Full opinion text
OPINION RESTANI, Judge. This matter is before the court on a motion for judgment based upon the agency record pursuant to USCIT Rule 56.2. The motion has been brought by AG der Dillinger Huttenwerke (“Dillinger”), EKO Stahl GmbH, Salzgitter AG Stahl und Technologie, Stahlwerke Bremen GmbH and Thyssen Krupp Stahl AG (collectively “Plaintiffs”), respondents in a countervailing duty (“CVD”) investigation. See Certain Steel Products from Germany, 58 Fed.Reg. 37,315 (Dep’t Comm.1993) (final determ.) [hereinafter “Final Determination”]. At issue is the final determination by the Department of Commerce (“Commerce” or the “Department”) pursuant to sunset review under 19 U.S.C. § 1675(c). See Certain Corrosion-Resistant Carbon Steel Flat Products; Cold-Rolled Carbon Steel Flat Products; and Cutr-to-Length Carbon Steel Plate Products from Germany, 65 Fed.Reg. 47,407 (Dep’t Comm.2000) (final sunset rev.) [hereinafter “Sunset Determination”]. Plaintiffs challenge the Sunset Determination principally on the following grounds: (1) Commerce improperly shifted the burden of proof to Plaintiffs to prove that the benefits under certain programs were not likely to continue; and (2) Commerce erred in not taking into account changes in law that took place subsequent to the initial investigation. JURISDICTION & STANDARD OF REVIEW The court has jurisdiction pursuant to 28 U.S.C. § 1581(c) (1994). The court will uphold Commerce’s determination in countervailing duty investigations unless it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(l)(B). FACTUAL & PROCEDURAL BACKGROUND On June 30, 1992, the domestic steel industry, Bethlehem Steel Corporation and United States Steel LLC (collectively “domestic producers” or “petitioners”) filed petitions with Commerce alleging that the Government of Germany (“Germany”) was providing countervailable subsidies to its steel industry through various subsidy programs. Commerce subsequently conducted a countervailing duty investigation, and on July 9, 1993, issued a final affirmative determination that countervailable benefits had in fact been provided by Germany to its steel industry. See Final Determination, 58 Fed.Reg. at 37,315. The period of investigation (“POI”) for which subsidies were measured was 1991. On September 1, 1999, Commerce initiated sunset reviews of the countervailing duty orders on corrosion-resistant and cut-to-length steel products from Germany. See Initiation of Five-year (“Sunset”) Reviews of Antidumping and Countervailing Duty Orders or Investigations of Carbon Steel Plates and Flat Products, 64 Fed. Reg. 47,767, 47,768 (Dep’t Comm.1999); see also Transition Orders; Final Schedule and Grouping of Five-Year Reviews, 63 Fed.Reg. 26,779, 26,786 (Dep’t Comm. 1998). On September 10, 1999, the domestic producers including Bethlehem Steel Corporation and U.S. Steel Group, a unit of USX Corp. (collectively “Defendant Inter-venors”), filed notices of intent to participate in the sunset reviews. See P.R. Docs. 531, 533. The German producers filed notices of intent to participate on September 28, 1999. See P.R. Docs. 552, 553. On September 30,1999, the Commission of the European Communities and the Government of Germany submitted substantive responses to the notice of initiation. See P.R. Docs. 577-78, 603. On October 1, 1999, substantive responses were submitted by the German producers and the domestic producers. See P.R. Docs. 555, 556, 558, 560, PLApp. Tabs 3, 4. Rebuttals to the substantive responses were filed by the various parties on October 15, 1999. See P.R. Docs. 593-602, 608. Having deemed the responses “adequate,” Commerce decided to conduct a “full sunset review” rather than an abbreviated, expedited sunset review based on the facts available. See Sunset Determination at cmt. 1. On November 9, 1999, the domestic producers filed a submission contesting Commerce’s decision to conduct a full sunset review. See P.R. Doc. 632. On March 14, 2000, the German producers filed a submission discussing a pending decision of the WTO dispute settlement panel and a decision issued by the U.S. Court of Appeals for the Federal Circuit, both of which related to Commerce’s change-in-ownership methodology. On March 16, 2000, the domestic producers submitted comments in response to the claims made in the German producers’ March 14 submission. See P.R. Doc. 723. On March 17, 2000, the German producers filed a submission regarding the allocation period from the German depreciation schedule, which had been used in Steel Wire Rod from Germany, 62 Fed.Reg. 54,990 (Dep’t Comm. Oct. 22,1997) (final determ.) [hereinafter “Steel Wire Rod”]. On March 20, 2000, Commerce issued its preliminary results of the sunset review. Certain Corrosiom-Resistant Carbon Steel Flat Products; Cold-Rolled Carbon Steel Flat Products; and Cub-to-Length Carbon Steel Products from Germany, 65 Fed. Reg. 16,176 (Dep’t Comm. March 27, 2000) (prelim, sunset determ.) [hereinafter “Preliminary Sunset Determination”]. In the Preliminary Sunset Determination, Commerce determined that revocation of the countervailing duty orders would be likely to lead to continuation or recurrence of a countervailable subsidies. Id. Commerce indicated that it relied on rates determined in the original investigation because no administrative review of the orders had been conducted. Id. at § II. Commerce found that grants were made under the Capital Investment Grants (“CIG”) program after 1986, producing a benefit stream that would last beyond the end of the sunset review. Id. at § 1.1. After the preliminary results, counsel for the German producers contacted Commerce to request that the calculation mem-oranda from the original investigation be made part of the administrative record in the sunset reviews to clarify the issue of whether benefits were given after 1986. On April 13, 2000, counsel for the German producers made this request in writing. See Pl.App. Tab 9. In late April of 2000, the Government of Germany submitted its questionnaire response from the original investigation and verified information from the Steel Wire Rod investigation explaining Germany’s regional assistance programs. See Pl.App. Tab 10. On April 25, 2000, Commerce issued a letter returning the March 11 and March 17, 2000 submissions and striking them from the record, on the ground that it “normally will not accept or consider any additional information from a party after the time for filing rebuttals has expired unless the Secretary requests additional information from a party after determining to proceed to a full sunset review.” See P.R. Docs. 742-43, Pl.App. Tab 11 at 1 (citing 19 C.F.R. § 351.218(d)(4)). On April 28, 2000, the domestic producers submitted its response to the German producers’ request to place the calculation memo-randa on the sunset review record. See P.R. Doc. 763, Pl.App. Tab 12. The response included portions of a questionnaire response of one of the German companies from the original investigation. Id. On May 2, 2000, counsel for the German producers met with Commerce staff and again requested the inclusion of the calculation memoranda in the administrative record. In May 2000, the Government of Germany and the rest of the interested parties timely submitted case briefs. In early June 2000, all parties timely submitted post-preliminary determination rebuttal briefs, and a hearing was held on June 26, 2000. On July 11, 2000, Commerce issued a letter striking the portion