence. (5) Micron counters that there is not an established practice because Commerce declined to treat royalties as a cost of manufacture in a 1988 determination. See Color Televisions from Korea; Final Results of Administrative Review, 53 Fed. Reg. 24975 (July 1, 1988) (" Color TVs"). Yet, as Commerce correctly points out, unlike the case at bar, it was unclear in this earlier determination whether the royalty expenses were for production technology or other obligations. Thus, the lone Color TVs determination does not serve to undermine the more recent practice established in Rubber Thread and Canadian Steel. C. Allocation of Home Market Indirect Selling Expenses Typically, Commerce requires indirect selling expenses to be allocated on a sales value basis. See Carbon Steel Butt-Weld Pipe Fittings from Thailand, 57 Fed. Reg. 21065, 21067 (May 18, 1992); Sweaters of Man-made Fibers from Taiwan, 55 Fed. Reg. 34585, 34596 (Aug. 23, 1990); see also Def. 's Br. In Opp. to Mot. for J. on the Agency R., at 17 n. 6 (acknowledging that Commerce usually allocates indirect selling expenses on a sales value basis). In this case, LG Semicon identified its home market selling expenses by subdivision of the company's overall sales division, and reported these expenses on a sales value basis. LG Semicon also reported certain "common expenses," which were not broken out by subdivision. For these additional expenses, LG Semicon offered alternative allocation methodologies to account for the indirect selling expenses. Commerce accepted LG Semicon's alternative allocation methodologies. Micron contends that Commerce erred when it allowed LG Semicon to report certain indirect selling expenses using allocation methodologies other than sales value. Micron particularly complains that a respondent should not be allowed to "pick and choose among allocation methodologies." Pl. 's Br. In Supp. of Mot. for J. on Agency R., at 24. Responding to Micron's challenge on this issue at the administrative level, Commerce stated as follows: It is not our policy to require allocation of indirect selling expenses based upon relative sales value in every instance. More specifically, in the final results of the less-than-fair-value investigation we clearly noted that we would accept an allocation basis other than relative sales value provided the methodology was reasonable. Moreover, we note that Hyundai and LGS used three separate bases of allocation for different selling expenses, one of which was relative sales value. In addition, Hyundai used manpower hours in allocating labor expenses and the number of invoices in allocating accounting department expenses. LGS used a similar methodology to allocate its indirect selling expenses that were not identified by subdivision. We believe that it is more appropriate to allocate human resource and accounting department expenses on the basis of manpower and number of invoices than on the basis of sales value because human resource is a function of the number of employees, and accounting department expense is a function of the volume of invoices prepared. Thus, we believe that these allocation bases are reasonable and have continued to accept them for purposes of these final results of review. Furthermore, we verified HEA[ 's] and LGS's allocation bases for its [sic] indirect selling expenses during our U. S. sales verifications and found no discrepancies or inaccuracies in Hyundai[ 's] or LGS's allocation methodology. Final Results, 61 Fed. Reg. at 20217 (internal citations omitted). From this statement, it is plain that Commerce did not blindly accept LG Semicon's alternative methodologies. Rather, Commerce explained that the circumstances in this case made it more appropriate to allocate certain indirect selling expenses using methodologies other than sales value. The Court agrees. While it is true that Commerce typically allocates indirect selling expenses based on sales value, "[ Commerce] is given discretion in its choice of methodology as long as the chosen methodology is reasonable and [Commerce's] conclusions are supported by substantial evidence in the record." Federal-Mogul Corp. v. United States, 18 CIT 785, 807-08, 862 F. Supp. 384, 405 (1994), aff'd, 810 F. 2d 1137 (Fed. Cir. 1987) (upholding the acceptance of an alternative allocation methodology used to account for indirect selling expenses). As Commerce noted in the Final Results and as evidenced by the confidential record, the allocation methodologies offered by LG Semicon are more appropriate. In particular, each alternative allocation methodology bears a direct relation to the manner in which the common expense is incurred. See LG Semicon's Section V Resp. (Aug. 29, 1994), C. R. Doc. 14, at 31; LG Semicon's Home Market Verification Report (Apr. 13, 1995), C. R. Doc. 90, at 13. In addition, when Commerce verified LG Semicon's alternative methodologies, it "tested the arithmetic accuracy of the allocation, and found no discrepancies." See id.; see also INA Walzlager Schaeffler KG v. United States, 21 CIT __, __, 957 F. Supp. 251, 275 (1997) (upholding the reporting of indirect selling expenses where Commerce verified the accuracy of the reporting and where there was no indication that the verification was deficient or incomplete). The Court