Full opinion text
OPINION RIDGWAY, Judge. This action challenges a July 1999 agreement between the U.S. Department of Commerce (“Commerce”) and three Brazilian steel exporters (“Brazilian Exporters”), suspending at the eleventh hour the investigation into the alleged dumping in the United States of certain steel products from Brazil. See HoNRolled Flat-Rolled Carbon-Quality Steel Products from Brazil, 64 Fed.Reg. 38,792 (Dep’t Commerce 1999) (suspension of antidump-ing investigation and entry of suspension agreement) (Public Administrative Record Document (“P.R.Doc.”) No. 295) (“Notice of Suspension” or “Agreement”). That investigation was initiated at the behest of, among others, the plaintiff domestic steel producers here (“Domestic Producers”). See Certain Hot-Rolled Flat-Rolled Carbon-Quality Steel Products From Brazil, Japan, and the Russian Federation, 63 Fed.Reg. 56,607 (Dep’t Commerce 1998) (initiation of antidumping investigations). The Domestic Producers contend that the July 1999 Suspension Agreement is the product of a high stakes “bait and switch” maneuver (Tr. at 26), and that the Agreement is unlawful. As the Domestic Producers have observed, a suspension agreement is essentially a unique form of settlement agreement: a settlement agreement to which the complainant — that is, the domestic industry — is not a signatory. Tr. at 4; see also 125 Cong. Rec. 20,168 (1979). While Congress has authorized Commerce to enter into suspension agreements, Congress has emphasized the limited circumstances in which such agreements are appropriate, and has established rigorous procedural safeguards to ensure that they are not abused to the detriment of domestic interests. The law on suspension agreements requires, among other things, that Commerce afford the domestic industry the opportunity to review and comment on any proposed agreement. 19 U.S.C. § 1673c(e)(1994). But the Domestic Producers note that the Proposed Agreement that Commerce proffered for review and comment in this case was materially different from the final Suspension Agreement that Commerce executed. See Proposed Agreement Suspending the Antidumping Investigation on Hot-Rolled FlaWRolled Carbon-Quality Steel from Brazil (June 6, 1999) (P.R. Doc. No. 267) (“Proposed Agreement”). The Domestic Producers argue that Commerce’s “bait and switch” of agreements failed to satisfy its notice, comment and consultation obligations under the applicable law. The Domestic Producers further assert that, under the facts of this case, Commerce cannot make the substantive determinations required to justify the Suspension Agreement. Pending before the Court is Plaintiffs’ Motion for Judgment on the Agency Record (filed under USCIT Rule 56.2), in which the Domestic Producers request that the Court declare the Suspension Agreement null and void, and instruct Commerce to enter an antidumping duty order on the subject merchandise. Plaintiffs’ motion is opposed by Defendant, the United States (“the Government”), as well as the Brazilian Exporters, who argue that Commerce’s determination to suspend the antidumping investigation should be sustained in all respects. Plaintiffs motion is granted in part. For the reasons discussed below, this case is remanded to the Department of Commerce to enable it to comply with the notice, comment and consultation requirements of the statute (19 U.S.C. § 1673c(e)), to enable it to articulate a legal standard for making its public interest determination or otherwise explain the connection between the facts found and the choice made pursuant to the statute (19 U.S.C. § 1673c(d)(l)), and — if appropriate in the case of a subsection (c) agreement, which requires a finding by Commerce that the investigation is “complex”) — to enable it to interpret the phrase “large number” of transactions or adjustments in the context of the “extraordinary circumstances” requirement of the statute (19 U.S.C. § 1673c(c)(2)(B)(i)). I. Background A. The Suspension Agreement Statute The statutory provisions authorizing suspension agreements in antidumping duty eases were added to the Tariff Act of 1930 as part of the Trade Agreements Act of 1979 (“the Act”), as a means of limiting the Government’s discretion to suspend investigations. Although there was no specific statutory authorization for doing so, it was the practice of the Secretary of the Treasury — prior to the Act — to discontinue antidumping duty investigations at any time if it was determined that (1) possible margins of dumping were minimal relative to export volume, price revisions had been made to eliminate any likelihood of less than fair value (“LTFV”) sales, and there were assurances that there would be no LTFV sales in the future; (2) sales to the United States had terminated and there were assurances that they would not resume; or (3) there were other circumstances making it inappropriate to continue the investigation. H. Rep. No. 96-317 at 67; see also S.Rep. No. 96-249 at 67, reprinted in 1979 U.S.C.C.A.N. 381 at 453 (1979) (briefly summarizing Treasury Department practice at the time). See generally Letter from Dewey Ballantine to Clerk of Court of International Trade (March 9, 2001) at 1 (summarizing use of suspension agreements prior to Trade Agreements Act). The suspension agreement provisions were adopted to impose “strict limits on discontinuing or suspending investigations pursuant to deals with foreign governments or producers.” 125 Cong. Rec. 20,-163 (1979). The provisions were intended to make “major changes in ... [the then-existing] law and practice concerning the suspension of an investigation based on exporter agreements” by, inter alia, replacing certain general, subjective criteria then in use (e.g., “minimal” dumping margins) with “specific criteria and requirements”; eliminating the Secretary’s “unfettered discretion” to terminate investigations; and “improving] the procedural safeguards ... by providing increased participation by the petitioner.” H. Rep. No. 96-317 at 63, 67. With the Trade Agreements Act, Congress expressly authorized Commerce to suspend an antidumping duty investigation in certain circumstances where foreign exporters of the merchandise agree to take remedial action. 19 U.S.C. §§ 1673c(b)-(m) (1994). In doing so, Congress recognized the potential advantages of suspension agreements, as a “means of achieving the remedial purposes of the [antidump-ing] law in as short a time as possible and with a minimum expenditure of resources by all parties involved.” H. Rep. No. 96-317 at 63. Accord, S.Rep. No. 96-249 at 71, reprinted in 1979 U.S.C.C.A.N. 381, 457 (suspension agreements “permit rapid and pragmatic resolutions of antidumping duty cases”). But Congress was equally mindful of the potential for abuse of suspension agreements. To ensure that such agreements are not entered into to the disadvantage of a petitioning domestic industry (for foreign policy or other political reasons), the statute is replete with stringent requirements that must be met before an agreement can be concluded. Congress emphasized that “the authority to suspend investigations [is to] be exercised within the carefully • circumscribed limits” set forth in the law. H. Rep. 96-317 at 63; see also S.Rep. No. 96-249 at 71, reprinted in 1979 U.S.C.C.A.N. 381, 457 (to ensure that suspension agreements are used only in those cases where they “serve[ ] the interest of the public and the domestic industry affected,” agency authority to suspend investigations is “narrowly circumscribed”); 125 Cong. Rec. 20,-168-69 (Senator Heinz’s understanding that suspension agreements permitted only “under certain narrowly constrained circumstances” confirmed by bill managers Senators Ribicoff and Roth). Congress further cautioned that “suspension is an unusual action which should not become the normal means for disposing of [anti-dumping] cases.” S.Rep. No. 96-249 at 71, reprinted in 1979 U.S.C.C.A.N. 381, 457. There are two distinct types of suspension agreements, both of which are relevant here — so-called “subsection (b) agreements” and “subsection (c) agreements.” Subsection (b) agreements eliminate all sales at less than fair value. Under such agreements, exporters must agree either to cease all exports, or to revise their prices to eliminate completely LTFV sales. 