Rubberflex Sdn. Bhd. v. United States

59 F. Supp. 2d 1338, 23 Ct. Int'l Trade 461, 23 C.I.T. 461, 21 I.T.R.D. (BNA) 1613, 1999 Ct. Intl. Trade LEXIS 63

• Court: United States Court of International Trade
• Decided: July 23, 1999
• Docket: Slip Op. 99-68; Court 97-12-02180
• Status: Published

Opinion

OPINION

GOLDBERG, Judge.

In this action, the Court reviews challenges to the Department of Commerce’s (“Commerce”) final determination for the second administrative review of an anti-dumping duty order covering extruded rubber thread from Malaysia. Extruded Rubber Thread From Malaysia; Final Results of Antidumping Duty Administrative Review, 62 Fed.Reg. 62,547 (Nov. 24, 1997) {“Final Results”). Plaintiff Rubberflex Sdn. Bhd. (“Rubberflex”) argues that (1) the manner in which Commerce conducted the second administrative review, particularly verification, prejudiced Rubberflex; (2) Commerce was able to verify most of Rubberflex’s responses and therefore should not have used total “best information available” (“BIA”) to assign Rubberflex a dumping margin; and (8) the duty rate of 29.83% assigned to Rubberflex in the second review was arbitrary in light of the 20.38% rate assigned to Rubberflex in the third review. The Court addresses these arguments in Section I.

Plaintiff Rubfil Sdn. Bhd. (“Rubfil”) argues that, in calculating the net price for Rubfil’s U.S. sales, Commerce erroneously failed to convert Rubfil’s ocean freight expenses from Malaysian ringgits to U.S. dollars. The Court addresses this argument in Section II.

The Court exercises jurisdiction over this matter pursuant to 28 U.S.C. § 1581(c) (1994). The Court remands the Final Results.

I.

RUBBERFLEX

A. Background

1. Summary of the Second Administrative Review

On October 7, 1992, Commerce published an antidumping order covering extrud ed rubber thread from Malaysia. See Antidumping Duty Order and Amendment of Final Determination of Sales at Less Than Fair Value: Extruded Rubber Thread from Malaysia, 57 Fed.Reg. 46,-150 (Oct. 7, 1992). Rubberflex was one of the two respondents in the underlying an-tidumping investigation. See id. at 46, 151. During the second anniversary month of the order, Rubberflex, three other Malaysian producers and exporters of extruded rubber thread, and petitioner North American Rubber Thread requested an administrative review of the order, which Commerce agreed to undertake. See Initiation of Antidumping and Countervailing Duty Administration Reviews, 59 Fed.Reg. 56,459 (Nov. 14, 1994) (covering the period from October 1, 1993 through September 30,1994). 1

Commerce issued Rubberflex a questionnaire for the second review on or around February 8, 1995. 2 See Letter from APO Specialist to White & Case Transmitting Questionnaire (Feb. 8, 1995), Pub. Admin. R., Fiche 2, Frame 25. On April 17, 1995, Rubberflex timely submitted its response to Commerce’s questionnaire. See Letter from White & Case to Secretary of Commerce (Apr. 17, 1995), Conf. Admin. R., Fiche 30—37, Frame 1.

On or around September 3, 1996, Commerce issued Rubberflex a supplemental questionnaire 3 for the second administrative review. 4 . Rubberflex requested an extension of time in which to file its response to the supplemental questionnaire. See Letter from White & Case to Secretary of Commerce Requesting Extension to File Supplemental Questionnaire Response (Sept. 11, 1996), Pub. Admin. R., Fiche 19, Frames 69-71. Commerce granted the request and allowed Rubberflex two additional business days to file. See Letter from Program Manager/IA to White & Case Granting Extension of Time (Sept. 13, 1996), Pub. Admin. R., Fiche 19, Frame 72. On September 17, 1996, Rub-berflex timely filed its response to Commerce’s supplemental questionnaire. See Letter from White & Case to Secretary of Commerce Transmitting Rubberflex Supplemental Questionnaire Response (Sept. 17, 1996), Pub. Admin. R., Fiche 20, Frame 1, Conf. Admin. R., Fiche 51-52, Frame 1.

On Wednesday, September 18, 1996, the day after receiving Rubberflex’s supplemental response, Commerce issued Rub-berflex a verification outline. See Letter from Program Manager/IA to White & Case Transmitting Verification Outline for Rubberflex (Sept. 18, 1996) (“Verification Outline”), Pub. Admin. R., Fiche 20, Frame 67. And on Monday, September 23, 1996, Commerce began verification of Rubberflex’s sales and cost data in Malaysia. Commerce completed verification on October 5, 1996. Commerce also conducted verification of U.S. sales data in North Carolina at Rubberflex’s affiliated reseller, Flexfll Corporation, from October 16 through October 18,1996.