has reviewed the alternative methodologies in the confidential record and finds them reasonable in light of the nature of the common expenses at issue. Accordingly, because Commerce's decision to accept LG Semicon's indirect expenses was reasonable and properly verified, the Court sustains the determination as supported by substantial evidence and in accordance with law. D. Adjustment for Loan Guarantee Fees Micron next claims that Commerce failed to account for all the costs associated with LG Semicon's loan guarantees. In particular, Micron argues that in the absence of loan guarantee fees paid by LG Semicon, Commerce should have imputed expenses for hypothetical costs associated with the loans. Micron contends this error was not excused by the fact that under Korean law, a guarantor is not required to charge a fee for related company guarantees unless there is a default. By allowing LG Semicon to report its costs in accordance with the Korean law at issue, Micron maintains that Commerce erroneously elevated a "dubious" foreign law over the strictures of U. S. antidumping law. Pl. 's Br. In Supp. of Mot. for J. on Agency R., at 29. The Court does not agree. First, Commerce extensively verified LG Semicon's financial expenses and determined that LG Semicon made no payments with respect to loan guarantees. See Final Results, 61 Fed. Reg. at 20218; Lg Semicon's Home Market Verification Report (Apr. 13, 1995), C. R. Doc. 90, at 18. And, as Micron noted, Commerce found that Korean law does not require the guarantor of a loan to charge a fee for related party guarantees unless there is a default. Id. Commerce also verified that LG Semicon complied with the Korean law in reporting its loan guarantees. Id. Commerce's decision is therefore supported by substantial evidence. Second, the notion that Commerce should reject LG Semicon's reporting because the Korean law is at odds with U. S. antidumping law is without merit. In this instance, there is no evidence to suggest that LG Semicon actually incurred expenses for the loan fees or would have incurred such expenses were it not for the law. As Commerce points out, without some evidence that actual expenses were incurred or even might have been incurred, Micron's request to impute costs for loan fees is entirely too speculative and is therefore unreasonable. Cf. Koenig & Bauer-Albert AG v. United States, 22 CIT __, __, 15 F. Supp. 2d 834, 848 (1998) (noting that the fundamental purpose of the antidumping statute is to ensure the accurate calculation of dumping margins, and in pursuing this goal, Commerce has the discretion to reject information that does not reflect actual costs). Commerce's decision to accept the reporting of the loan guarantee data is therefore not at odds with the antidumping law. Accordingly, the Court finds that Commerce's accounting of the costs associated with loan guarantees was supported by substantial evidence and otherwise in accordance with law. E. Adjustment to U. S. Price for Trading Company Expenses Finally, Micron claims that Commerce ignored the role that a U. S. trading company played in processing sales for LG Semicon. Micron maintains that Commerce must account for the trading company's role by making a deduction to U. S. price for expenses LG Semicon incurred as a result of the relationship. Again, Micron's claim is without merit. At the administrative level, Commerce addressed Micron's concerns in the following passage: DOC Position: We [] examined LGS's relationship with its trading company. See LGS Home Market Sales Verification Report, pp. 18-19. We verified that LGS did not incur costs for the use of its trading company's name. Moreover, we verified that this trading company did not provide any services to sales of subject merchandise to LGS. Final Results, 61 Fed. Reg. at 20218. Micron does not contest the veracity of the verification report on this issue. Instead, Micron essentially insists that, notwithstanding verification, the trading company played a more integral role in the U. S. sales process than LG Semicon acknowledges, and it is therefore inconceivable that LG Semicon did not incur additional costs relating to the relationship. The record again belies Micron's assertion. At verification, Commerce and LG Semicon engaged in an extensive dialogue pertaining to the role of the trading company, and Commerce verified the nature of this stated relationship. See LG Semicon's Home Market Verification Report (Apr. 13, 1995), C. R. Doc. 90, at 18-19. In addition, Commerce verified that LG Semicon incurred no costs for the use of the trading company's name and that the trading company provided no services related to the sale of subject merchandise. Id. In view of this uncontroverted and exhaustive record, it is unclear what more Micron would have Commerce verify. Accordingly, the Court finds that Commerce's decision on this issue is supported by substantial evidence. IV. CONCLUSION For the foregoing reasons, the Court remands the Final Results to Commerce to clarify its position with regard to calculation of R& D expenses, and sustains the Final Results in all other respects. A separate Order will be entered accordingly. _________________________________ Richard W. Goldberg JUDGE Dated: January 28, 1999 New York, New York.