19 U.S.C. § 1673c(b). In contrast, subsection (c) agreements do not completely eliminate dumping; rather, they eliminate only its injurious effect. 19 U.S.C. § 1673c(c). Prior to accepting either a subsection (b) or (c) agreement, Commerce must find both that “suspension of the investigation is in the public interest,” and that “effective monitoring of the agreement by the United States is practicable.” 19 U.S.C. § 1673e(d). Commerce is also required to notify petitioners of, and consult with them concerning, its intention to suspend the investigation. In addition, Commerce must provide petitioners with a copy of the proposed agreement, and accord them an opportunity to comment. 19 U.S.C. § 1673c(e). But there are additional requirements for subsection (c) agreements. Because such agreements, by definition, may permit exporters to continue to engage in a certain amount of dumping, Congress restricted subsection (c) agreements to cases involving “extraordinary circumstances”— cases where the suspension of the investigation is more beneficial to the domestic industry than its continuation, and where the investigation is “complex.” See S.Rep. No. 96-249 at 68, reprinted in 1979 U.S.C.C.A.N. 381, 454 (discussing the extraordinary circumstances requirement set .out in 19 U.S.C. § 1673c(c)(2)). Moreover, subsection (c) agreements must limit the margin by which normal value exceeds the export price, and prevent the suppression or undercutting of price levels of domestic products by imports. 19 U.S.C. § 1673(c)(1). As Congress intended, Commerce has invoked the suspension agreement provisions of the trade laws relatively infrequently — -at least until recently. Notably, prior to the suspension of the investigation at issue here, Commerce had accepted only two other subsection (c) agreements in antidumping cases. See Potassium Chloride from Canada, 53 Fed.Reg. 1393 (Dep’t Commerce 1988) (suspension of antidump-ing investigation and entry of suspension agreement); Fresh Tomatoes from Mexico, 61 Fed.Reg. 56618 (Dep’t Commerce 1996) (suspension of antidumping investigation and entry of agreement). B. The Facts of This Case On September 30, 1998, the Domestic Producers who are plaintiffs here — among others — petitioned Commerce and the International Trade Commission (“ITC”), seeking the imposition of antidumping duties on certain steel products from Brazil. See Letter from Dewey/Schag-rin/Skadden to Commerce and ITC, and enclosed Petition (Sept. 30, 1998) (P.R. Doc. No. 1). That same day, the same petitioners filed a parallel petition, seeking the imposition of countervailing duties on the same products. See Certain Hob-Rolled Flat-Rolled Carbon-Quality Steel Products from Brazil, 63 Fed.Reg. 56,623 (Dep’t Commerce 1998) (initiation of countervailing duty investigation) (referring to petition); Certain Hot-Rolled Steel Products From Brazil, Japan, and Russia, 63 Fed.Reg. 53,926 (ITC 1998) (institution of countervailing duty and antidumping investigations) (referring to petition). Both the antidumping and countervailing duty petitions were accepted, and the requested investigations were initiated on October 15, 1998. See Certain HoNRolled Flat-Rolled Carbon-Quality Steel Products From Brazil, Japan, and the Russian Federation, 63 Fed.Reg. 56,607 (Dep’t Commerce 1998) (initiation of antidumping investigations); Certain Hot-Rolled Flat-Rolled Carbon-Quality Steel Products From Brazil, 63 Fed.Reg. 56,623 (Dep’t Commerce 1998) (initiation of countervailing duty investigation). One month later, the ITC notified Commerce of its preliminary determination, finding in both investigations that “there [was] a reasonable indication that an industry in the United States [was] threatened with material injury by reason of imports” of the Brazilian steel. See Certain Hot-Rolled Steel Products From Brazil, Japan, and Russia, 63 Fed.Reg. 65,221 (ITC 1998) (preliminary injury determinations in antidumping and countervailing duty investigations). On February 12, 1999, Commerce made its preliminary determination in the anti-dumping duty investigation, finding that the Brazilian steel was being, or was likely to be, sold in the United States at less than fair value. See Hot-Rolled Flat-Rolled Carbon-Quality Steel Products from Brazil, 64 Fed.Reg. 8,299 (Dep’t Commerce 1999) (P.R. Doc. No. 184). Simultaneously, in the countervailing duty investigation, Commerce made a preliminary determination that eountervailable subsidies were being provided to the Brazilian Exporters. See Certain Hob-Rolled FlaiARolled Carbon-Quality Steel Products From Brazil, 64 Fed.Reg. 8,313 (Dep’t Commerce 1999) (preliminary determination of sales at LTFV). On June 6, 1999, literally one month to the day before the deadline for its final determinations, Commerce sent the Domestic Producers (among others) initialed proposed agreements to suspend both the antidumping and countervailing duty investigations, and requested their comments by June 28, 1999. The proposed agreement in the antidumping investigation — the investigation directly at issue in this case — was a subsection (b) agreement. See Letters from Commerce to Interested Parties (June 6, 1999) (P.R. Doc. No. 267) (requesting comments on enclosed Proposed Agreement). The Domestic Producers filed timely comments, detailing their many objections to the proposed suspension agreements. See Letter from Dewey/Skadden/Schagrin to Commerce (June 28, 1999) (P.R. Doc. No. 270); Certain Hot-Rolled Flat-Rolled Carbon-Quality Steel Products From Brazil, 64 Fed.Reg. 38,797 (Dep’t Commerce 1999) (suspension of countervailing duty investigation and entry of suspension agreement) (acknowledging petitioners’ comments on proposed suspension agreement in countervailing duty case). Nevertheless, on July 6, 1999, the deadline for its final determinations in both the antidumping and countervailing duty investigations, Commerce entered into final agreements suspending those investigations. See Hot-Rolled Flat-Rolled Carbon-Quality Steel Products From Brazil, 64 Fed.Reg. 38,792 (Dep’t Commerce 1999) (suspension of antidumping investigation and entry of suspension agreement); Certain Hob-Rolled FlaNRolled Carbon-Quality Steel Products from Brazil, 64 Fed.Reg. 38,797 (Dep’t Commerce 1999) (suspension of countervailing duty investigation and entry of suspension agreement). The final suspension agreement in the antidumping duty investigation was based not on subsection (b), but on subsection (e). That final agreement thus permits a certain amount of continued dumping — in contrast to the proposed