On December 12, 1996, Commerce issued an internal memorandum concluding that Rubberflex had failed verification. See Memo From Office DIR/IA to DAS/IA; Rubberflex — Reasons for Failed Verification — Recommendation Memo for Using Facts Available (Dec. 12, 1996), Pub. Admin. R., Fiche 25, Frame 30. In response to the memo, Rubberflex complained that the case handlers had confused issues between the second and the third reviews in making their determination. Consequently, Commerce agreed to separate its findings by review period. See Verification Memo, Pub. Admin. R., Fiche 25, Frame 62.

On February 13, 1997, Commerce published its preliminary determination for the second administrative review. Because Commerce “found that responses provided by Rubberflex could not be verified,” it resorted to total BIA and assigned Rubberflex a dumping margin of 29.83%. 5 Notice of Preliminary Results of Anti-dumping Duty Administrative Review: Extruded Rubber Thread from Malaysia, 62 Fed.Reg. 6758, 6759 (Feb. 13, 1997) (‘Preliminary Results”).

On March 10, 1997, Rubberflex submitted a case brief challenging various aspects of the Preliminary Results. See Brief from White & Case to Secretary of Commerce Regarding Rubberflex (Mar. 10, 1997) ("Case Brief’), Pub. Admin. R., Fiche 27, Frame 1, Conf. Admin. R., Fiche 61, Frame 1. Rubberflex argued that (1) despite Commerce’s claims to the contrary, Commerce was able to verify Rub-berflex’s responses at verification; (2) Commerce had no legal basis for using total BIA to determine Rubberflex’s dumping margin and, at most, should have used BIA for specific responses (“partial BIA”); and (3) the preliminary results, the verification report, and other Commerce documents contained numerous errors and inconsistencies. See id. passim. Rubber-flex emphasized in its Case Brief that it had cooperated to the best of its ability throughout a difficult and disorganized review process, and argued that Commerce should bear primary responsibility for any problems that arose in connection with verification. See Pub. Admin. R., Fiche 27, Frames 42-45, Conf. Admin. R., Fiche 61, Frames 42-45.

To further air its concerns, Rubberflex scheduled a meeting with the Commerce officials responsible for the second administrative review. 6 The meeting was held on April 14, 1997, and included Mr. Jeffrey P. Bialos, Principle Deputy Assistant Secretary for Import Administration, and the case handlers responsible for the second and third reviews. See Pl.’s Br. in Supp. of Mot. For J. Upon the Agency R..(“Pl.’s Br.”), at 8.

On November 12, 1997, Commerce issued a memo to file that analyzed the business proprietary information in Rub-berflex’s Case Brief. See Memo from Analyst/IA to File: Final Results of Administrative Review (Nov. 12, 1997), Pub. Admin. R., Fiche 28, Frame 34, Conf. Admin. R., Fiche 62, Frame 18. A portion of the analysis compared Rubber-flex’s questionnaire responses to the revised information Rubberflex attempted to submit at verification. Finally, on November 24, 1997, Commerce issued the Final Results. As in the Preliminary Results, Commerce relied on total BIA and assigned Rubberflex a margin of 29.88%. See Final Results, 62 Fed.Reg. at 62,548, 62,558.

2. Rubberflex’s Allegations

Rubberflex offers, in addition, its own view of the facts relevant to this suit. Specifically, Rubberflex alleges that (1) Commerce conducted the entire second administrative review in an extremely disorganized, irresponsible, and prejudicial manner; (2) Commerce denied Rubber-flex’s request to reschedule verification to a mutually convenient time; (3) Commerce gave Rubberflex only two business days to prepare for verification; (4) at verification, Commerce tacitly accepted Rubberflex’s proposal to submit all corrections at the conclusion of verification and then refused to accept such submissions; (5) Commerce case handlers conducted verification according to an undisclosed methodology; (6) Commerce case handlers were not adequately prepared for verification, thereby causing delay; and that (7) despite such delays, Commerce case handlers refused to work past 5:00 p.m. or on weekends.

Rubberflex first notes that after initiating the second review and issuing the questionnaire, Commerce did nothing for 18 months; then, when Commerce resumed work on the review, it compressed all remaining activities for two periods of review into a 45-day period. By Rubber-flex’s account, it contacted the case handler assigned to the second review several times during the 18-month period of inactivity to determine when Commerce planned to issue a supplemental questionnaire and conduct verification. Rubberflex alleges it was told that “it was unclear who would be handling the review and, therefore, no work was being performed on the Responses.” Case Brief Pub. Admin. R., Fiche 27, Frame 42, Conf. Admin. R., Fiche 61, Frame 42.

According to Rubberflex, the case handler finally contacted Rubberflex on or around August 30, 1996, one year and nine months after the initiation of the second review, to inform it that verification for both the second and third reviews would be conducted in late September and early October. See Case Brief, Pub. Admin. R., Fiche 27, Frame 43, Conf. Admin. R., Fiche 61, Frame 43. Rubberflex claims it was not even asked whether its counsel would be available during that time frame or whether those dates were feasible for the company. See Pl.’s Br., at 5. At this point, Rubberflex claims it informed the case handler that the company had only two accountants and therefore would likely be unable to prepare supplemental responses for both reviews and prepare for verification in the time frame allotted by Commerce. Rubberflex alleges that to foreclose this potential problem it requested a two-week postponement of verification, but that Commerce denied the request. 7 See Case Brief, Pub. Admin. R., Fiche 27, Frame 43, Conf. Admin. R., Fiche 61, Frame 43.

