Hynix Semiconductor, Inc. v. United States

OPINION

The Court reviews the U.S. Department of Commerce’s (“Commerce”) Final Results of Redetermination Pursuant to Court Remand (June 6, 2003) (Def. Conf. App. Ex. 1) (“Remand Results”), filed with the Court in response to its opinion and order in Hynix Semiconductor, Inc. v. United States, 248 F. Supp. 2d 1297 (Ct. Int’l Trade 2003), pursuant to 28 U.S.C. § 1581(c) (2000). The Court directed Commerce to reconsider and further explain its decision to recalculate Plaintiffs’ reported research and development (“R&D”) costs and its decision to reject Plaintiffs’ accounting adjustments for the average useful lives (“AULs”) of Plaintiffs’ semiconductor equipment in Dynamic Random Access Memory Semiconductors of One Megabit or Above From the Republic of Korea: Final Results of Anti-dumping Duty Administrative Review, 66 Fed. Reg. 52,097 (Oct. 12, 2001) (“Final Results”). See Hynix Semiconductor, Inc. v. United States, 248 F. Supp. 2d 1297 (“Hynix”).1 Specifically, the Court ordered Commerce to:

1. Reconsider and further explain why the use of Plaintiffs’ amortized R&D costs would not reasonably reflect Plaintiffs’ actual R&D expenses for this period of review, and to identify what distortions, if any, would arise in the COP calculation if amortized R&D costs were used; and to reconsider and address Plaintiffs’ assertion that all 1996 R&D costs that should have been carried forward into this period of review, if amortized, were fully taken into account prior to or within the Fifth Administrative Review, when Commerce used expensed R&D costs in the cost of production calculation. Id. at 1312-1313.

2. Reconsider and further explain why Plaintiffs’ deferral of certain R&D costs does not reasonably reflect the R&D costs related to the subject merchandise. Id. at 1313.

3. Further explain whether the subject merchandise has benefitted from R&D activities for non-memory products and identify substantial evidence in the record to justify this conclusion. Id. at 1317.

4. Explain how the revised average useful lives reported by Plaintiffs are not standard industry practice; how and where in the record Plaintiffs’ reported AULs were overstated; and whether the use of Plaintiffs’ reported AULs would not reasonably reflect the cost of production. Id. at 1319.

*1721Commerce filed the Remand Results on June 6, 2003. Plaintiffs filed comments on the Remand Results, and Defendant and Defendant-Intervenor subsequently submitted responses to Plaintiffs’ comments.

Standard of Review

The Court will sustain the Remand Results, unless they are “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(l)(B). While substantial evidence “is something less than the weight of the evidence,” Consolo v. Federal Maritime Comm’n, 383 U.S. 607, 620 (1966) (citations omitted), it consists of “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Universal Camera Corp. v. NLRB, 340 U.S. 474, 477 (1951) (quoting Consol. Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)). “[T]he possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Consolo, 383 U.S. at 620 (citations omitted). In fact, “[the] court may not substitute its judgment for that of the [agency] when the choice is ‘between two fairly conflicting views, even though the court would justifiably have made a different choice had the matter been before it de novo.’ ” American Spring Wire Corp. v. United States, 590 F. Supp. 1273, 1276 (Ct. Int’l Trade 1984) (alteration in original) (quoting Penntech Papers, Inc. v. NLRB, 706 F.2d 18, 2223 (1st Cir. 1983) (citation omitted)); see also, Ceramica Regiomontana, S.A. v. United States, 636 F. Supp. 961, 966 (Ct. Int’l Trade 1986), aff’d, 810 F.2d 1137 (Fed. Cir. 1987) (“As long as the agency’s methodology and procedures are reasonable means of effectuating the statutory purpose, and there is substantial evidence in the record supporting the agency’s conclusions, the court will not impose its own views as to the sufficiency of the agency’s investigation or question the agency’s methodology.”).

Discussion

I. Commerce’s Decision to Reject Plaintiffs’ Method of Accounting for Research and Development Expenses On Remand is Remanded in Part and Sustained in Part.