agreement, which would have either required the complete elimination of dumping, or cut off exports altogether. In short, not only were the suspension agreements in general an eleventh hour development; there was also a last minute substitution of a different type of suspension agreement in the antidump-ing duty case. Under the terms of the final agreement suspending the antidumping duty investigation (the agreement at issue here), the Brazilian Exporters are prohibited from exporting the steel at issue to the United States for less than the negotiated Reference Price. Agreement, Part IV, 64 Fed.Reg. at 38,794. In addition, the Brazilian Exporters agree to limit dumping to 15% (as required under 19 U.S.C. § 1673c(c)(l)(B)). Agreement, Part IV, at id. The Agreement obligates the Brazilian Exporters to report certain data concerning U.S. sales, and includes a “catch-all” requirement that each Brazilian Exporter “supply to [Commerce] 30 days after the end of each Quarter all information that [Commerce] determines is necessary to ensure that the [Exporter] is in full compliance with the terms of [the] Agreement.” Agreement, Part V, 64 Fed.Reg. at 38,794. The Agreement also authorizes Commerce to share business propriety information obtained from the Brazilian Exporters with the domestic parties to the underlying antidumping duty investigation, under appropriate administrative protective orders. Agreement, Part VI, 64 Fed.Reg. at 38,-794. The monitoring provisions of the Agreement provide that Commerce is to ensure compliance with the Agreement by reviewing, inter alia, “publicly-available data and other official import data, including, as appropriate, records maintained by the U.S. Customs Service.” Agreement, Part VII, 64 Fed.Reg. at 38,794. In addition, Commerce is authorized to require any Brazilian Exporter to provide documentation to confirm that the price received on any sale was not less than the established Reference Price. The Agreement also permits Commerce to obtain from each Brazilian Exporter a report of each sale of steel subject to the Agreement (including each adjustment applicable to each sale). Finally, the Brazilian Exporters consent to permit “review and on-site inspection of all information deemed necessary” by Commerce to verify the reported information. Agreement, Part VII, 64 Fed.Reg. at 38,-794. The Agreement further provides that Commerce may conduct administrative reviews (upon request or at its own initiative), and includes certain provisions designed to ensure that the Agreement is not circumvented. Agreement, Parts VIII, IX, 64 Fed.Reg. at 38,794-95. In the event of suspected noncompliance or violation of the Agreement, Commerce is authorized to request that the subject Brazilian Exporter provide “all information relating to the allegation, including all sales information pertaining to covered and non-covered merchandise [that the Exporter has] manufactured or sold.” Agreement, Part X, 64 Fed.Reg. at 38,795. Any party to the underlying antidumping duty investigation may then submit comments on that information. Id. If Commerce determines that the Agreement is being or has been violated or no longer meets the relevant requirements of the suspension agreement statute, the Agreement provides that Commerce “shall take whatever action it deems appropriate” under the relevant provision of the suspension agreement statute (i.e., 19 U.S.C. § 1673c(i)) and the applicable regulations. Agreement, Part XI, 64 Fed.Reg. at 38,-795. Concurrent with its execution of the suspension agreements in the parallel anti-dumping duty and countervailing duty proceedings, Commerce separately circulated three decision memoranda setting forth the bases for the determinations supporting its decision to suspend the two investigations. See Memoranda from ITA Office of Policy to Ass’t Sec. for Import Administration (July 6, 1999) (P.R. Doc. Nos. 281, 282, and 283) (addressing, respectively, the prevention of price suppression and undercutting, the existence of extraordinary circumstances, and the public interest). That same day, Commerce also issued final affirmative determinations in both the antidumping duty investigation and the counteivailing duty investigation. See Certain Hot-Rolled Flat-Rolled Carbon-Quality Steel Products From Brazil, 64 Fed.Reg. 38,756 (Dep’t Commerce 1999) (final determination of sales at LTFV); Certain Hot-Rolled Flat-Rolled Carbon-Quality Steel Products From Brazil, 64 Fed.Reg. 38,742 (Dep’t Commerce 1999) (final affirmative countervailing duty determination). Both investigations were continued, notwithstanding the suspension agreements, pursuant to the request of petitioners in the underlying proceedings. See Letter from Skadden/Dewey/Schagrin to Commerce and ITC (July 2, 1999) (P.R. Doc. No. 271) at 1. Commerce found the final dumping margins for CSN, USIMI-NAS/COSIPA, and all others to be 41.27%, 43.40%, and 42.12%, respectively. Similarly, the net subsidy rates for CSN, USIMINAS/COSIPA, and all others were determined to be 6.35%, 9.67%, and 7.81%, respectively. The ITC’s final determinations — issued August 24, 1999 — confirmed its affirmative preliminary findings on material injury in both the antidumping duty and countervailing duty cases. See Certain Hot-Rolled Steel Products From Brazil and Russia, 64 Fed.Reg. 46,951 (ITC 1999) (final injury determinations in antidumping and countervailing duty investigations). II. Jurisdiction and Standard of Review Jurisdiction in this matter is predicated on 28 U.S.C. § 1581(c) (1994). Commerce’s decision to suspend the anti-dumping duty investigation at issue is reviewable pursuant to 19 U.S.C. § 1516a(a)(2)(B)(iv) (1994), and must be sustained 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) (1994). “Substantial evidence” is “something less than the weight of the evidence.” Consolo v. Federal Maritime Comm’n, 383 U.S. 607, 620, 86 S.Ct. 1018, 16 L.Ed.2d 131 (1966). Moreover, the possibility of drawing two inconsistent conclusions from the same evidence does not mean that substantial evidence is lacking. Id. In other words, Commerce’s determination cannot be overturned merely because the plaintiff “is able to produce evidence ... in support of its own contentions and in opposition to the evidence supporting the agency’s determination.” Torrington Co. v. United States, 14 CIT 507, 514, 745 F.Supp. 718, 723 (1990) (internal quotation omitted), aff'd, 938 F.2d 1276 (Fed.Cir.1991). Nevertheless, Commerce’s decision must be supported by “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Consolidated Edison v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 83 L.Ed. 126 (1938); accord, Matsushita Elec. Indus. Co. v. United States, 750 F.2d 927, 933 (Fed.Cir.1984). Its determination must be supported by the record as a whole, including whatever fairly detracts from the substantiality of the evidence. Tianjin Mach. Imp. & Exp. Corp. v. United States, 16 CIT 1020, 1023 (1992) (citing Atlantic Sugar, Ltd. v. United States, 744 F.2d 1556, 1561 (Fed.Cir.1984)). Commerce’s conclusions must be “reached by ‘reasoned decisionmaking,’ including an examination of the relevant data and a reasoned explanation supported by a stated connection between the facts found and the choice made.” Electricity Consumers Resource Council v. Federal Energy Regulatory Comm’n, 747 F.2d 1511, 1513 (D.C.Cir.1984), citing Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 