Rubberflex also notes that Commerce gave it only two business days to prepare for verification. See Case Brief, Pub. Admin. R., Fiche 27, Frame 43-44, Conf. Admin. R., Fiche 61, Frame 43-44. As the record shows, Rubberflex submitted its supplemental questionnaire response on September 17, 1996, Commerce issued Rubberflex a verification outline on Wednesday, September 18, and Commerce began verification on Monday, September 23, 1996. As a result, Rubberflex claims it had only Thursday, September 19 and Friday, September 20 to prepare for verification.

While preparing for verification, Rub-berflex, by its own account, discovered errors in its questionnaire responses. Rubberflex claims it could not prepare corrected worksheets by the start of verification, however, because it only had two business days to prepare. To address the situation, Rubberflex alleges it informed Commerce of the errors at the start of verification, and proposed to prepare corrected worksheets each night after the case handlers left and then present them as the need arose during verification. See Pl.’s Br., at 6; see generally Case Brief, Pub. Admin. R., Fiche 27, Frames 40-41, Conf. Admin. R., Fiche 61, Frames 40-41. In addition, Rubberflex alleges that it offered to submit one collection of revisions to the case handlers at the close of verification. See Case Brief, Pub. Admin. R., Fiche 27, Frame 49, Conf. Admin. R., Fiche 61, Frame 49; Pl.’s Br., at 6-7. According to Rubberflex, the case handlers did not object, and consequently Rubberflex proceeded with that understanding. Rubberflex alleges, however, that despite this tacit agreement, the case handlers refused to accept Rubberflex’s corrected worksheets on the last day of verification. 8 See id.

Rubberflex also asserts that the case handlers conducted verification according to “their own, personal methodology” instead of following the outline furnished by Commerce. PL’s Br., at 7. The consequence of this, as the Court understands it, is that Rubberflex prepared worksheets prior to the start of verification in the order of the verification outline; yet, be cause the case handler’s new methodology-changed the order of verification, the process was held up a number of times -while Rubberflex prepared worksheets it had anticipated needing later on in the process. Tr. of Oral Argument, at 19-20.

Finally, Rubberflex asserts that Commerce’s case handlers were not familiar with Rubberflex’s responses, particularly those relating to costs, at the start of verification. Pl.’s Br., at 14. Rubberflex believes that, as a result, it wasted a great deal of time during verification familiarizing the case handlers with its data. Rub-berflex complains that in spite of the wasted time, the case handlers refused to work past 5:00 p.m. or on the weekends. See Pl.’s Br., at 7,14.

B. Standard of Review

The Court will sustain Commerce’s Final Results if they are supported by substantial evidence on the record and are otherwise in accordance with law. See 19 U.S.C. § 1516a(b)(1)(B) (1994).

By law, Commerce is required to “verify all information relied upon in making ... a review and determination.” 19 U.S.C. § 1677e(b)(3) (1988). Outside of this mandate, however, Congress has- not prescribed any verification procedures. See Micron Tech., Inc. v. United States, — Fed. Cir. (T) -, -, 117 F.3d 1386, 1394 (1997). Instead, “Congress has implicitly delegated to Commerce the latitude to derive verification procedures ad hoc.” Id. at-, 117 F.3d at 1396. At the same time, Commerce has not defined the requirements or procedures associated with verification in any formal way. See id. at-, 117 F.3d at 1395. Thus, when reviewing the procedures' Commerce uses at verification, the Court does not look to “previously-set standards.” Id. at -, 117 F.3d at 1396. Rather, it “reviewfs] verification procedures employed by Commerce in an investigation for abuse of discretion.” Id. ■

The Federal Circuit and this court have routinely sustained Commerce’s administration of reviews and verifications as within its discretion. 9 In spite of, or perhaps because of, the wide latitude given Commerce in conducting reviews and verifications, however, “[t]he Court must be ever vigilant of abuse of discretion by the agency.” Wheatland Tube Corp. v. United States, 17 CIT 1230, 1236, 841 F.Supp. 1222, 1227 (1993).

C. Discussion

Rubberflex argues three major points in its papers. It challenges (1) the manner in which Commerce conducted its review and verification; (2) Commerce’s use of totál BIA; and (3) the duty rate assigned to Rubberflex. In the following discussion, the Court first examines Commerce’s administration of the second administrative review and concludes that Commerce abused its discretion. Commerce’s conduct during the review, specifically its issuance of a verification outline to Rubberflex two business days prior to verification, effectively prejudiced Rubber-flex. Commerce’s abuse of discretion ser iously compromises the validity of its verification results in this case, and impairs this Court’s ability conduct a meaningful review. Accordingly, the Court takes the extraordinary step of remanding this case to Commerce with instructions to repeat verification. Given the remedy ordered by the Court, there is no need to reach Rubberflex’s second and third issues at this time.