of the submissions from the record pertaining to “new factual information and/or materials ... submitted in another proceeding.” P.R. Doc. 849, PLApp. Tab 15. In the same letter, Commerce indicated that it decided to keep in the sunset review record “the portion [of the submission] which were parts of the record in the original investigation of the orders, because they were submitted in connection with the original investigation and it is our practice to use information contained [in] the record of the investigation, where appropriate.” Id. On July 27, 2000, Commerce issued the final results pursuant to sunset review. Sunset Determination, 65 Fed.Reg. 47,407. In the Sunset Determination, Commerce determined that revocation of the countervailing duty orders would be likely to lead to continuation or recurrence of eounter-vailable subsidies. Id. at 47,408. Commerce maintained the rates arrived at in the Final Determination for cut-to-length carbon steel products. See Sunset Determination at Background Section. With respect to corrosion-resistant carbon steel flat products, however, Commerce made adjustments to the net subsidy rate for this class of subject merchandise by deducting the subsidy rates attributable to these programs. On December 15, 2000, Commerce published notice of the continuation of countervailing duty order. See Certain Carbon Steel Products from Germany, 65 Fed. Reg. 78,469 (Dep’t Comm.2000). DISCUSSION I. Scope of Commerce’s Inquiry pursuant to Sunset Review Under the Capital Investment Grants program, the Government of Germany provided grants to the iron and steel industry amounting to 20% of the acquisition cost of assets purchased or produced prior to January 1, 1986, and ordered or produced after July 80, 1981. Under the Investment Premium Act (“IPA”), grants (or “investment allowances”) were provided to companies investing in three specific regions of Germany, calculated as a percentage of eligible expenditures in these areas. Companies were eligible for grants under the IPA only for investments made prior to January 1, 1991. In the Final Determination, after finding the grants given under these programs countervaila-ble, Commerce deemed the benefits received “nonrecurring,” and therefore allo-cable over fifteen years in accordance with 19 C.F.R. § 351.524. Final Determination at 37,316. Pursuant to the sunset review, Commerce determined that benefit streams from the CIG and IPA programs continue beyond the end of sunset review. Commerce agreed with the domestic interested parties that certain manufacturers of the subject merchandise received “some benefits” from both the CIG and the IPA after January 1, 1985. Sunset Determination at cmt. 7. The German producers had argued that the benefits received under the CIG and/or IPA after that date were so small that they should be expensed in the year they were received. Commerce rejected this argument on the ground that “the record of these sunset reviews is not sufficient for us to definitively conclude whether [those benefits] were less than 0.5% of the corresponding beneficiary’s annual net sales.” Id. Commerce further found that “since no administrative reviews of the orders were conducted, we are unable to determine whether any additional benefits under these programs were received subsequent to the period of investigation.” Id. Based on these findings, Commerce concluded that “the benefit streams from the Capital Investment Grants and Investment Premium Act Programs were likely to continue beyond the end of sunset review, and that, therefore, a eountervailable subsidy from the CIG and IPA to manufacturers of subject merchandise would be likely if the orders were revoked.” Id. Plaintiffs dispute this determination on the ground that Commerce impermissibly shifted the burden of proof to the Plaintiffs to show that the countervailable subsidies were not likely to continue or recur. Plaintiffs argue that by ignoring evidence of actual amounts given under the programs and of amortization of non-recurring benefits calculated in the original investigation, Commerce erroneously presumed that the benefit of almost all of the programs examined in the original investigation would continue undiminished after the end of the sunset review (July 27, 2000). Plaintiffs contend that Commerce erred in refusing to consider the following evidence: (1) information from a questionnaire submitted in the original investigation regarding amounts actually given to Preussag (predecessor to Salzgitter AG Stahl und Technologie) under the two programs after January 1, 1985; and (2) the calculation memoranda supporting the original CVD rate determination showing’ “the precise subsidy amount received by each producer in every year up to and including the 1991 period of investigation,” as well as “annual net sales and ... results of the expense test.” PI. Br. at 19. Commerce counters that it did not presume or even determine that the non-recurring benefits at issue would continue undiminished, as such a determination would require calculations outside the scope of a sunset review. Commerce asserts that the legislative history and the regulations prevent it from making adjustments to the original CVD rate except under “extraordinary circumstances,” which do not apply in this case. Commerce adds that even if it were permitted to make adjustments to the original CVD rate in this case, the record evidence was not sufficient to enable Commerce to do so. Defendant Intervenors further assert that there is a presumption that subsidization will continue or recur at the original CVD rate, and that Plaintiffs failed to rebut the presumption. Def.-Int. Br. at 9. The court finds that Commerce misconstrues: (A) the evidentiary burden applicable to its likelihood determination; and (B) the scope of its discretion to make adjustments based on that evidence. A. Evidence Supporting the Likelihood Determination pursuant to Sunset Review Pursuant to sunset review, Commerce must determine the likelihood of continuation or recurrence of a subsidy deemed countervailable in the original investigation if the order issued pursuant thereto were to be revoked. In the absence of an affirmative determination in this regard, the statute directs that the countervailing duty order be revoked. See Section 1675(d)(2) (“In a [sunset review], the administering authority shall revoke a countervailing order ... unless (A) the administering authority makes a determination that ... a countervailable subsidy ... would be likely to continue or recur, and (B) the Commission makes a determination that material injury would be likely to continue or recur... ,”). The statute, however, does not charge any interested party with the ultimate burden of persuasion, or otherwise create a presumption that a countervailable subsidy is or is not likely to continue or recur if the order is revoked. See Eveready Battery Co., Inc. v. United States, 77 F.Supp.2d 1327, 1330 (CIT 1999) (“The only difference of significance ... is that [the party seeking revocation] would bear the burden of persuasion in a changed circumstances review, but has no such burden in a sunset review.”); see also 19 U.S.C. § 1675(b)(3)(A) (in changed circumstances review, party seeking revocation of an order shall have the burden of persuasion). Because there is no presumption as to the likelihood of continuation or recurrence, it follows that there is no presumption that the counter-vailable