9 L.Ed.2d 207 (1962). In determining whether Commerce’s interpretation and application of the anti-dumping statute is “in accordance with law,” “[fjirst, always, is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.” Chevron U.S.A. Inc. v. Natural Resources Defense Council, 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). If the statute is silent or ambiguous, the degree of respect accorded an agency’s interpretation depends on whether that interpretation is the product of a process that gives it “the force of law” — for example, a formal adjudication or notice-and-comment rulemaking. See Christensen v. Harris County, 529 U.S. 576, 120 S.Ct. 1655, 1662, 146 L.Ed.2d 621 (2000). If it is, the court must accord Chevron deference, and uphold any reasonable agency interpretation. Id.; see also Chevron, 467 U.S. at 842-43, 104 S.Ct. 2778. In contrast, agency “[ijnterpretations ... in opinion letters — like interpretations contained in policy statements, agency manuals, and enforcement guidelines, all of which lack the force of law — do not warrant Chevron-style deference.” Christensen, 120 S.Ct. at 1662, 120 S.Ct. 1655. Agency interpretations which lack the force of law are “entitled to respect ... but only to the extent that those interpretations have the ‘power to persuade’.” Id. at 1663, 120 S.Ct. 1655, citing Skidmore v. Swift & Co., 323 U.S. 134, 140, 65 S.Ct. 161, 89 L.Ed. 124 (1944). III. Discussion A. Notice, Comment and Consultation The Domestic Producers here argue that Commerce failed to comply with the notice, comment and consultation requirements of the suspension agreement statute because they were denied a meaningful opportunity to submit comments and information for the record prior to the suspension of the antidumping investigation. In essence, the Domestic Producers contend that affording them an opportunity to comment on a proposed subsection (b) agreement cannot satisfy the notice and comment obligations as to the subsection (c) agreement that Commerce and the Brazilian Exporters executed. Memorandum of Points and Authorities in Support of Plaintiffs’ Rule 56.2 Motion for Judgment Upon the Agency Record (“Plaintiffs’ Memo”) at 24-26. The notice, comment and consultation provisions of the statute require that, before entering into a suspension agreement, Commerce must: (1) notify the petitioner of, and consult with the petitioner concerning, its intention to suspend the investigation ... not less than 30 days before the date on which it suspends the investigation. (2) provide a copy of the proposed agreement to the petitioner ... together with an explanation of how the agreement will be carried out and enforced, and of how the agreement will meet the requirements of subsections (b) and (d) or (c) and (d) of [the statute], and (3) permit all interested parties ... to submit comments and information for the record before the date on which notice of suspension of the investigation is published.... 19 U.S.C. § 1673c(e). The legislative history of the statute highlights the importance of those provisions, emphasizing that “the requirement that the petitioner be consulted will not be met by pro forma communications. Complete disclosure and discussion is required.” S.Rep. No. 96-249 at 71, reprinted, in 1979 U.S.C.C.A.N. 381, 457. The Government argues that the procedure that Commerce followed complied fully with the letter of the law: Commerce provided the petitioners below with a copy of the proposed agreement; Commerce afforded the petitioners an opportunity to-comment on that proposed agreement; and Commerce responded directly to the petitioners’ comments on the proposed agreement — albeit by executing a subsection (c) agreement in place of the subsection (b) agreement that Commerce had proposed. Defendant’s Response in Opposition to Plaintiffs’ Motion for Judgment Upon the Agency Record (“Defendant’s Memo”) at 55-57. See also Defendant-Intervenors’ Memorandum of Points and Authorities (“Defendant-Intervenors’ Memo”) at 42-4A (asserting that “[t]he notice and comment provisions, in short, worked: they caused the Department to alter the basis of the agreement to accommodate Petitioners’ concerns”). The Government maintains that Commerce “is not required to afford interested parties an unlimited opportunity to comment on each modification of the agency’s practice or procedure.” Defendant’s Memo at 57, citing British Steel, plc v. United States, - 19 CIT 176, 255, 879 F.Supp. 1254, 1317 (1995), aff'd in part, rev’d on other grounds, 174 F.3d 1359 (Fed.Cir.1999). True enough. But, in this case, the Domestic Producers had no opportunity to comment on several of the most important aspects of the Agreement that Commerce actually executed — a subsection (c) agreement. As outlined in section I.A above, the requirements for subsection (b) agreements differ from those for subsection (c) agreements. In addition to the requirements applicable to both (b) and (c) agreements — i.e., the requirements that an agreement be in the public interest and able to be effectively monitored, a subsection (b) agreement requires only that the agreement provide either for the complete cessation of exports of the subject merchandise or the total elimination of dumping. 19 U.S.C. § 1673c(b). In contrast, because such agreements may permit a limited amount of dumping to continue, the requirements for subsection (c) agreements are more stringent. Specifically, a subsection (c) agreement requires not only the complete elimination of the injurious effects of dumping, the prevention of suppression or undercutting of domestic price levels, and a 15% dumping limitation, but also the presence of “extraordinary circumstances” (which are, in turn, defined as circumstances where the investigation is complex and where suspension of the investigation will be more beneficial to the domestic industry than its continuation). 19 U.S.C. § 1673c(c). Because the proposed agreement offered to the Domestic Producers for comment was a subsection (b) agreement, they had no opportunity to comment on those requirements which are unique to subsection (c) agreements. Indeed, even as to those requirements which are common to both subsection (b) and (c) agreements — i.e., the § 1673c(d) conditions (1) that suspension of the investigation be in the public interest, and (2) that effective monitoring be practicable' — a petitioner’s analysis might differ, depending on the type of agreement. It is thus impossible to say, even as to those factors, that the Domestic Producers here had adequate notice and opportunity to comment. Similarly, the statute mandates that Commerce explain to petitioners how a proposed agreement will meet the requirements of either subsections (b) and (d) or subsections (c) and (d), prior to suspending an investigation. 