The Court ordered Commerce to reconsider its treatment of Plaintiffs’ R&D costs in two respects. First, the Court ordered Commerce to reconsider and further explain why the use of Plaintiffs’ amortized R&D costs would not reasonably reflect Plaintiffs’ actual R&D expenses for this period of review, and to identify what distortions, if any, would arise in the cost of production (“COP”) calculation if amortized R&D costs were used. Hynix, 248 F. Supp. 2d at 1312-1313. The Court ordered Commerce to address in its explanation Plaintiffs’ assertion that all R&D costs incurred prior to and including 1996 *1722that should have been carried forward into this period of review through amortization accounting, were fully taken into account prior to or within the Fifth Administrative Review, when Commerce used expensed R&D costs in the cost of production calculation. Id. at 1312. The Court also ordered Commerce to reconsider and further explain why Plaintiffs’ deferral of certain R&D costs does not reasonably reflect the R&D costs related to the subject merchandise. Id. at 1313. Second, the Court ordered Commerce to further explain whether the subject merchandise has benefitted from R&D activities for non-memory products and identify substantial evidence in the record to justify this conclusion. Id. at 1317. The two issues will be addressed separately in the discussion below.

1. Commerce’s Rejection of Plaintiffs’Amortized R&D Costs and Plaintiffs’ Deferral of Certain R&D Costs

A. Rejecting Plaintiffs’Amortized R&D Costs

Commerce offers two hypotheticals in response to the Court’s order requesting an explanation for rejecting Plaintiffs’ amortized R&D costs. The first hypothetical assumes that Plaintiffs amortized their R&D costs over a three-year period consecutively for seven years, holding the annual R&D expenses constant at $150,000. Remand Results at 6. In the second hypothetical, Commerce presents a situation in which Plaintiffs changed accounting methodologies annually from expensing to amortization and back over a six-year period. Id. at 7. Commerce asserts that these hypotheticals demonstrate that “switching of methodologies can lead to distortions for antidumping purposes because the fluctuating costs tend to overstate per unit amounts in one period and understate these amounts in other periods.” Id.

Commerce also submits possible explanations for why a company may decide to amortize costs in one period of review while deciding to expense in another period of review. Id at 3-5. Commerce states that the possible reason for amortization “is that R&D will benefit future years and the sum of each previous year’s amortized price forms a whole. [T]he theory behind expensing the full amount of R&D in the year incurred is that there is no certainty that the R&D performed in the current year will benefit future years; thus, being conservative, a company expenses the full amount of its R&D costs in the year incurred.” Id. at 4.

The Court finds that Commerce’s use of hypotheticals, generalizations as to why companies may choose one accounting method over another, and conditional language suggesting possible distortions in antidumping calculations offer conjecture rather than a reasoned explanation founded on substantial evidence for its decision to reject Plaintiffs’ amortized costs in this case.

While Commerce is fixated on the undisputed fact that Plaintiffs have changed accounting methods, the Court is concerned with how *1723this change results in distortions to the antidumping calculations, given the particular characteristics of the two accounting methods at issue. Commerce has failed to establish through a reasoned explanation founded on substantial evidence on the record that a change from one permissible accounting method to another necessarily creates a distortion in the cost of production calculation in this period of review for these Plaintiffs under these facts.

Commerce presents the second hypothetical as evidence of the distortions that it claims occurred in this period of review when Plaintiffs changed their accounting methodologies in 1996 from expensing to amortizing. Remand Results at 7. However, Commerce’s second hypothetical is significantly distinguishable from the facts of this case and, as such, is not helpful. Specifically, Plaintiffs have not changed accounting methodologies annually over the course of the administrative review. Hynix, 248 F. Supp. 2d at 1312. Prior to the initiation of administrative reviews, while Plaintiffs operated as Hyundai2 , they historically amortized R&D costs. Id. at 1306; see also, Micron Tech., Inc. v. United States, 893 F. Supp. 21, 28-29 (Ct. Int’l Trade 1995) (‘‘Micron F) (ordering Commerce to use Plaintiffs’ reported amortized R&D costs in its calculations of sales at less than fair value of the subject merchandise). Hyundai expensed R&D costs during the First to the Fourth Administrative Reviews. Remand Results at 3. LG Semicon, the company that Plaintiffs acquired in 1999, expensed its R&D costs from 1993 to 1997. Hynix, 248 F. Supp. 2d at 1306; Micron Tech., Inc. v. United States, 23 Ct. Int’l Trade 55, 58-60 (1999). In 1997, during the Fifth Administrative Review, Hyundai returned to amortization, citing business reasons. Hynix, 248 F. Supp. 2d at 1306. Plaintiffs have since amortized their R&D costs. Id. Given these facts, Commerce’s second hypothetical seems to present an exaggerated and unrealistic situation, and does not assist in explaining Commerce’s determination in this case. See Micron Tech., Inc. v. United States, 23 Ct. Int’l Trade 380, 382 (1999) (“Micron II") (“It makes little, if any, business sense to switch accounting systems on an annual basis.”).