subsidies will continue at the specific rate determined in the original investigation. Rather than place a burden of proof on either the foreign or the domestic interested parties, the statute provides that parties may submit information in their response to Commerce’s notice of initiation of review. See 19 U.S.C. § 1675(c)(2) (notice of initiation requests interested parties to express their willingness to participate in the review, state the likely effects of revocation, and provide any “other information or industry data” as the agency may specify). If no interested party responds, the order is revoked. See 19 U.S.C. § 1675(c)(3)(A). The regulations further provide that in submitting a “substantive response,” the parties must include, inter alia, “a statement regarding the likely effects of revocation of the order under review, which must include any factual information, argument, and reason to support such statement”; and “factual information, argument, and reason concerning the ... countervailing duty rate ... that is likely to prevail if the Secretary revokes the order ... that the Department should select for a particular interested party.” 19 C.F.R. § 351.218(d)(3)(ii)(F), (G). Under 19 C.F.R. § 351.218(d)(3)(iv)(B), “[a] substantive response from an interested party ... also may contain any other relevant information or argument that the party would like the Secretary to consider.” If a response is deemed “inadequate,” the likelihood determination may be issued based on “facts available.” 19 U.S.C. § 1675(c)(3)(B). In this case, the responses were deemed adequate and Commerce proceeded to a “full sunset review.” The SAA distinguishes between a “full-fledged review” (or “full sunset review”) involving fact gathering, and an “expedited review based on facts available.” The SAA explains the distinction as follows: The facts available may include prior agency determinations involving the subject merchandise as well as information submitted on the record by parties in response to the notice of initiation .... [T]he agencies may decide separately whether responses are inadequate and whether to issue a determination based on the facts available without further fact-gathering. [Section 1675(c)(3) ] is intended to eliminate needless reviews.... If parties provide no or inadequate information in response to a notice of initiation, it is reasonable to conclude that they would not provide adequate information if the agencies conducted a full-fledged review. However, when there is sufficient willingness to participate and adequate indication that parties will submit information requested throughout the proceeding, the agencies will conduct a full review. SAA at 879-80. Thus, even in an “expedited review based on facts available,” Commerce is to rely on information from prior determinations as well as from submissions by the parties in the sunset proceedings. See also 19 C.F.R. § 351.308(f) (“[w]here the Secretary determines to issue final results of sunset review on the basis of facts available, the Secretary normally will rely on: (1) calculated countervailing duty rates or dumping margins, as applicable, from prior Department determinations; and (2) Information contained in the parties’ substantive responses to the Notice of Initiation... ,”). It stands to reason, then, that in a “full review,” Commerce must engage in an analysis that is at least somewhat more searching than simply continuing to apply the CVD rate determined in the original investigation for particular subsidies without considering evidence proffered by the parties that would support making adjustments thereto. In addition, pursuant to its “fact-gathering” obligation in a full sunset review, Commerce may solicit more information as necessary. The court finds that Commerce did not fulfil its obligations pursuant to a full sunset review because it failed to consider adequately the evidence on the record, or to seek additional evidence necessary to make its determination. 1. Allocation of Benefits under the CIG: Evidence relating to Preussag In the Preliminary Sunset Determination, Commerce stated: Because the Department determined that CIG is a non-recurring program and that Preussag received the grant as late as in 1990, regardless of [Germany’s] claim that CIG was terminated in 1985, or the EC’s claim that the program no longer exists, based on the 15 year allocation period used by the Department in its investigations, we preliminarily determine that benefit streams from this program continue beyond the end of this sunset review and that, therefore, a countervailable subsidy from CIG to manufacturers/exporters of subject merchandise will continue to exist. Preliminary Sunset Determination at § 1.1. In the sunset determination, Commerce determined that “certain manufacturers of the subject merchandise received some benefits from both the CIG and IPA after January 1, 1985.” Sunset Determination at cmt. 7 (citing Dillinger’s October 15,1999 Rebuttal Brief). Commerce’s reasoning seems to be as follows: given that the program had been deemed nonrecurring, and Plaintiffs concede that amounts were given under the program as late as 1990; it follows that allocating these amounts over 15 years would cause benefits to continue beyond the end of sunset review. The issue raised in the Rebuttal Brief, however, was whether the amounts given after 1985 were more properly ex-pensed in the year they were received. Commerce did not reach this issue because it determined that it lacked the information necessary to determine the exact amounts given. The information, however, was readily available in Preussag’s questionnaire submitted in the original investigation, and Plaintiffs satisfied their eviden-tiary burden by requesting that Commerce consider the amounts specified therein. Therefore, the court rejects Commerce’s explanation that it lacked the information necessary to determine whether the amounts given after 1985 were de minim-is, and instructs Commerce to consider the information on the record to determine whether the amounts should be allocated over time or expensed in the year received. 2. De Minimis Net Subsidy Rate: Calculation Memoranda Plaintiffs contend that with respect to corrosion-resistant carbon steel flat products, the calculation memoranda from the original investigation show that under Commerce’s methodology for allocating non-recurring subsidies, and applying a discount rate of 8%, the benefit of the subsidies given under the CIG would decline by at least 3.6% annually, diminishing to a rate of 0.26% by 2000 (rather than the 0.39% calculated by Commerce in the original investigation and retained by Commerce in the sunset review). See 19 C.F.R. § 351.524(d) (providing a formula for allocating a non-recurring benefit over time and determining the annual benefit amount that should be assigned to a particular year). Under Plaintiffs’ calculations, adjusting the amount attributable to the CIG program in the original investigation would result in a de minimis net subsidy rate of 0.41% in the sunset review for all programs combined. Similarly, Plaintiffs maintain that if Commerce applied its methodology for allocating benefit streams over time, it would have found that the CVD rate attributable to the IPA for cut-to-length carbon steel flat products (0.06%) would decline by over 32% by 2000. In the Preliminary Sunset Determination, Commerce indicated that there were inconsistencies in the evidence