19 U.S.C. § 1673c(e). Because the Proposed Agreement here was a subsection (b) agreement, the explanation accompanying it addressed only the requirements of subsections (b) and (d). See Letters from Commerce to Interested Parties (June 6, 1999) (P.R. Doc. No. 267) (requesting comments on enclosed Proposed Agreement). Commerce made no attempt to explain to the Domestic Producers how the Agreement would meet the requirements of subsection (c) until after it had suspended the investigation. See Notice of Suspension, 64 Fed.Reg. at 38,793 0citing Extraordinary Circumstances Memo and Price Suppression Memo). Moreover, Commerce’s explanation as to how the Proposed Agreement would satisfy the requirements of subsection (d) was rendered moot when Commerce decided to enter into a fundamentally different suspension agreement — the subsection (c) Agreement. Again, Commerce offered its first explanation of the application of subsection (d) to that subsection (c) Agreement only after the investigation was suspended. See Notice of Suspension, 64 Fed.Reg. at 38,793 (citing Price Suppression Memo). In short, Commerce defaulted on its obligation to explain to the Domestic Producers — in advance — how the Agreement meets the applicable requirements of the statute. In essence, Commerce’s actions in this case “stacked the deck.” Judicial review is limited to the administrative record; but, here, that record was compiled selectively. Commerce effectively deprived the Domestic producers of any opportunity for meaningful comment on any agreement which remotely resembled the suspension agreement that Commerce actually signed— and thus limited the very record on which Commerce now seeks to stand. As a result of Commerce’s failure to comply with the notice, comment and consultation requirements of 19 U.S.C. § 1673c(e), the existing record in this matter cannot serve as the basis for a “substantial evidence” review of Commerce’s factual findings. Even as to the legal issues presented, the considerations of judicial economy and deference to agency autonomy and expertise that undergird the related doctrines of exhaustion and ripeness counsel remand here. Remand will allow all parties to fully exhaust their administrative remedies, and will afford Commerce the opportunity to consider the Domestic Producers’ comments, find facts, apply its expertise to the record, and explain the bases for its action. Remand also will protect agency autonomy, and allow Commerce to exercise the discretion granted it by Congress. Finally, by affording Commerce an opportunity to correct any errors it may have made, remand conceivably may obviate entirely the need for further judicial review. See generally 2 K. Davis & R. Pierce, Jr., Administrative Law Treatise §§ 15.2 (citing McKart v. United States, 395 U.S. 185, 193-95, 89 S.Ct. 1657, 23 L.Ed.2d 194 (1969)), 15.12 (3d ed.1994). Accordingly, for the reasons set forth and discussed above, and more fully below, this case must be remanded to allow Commerce to develop a complete record in accordance with the mandates of the statute and to reconsider its action. B. The 15% Dumping Limitation While the suspension agreement statute expressly contemplates that subsection (c) agreements may permit some dumping to continue, that dumping cannot exceed 15% of the weighted average amount by which the home market price (or cost) exceeded’ the U.S. market price for the goods. Specifically, the statute requires that: for each entry of each exporter the amount by which the estimated normal value exceeds the export price ... will not exceed 15 percent of the weighted average amount by which the estimated normal value exceeded the export price ... for all less-than-fair-value entries of the • exporter examination [sic; examined] during the course of the investigation. 19 U.S.C. § 1673c(c)(1)(B). The Domestic Producers contend that the Agreement fails to comply with the 15% limit because — to effectively enforce that limitation — Commerce must obtain and verify information that permits it to determine normal value, so that normal value can be compared to prices in the United States. However, the Agreement says nothing about the reporting or systematic collection of Brazilian market price and cost data — the components of normal value. Plaintiffs’ Memo at 9-11; Reply Memo at 7-9,13-14. The Brazilian Exporters contend that the Agreement on its face complies with the 15% limit, emphasizing that it not only invokes, but indeed mirrors, the exact wording of the statute: In order to satisfy the requirements of section 734(c)(1)(B) of the Act, each Signatory agrees that, for each entry of Hot-Rolled Steel subject to this Agreement, the amount by which the estimated normal value exceeds the export price (or the constructed export price) will not exceed 15 per cent of the weighted average amount by which the estimated normal value exceeded the export price (or the constructed export price) for all less-than-fair-value entries of the Signatory examined during the investigation. Agreement, Part IV.E, 64 Fed.Reg. at 38,-794. Defendant-Intervenors’ Memo at 35. See also Defendant’s Memo at 34. Moreover, while the Government concedes that the Agreement does not specifically provide for the reporting and collection of Brazilian price and cost data, it points to something that it claims is even better: a catch-all provision requiring the Brazilian Exporters to provide “all information that the Department determines is necessary to ensure that [each signatory] is in full compliance.” Defendant’s Memo at 28-32 (citing Agreement, Part V.A, 64 Fed.Reg. at 38, 794). The Domestic Producers argue that the catch-all reporting provision is insufficient to satisfy the statute, because Commerce may never require the Brazilian Exporters to provide the necessary information. Reply Memo at 9-12. And the Domestic Producers take no comfort in their ability to enforce the 15% limit by requesting an administrative review, which is — they note — cumbersome and expensive. In any event, they assert, it is the responsibility of Commerce in the first instance to ensure compliance with the 15% limit. Reply Memo at 13. For similar reasons, they contend that it is not enough that the Brazilian Exporters have committed in the Agreement to comply with the 15% limit. According to the Domestic Producers, Commerce’s reliance on that commitment amounts to an impermissible delegation of the agency’s duty to establish compliance with the mandates of the statute. Reply Memo at 12-13. The Domestic Producers assert that Commerce does not ordinarily leave compliance with pricing requirements unenforced. Plaintiffs’ Memo at 11. Indeed, they note, the Proposed Agreement required that the Brazilian Exporters provide detailed home market information to permit Commerce to satisfy itself that dumping would be eliminated (as required for a subsection (b) agreement). Plaintiffs’ Memo at 11; Proposed Agreement, Part V. The crux of the problem here is that, while the Domestic Producers had notice of and an opportunity to comment on the Proposed Agreement’s provisions for reporting and collection of data to verify the elimination of dumping, they had no such notice or opportunity as to the provisions for verifying compliance with the 15% dumping limit, because that limit applies only to subsection (c) agreements. Although the issue of the 15% limit has now been fully briefed before the Court, the Court is obligated to decide the case on the basis of the administrative record— a record which is presently defective. As discussed in section III.A above, the ease must be remanded to Commerce for further proceedings. On remand, assuming that Commerce issues a subsection (c) agreement for comment, the Domestic Producers will have the opportunity to make their case on the 15% limit directly to Commerce, and Commerce will have the benefit of the Domestic Producers’ comments on the administrative record in making its findings. C. The Practicability of Effective Monitoring The statute permits suspension agreements only where “effective monitoring of the agreement by the United States is practicable.” 