1. Evidence on the record

The Court’s statutory mandate dictates that it review the “evidence on the record.” 19 U.S.C. § 1516a (b)(1)(B)(i) (1994); see also Koyo Seiko Co., Ltd. v. United States, 21 CIT-, -, 955 F.Supp. 1532, 1544 n. 9 (1997) (“The Court’s review of a final determination is limited to a review of the administrative record.”). Accordingly, the Court in this case may address Rubber-flex’s allegations only if they are part of the administrative record. Thus, as an initial matter, the Court must ensure that the evidence Rubberflex presents to support its claims is part of the administrative record.

Generally, the administrative record includes all information presented to or obtained by the Secretary during the review, copies of all determinations and notices published in Federal Register, and transcripts or records of all conferences, hearings, and ex parte meetings. See 19 U.S.C. § 1516a (b)(2)(A) (1994). Thus, the administrative record in this case includes, for example, Commerce’s questionnaires, Rubberflex’s responses, and verification exhibits. Importantly, Rubberflex’s Case Brief is also part of the administrative record. The Case Brief is significant because it was the company’s first formal, written opportunity to contest Commerce’s preliminary determination. Yet, Rubber-flex’s Case Brief does not reflect all of the allegations it currently levies against Commerce. In particular, Rubberflex does not allege in its Case Brief that (1) the case handlers conducted verification according to an undisclosed methodology, (2) the case handlers were not prepared for verification, or (3) the case handlers refused to work past 5:00 p.m. or on the weekends. And, no other evidence on the record documents those claims. Therefore, because the Court’s review is limited to the administrative record, there is no basis to address these three allegations.

In contrast, however, Rubberflex does assert in its Case Brief that Commerce (1) conducted the second administrative review in a negligent manner, (2) refused to reschedule verification to a mutually convenient time, (3) allowed Rubberflex only two business days to prepare for verification, and (4) contravened the agreement for submission of corrections. And, Commerce responded to all but one of these allegations in its Final Results. Thus, in reviewing whether Commerce’s conduct during the review and verification prejudiced Rubberflex, the Court considers these four allegations.

2. Commerce’s issuance of a verification outline two business days pri- or to the start of verification constitutes an abuse of discretion

The Court finds that Commerce’s conduct in this case constitutes an abuse of discretion. Commerce’s tardy issuance of a verification outline precluded Rubberflex from having a meaningful opportunity to participate in the review process. The effects of this action reverberated through verification, and thus the Court cannot sustain Commerce’s Final Results.

As a general matter, Commerce justifies its administration of the second review and verification as consistent with its discretionary authority. In response to Rubber-flex’s numerous allegations, Commerce stated that

Rubberflex was given sufficient notice of the timing of verification, and the Department followed the same standard procedures, and issued a standard verification outline which was substantially similar for the verification of information in both the 1993-1994 and 1994-1995 reviews. These procedures were similar to those followed in the original investí- gation, when Rubberflex underwent verification. Thus, there is little evidence that the Department’s conduct of the case placed an unreasonable burden on Rubberflex. Rather, in this case, as in virtually every case the Department conducts, the burden on respondents is to provide accurate and timely data which can be verified. To the greatest extent possible, the Department strives to be flexible with deadlines for respondents; ultimately, however, it is respondents’ responsibility to meet this burden. Nevertheless, we took into account Rubberflex’s level of cooperation in this case in our selection of the appropriate BIA rate for Rubberflex’s dumping margin.

Final Results, 62 Fed.Reg. at 62,555. The Court finds Commerce’s attempt to rationalize its behavior unacceptable.

Commerce’s statutory mandate is to determine dumping margins as accurately as possible. See Rhone Poulenc, Inc. v. United States, 8 Fed. Cir. (T) 61, 67, 899 F.2d 1185, 1191 (1990). To determine accurate margins, Commerce must gather accurate data from respondents. And to do that, Commerce must give respondents a reasonable opportunity to participate in the review and verification process. See Bowe-Passat v. United States, 17 CIT 335, 339 (1993) (noting that “[t]he review process is [supposed to be] bilateral and interactive”). This includes giving respondents advance notice of on-site activities, adequate time to complete requests for information, and predictable scheduling of the review and verification. Thus, while Commerce generally has discretion to schedule a review and allocate resources to a review as it sees fit, it must do so within the confines of its statutory mandate. When its actions compromise the accuracy of dumping margins, then it has abused its discretion.

In particular, Commerce has an obligation to issue a verification outline a reasonable amount of time prior to the start of verification. An examination of Rubberflex’s Case Brief, Pub. Admin. R., Fiche 27, Frames 43-44, Conf. Admin. R., Fiche 61, Frames 43^44, as well as other evidence in the record, demonstrates that Commerce issued a verification outline to Rubberflex on Wednesday, September 18, and began verification on Monday, September 23, two business days (Thursday, September 19 and Friday, September 20) later.