regarding whether the CIG would continue to provide countervailable benefits beyond the end of sunset review. Preliminary Sunset Determination at § 1.1. The German producers requested that Commerce make the calculation memoranda from the original investigation part of the administrative record of the sunset review. Commerce rejected this request as untimely. Commerce stated that “[ijnsofar as Dillinger could have submitted some version [e.g., a public version] of the memoranda in the sunset reviews, we determined that ... Dillinger did not file the information in a timely manner.” Sunset Determination at cmt.4. The record from the original investigation may be incorporated into the record for purposes of the sunset review. See Floral Trade Council v. United States, 13 CIT 242, 243, 709 F.Supp. 229, 230 (1989) (“in a case ... where the agency in its decision states without qualification that it has examined ‘the original investigations’ ... the court must assume that all relevant information from those previous investigations is before the agency for the purpose of the current decision.”). Further, in the past, even in an expedited sunset review Commerce apparently has accepted untimely submissions relating to “important factual information that is already on the record of this [sunset review] proceeding, i.e., in the [prior] administrative review segment.” See Iron Metal Castings from India, 64 Fed.Reg. 37,509, 37,511 (Dep’t Comm.1999) (amd’d final sunset review determ.). Thus, it is sufficient that Plaintiffs alerted the agency in their substantive response or rebuttal to the other interested parties’ substantive response that the countervailable benefits accruing from the nonrecurring subsidies would be de min-imis. The court finds that the issue was raised with sufficient clarity to put Commerce reasonably on notice in a timely manner that it needed to consider the data underlying the calculation of the original CVD rate. See Rebuttal Brief, P.R. Doc. 597, Pl.App. Tab 5, at 4 (“[I]t is clear that none of the recurring subsidies reviewed by the Department in the original investigation would provide any meaningful coun-tervailable benefit that continues after the end of this review.”). To the extent Commerce needed information beyond these calculation memoranda, it could have requested the information from the parties or from a third source. Accordingly, Commerce shall consider the calculation memo-randa as part of the record in this proceeding. B. Scope of Commerce’s Discretion to make Adjustments 1. Framework for Inquiry Commerce maintains that, even if it considers the evidence described above, it is restricted from making adjustments to the net countervailing duty rates determined in the original investigations where no administrative review has been conducted. As a preliminary matter, the statute does not explicitly prevent any adjustments from the original CVD rate. Rather, the statute sets up a framework for inquiry in making the likelihood determination. Section 1675a(b)(l) directs Commerce in a sunset review to consider: A. The net countervailable subsidy determined in the investigation and subsequent reviews; and B. Whether any change in the program which gave rise to the net counter-vailable subsidy described in subpara-graph A has occurred that is likely to affect that net countervailable subsidy. That Commerce shall “consider” the original CVD rate does not, by itself, indicate that Commerce may not make adjustments thereto. It does not necessarily follow from a statutory directive for an agency to consider certain facts, that the agency may not consider other facts. Indeed, the second required consideration provides that Commerce shall determine whether adjustments are necessary based on “changes in the program.” The SAA specifies, however, that where benefits are allocated over time, “Commerce will consider whether the fully allocated benefit stream is likely to continue after the end of the review, without regard to whether the program that gave rise to the long-term benefit continues to exist.” SAA at 889. See also Sunset Policy Bulletin at § III.A.4. Thus, a non-recurring subsidy, though terminated, may factor into Commerce’s likelihood determination where the benefits accruing therefrom are allocated beyond the sunset review period. This provision, however, does not absolve Commerce of its obligation to “consider” the original CVD rate in light of facts and arguments raised by the parties in exercising their right to participate in a sunset review. 2. Selection of CVD Rate The statute indicates that Commerce is to provide the Commission with the net countervailable subsidy that is “likely to prevail” if the order is revoked, and that Commerce “shall normally choose a net countervailable subsidy that was determined under [19 U.S.C. § 1671d regarding final determinations] or [19 U.S.C. § 1675(a) regarding administrative reviews, or § 1675(b)(1) regarding changed circumstances reviews].” 19 U.S.C. § 1675a(b)(3). The SAA explains that—in light of the directive under section 1675a(b)(3) that Commerce provide the Commission with the net countervailable subsidy that is “likely to prevail”—it may be more appropriate in certain instances for Commerce to choose a CVD rate other than that derived in the original investigation. The SAA provides the following guidance: The Commission may consider likely ... net countervailable subsidies to be relevant to its analysis of the likelihood of injury. [Section 1675a(b)(3) ] direct[s] Commerce to provide the Commission with the net countervailable subsidies ... likely to prevail in the event of revocation or termination. Commerce normally will select ... net countervaila-ble subsidies determined in the original investigation or in a prior review. The Administration intends that Commerce normally will select the rate from the investigation, because that is the only calculated rate that reflects the behavior of exporters and foreign governments without the discipline of an order or suspension agreement in place. In certain instances, a more recently calculated rate may be more appropriate. SAA at 890. That Commerce will “normally select” a net eountervailable subsidy calculated in the original investigation or a prior review does not in any way indicate that Commerce is “barred” from making adjustments thereto based on information gathered in a sunset review. In this case, Commerce does not argue that the original rate should be used because it better predicts future behavior. Rather, it emphasizes the inconvenience of conducting a broader sunset review instead of an ordinary administrative review to determine a proper rate. The SAA contemplates that “in certain instances” Commerce may select a rate other than the one derived in the original investigation. The SAA does not limit the instances in which another selection may be “more appropriate,” nor does it explicitly prevent adjustments from being made pursuant to a sunset review. Accordingly, Commerce has far more discretion than it admits. 