In the Remand Results, Commerce, relying on Micron II, repeats its argument that changing accounting methods resulted in an understatement of R&D costs. Remand Results at 5. Commerce is concerned with using what it calls the “full amount of R&D” costs in the COP calculations. Id. at 4-5. Commerce defines “full amount of R&D” costs to be either the expensed amount for a given year (i.e. 1999) or amortized R&D costs compiled from fractional amounts of *1724R&D amounts carried forward from preceding years in a fractional relationship equaling one (i.e. three-thirds, five-fifths). Id. at 3-5. Commerce asserts that the “full amount of R&D” costs can only be achieved through the consistent use of one accounting method, and that a change from expensing to amortizing necessarily results in an understatement of R&D costs. Id. at 5, 7.

Commerce has failed to establish through evidence on the record that an understatement of R&D costs has occurred in this period of review by the change of accounting methods, such that Plaintiffs’ reported and verified amortized R&D costs do not “reasonably reflect the costs associated with the production and sale of the merchandise.” 19 U.S.C. § 1677b(f)(l)(A). This is the factual premise that the Court ordered Commerce to establish, if it could, on remand. Furthermore, contrary to Commerce’s assertions, this Court is not persuaded that Micron II stands for the position that “actual production costs for a period of review” means that COP calculations require either the use of expensed R&D costs or the use of a fractional relationship equaling one if R&D costs are amortized. Micron II, 23 Ct. Int’l Trade at 382. Continued reliance on Micron II, which the Court has distinguished from the facts of this case, Hynix, 248 F. Supp. 2d at 1312, the use of hypotheticals, and general suggestions as to why parties may select one accounting method over another do not establish that Plaintiffs’ reported R&D costs would result in distortions to cost of production calculations and do not “reasonably reflect” the costs of production.

Commerce did not address the Court’s order to consider and explain whether all R&D costs that should have been carried forward into this period of review from 1996 and before, if amortized, were fully taken into account prior to or within the Fifth Administrative Review, when Commerce used Plaintiffs’ expensed R&D costs in its cost of production calculation. Sec Hynix, 248 F. Supp. 2d at 1312. This Court asked Commerce to consider the facts of these Plaintiffs for this period of review. This Court understands that Commerce is seeking a “whole” (i.e., three-thirds or five-fifths) R&D amount and that its argument emphasizes Plaintiffs’ change in accounting methods. However, this Court directs Commerce to provide a reasoned explanation, supported by substantial evidence, if it is able, that distortions in the cost of production calculations for this period of review necessarily arise, where Plaintiffs’ R&D costs which were previously accounted for through expensing, are now accounted for through amortization. This Court again orders Commerce to consider and explain whether Plaintiffs’ R&D costs prior to the Fifth Administrative Review were accounted for through the expensing of these costs, and if this expensing of R&D costs would leave nothing to carry forward to subsequent review periods. If Plaintiffs’ R&D costs were accounted for prior to the Fifth Administrative Review, then the Court orders Commerce to explain why Plaintiff’s reported *1725amortized R&D costs in this period of review are not fully inclusive of Plaintiffs’ R&D costs.

B. Rejecting Plaintiffs’ Indefinite Deferral of Certain R&D expenses

Commerce has provided a reasoned explanation for rejecting Plaintiffs’ indefinite deferral of certain R&D expenses. Remand Results at 7-9. Despite the fact that Korean GAAP permits the indefinite deferral of certain expenses, and this provision of the Korean GAAP is consistent with the International Accounting Standards No. 9 and the matching principle of accounting, Commerce may select an other methodology if it determines that the accounting method at issue does not accurately reflect the cost of producing the subject merchandise for this period of review. See Ad Hoc Comm. v. United States, 25 F. Supp. 2d 352, 363 (Ct. Int’l Trade 1998). In its earlier opinion, the Court noted that the theory of conservativism in accounting “does not supersede the matching principle, but is incorporated into it as a general quality found in all information used in financial statements.” Hynix, 248 F. Supp. 2d at 1313 (citation omitted). In the Remand Results, Commerce provided a reasoned explanation, supported by evidence on the record, demonstrating that its determination did not result in the matching principle being superseded by conservativism. Remand Results at 9. Commerce explained that the lack of evidence on the record of Plaintiffs objectively obtaining future revenues from the deferred R&D expenses made the application of the matching principle in this case inappropriate. Id. Commerce could not find evidence in the record that Plaintiffs’ deferred R&D costs would produce future revenue within an appreciable amount of time. Remand Results at 8. Plaintiffs argue that evidence of future revenue production exists on the record. (Pis.’ Cmts. on the Dep’t of Commerce’s Final Results of Redetermination (“Pis.’ Cmts.”) at 12 (referring to Resp. of Hyundai Elec. Indus. Co., Ltd., and Hyundai Elec. Am. to the Dep’t’s Supp. Req. for Information, vol. 2 (Pis.’ Conf. App. Ex. 4 at 2-7).) While there is a reference as to when products would produce revenue in terms of the generation of a given product, the Court could not find the timing in terms of months, years, etc., of expected revenue from the R&D cost. (See Resp. of Hyundai Elec. Indus. Co., Ltd., and Hyundai Elec. Am. to the Dep’t’s Supp. Req. for Information, vol. 2 (Pis.’ Conf. App. Ex. 4 at 2-7).)