In this case, two business days was not a reasonable amount of time in advance of verification for Commerce to issue an outline, and the verification outline itself demonstrates why. First, the outline requires respondents to report errors prior to verification. Specifically, the outline instructs Rubberflex to “[a]dvise verifiers of any data or informational errors found during preparation for verification prior to the start of verification.” 10 Verification Outline, Pub. Admin. R., Fiche 20, Frame 71 (emphasis in original). Commerce’s rationale for requiring a respondent to report errors prior to verification is obvious. The policy benefits both Commerce and respondent by ensuring that both parties are fully prepared for verification and that Commerce has an opportunity to review and digest changes to a respondent’s data before verification. The Court does not dispute the wisdom underlying this policy.

Once Commerce delivers this exhortation to respondents, however, it must give them a reasonable opportunity to comply. In this case, Commerce was well aware that counsel for Rubberflex were American attorneys based out of Washington, D.C. Traveling from Washington to Kuala Lumpur, Malaysia alone takes at least one full day of travel. And yet, by issuing Rubberflex a verification outline two business days prior to verification, Commerce effectively gave Rubberflex only two business days to review the verification outlines and check for errors in verification worksheets and exhibits. Therefore, it is at a minimum disingenuous for Commerce, on the facts of this case, to require Rub-berflex to advise it of errors “prior to the start of verification” and then begin verification almost immediately thereafter.

Indeed, when Commerce fails to issue a verification outline until the last moment, it is entirely unreasonable to expect a respondent to report errors found “during preparation for verification prior to the start of verification.” In this case, one reason Commerce refused to accept Rub-berflex’s corrected worksheets at verification is because “Rubberflex failed to provide written disclosure of changes made to its questionnaire response on the first day of verification,” but instead “provided [corrections] piecemeal, and late in the verification exercise.” Final Results, 62 Fed. Reg. at 62,552. Yet Rubberflex’s failure to provide all corrections at the start of verification resulted not from its own inadequacy, but from Commerce’s dilatory behavior in issuing the verification outline two business days prior to the start of verification. Thus, because Commerce “took the lateness of the information [Rub-berflex’s corrections] into account in concluding that Rubberflex’s response could not be verified,” Def.’s Br., at 42, the Court cannot sustain Commerce’s determination that Rubberflex failed verification.

The second reason Commerce must issue a verification outline a reasonable amount of time prior to verification is that the outline provides critical information so that a respondent may be, in Commerce’s words, fully “prepare[d] for the verification in advance.” Def.’s Br., at 23. As a general matter, the outline identifies the source documents required by Commerce, the subject matter areas to be covered during verification, the kinds of facilities Commerce wishes to tour, and the methods Commerce will employ to verify information. See, e.g., Pub. Admin. R., Fiche 20, Frame 79 (“The verification outline provides examples of the types of verification procedures which will be performed and general areas to be discussed.”). This general information is important because it defines the final scope and breadth of verification for a respondent. 11

More importantly, the verification outline also contains specific details without which a respondent cannot complete preparation for verification. For example, in this case, the cost of production/constructed value verification agenda Commerce issued Rubberflex “sets forth priorities of the items to be verified and specific products for which we will perform initial testing.” Verification Outline, Pub. Admin. R., Fiche 20, Frame 79 (emphasis added). And, the outline alerted Rubberflex to particular questionnaire responses Commerce wanted to cover during verification. For example, Commerce directed Rubberflex to “[e]xplain and document how you derived average purchase costs for packing materials as described in Exhibit B-8 of your April 15, 1996 QR [questionnaire response]. Demonstrate that you have not reused any packing materials.” Verification Outline, Pub. Admin. R., Fiche 20, Frame 75. Commerce also identified the home and U.S. market circumstance-of-sale adjustments it planned to analyze during verification, as well as specific portions of Rubberflex’s questionnaire response re lating thereto. See Verification Outline, Pub. Admin. R., Fiche 20, Frames 75, 78.

Importantly, the verification outline also provides a respondent its first glimpse at the transactions Commerce will “spot-check” at verification. For example, Appendix A of the outline Commerce issued Rubberflex lists the home market sales transactions pre-selected for verification and the supporting source documentation-required for verification. See Verification Outline, Pub. Admin. R., Fiche 20, Frame 73. Likewise, Appendix B listed the U.S. sales Commerce had pre-selected for verification. See Verification Outline, Pub. Admin. R., Fiche 20, Frame 76.

Commerce attempts to downplay the importance of both the verification outline and when it was issued in this case. To do so, it claims it followed “standard procedures, and issued [Rubberflex] a standard verification outline.” Final Results, 62 Fed.Reg. at 62,555. This argument is unavailing. First, as demonstrated above, the verification outline issued to Rubberflex contained not only “standard” verification guidance, but information specific to this verification. To be sure, if every outline Commerce issued was an exact copy of the last, there would be no point in issuing an outline. And, the Court doubts very much that Commerce would waste its efforts engaging in a meaningless exercise for every verification it conducts. 12 Moreover, Commerce has the authority to alter, and should alter, a so-called “standard” outline to address factors specific to a particular review period, respondent, or industry.