3. Extraordinary Circumstances The SAA cautions, however, that the requirement that Commerce provide the Commission with net eountervailable subsidy that is “likely to prevail” in the event of revocation is not to be construed as directing Commerce to calculate a “future net eountervailable subsidy.” The SAA reads as follows: In providing information to the Commission, the Administration does not intend that Commerce calculate future ... net eountervailable subsidies, because such an exercise would involve undue speculation _ Only under the most extraordinary circumstances should Commerce rely on ... net eountervailable subsidies other than those it calculated and published in its prior determinations. SAA at 890-91. See also 19 C.F.R. § 351.218(e)(2)(i) (“Even where the Department conducts a full sunset review, only under the most extraordinary circumstances will the Secretary rely on a CVD rate ... other than those it calculated and published in its prior determinations .... ”). The implications of this limitation are not as broad as Commerce urges. This passage in the SAA regarding “future net eountervailable subsidies” indicates that Commerce should not attempt to calculate a CVD rate based on subsidies not yet in existence. The final sentence directs Commerce not to consider subsidy allegations it has not yet passed on. It cannot be read as a limit on needed adjustments to an outstanding rate, because section 1675a(b)(l)(B) expressly requires consideration of program changes. The SAA’s limitations on findings as to the future existence of subsidies or unreviewed subsidies say nothing about restricting Commerce’s discretion to adjust the original CVD rate as envisioned by the general .principles immediately preceding them. By adjusting an existing CVD rate, Commerce does not cease to rely on it. It does not abandon it for an entirely new subsidy calculation. In fact, in accordance with the statute, Commerce adjusted certain rates here because it apparently deemed some recurring subsidies terminated (i.e., the Structural Improvement Aids and the Zonal Border Area programs), or found nonrecurring subsidies inapplicable to the classes of merchandise under review in this ease (i.e., the Ruhr District Action program). Indeed, the SAA indicates that Congress contemplated that Commerce is to engage in some fact gathering when conducting a full sunset review. See discussion, supra, at Section I.A. Such fact gathering would be pointless if Commerce were prohibited from considering those facts in order to determine whether adjustments to the original CVD rate are warranted. Accordingly, Commerce is not restricted by the statute or the SAA from making adjustments to the original CVD rate in the absence of a separate subsequent determination. 4. Policy Bulletin Commerce seeks support from the Sunset Policy Bulletin for its refusal to make adjustments in the absence of an administrative review. The Sunset Policy Bulletin contains a non-exclusive list of examples of circumstances in which Commerce may or may not make adjustments to the original net countervailable rate. Sunset Policy Bulletin at § III.B.3. Example (g) indicates that “[wjhere the Department has not conducted an administrative review of the order, or suspension agreement, as applicable, subsequent to the investigation, except as provided in paragraph III.C, the Department normally will not make adjustments to the net countervailable subsidy rate determined in the original investigation.” Id. at § III.B.3.g. To the extent that Commerce interprets Example (g) as completely foreclosing any adjustments other than those specified in § 1675a(b)(2)(A) or (B) simply because no administrative review has been conducted, the court finds no support for such a restriction in the statute, the regulations or the SAA. In fact, the court finds that such an interpretation of Example (g) would directly conflict with the SAA and the statutory scheme, and is therefore impermissible. See 19 U.S.C. § 1675a(b)(l)(B) and, supra, discussion I.B.2 & 3. Cf. Pesquera Mares Australes Ltda. v. United States, 266 F.3d 1372, 1380-84 (Fed.Cir. 2001) (according deference to Commerce’s interpretation of statute where interpretation arising out of a “relatively formal administrative procedure” constituted “a permissible construction of the statute.”). 5. Verification Commerce explains that it rejected the information submitted by the Plaintiffs in part because of a “preference” for verified information. Commerce stated that it typically does not conduct investigations in sunset reviews, and only rarely conducts on-site verifications. Commerce further explained that given the “unusually stringent” time constraints, it would be “unreasonable and burdensome” to require Commerce to conduct an investigation as extensive as that of an administrative review. DOC Br. at 23. Commerce’s arguments lack merit. Commerce does not dispute that the information in the calculation memoranda and Preussag’s questionnaire had been verified in the original investigation. Rather, Commerce’s concern seems to be that in a sunset review, it is impracticable to verify that no subsidies were given to companies other than Preussag under the CIG and/or IPA programs after the POI (1991), or that Preussag in fact received after this date only 0.6% of the total amount given. This concern seems unwarranted, given that Defendanh-Interve-nors have not pointed to any evidence that subsidies were given under these programs after 1991 to any company other than Preussag, or that Preussag in fact received more than Plaintiffs allege. Defendant-Intervenors merely speculate that “[ojther companies may have received more.” Def.-Int. Br. at 17. Commerce is also apparently concerned that it would be unable to verify any of the additional facts that may be necessary to determine a CVD rate under current laws and methodologies described in Section II, infra. Commerce cannot justify its failure to consider Plaintiffs’ arguments based on its conjecture that this information might be unavailable or otherwise incapable of being verified. Commerce is correct that it is not obligated in all sunset reviews to verify factual information relied upon in making its final determination. According to the regulations, it is within Commerce’s discretion to verify information prior to issuing the final results pursuant to sunset review, although once a determination to revoke is made, verification is mandatory. See 19 C.F.R. § 351.307 (“Prior to making a final determination in an investigation or issuing final results of review, the Secretary may verify relevant factual informa-tion_ [T]he Secretary will verify factual information upon which the Secretary relies in ... a revocation under [19 U.S.C. § 1675(d)].”) (emphasis added). See also 19 U.S.C. § 1677m(i)(2) (“The administering authority shall verify all information relied upon in making ... a revocation under section 1675(d).”). Cf. 19 U.S.C. § 1676m(i)(3) (requiring verification in an administrative review only if requested by an interested party, and, unless good cause is shown, if no verification was made during the two preceding administrative reviews). The regulations further indicate that it is within Commerce’s discretion to verify information it receives in a full sunset review if it determines that such verification is “needed.” See 19 C.F.R. § 351.218(f)(2)(i) (Commerce “will verify factual information relied upon in making its final determination normally only in a full sunset review ... and only where needed.”). Although the regulations describe limited circumstances in which Commerce will typically verify information, Commerce does not contend that a party might be prejudiced if it chooses to conduct a verification when these circumstances are not present. Even if Commerce were prohibited from conducting verification in a full sunset