While Plaintiffs present an alternative explanation to support their indefinite deferral of certain costs, this Court must sustain Commerce’s decision to reallocate those costs, noting that Commerce has sufficiently complied with the remand order. See American Spring Wire, 590 F. Supp. at 1276. The Court holds that Commerce’s explanation for rejecting Plaintiffs’ indefinite deferral of certain R&D expenses is supported by substantial evidence and is otherwise *1726in accordance with law. Recalculation of these deferred R&D expenses may be required pursuant to resolution of the issue of Commerce’s rejection of Plaintiffs’ amortized R&D costs.

2. Commerce’s Decision to Reject Product-Specific R&D Costs (Cross-fertilization of R&D) is Remanded

The Court ordered Commerce to explain its conclusion that R&D for the subject merchandise benefitted from R&D activities for non-subject merchandise products (“cross-fertilization”) is established through substantial evidence on the record. Hynix, 248 F. Supp. 2d at 1317. In the Remand Results, Commerce reiterated its theory of the existence of cross-fertilization of R&D in the semiconductor industry and its application of this theory in the administrative reviews of the subject merchandise in this proceeding and in other semiconductor proceedings. Remand Results at 10. Commerce again cites to the memorandum of Dr. Jhabvala in support of the cross-fertilization theory. Id. Commerce then discusses Plaintiffs’ failure to establish on the record that the subject merchandise does not benefit from R&D activities for non-subject merchandise. Id. at 10-11. In support of the application of the cross-fertilization theory to the subject merchandise in this case, Commerce points to six projects in Plaintiffs’ non-memory R&D lab. Id. at 11. Commerce argues that “[o]n its face, it appears that the R&D at Hynix’s non-memory lab could provide benefits to the production of Hynix’s memory products, as these titles appear to reference memory products.” Id.

Commerce has not complied with the Court’s remand order. Commerce was ordered to point to substantial evidence on the record to establish the existence of cross-fertilization of R&D in this case to justify its decision to reject Plaintiffs’ R&D costs reported on a project-basis. Hynix, 248 F. Supp. 2d at 1317. In response, Commerce provided the Court with a repetition of its theory based on Dr. Jhabvala’s memorandum and speculation that cross-fertilization exists in this case because of the names of six R&D projects in Plaintiffs’ non-memory lab. This is not substantial evidence. This Court held that Dr. Jhabvala’s memorandum does not provide substantial evidence establishing the existence of cross-fertilization of R&D in this case. Id. at 1316. The simple recitation of the titles of Plaintiffs’ six projects does not provide substantial evidence to establish the existence of cross-fertilization either. Commerce is concerned that Plaintiffs have “not demonstrated that the R&D for its subject merchandise and non-subject merchandise do not enjoy a mutually beneficial relationship.” Remand Results at 11. This concern is misplaced. Commerce should instead have focused on finding substantial evidence on the record to support its application of the cross-fertilization theory in this case. See Micron I, 893 F. Supp. at 27 (“[T]he factual premise upon which Commerce bases its choice of methodology must be supported by substantial evidence on the *1727record.”). The Remand Results before the Court do not establish through substantial evidence Commerce’s conclusion that the subject merchandise derives benefits from R&D activities involving non-subject merchandise.

This issue is remanded again to Commerce to establish, if it can, through substantial evidence on the record that the six non-subject merchandise projects Commerce mentions or other non-subject merchandise projects provide benefits to the R&D activities of the subject merchandise. In the alternative, Commerce is ordered to recalculate Plaintiffs’ R&D costs, excluding R&D costs for non-subject merchandise.