19 U.S.C. § 1673c(d)(2). The statute’s legislative history underscores the importance of effective monitoring provisions: The committee intends that no agreement be accepted unless it can be effectively monitored by the United States. This will require establishment of procedures under which entries of merchandise covered by an agreement can be reviewed by the authority and by interested parties. Adequate staff and resources must be allocated for monitoring to insure that relief under the agreement occurs. S.Rep. No. 96-249 at 71 (1979), reprinted in 1979 U.S.C.C.A.N. 381, 457. The four relevant monitoring provisions appear in Part VII of the Agreement. Under the first, Commerce is to monitor entries covered by the agreement utilizing, among other means, official import data and records to determine whether there have been imports that are inconsistent with the agreement. Second, Commerce is authorized to require — and the Brazilian Exporters have agreed to provide — confirmation that the price on any sale covered by the Agreement is not less than the established reference price. Third, Commerce is authorized to require — and the Brazilian Exporters have agreed to provide — information concerning each sale on computer disk, including each adjustment applicable to each sale. Finally, the Brazilian Exporters have agreed to permit onsite inspection of all information for purposes of verification, as Commerce deems necessary. Agreement, Part VII, 64 Fed.Reg. at 38,794. The Domestic Producers mount two attacks on the monitoring provisions of the Agreement. Their first challenge generally echoes their arguments concerning the Agreement’s provisions concerning the 15% limitation on dumping, outlined in section III.B above. Specifically, the Domestic Producers argue that, in the absence of specific provisions requiring the reporting of Brazilian cost and price data, it is impossible for Commerce to monitor compliance with the 15% limit on dumping. Plaintiffs’ Memo at 23. The Domestic Producers also criticize Commerce’s failure both to define the term “estimated normal value” in the Agreement and to explain how that value should be calculated. According to the Domestic Producers, the Brazilian Exporters’ “agreement” to comply with the 15% limit is meaningless in the absence of such a definition. Plaintiffs’ Memo at 23. The Government defends the decision not to define “estimated normal value” in the Agreement, emphasizing that the statute does not define the term either. The Government further contends that, should Commerce determine that it needs normal value data, it is entitled to obtain that data from the Brazilian Exporters under Part V of the Agreement (which requires the Brazilian Exporters to provide “all information that the Department determines is necessary” to verify compliance). Defendant’s Memo at 28-32, 52-53. The Domestic Producers’ second attack on the monitoring provisions of the Agreement targets the provisions for disclosure, comment on and verification of domestic sales prices. Dissatisfied with Parts VI and VII.B. of the Agreement, which provide that Commerce “may” make information available to them, the Domestic Producers assert that “[fjundamental fairness and public interest” instead require that any Agreement expressly permit their participation in the monitoring and enforcement process. Plaintiffs’ Memo at 24. The Government dismisses the Domestic Producers’ argument out of hand, pointing to the language of the statute, which.requires the practicability of “effective monitoring of the agreement by the United States. ” Defendant’s Memo at 53-54 (citing 19 U.S.C. § 1673c(d)(2)) (emphasis in the original). Moreover, the Government notes, the Domestic Producers are free to enhance their “participation in the enforcement and monitoring of Brazilian exports” by requesting an administrative review. Defendant’s Memo at 54 (quoting Plaintiffs Memo at 24, which actually refers to “Brazilian exporters”). Although the requirement for practicable, effective monitoring is common to both types of suspension agreements, it seems clear that the nature of the monitoring required to ensure compliance might vary depending on whether the agreement provides for the complete cessation of exports or the total elimination of dumping (a subsection (b) agreement) or merely the elimination of its injurious effects (a subsection (c) agreement). Thus, affording the Domestic Producers in this case the opportunity to review and comment on the monitoring provisions of the proposed subsection (b) agreement did not obviate the need to give them the same opportunity to review and comment on the monitoring provisions in any subsection (c) agreement that Commerce decided to consider as an alternative. It may be that the monitoring provisions in the Agreement at issue here — e.g., the parroting of the 15% limit and the “catch all” provision — are a product of the timing of Commerce’s decision to substitute a subsection (c) agreement for the proposed subsection (b) agreement, and the absence of input from the Domestic Producers. The Domestic Producers point out, for example, that the proposed subsection (b) agreement required the Brazilian Exporters to provide detailed home market information to permit Commerce to verify that dumping would be eliminated. Plaintiffs’ Memo at 11; Proposed Agreement, Part V. Indeed, as the Domestic Producers note, Commerce has routinely included requirements for the systematic reporting of home market sales and cost data in subsection (b) agreements. Reply Memo at 10. Assuming that Commerce on remand does not abandon the concept of using a suspension agreement in this case, Commerce will have the opportunity — if it wishes — to explore the possibility of another suspension agreement with different monitoring requirements that may be more responsive to the Domestic Producers’ concerns. At a minimum, the Domestic Producers will have an opportunity to make their case on monitoring requirements directly to Commerce, and Commerce will have to confront the Domestic Producers’ comments and build a proper administrative record on its interpretation of the monitoring provisions of the statute and their application to the facts of this case. D. Extraordinary Circumstances Subsection (c) agreements are limited to cases involving “extraordinary circumstances”- — that is, “circumstances in which- — (i) suspension of an investigation will be more beneficial to the domestic industry than continuation of the investigation, and (ii) the investigation is complex.” 