review, Commerce still must analyze the submitted evidence to determine the proper weight it should be given. See Floral Trade Council v. United States, 20 CIT 595, 601 (1996). The lack of an obligation to conduct verification in a sunset review does not relieve Commerce of its obligation to consider and weigh information submitted by the parties in support of arguments reasonably made as to why revocation would or would not be likely to continue or recur beyond the end of sunset review. Accordingly, Commerce on remand must consider the evidence proffered by both parties and the Government of Germany. II. Adjustments Pursuant to Changes in Law and Methodology Plaintiffs claim that Commerce erred in failing to consider any changes in U.S. or international law occurring since the original investigation that affect the previously investigated subsidies. Plaintiffs contend that failure to consider these changes contravenes the purpose of a sunset review, i.e., determining the net countervailable subsidy that is likely to prevail if the countervailing duty order is revoked. Specifically, Plaintiffs assert that Commerce erred in not considering changes in law relating to: (1) a Commission of the European Communities decision putting limitations on aid to the steel industry; (2) the passage of the Uruguay Round Agreements Act with provisions rendering certain subsidies non-actionable; (3) a change in U.S. regulations governing contingent liability interest-free loans; and (4) a GATT dispute settlement panel decision regarding private bank debt forgiveness. Plaintiffs also allege that Commerce should have taken into account the following changes in Commerce’s methodologies regarding: (1) the calculation of “average useful life”; and (2) change-in-ownership. A. Changes in Law As discussed in Section I.B, supra, with respect to whether adjustments could be made to the original CVD rate in light of facts and arguments raised by the parties, Commerce determined that an administrative review, rather than a sunset review, was the appropriate context within which Plaintiffs could seek relief as to a change in the duty rate calculated in accordance with changes in applicable laws and regulations. In the Sunset Determination, Commerce stated: [Although the sunset segment is created by the WTO agreement and thus follows post-WTO provisions, laws, or regulations, we do not consider that it is appropriate for the Department to retroactively change its previous results (especially those of the original investigation) based on new provisions, laws, or regulations. If respondents had desired that post-WTO changes in the law regarding the identification and quantification of subsidies be taken into account, they could have requested an administrative review subsequent to changes in the applicable laws and regulations, but they did not. Therefore, we determine that there is no basis to reject the Department’s earlier findings in an investigation due to subsequent changes in methodology because such changes do not invalidate the Department’s findings under the prior methodology especially where, as here, there has been no administrative review. Sunset Determination at cmt. 14. As Commerce misapprehends the purpose and scope of a sunset review, its concerns regarding the consideration of changes in law are misplaced. Under a sunset review, Commerce is to determine as accurately as possible whether countervailable subsidies are likely to continue or recur. Sunset reviews are “inherently prospective.” See Ad Hoc Comm. of Domestic Uranium Producers v. United States, 162 F.Supp.2d 649, 664 (CIT 2001) (likening ITC’s sunset review to threat of material injury determination). See also Matsushita Elec. Indus. Co. v. United States, 750 F.2d 927, 933 (Fed.Cir.1984) (finding that given “inherently predictive” nature of review investigation, “Commission must assess, based on currently available evidence and on logical assumptions and extrapolations flowing from that evidence, the likely effect of revocation of the antidumping order on the behavior of the importers.”). By requesting that Commerce consider changes affecting the original CVD rate, Plaintiffs are not asking Commerce to “reject” its findings under the original investigation, or to change “retroactively” the results of the original or subsequent investigations. A revocation order issued pursuant to sunset review would apply to imports entered after the effective date of such order. This is not a retroactive application. As a result, this case does not present the type of retroactivity problems at issue in Melex USA v. United States, 19 CIT 1130, 899 F.Supp. 632 (1995). There, the court held that assessment of antidumping duties for entries that took place before revocation of the injury determination was authorized under the antidumping statute, but that methodologies under the amended statute should not have been retroactively applied to prior entries. In that case, the Department of Treasury made a finding of dumping in 1975. Melex, 19 CIT at 1131, 899 F.Supp. at 634. Later, the Commission conducted a changed circumstances review, and on June 11, 1980, determined that the domestic industry would not be threatened with material injury if the antidumping order were revoked. Id. Commerce subsequently issued a notice of revocation, but specified that unappraised entries made prior to June 11, 1980 were unaffected. Id. In 1991, the exporters requested an administrative review regarding entries made pri- or to June 10, 1980. With respect to these entries, Commerce determined that the exporters sold the merchandise at less-than-fair-value. Melex, 19 CIT at 1132, 899 F.Supp. at 635. The exporters appealed, claiming that because the anti-dumping duty order was revoked in 1980, Commerce could not assess duties on the pre-1980 entries. Melex, 19 CIT at 1133, 899 F.Supp. at 636. The court rejected this argument, holding that “a revocation of a determination of injury is prospective in that it does not revoke the determination of injuries for entries prior to the effective date of the revocation.” Id. Thus, the court allowed for a review of the pre-1980 unliquidated entries, notwithstanding the 1980 revocation order. Nevertheless, the court found that Commerce erred in applying current law in the 1992 administrative review of the pre-1980 entries. The court reasoned that “the application of the 1979 and 1984 amendments to entries made prior to the effective dates of those amendments would be the retroactive application of those amendments which is not clearly required by statutory language, legislative history or administrative practice.” Melex, 19 CIT at 1138, 899 F.Supp. at 639. In this case, Plaintiffs are not seeking application of the current legal regime to imports already entered into the U.S. to obtain an adjustment of duties already paid. In this case, the likelihood determination involves whether subsidies would be likely to continue or recur, thereby having an effect on imports entered after the effective date of the potential revocation order. The statute directs Commerce to consider “any change in the program” that is likely to affect the original CVD rate. 19 U.S.C. § 1675a(b)(l)(B). Although the statute leaves undefined the term “change in the program,” the SAA indicates that Commerce is to consider whether the subsidy programs have been “continued, modified, or eliminated.” SAA at 888. The SAA thus contemplates that pursuant