II. Commerce’s Decision to Reject the Average Useful Lives (“AULs”) for Hynix’s Fixed Assets is Remanded.

The Court ordered Commerce to explain the following: (1) how the revised recommendations for Plaintiffs’ AUL’s provided by an independent appraiser are not standard industry practice; (2) how and where in the record the AULs were overstated; (3) how the use of the revised AULs would not reasonably reflect Plaintiffs’ cost of production. See Hynix, 248 F. Supp. 2d at 1319. In the Remand Results, Commerce repeats its assertion contained in the Final Results that Plaintiffs’ revised AULs do not reasonably reflect the cost of producing the subject merchandise and the revised AULs were not historically used. Remand Results at 13. Commerce also offers hypotheticals to illustrate “its concerns that Hynix’s accounting revisions distort! ] its production costs during the [period of review] in question.” Id. at 14-15. Commerce notes that Plaintiffs revised their AULs in 1996 and Commerce accepted this revision. Id. at 13. Commerce, however, rejected the revision of AULs for this period of review because the adjustment was “based solely upon the information contained in contained in the appraisers’ report,” and Plaintiffs only supplied Commerce with a portion of a partially translated report and failed to provide information “to establish the authority or expertise of the independent appraisers.”Id.

In its earlier opinion, the Court held that Commerce failed to provide reasoned analysis for its decision to reject Plaintiffs revised AULs. Hynix, 248 F. Supp. at 1319. “It is well established that an agency’s action must be upheld, if at all, on the basis articulated by the agency itself.” Motor Vehicle Mfrs. Ass’n. v. State Farm, 463 U.S. 29, 50 (1983). The explanation provided in the Remand Results, dependant on hypotheticals and challenges to the qualifications of Plaintiffs’ appraisers and the submitted appraisers report falls short of being a reasoned explanation.

First, Commerce does not clearly articulate why it accepted Plaintiffs’ 1996 AUL revisions, yet rejected AUL revisions for this period of review. It would appear disingenuous for Commerce to reject the new revisions based on an argument that Plaintiffs are “continually” *1728revising AULs, without articulating a reasoned explanation for distinguishing its acceptance of Plaintiffs’ previous revision, particularly when there appears to be nothing in the record to support an inference that Plaintiffs have changed their AULs more often than the documented 1996 revision. Furthermore, Plaintiffs raise an interesting assertion in their comments to the Remand Results. Plaintiffs assert that Commerce characterized Plaintiffs’ 1996 AUL revision as “represent[ing] only a change- in an accounting estimate. It does not constitute a change in depreciation methodology.” (Pls. Cmts. at 26 (citing Dynamic Random Access Memory Semiconductors of One Megabit and Above From the Republic of Korea: Final Results of Antidumping Duty Administrative Review, Partial Rescission of Administrative Review and Notice of Determination Not to Revoke Order, 63 Fed. Reg. 50,867, 50,871 (Sept. 23, 1998)).) The Court is unclear whether Commerce characterized the two AUL revisions differently, as Plaintiffs contend, or whether Commerce considered the AUL revisions characteristically the same, and accepted one but not the other.

Defendant argues that Commerce substantively “considered the information that [Plaintiffs] submitted to demonstrate that broader AULs are claimed by other producers, but found such information unconvincing.” (Def.’s Resp. to Pis.’ Cmts. at 22 (referring to “Hyuiidai’s Resp. to Second Supp. Questionnaire,” Ex. SS-14 (Mar. 6, 2001) (Pis.’ Conf. App. Ex. 5) (Conf. Corrected App. to Def.’s Resp. To Pis.’ Cmts. to the Remand Determination Ex. 5).) Commerce, however, makes no mention of considering the information Plaintiffs introduced to support the appraisers’ recommended revision of the AULs. See Remand Results at 11-15, 24. It is well settled that “counsel’s post hoc rationalization” cannot bé used to provide a rationalization for Commerce’s determination. Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168 (1962).

Commerce also challenges the quality of Plaintiffs’ appraisers and the adequacy of the report submitted by Plaintiffs as related to the AULs for the first time in the Remand Results, Remand Results at 13. The Court finds it unusual that the form of the report and the qualification of the appraisers are raised for the first time in the Remand Results. See id. at 24. This new challenge seems arbitrary, given the fact that Commerce used “a portion of the [same] appraisers’ findings in the Final Results, [but] it was only with respect to revaluation of assets, which [were] discussed in a separate portion of the appraisers’ report.” Id. at 24 (citing Final Decision Memorandum at 15-18 (Def.’s Pub. App. Ex. 7)). Commerce adds that the appraisers’ report with respect to the revaluation of assets was “appropriate, given the widely-known economic circumstances that affected the [period of review].” Id.