Commerce also maintains it did not act unreasonably because the verification procedures it used in the second review “were similar to those followed in the original investigation.” 13 Final Results, 62 Fed. Reg. at 62,555. This is simply unacceptable. Having chosen to provide an outline that creates obligations on a respondent’s part, Commerce is required to issue a verification outline a reasonable amount of time prior to the start of verification. Such obligation does not change simply because Commerce has conducted a prior review. First, each administrative review is a separate proceeding, see 19 U.S.C. § 1675(a)(1), and thus neither Commerce, nor a respondent, can rely on documents issued in a prior proceeding. Second, as shown, there are portions of the Verification Outline that Rubberflex could not have anticipated, even based on the original investigation.

Nonetheless, Commerce also argues that the compressed time frame in which it issued Rubberflex an outline and commenced verification had no effect on the outcome of verification because Rubberflex was given “sufficient notice of the timing of verification.” Id. Commerce’s theory is that Rubberflex submitted its original questionnaire response in April, 1995 and thus had over one year to prepare for verification. Def.’s Br., at 23. This misses the point. In April, 1995, Commerce had not yet issued Rubberflex a verification outline. And without having received the verification outline, Rubberflex could not have definitively known the scope of verification, specific responses on which Commerce would focus, or the transactions Commerce preselected, and therefore could not have prepared fully for verification.

Commerce’s claim that verification was not compromised by the tardy issuance of the outline is also belied by the administrative and logistical tasks that a respondent must perform prior to verification. For instance, Commerce requires respondents to make two copies of all relevant documents for the case handlers. See Verification Outline, Pub. Admin. R., Fiche 20, Frame 68. Clearly, it would be unreasonable and wasteful for a respondent to make copies of all. documents relating to costs, for example, and thus the verification outline is required to narrow the task. And, in this case, Rubberflex was required, “[flor each of the pre-selected sales,” to make “copies of the ... [supporting] documents with English translations ... [to] be made available for the analysts at the beginning of sales verification.” Verification Outline, Pub. Admin. R., Fiche 2Ó, Frame 74. Again, before receiving the outline a respondent cannot know which transactions Commerce has pre-selected, and therefore which documents to translate and copy. The Court imagines that translating and copying might take significant time, depending on the scope of verification and the technology available on-site, further justifying the finding that a verification outline must be issued a reasonable amount of time in advance of verification.

For all of the foregoing reasons, the Court holds that Commerce’s issuance of a verification outline to Rubberflex two days prior to the start of verification was an abuse of discretion.

3. Remedy

Commerce’s conduct, which constituted an abuse of discretion, “[d]id not promote cooperation or accuracy or reasonable disclosure by cooperating parties intended to result in realistic dumping determinations,” Bowe-Passat, 17 CIT at 343, and compromised the validity of the verification process from the outset. Thus, the Court takes the extraordinary step of ordering the parties to repeat verification.Cf . Wheatland, Tube Corp. v. United States, 17 CIT 1230, 1235-36, 841 F.Supp. 1222, 1227 (1993) (noting that an exercise of discretion will not- be sustained when it contravenes statutory objectives); Tehnoimportexport v. United States, 15 CIT 250, 259, 766 F.Supp.-1169; 1178 (1991) (noting' that a court may remand when an act is an abuse of discretion and “undermined the interests of justice”).

Accordingly, the Court orders the parties to confer and file a scheduling order within ten days of the date of publication of this opinion. That scheduling order shall list the dates on which (1) Commerce will issue Rubberflex a verification outline, (2) Commerce will conduct verification, and (3) Commerce will file remand results with .the Court. Remand results must be filed with the Court no later than ninety (90) days after the date of publication of this opinion, The Court emphasizes that it ■is allowing the parties more time than for the typical -remand and does not anticipate any requests for extensions.

■ The Court expects the parties to approach remand verification in a rational, reasonable manner, and in the spirit of cooperation and compromise. For example, the parties should confer and decide how to resolve the gaps in the administrative record, i.e., the missing questionnaires. And, the Court encourages the parties to expedite the verification process by exploring the possibility of stipulating to the reliability of selected data.

The best use of this Court’s resources is not to referee this dispute. The Court senses that verification was hindered by mistakes and acrimony on the part of both parties. The Court cannot predict what would have been- different had Commerce issued Rubberflex a verification outline a reasonable amount of time prior to verification, and had verification been conducted in a more cooperative and less contentious manner. And in remanding the case, the Court makes no judgment about the substantive outcome of the review. Commerce may well have been correct'in resorting to BIA and assigning Rubberflex a dumping margin of 29.83%. Or Rubberflex may be correct in asserting its margin should instead be less than 1%. See Pl.’s Br., at 3. The Court is confident, however, that Commerce did not arrive at the current margin in a fair and reasonable manner. Commerce assigned a dumping margin using BIA, and based its use of BIA on its finding that Rubberflex had failed verification. Yet, verification was prejudiced. Thus, the Court remands the case for Commerce to reach its determination in a fair and reasonable manner.