to a sunset review, Commerce is to consider changes in foreign law that affect the program. By its nature, then, a sunset review is designed to account for changes in law that have a bearing on whether counter-vailable subsidies will continue or recur. A sunset review does not provide for recalculation of the original CVD rate such that duties on entered imports that were subject to the order must be revised retroactively. It stands to reason, however, that how Commerce views a particular subsidy under current practices and regulations will bear on its determination of the likelihood that the subsidy will continue or recur beyond the end of sunset review. In addition, the court finds no support for Commerce’s tacit imposition of a requirement that an interested party participating in a sunset review must have first requested and completed an administrative review. Administrative reviews may not be available for several reasons. For example, under 19 C.F.R. § 351.213(d)(3), an administrative review may be rescinded based on a finding that “during the period covered by review, there were no entries, exports, or sales of the subject merchandise.” In this case, Commerce does not dispute that none of the German producers, except Dillinger, made any shipments of subject merchandise since the issuance of countervailing duty orders in 1993, or that Dillinger’s last shipment in 1995 pre-dated the changes in law at issue in this case. Commerce suggests that the Plaintiffs could have made a token shipment to the United States in order to avoid rescission under 19 C.F.R. § 351.213(d)(3). Nothing in the statute requires such machinations in order to obtain a meaningful sunset review. Thus, Commerce is not barred from considering changes in the foreign or U.S. laws described above, or from considering facts underlying these changes, which may or may not have taken place after the imposition of the original CVD order. Commerce must consider and give a reasoned explanation in response to material and reasonable arguments as to why a change in U.S. or foreign law would or would not have an impact on the likelihood of continuance or recurrence of the subsidies under review. See SAA, at 892, reprinted in 1994 U.S.C.C.A.N. at 4216 (“the agencies must specifically reference in their determinations factors and arguments that are material and relevant, or must provide a discussion or explanation in the determination that renders evidence the agency’s treatment of a factor or argument.”). If Commerce chooses to exercise its discretion not to apply these changes in law, or decide that the facts do not warrant an adjustment, it must state its reasons in its likelihood determination. It is insufficient to decline consideration of any change in law subsequent to the original determination simply based on the supposed limitation on the scope of a sunset review, as such considerations are inherent in making the likelihood determination required in the sunset review. Commerce will therefore determine on remand whether, if at all, adjustments are warranted based on the points raised by the Plaintiffs in their substantive response and subsequent submissions. The court takes no position on the cited changes as these matters are not ripe for review. B. Current Methodologies Commerce concedes that it applies current law in administrative reviews, but that an exception should be made for sunset reviews. There is no reason for making such a distinction. Under the historical and statutory notes to 19 U.S.C. § 1671, the 1994 URAA amendments “shall take effect on [January 1,1995], and apply with respect to ... reviews initiated under section 1675,” which includes administrative, changed circumstances and sunset reviews. Furthermore, there is no indication in these notes that Commerce or the Commission should distinguish between review of an order entered pursuant to an investigation that took place after the URAA amendments, and those that took place before. The general principle seems to be, therefore, that current law governs any review under section 1675 irrespective of the date of issuance of the order under review. It follows from this principle that Commerce is not barred from applying current methodologies in a sunset review. Indeed, Defendant-Intervenors concede that “if [Commerce] must consider subsequent changes in law in a sunset review, then [it] must consider its new regulations as well.” Def.-Int. Br. at 33. 1. AUL Calculation Methodology Commerce maintains that even if it were appropriate to consider its current methodologies in a sunset review, it would not change its average useful life (AUL) calculation from the original investigation. Commerce asserts that according to its past practice, where a company’s AUL previously has been calculated in an investigation, it will use the same AUL calculation in a subsequent review notwithstanding any change in methodology, provided the same subsidies are again being investigated, and in the absence of any new evidence. Plaintiffs argued in the sunset proceedings that Commerce should not continue to apply a 15-year allocation period from the IRS Class Life Depreciation Tables rather than the more specific 11-year period contained in the German depreciation schedule. Plaintiffs urged Commerce to reevaluate the 15-year allocation period in light of the following: (1) Commerce’s current regulations require it to use an allocation period based upon a country-wide depreciation rate if that rate differs from the IRS tables by more than one year; and (2) findings in an investigation that the applicable allocation period from the German depreciation schedule for productive assets in the German steel industry was found to be 11-years. See Steel Wire Rod, 62 Fed.Reg. at 54,991. Administrative practice does not detract from the general principle described above. Commerce has applied methodologies in administrative reviews that were instituted in accordance with the court’s decisions issued subsequent to the original investigation. Commerce has also refrained from doing so where circumstances of a case warrant, such as when the application of current law would not lead to more accurate results. For example, the investigations in Certain Hot-Rolled Lead and Bismuth Carbon Steel Products from the United Kingdom, 63 Fed.Reg. 18,367, 18,369 (Dep’t Comm.1998) (final adm. review), stand in contrast to those in Certain Cutr-to-Length Carbon-Quality Steel Plate from France, 64 Fed.Reg. 73,277 (Dep’t Comm.1999) (final adm. review). In the former case, Commerce applied an allocation methodology based on company-specific average useful life data in accordance with British Steel v. United States, 929 F.Supp. 426, 439 (CIT 1996). Commerce specifically rejected the respondent’s request to apply the method for calculating average useful life according to the IRS tables as had been done in the original investigation. Commerce chose the rate derived under the current methodology, reasoning that applying different allocation periods for the same subsidies in two different proceedings involving the same company would generate significant inconsistencies. Id. In the latter case, by contrast, Commerce continued to allocate subsidies that had been countervailed in prior cases according to the company-specific AUL assigned to the same subsidies in those prior cases based on the POI. The respondent had requested that Commerce apply the current methodology of calculating the company’s AUL in each year that a subsidy is provided. 