Commerce claims that it was unable to evaluate Plaintiffs’ report because only portions of the report were translated, thus Plaintiffs *1729did not comply with 19 C.F.R. § 351.303(e)’s requirement for supplying a fully translate document or obtaining permission from Commerce before submitting a partially translated document. Id. at 24; 19 C.F.R. § 351.303(e). However, in the General Instructions accompanying Commerce’s letter soliciting information from Plaintiffs for this administrative review, Plaintiffs were instructed to “[p] repare [their] responses in typed form and in English [and to] [i]nclude an original and translated version of all pertinent portions of non-English language documents that accompany [the] response, including the financial statements.” (Letter from Ronald Trentham, Acting Program Manager, Dep’t of Commerce, Int’l Trade Admin., to Lawrence R. Walders, Esq. of 7/18/00 (“Commerce July 18, 2000 Letter”) (Pls.’ Supp. App. Ex. 8 at 8.) (emphasis added).) Based on these instructions, it is far from obvious, as Commerce contends, that Plaintiffs “must have been fully aware of [the requirement to provide fully translated documents,] as [they were] informed of this requirement in the Department’s standard questionnaire.” Remand Results at 24 (citing General Instructions, attached to Commerce July 18, 2000 Letter (Pis.’ Supp. App. Ex. 8 at 8)). At the very least, Commerce’s instructions to submit “pertinent portions of non-English language documents” might lead Plaintiffs to believe that their submission would be sufficient. (General Instructions, attached to Commerce July 18, 2000 Letter (Pls.’ Supp. App. Ex. 8 at 8).) Additionally, Commerce has, in the past, notified respondents that their submissions did not meet 19 C.F.R. § 351.303(e)’s requirement and offered respondents an opportunity to correct their submissions before Commerce would rely on adverse facts available. See Ocean Harvest Wholesale, Inc. v. United States, No. 00-05-00231, 2002 Ct. Intl. Trade LEXIS 31, at *15-*16 (Ct. Int’l Trade Mar. 20, 2002). The Court finds that Commerce’s attempt to distinguish the use of the portions of the appraisers’ report referring to asset valuation based on the “widely-known economic circumstances” of the Korean won fluctuations does not fully provide an explanation as to why the “authority or expertise of the independent appraisers” was established for the asset valuation portion of the report, but not for the revision of the AULs.

Commerce’s new challenge of Plaintiffs’ appraisers and the appraisal report raises a significantly different issue before the Court. The Court has been under the impression that Commerce based its decision to reject the revisions of the AULs on substantive grounds, through an evaluation of the information contained in the appraisers’ report and verified by Commerce. See Final Decision Memorandum at 17-18 (Pub. Doc. No. 72) (Def.’s Pub. App. Ex. 7); Remand Results at 13-15. However, Commerce has now introduced what appears to be a reason for rejecting Plaintiffs AULs based on form, not substance. As a result, the Court is unable to determine what evi*1730dence on the record Commerce evaluated to reach its decision to reject Plaintiffs’AUL revision.

The Court again remands this issue to Commerce to provide a reasoned explanation for rejecting Plaintiffs’ revised AULs. This explanation must include: (1) a discussion of why Commerce accepted Plaintiffs’ 1996 AUL revision, and whether Commerce characterized the 1996 AUL revision and this period of review’s AUL revisions differently; (2) a clarification of what information Commerce evaluated in reaching its determination to reject Plaintiffs’ revised AULs; (3) a clarification of whether Commerce did, in fact, consider Plaintiffs’ information demonstrating industry-wide AUL ranges, and if not, to do so now; (4) an explanation addressing why Commerce accepted Plaintiffs’ appraisers’ report for asset revaluation, while rejecting the same report for AUL revision; this explanation should compare the quality of the two sections of the report, including whether all pages of the asset revaluation section were translated and why the qualifications of the appraisers were acceptable for the asset revaluation and not for the AUL section. Should Commerce find it necessary to make recalculations, it is ordered to so do.

Conclusion

Upon consideration of the Remand Results, Plaintiffs’ and Defendant-Intervenor’s Comments and Defendant’s Response, the Remand Results are affirmed in part and remanded in part. The Court remands to Commerce for reconsideration and further explanation its decision to reject Plaintiffs’ reported R&D costs and Plaintiffs’ revised AULs.