19 U.S.C. § 1673c(c)(l), (c)(2). The Domestic Producers contest both parts of Commerce’s determination that extraordinary circumstances are present in this case. 1. Whether the Agreement Is More Beneficial to the Domestic Industry than Continuation of the Investigation As a threshold matter, the Domestic Producers contend that allowing any dumping to continue “means that the Agreement is not more beneficial” to the domestic industry than would be an anti-dumping order. Plaintiffs’ Memo at 12 (emphasis in the original). But, as both the Government and the Brazilian Exporters note, that proposition cannot be squared with the very existence of subsection (c), which' — by definition — contemplates suspension agreements that permit a limited amount of dumping to continue. Defendant’s Memo at 38-39; Defendant-Inter-venors’ Memo at 13-15. Commerce’s determination in this ease that suspension is more beneficial to the domestic industry than continuation of the investigation rests on its findings that the Agreement provides greater relief and greater certainty than would an antidump-ing order. Defendant’s Memo at 35 (citing Extraordinary Circumstances Memo). In making those findings, Commerce relied solely on the analysis set out in its Public Interest Memorandum. See Defendant’s Memo at 35 n. 25 (citing Public Interest Memo). According to Commerce, the Agreement affords the Domestic Producers greater relief than an antidumping duty order because it protects them from “future exchange rate-driven surges of Brazilian hot-rolled steel.” Defendant’s Memo at 35 0citing Public Interest Memo). But the Domestic Producers dispute the fundamental premise of Commerce’s analysis. The Domestic Producers maintain that the surges of Brazilian steel are attributable not to exchange rate fluctuations, but rather to excess production in Brazil and the unavailability of other markets for the steel. Tr. at 10, 16. In short, the Domestic Producers charge, Commerce purports to have been trying to solve a problem that does not exist. Pressed for record evidence that exchange rate fluctuations result in surges of Brazilian steel, the Government argues that Commerce properly made “logical assumptions and extrapolations” flowing from petitioners’ comments on the proposed subsection (b) agreement, which referred to “the sudden decrease in the value of the real that occurred during the first few months of 1999.” Defendant’s Memo at 36 (citing Letter from Dewey/Skad-den/Schagrin to Commerce) (June 28, 1999) (P.R. Doc. No. 270) at 10-12 and Matsushita Elec. Indus. Co. v. United States, 750 F.2d 927, 933 (Fed.Cir.1984). But, as the Domestic Producers note, the comment on which Commerce relies does not suggest that surges in Brazilian steel result from exchange rate fluctuations. Rather, the comment is a complaint about the exchange rate mechanism in the proposed subsection (b) agreement, advocating greater flexibility in light of exchange rate fluctuations (particularly the sudden devaluation of the Brazilian currency, the real ). Tr. at 94-95. That is indeed a slender reed on which to base “assumptions and extrapolations” concerning the causal relationship (if any) between exchange rate fluctuations and dumping. “[Assumptions and extrapolations” are also the basis for Commerce’s finding that the Agreement provides greater certainty — i.e., “a set level of relief that would be unavailable pursuant to an antidumping duty order” — and is thus more beneficial to the domestic industry. Defendant’s Memo at 36-37 (citing Public Interest Memo). The Government maintains that, under an antidumping duty order, “the foreign signatories would be free to set their prices below the prevailing U.S. market price.” Defendant’s Memo at 37. In contrast, under the Agreement, they have agreed not to sell below reference prices. Defendant’s Memo at 37; Agreement, Part IV, 64 Fed.Reg. at 38,794. See also Defendant-Intervenors’ Memo at 17-20. But there is a certain unassailable logic to the Domestic Producers’ observation that certainty is not, in and of itself, a virtue — that is, that certainty is not al-ivays better than uncertainty. See Tr. at 14-15. Moreover, any suspension agreement will, by definition, provide certainty. Thus, if mere certainty suffices to make a suspension agreement “more beneficial” to the domestic industry than continuation of an investigation, a suspension agreement would be permissible in any antidumping proceeding. Clearly, that was not the intent of Congress. See S.Rep. No. 96-249 at 71, reprinted in 1979 U.S.C.C.A.N. 381, 457 (“suspension is an unusual action which should not become the normal means for disposing of cases”). Subsection (c) agreements, in particular, are reserved for cases involving “extraordinary circumstances.” 19 U.S.C. § 1673c(c)(2); S.Rep. No. 96-249 at 71, reprinted in 1979 U.S.C.C.A.N. 381, 457 (subsection (c) agreements to be accepted only “rarely”). In short, Commerce cannot logically rely on a fact that is true with respect to any investigation as a basis for its determination that a particular case involves extraordinary circumstances. The same criticism can be leveled at Commerce’s finding that the Agreement is “more beneficial” because it provides “a set level of relief’ compared to an anti-dumping duty order. See Tr. at 14-15 (■referring to Public Interest Memo). Commerce’s statement simply proves too much. It is true as to every antidumping duty order, because duties always are assessed retroactively (and thus are not “set”). If a “set level of relief’ were enough to make a suspension agreement “more beneficial” than an order, the “more beneficial” test would be read out of the statute, because it would be met in every antidumping case. Tr. at 14-15. As the House Committee on Ways and Means emphasized: [T]he [“more beneficial”] provision is not intended to be so general as to be meaningless. For example, the expenses saved because of prompt settlement of a case or the certainty of prompt relief may make settlement more beneficial than continuation of the investigation. However, every suspension of an investigation results in prompt, certain relief and reduced expenses. The Committee does not intend that for this reason every agreement be deemed more beneficial to domestic industry. H.R.Rep. No. 96-317 at 65 (1979) (emphasis supplied). ' Finally, the Domestic Producers condemn as pure post hoc rationalization by counsel the Government’s assertion that, under an antidumping order, “the foreign signatories would be free to set their prices below the prevailing U.S. market price.” The Domestic Producers argue that Commerce itself made no such finding, and that there is nothing in the administrative record which would support such a finding. Tr. at 15-16 (referring to Defendant’s Memo at 37). In addition to the benefits alleged by Commerce as bases for its “more beneficial” determination, the Brazilian Exporters tout certain other asserted benefits to the domestic industry (see generally Defendant-Intervenors’ Memo at 20-24), which the Domestic Producers basically dismiss. See generally Reply Memo at 17-18. For example, while the Brazilian Exporters emphasize that the Agreement limits the types of Brazilian hot-rolled steel products that can be exported to the United States to those for which reference prices have been fixed, the Domestic Producers contend that Commerce has “already expanded the coverage of the Agreement to [include] ... additional products” and likely will continue to do so “quite possibly at reference prices which are well below U.S. market levels.” Reply Memo at 17-18 (footnotes omitted). Compare Defendant-Intervenors’ Memo at 20-22. The Domestic Producers also contend that “there have been several instances in which the Brazilian signatories appear to have sold hot-rolled products in the U.S. market at prices below the reference prices stipulated in the Agreement.” Reply Memo at 18. Whatever the merits of the Brazilian Exporters’ arguments, it is well-settled law that “[t]he grounds upon which an administrative order must be judged are those upon which the record discloses that its action was based.” Securities and Exchange Comm’n v. Chenery Corp., 318 U.S. 80, 87, 63 S.Ct. 454, 87 L.Ed. 626 (1943). Commerce’s “more beneficial” determination therefore could not be sustained on the basis of asserted benefits on which it did not rely. In any event, as discussed in section III.A above, the existing administrative record on the “more beneficial” issue—like the other issues in this case—is insufficient to serve as a basis for judicial review. The Domestic Producers simply had no opportunity to build a record on the “more beneficial” requirement. Because the Proposed Agreement was a subsection (b) agreement, the “more beneficial” requirement was irrelevant; accordingly, the Domestic Producers did not address it in their comments. And they had no notice of the subsection (c) Agreement, and thus no opportunity to comment on whether or not it was more beneficial than continuation of the investigation. For the reasons set forth in section III.A, this matter is being remanded to remedy Commerce’s failure to comply with the notice, comment and consultation requirements of the suspension agreement statute. All parties then will have the opportunity to develop a proper administrative record. For example, assuming that Commerce issues for comment a subsection (c) agreement, the Domestic Producers may seek to put on the record for Commerce’s consideration any information they may have concerning the Brazilian Exporters’ compliance to date with the terms of the existing Agreement. Commerce, for its part, will have ample opportunity to make a proper factual record, to preclude (or at least minimize) the need to rely on “assumptions and extrapolations,” and to avoid the risk of being accused of engaging in “post hoc rationalization.” Commerce also may, if it wishes, adopt— or reject — any or all of the Brazilian Exporters’ analysis of the existing Agreement’s benefits to the domestic industry. Most importantly, Commerce will have the opportunity to detail precisely why its agreement is “more beneficial” to the domestic industry than an antidumping order — even though a substantial majority of the domestic industry believes that it is not. In doing so, Commerce should bear in mind the history and intent of the “more beneficial” requirement. Subsection (c) agreements are intended for use only in rare cases such as those where “the value of settling the case quickly or the certainty of prompt relief the settlement provides” outweighs the benefits of continuing the investigation. See, e.g., Statements of Administrative Action for Trade Agreements Act of 1979, H.R. Doc. No. 96-153, Part II at 420 (1979). As the Domestic Producers note: A suspension agreement is a tool to be used in lieu of completing an investigation, without taking the time and resource-consuming burdens of completing an extraordinarily complex investigation. In this case, the Department had completed its investigation by the time that it signed the Agreement, and issued its final determination simultaneously with the Agreement. In fact, the final determination identified, addressed at length, and resolved all issues. The Department did not use the suspension agreement in the manner intended by ... the Administration ..., or by the Congress. Plaintiffs’ Memo at 13. Moreover, while it is true that Congress entrusted the “more beneficial” determination to Commerce, and did not expressly accord the domestic industry a veto power, any decisionmaker should be chary of concluding that a particular course of action is in the best interests of an industry that generally opposes it. As Senator Heinz observed during floor debate on the suspension agreement statute, “I would find it very difficult to believe a judgment that the domestic industry would benefit more from a suspension than a completed investigation if that industry had expressed its opposition to such an action.” 125 Cong. Rec. 20,168 (July 23, 1979). 2. Whether the Investigation Was Complex Even if Commerce properly concluded that the Agreement is more beneficial to the domestic industry than continuation of the investigation, that would not be enough to constitute the “extraordinary circumstances” required to justify a subsection (c) agreement. Commerce must also determine that the investigation is “complex.” 19 U.S.C. § 1673c(c)(2)(A)(ii). The Domestic Producers vigorously dispute that determination in this case. For purposes of the suspension agreement statute, an investigation is deemed “complex” if “(i) there are a large number of transactions to be investigated or adjustments to be considered, (ii) the issues raised are novel, or (iii) the number of firms involved is large.” 19 U.S.C. § 1673c(e)(2)(b). Commerce relied on the first two criteria in determining that the investigation here at issue was complex. See Extraordinary Circumstances Memo. All parties agree that the investigation involved numerous transactions and adjustments. Plaintiffs’ Memo at 15; Defendant’s Memo at 40-41; Defendant-In-tervenor’s Memo at 26. However, the Domestic Producers contend that, in the context of the statute’s “extraordinary circumstances” requirement, “large” means large relative to other investigations, “i.e., an unusual number which is so burdensome and which renders the investigation so complex that the domestic industry is better served by an agreement suspending the investigation.” Plaintiffs’ Memo at 15; Reply Memo at 22. The statute itself is silent on this point. While “large” is a relative term, it is unclear whether it should be interpreted to mean a large number of transactions and adjustments relative to other antidumping duty investigations generally, or large in relation to other antidumping duty investigations involving the steel industry, or large in some other way. In light of that ambiguity, analysis would normally proceed to the second prong of the Chevron /Christensen test, to evaluate the reasonableness or persuasiveness of Commerce’s interpretation of the phrase “large number.” However, Commerce here simply asserted that the investigation “involves a large number of transactions and multiple adjustments.” Extraordinary Circumstances Memo. Absent any explanation of Commerce’s interpretation of the statute, it would be impossible to determine whether “the agency has exercised a reasoned discretion, with reasons that do not deviate from or ignore the ascertainable legislative intent.” See Greater Boston Television Corp. v. FCC, 444 F.2d 841, 850 (D.C.Cir.1970). For the reasons discussed in section III.A above, this entire case is being remanded to allow the parties to develop a proper administrative record. Assuming that, on remand, Commerce issues this or some other proposed subsection (c) agreement for comment, the Domestic Producers will be able to make their case on this criterion directly to the agency (for the first time, since it is unique to subsection (c) agreements); and Commerce will have the opportunity — with the benefit of the Domestic Producers’ comments — to explain the rationale for whatever decision it may make. As an alternative ground for its determination that the investigation here was “complex,” Commerce found that the investigation involved “novel issues.” Extraordinary Circumstances Memo. The Government argues that the “affiliation/collapsing analysis” in the investigation was novel because certain aspects of the analysis must be resolved on a case-by-case basis. Defendant’s Memo at 41-42. While it may be true that “the unique facts before it required [Commerce] to visit the