64 Fed.Reg. at 73,293 cmt. 13. Commerce rejected this argument, reasoning that recalculating the AUL under the current methodology would be “extremely burdensome” and not necessarily accurate where the company-specific AUL proposed by the respondents derived from an on-going investigation. Id. In contrast, Commerce found that the AUL from the prior investigation was “reasonable and accurate as possible without being burdensome.” Id. At the same time, Commerce applied its new discount policy under 19 C.F.R. § 351.524 because the discount rate derived therein was “more realistic” and would result in a “better measure of the subsidy.” Id. Thus, under its own practices, Commerce may choose or not choose to apply current law in a review as circumstances warrant. In this case, Commerce has made no findings pursuant to sunset review with respect to whether application of current methodologies as applied in Steel Wire Rod would result in a more accurate CVD rate. In the Sunset Determination, Commerce refused to apply the methodology and findings in Steel Wire Rod because “in this review we are dealing with different companies and a different set of subsidies.” Sunset Determination at cmt. 7, n. 32. On appeal, Commerce reiterates this statement without making factual findings relating to Plaintiffs’ assertions that (1) Steel Wire Rod involved the SVK assistance at issue in this case, and (2) Commerce had found in the Preliminary Sunset Determination that “DHS and Dillinger remained, for all intents and purposes, the same entities as the pre-privatization Saar-stahl/DHS.” Preliminary Sunset Determination at 1.10-13. As it is not the court’s role to make findings in this regard, Commerce must do so on remand. 2. Change-m-Ownershvp Methodology Plaintiffs argue that Commerce erred in continuing to apply its change-in-ownership methodology found “illegal” by the U.S. Court of Appeals for the Federal Circuit and the WTO Dispute Settlement Body. Plaintiffs contend that had Commerce applied its current change-in-ownership methodology in this case, the countrywide subsidy rate in the Cut-to-Length Plate proceeding would have fallen from 14.84% to 0.21%. Commerce determined that it was “not appropriate” to reach the privatization issue for purposes of sunset review. Commerce explained that: Both the Appellate Body and CAFC decisions were issued relatively late in this proceeding. A sunset review is not the appropriate proceeding in which to examine a complicated privatization transaction and to consider a new privatization methodology. In light of the complexity and fact-intensive nature of this issue, it is imperative that the issues be fully developed on the record. Commerce argues that the sunset review schedule does not allow it time to analyze the complex factual issues that would be revisited by an analysis of a change in law. Commerce ignores the statutory provision that it may seek an extension of time for “extraordinarily complicated” sunset reviews. See 19 U.S.C. § 1675(c)(5)(C). See also 19 C.F.R. § 351.218(f)(3)(ii) (“If the Secretary determines that a full sunset review is extraordinarily complicated under section 751(c)(5)(C) of the Act, the Secretary may extend the period for issuing final results by not more than 90 days.”). Indeed, the statute specifically provides that Commerce may treat a review as “extraordinarily complicated” if it is a review of a transition order. See 19 U.S.C. § 1675(c)(5)(C)(v). In any event, as Plaintiffs proposed that a change-in-ownership methodology consistent with current law be applied to facts already on the record, it is not apparent to the court why Commerce was unable to make any necessary adjustments within the time requirement of the statute. If Commerce deemed the issues raised by the Plaintiffs to warrant a more thorough analysis or more extensive fact gathering, it could have sought additional time, but did not do so. Commerce cannot justify its decision not to address the issues raised by the parties simply on the basis that the issues arise in the context of a sunset review and the stringent time limits involved, especially when Commerce does not avail itself of the mechanisms provided for by its own regulations. 3. Evidence relating to Methodologies Plaintiffs also argue that Commerce abused its discretion in removing from the record the German producers’ March 14 and March 17 submissions. Plaintiffs make the following contentions: (1) Commerce acted inconsistently in not rejecting the domestic producers’ November 9,1999, and March 16, 2000 post-rebuttal comments, and allowing the domestic producers to submit “new factual information” (namely, copies of questionnaire responses from the original investigation) as late as April 28, 2000, see P.R. Doc. 763, PLApp. Tab 12; and (2) Commerce had already accepted the submissions, as is evidenced by Commerce’s references in the Preliminary Sunset Determination to the issues raised therein. Commerce responds that there were no inconsistencies in its rejection of submissions where: (1) its return of the March 14 submission rendered the domestic producers’ March 16, 2000 response “moot and irrelevant” and it did not consider it for the purposes of the Sunset Determination; and (2) the domestic producers’ accepted submission dated November 9, 1999, related to its adequacy determination and was therefore timely in fact. DOC Br. at 42-43. Commerce also explained in the Sunset Determination that in fact it did not rely on the March submissions in the Preliminary Sunset Determination. There is no statutory “deadline” for Commerce to reject a submission in a sunset review. Nevertheless, once Commerce has accepted a submission, in certain cases it may be improper and/or prejudicial for it to remove the submission from the record. Commerce is presumed to consider the evidence on the record in making its determinations. At the time of the preliminary sunset determination in this case, Commerce had not yet rejected the German producer’s submission. Thus, the court must presume that it considered the German producers’ March 14, 2000, and March 17, 2000 submissions as well as the domestic producers’ March 16 submission in making its preliminary sunset determination. The court further finds an inexplicable inconsistency in Commerce’s subsequent rejection of the German producers’ submission while failing to do the same with the domestic producers’ March 16, 2000, and April 28, 2000 submissions. The argument that the unequal treatment of the parties concerns a moot issue is a post hoc argument and does not explain Commerce’s behavior. In evaluating the changes in methodology described above, Commerce shall consider the evidence submitted by the parties relating thereto. Conclusion Accordingly, Commerce’s determination pursuant to sunset review is remanded. Commerce cannot act irrationally and arbitrarily. It cannot accept Domestic Parties’ “late” submissions and reject those of respondents. It cannot make some adjustments to an original CVD rate and then state it will not consider evidence of other adjustments because all adjustments are barred. It cannot avoid applying changes in the law because it is burdensome or inconvenient. Certainly, Commerce has a difficult task in deciding whether subsidies will continue or recur. It can streamline its procedures to some degree, but it must conduct meaningful sunset reviews. It has great