II.

RUBFIL

Rubfil argues that Commerce’s calculation of its U.S. price is unsupported by substantial evidence and contrary to law. According to Rubfil, Commerce’s questionnaire required Rubfil to report expenses in the currency in which those expenses were incurred. Therefore, Rubfil reported the ocean freight expenses associated with its U.S. sales in Malaysian ringgits, the currency in which those expenses were incurred. Rubfil complains, however, that when Commerce deducted ocean freight expenses from the gross price of Rubfil’s U.S. sales, Commerce did not convert the expenses from ringgits to U.S. dollars. Commerce’s mistake, Rubfil claims, resulted in an artificially low net U.S. price, which in turn, led to a higher dumping margin.

Commerce agrees that it never converted Rubfil’s ocean freight expenses from Malaysian ringgits to U.S. dollars, and requests a remand to correct its mistake. Accordingly, the Court remands this case, as it relates to Rubfil, in order for Commerce to convert Rubfil’s ocean freight expenses to U.S. dollars and to recalculate Rubfil’s U.S. price.

III.

CONCLUSION

For the foregoing reasons, the Court remands the Final Results to Commerce.

1 . The underlying administrative review was conducted prior to January 1, 1995. Consequently, the applicable law in this case is the antidumping statute as it existed prior to the amendments made by the Uruguay Round Agreements Act, Pub.L. No. 103-465, 108 Slat. 4809 (1994). See Torrington Co. v. United States,-Fed. Cir. (T)-,-, 68 F.3d 1347, 1352 (1995).

2 . The administrative record contains only the first page of Commerce's letter transmitting a questionnaire to Heveafil Sdn. Bhd., another respondent in the review. Notably, the record does not contain the actual questionnaire sent to any respondent in the review, including Rubberflex.

A gap in the administrative record such as this potentially compromises the Court’s ability to conduct a meaningful review of the Final Results. For example, in contesting Commerce's use of BIA, Rubberflex contends it used the proper date of sale methodology to report U.S. sales. Commerce disagrees, claiming Rubberflex did not report "all of its sales to the United States that were required by the questionnaire." Final Results, 62 Fed. Reg. at 62,551 (emphasis added). Without the questionnaire, however, it is impossible for the Court to know precisely what information Commerce solicited from Rubberflex, and whether Rubberflex in fact complied with that request. Although the Court does not reach the merits of this issue, see discussion Section 1(C) infra, the parties will nevertheless be forced to readdress this issue on remand. In doing so, the Court urges them to reach an agreement as to the precise terms of the questionnaire that should be part of the administrative record.

3 . The record does not contain the supplemental questionnaire Commerce issued to Rub-berflex either. The Court emphasizes once again that its ability to conduct a meaningful review is dependant upon the record. It will not condone speculation on the part of Commerce or Rubberflex, and therefore, will not itself engage in speculation. Thus, although the record contains the supplemental questionnaire Commerce sent to Rubfil on September 3, 1996, the Court cannot and will not assume the questionnaire Commerce issued to Rubberflex is the same.

4 . At the same time, Commerce issued Rubber-flex a supplemental questionnaire for the third administrative review, which Commerce had initiated in November, 1995. See Initiation of Antidumping and Countervailing Duty Administrative Reviews, 60 Fed.Reg. 57573 (Nov. 16, 1995) (Extruded Rubber Thread from Malaysia). Beginning with the supplemental questionnaire, Commerce conducted the second and the third reviews in tandem, including, most importantly, simultaneous verification for the second and the third reviews in Malaysia. See, e.g., Memo from Analyst/IA to File Regarding Cost and Sales Verification of Rubberflex and Confusion of Review Periods (Feb. 14, 1997) ("Verification Memo"), Pub. Admin. R., Fiche 25, Frame 62. For the Court's analysis of the results of the third administrative review, see Heveafil Sdn. Bhd. v. United States, - F.Supp.2d -, 1999 WL 566162 (CIT July 23, 1999).

5 . Commerce is required to use BIA when it is unable to verify a respondent’s information. See 19 U.S.C. § 1677e (1988); 19 C.F.R. § 353.37(a) (1994). In practice. Commerce applies either "total” BIA or “partial” BIA. "Commerce applies total BIA when a respondent has failed to submit information in a timely manner, or when part of the submitted data is sufficiently flawed, so that the response as a whole is rendered unusable.” National Steel Corp. v. United States, 18 CIT 1126, 1131, 870 F.Supp. 1130, 1135 (1994). It "applies partial BIA,” on the other hand, "when only part of the submitted information is deficient.” Id.

Further, when applying total BIA, Commerce employs a two-tiered methodology. See Allied-Signal Aerospace Co. v. United States, -Fed. Cir. (T)-, -, 996 F.2d 1185, 1188 (1993) (upholding the two-tier BIA methodology as reasonable). Essentially, when a respondent refuses to cooperate or impedes the review process, Commerce applies first-tier total BIA. When a respondent is cooperative, however, Commerce applies second-tier total BIA, which is generally more favorable to a respondent. See id. In this case, Commerce assigned Rubberflex a dumping margin using total, second-tier BIA, or the highest rate assigned to any firm in the second review.

6 . The index to the administrative record does not contain an entry for this meeting, nor is the Court aware of any record transcript of the proceedings. Thus, the Court assumes that this meeting does not qualify as an "ex parte meeting between ... interested parties or other persons providing factual information in connection with a proceeding,” 19 U.S.C. § 1677f(a)(3), such that it is required to be part of the record. See 19 U.S.C. § 1516a(b)(2)(A)(i).

7 . Counsel for Rubberflex represented to the Court during oral argument that Commerce refused to reschedule verification for the second and the third reviews because it needed to comply with the newly-imposed statutory deadlines governing the third review. Tr. of Oral Argument (Mar. 25, 1999), at 57. Under current regulations, the deadline for completing a review is a little over one year. See 19 C.F.R. Pt. 351, Annex IV (1998).

8 . Following verification, the record shows that Rubberflex attempted to submit the corrections presented to case handlers during verification, as well as several documents the case handlers allegedly had not accepted on grounds that there was insufficient time to verify the data. See Letter from White & Case Submitting Two Exhibits Not Accepted in Malaysia Due to Time Constraints (Oct. 22, 1996), Pub. Admin. R., Fiche 24, Frame 78, and Letter from White & Case Submitting Summary of Corrections Presented During U.S. Verification (Oct. 23, 1996), Pub. Admin. R., Fiche 24, Frame 73.

9 . See, e.g., American Alloys, Inc. v. United States, - Fed. Cir. (T) -, -, 30 F.3d 1469, 1475 (1994) (noting that "the statute gives Commerce wide latitude in its verification procedures”); PPG Indus., Inc. v. United States, - Fed. Cir. (T)-, -, 978 F.2d 1232, 1238 (1992) (recognizing ■ Commerce's "authority to determine the extent of investigation and information it needs”); Emerson Power Transmission Corp. v. United States, 19 CIT 1154, 1160, 903 F.Supp. 48, 54 (1995) (noting Commerce's discretion whether to request post-verification information); Persico Pizzamiglio, S.A. v. United States, 18 CIT 299, 307 (1994) (finding Commerce’s decision to conduct a 3-day overseas verification reasonable); NSK Ltd. v. United States, 16 CIT 745, 750, 798 F.Supp. 721, 725 (1992) (upholding Commerce's rejection of untimely factual information); Mons anto Co. v. United States, 12 CIT 937, 944, 698 F.Supp. 275, 281 (1988) (upholding Commerce's acceptance of post-verification information which corrects or clarifies original response and recognizing Commerce’s discretion to determine the adequacy of a response); Hercules, Inc. v. United States, 11 CIT 710, 726, 673 F.Supp. 454, 469 (1987) (recognizing Commerce’s sole discretion "to select a particular verification methodology”).

10 . In contrast, the outline states that Commerce would accept new information at verification "when that information makes minor corrections to information already on the record.” Verification Outline, Pub. Admin. R., Fiche 20, Frame 68. This suggests a construct whereby a respondent must notify Commerce of "errors” in data prior to verification, but can submit "corrections” at verification. The Court is not sure whether Commerce distinguishes between "errors” and "corrections,” and if so, where it draws the line between the two. In any event, the Court understands the outline as requiring a respondent to submit significant changes prior to verification, and as putting respondent on notice that Commerce’s case handlers would accept only minor changes during verification.

11 . The scope of verification may not be conterminous with the scope of inquiry in Commerce's original and supplemental questionnaires. Indeed, "Verification is a spot check and is not intended to be an exhaustive examination of the respondent’s business. ITA has considerable latitude in picking and choosing which items it will examine in detail.” Monsanto v. United States, 12 CIT 937, 944, 698 F.Supp. 275, 281 (1988).

12 . Moreover, this argument strikes the Court as somewhat inconsistent with Commerce’s representations at oral argument. Counsel for Commerce strenuously argued to this Court that Commerce has no “standard” verification procedures. See Tr. of Oral Argument (March 25, 1999), at 35 ("[tjhere is no strict methodology in terms of how a verification occurs”).

13 . It seems to the Court that Commerce wants to have it both ways in this case. Commerce implies here that respondents like Rub-berflex should anticipate that Commerce will conduct verification exactly as it has in the past, and in advance and without notice, prepare under that assumption. Yet, at the same time. Commerce "has not restricted itself by clarifying ... verification requirements],” Micron Tech, - Fed. Cir. (T) at -, 117 F.3d at 1396, and fiercely defends its discretion to conduct reviews and verification using ad hoc procedures.