Yangzhou Bestpak Gifts & Crafts Co. v. United States

United States Court of Appeals for the Federal Circuit ______________________ YANGZHOU BESTPAK GIFTS & CRAFTS CO., LTD., Plaintiff-Appellant, v. UNITED STATES, Defendant-Appellee, AND BERWICK OFFRAY LLC, Defendant-Appellee. ______________________ 2012-1312 ______________________ Appeal from the United States Court of International Trade in No. 10-CV-0295, Senior Judge Judith M. Barzi- lay. ______________________ Decided: May 20, 2013 ______________________ NED H. MARSHAK, Grunfeld, Desiderio, Lebowitz, Sil- verman & Klestadt, LLP, of New York, New York, argued for plaintiff-appellant. With him on the brief were BRUCE M. MITCHELL, MARK E. PARDO and ANDREW T. SCHUTZ. 2 YANGZHOU BESTPAK GIFTS v. US RENEE GERBER, Trial Attorney, Commercial Litigation Branch, Civil Division, United States Department of Justice, of Washington, DC, argued for defendant- appellee, United States. With her on the brief was STUART F. DELERY, Acting Assistant Attorney General, JEANNE E. DAVIDSON, Director, and PATRICIA M. MCCARTHY, Assistant Director. Of counsel on the brief was SCOTT D. MCBRIDE, Senior Attorney, Office of the Chief Counsel for Import Administration, United States Department of Commerce, of Washington, DC. GREGORY C. DORRIS, Pepper Hamilton LLP, of Wash- ington, DC, argued for defendant-appellee, Berwick Offray LLC. ______________________ Before RADER, Chief Judge, MAYER, and PROST, Circuit Judges. RADER, Chief Judge. Yangzhou Bestpak Gifts & Crafts Co., Ltd. (Bestpak) appeals from a final judgment of the United States Court of International Trade concerning its importation of narrow woven ribbons with woven selvedge from China. Yangzhou Bestpak Gifts & Crafts Co. v. United States, 825 F. Supp. 2d 1346 (Ct. Int’l Trade 2012) (Bestpak II). The Court of International Trade sustained the United States Department of Commerce’s (Commerce) calculation of Bestpak’s separate rate margin using a simple average of a de minimis and an adverse facts available margin, yielding a rate of 123.83%. Because substantial evidence does not support the 123.83% rate, this court vacates and remands. YANGZHOU BESTPAK GIFTS v. US 3 I. Commerce imposes antidumping duties upon import- ed products that it determines have been “dumped,” or sold in the United States at less than fair value. See 19 U.S.C. § 1673. An antidumping duty reflects the amount by which the normal value exceeds the export price of a foreign exporter’s merchandise. §§ 1673e(a)(1), 1677(35). This excess amount becomes the “dumping margin.” Commerce must determine individual dumping mar- gins for each known exporter or producer of the subject merchandise within a twelve-month period. §§ 1675, 1677f-1(c)(1). Commerce calculates a dumping margin specific to each respondent based upon analysis of sales and cost data collected from the respondent via an anti- dumping questionnaire that may total thousands of pages of extensive narrations and exhibits. Appellee United States’ Br. at 22. However, if this process is not practica- ble because of the large number of respondents involved in the investigation, Commerce may select a more reason- able number of mandatory respondents for these individ- ual investigations. § 1677f-1(c)(2). Commerce often limits mandatory respondents to those with the largest volume of exports and/or shipments of subject merchandise dur- ing the period of investigation, or a statistically valid sample among all known respondents. § 1677f-1(c)(2)(A)– (B). For the remaining non-mandatory respondents, Commerce calculates an “all others” rate, usually by taking the weighted average of all mandatory respond- ents’ rates, excluding any zero or de minimis rates and rates based entirely on adverse facts available (AFA). However, when all dumping margins established are only either de minimis or AFA rates, Commerce applies the exception found in § 1673d(c)(5)(B). In such cases, Com- merce “may use any reasonable method to establish the estimated all others rate for exporters and producers not individually investigated, including averaging the esti- mated weighted average dumping margins determined for 4 YANGZHOU BESTPAK GIFTS v. US the exporters and producers individually investigated.” 19 U.S.C. § 1673d(c)(5)(B). The Statement of Administrative Action (SAA), recog- nized by Congress as an authoritative expression concern- ing the interpretation and application of the Tariff Act under 19 U.S.C. § 3512(d), provides more guidance on the methodology Commerce should apply under the exception to the general rule: In such situation, Commerce may use any reason- able method to calculate the all others rate. The expected method in such cases will be to weight average the zero and de minimis margins and margins determined pursuant to the facts availa- ble, provided that volume data is available. How- ever, if this method is not feasible, or if it results in an average that would not be reasonably reflec- tive of potential dumping margins for non- investigated exporter or producers, Commerce may use other reasonable methods. SAA, accompanying the Uruguay Round Agreements Act, H.R. Doc. No. 103-316, at 873 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4200. Non-mandatory respondents also have the option of voluntarily completing the antidumping questionnaire to seek individual investigation. However, even after the voluntary respondent timely submits its response to Commerce’s questionnaire, Commerce may decline to fully investigate the voluntary respondent. This occurs when Commerce determines that the number of exporters or producers who have submitted such information is so large that individual examination of such exporters or producers would be unduly burdensome and inhibit the timely completion of the investigation. § 1677m(a)(2). Proceedings involving a nonmarket economy, such as China, are slightly different. Although Commerce selects YANGZHOU BESTPAK GIFTS v. US 5 mandatory respondents to individually investigate, Com- merce begins with a rebuttable presumption that all respondents in the investigation are under foreign gov- ernment control and thus should receive a single country- wide dumping rate. Sigma Corp. v. United States, 117 F.3d 1401, 1405 (Fed. Cir. 1997). In many cases, the country-wide rate is based on AFA. 1 Antidumping & Countervailing Duty Laws Appendix B. Commerce may use adverse inferences when calculating a rate if an investigated respondent refuses to cooperate by impeding the investigation or not properly providing information. Commerce typically concludes that some part of the country-wide entity has not cooperated in the proceeding because those that have responded do not account for all imports of the subject merchandise. § 1677e(b). Com- merce is required to corroborate chosen AFA rates to ensure that they fall within the purportedly acceptable range of margins determined. § 1677e(c). In order to secure a separate rate from the country- wide rate, respondents in a nonmarket economy must establish an absence of de jure and de facto government control. Id. The mandatory respondents’ antidumping questionnaire allows a respondent to assert independence from the country-wide entity. All other respondents seeking eligibility for a separate rate must complete a separate rate application that is about thirty pages of responses and attached exhibits. Appellee United States’ Br. at 22. The separate rate for eligible non-mandatory re- spondents is generally calculated following the statutory method for determining the “all others rate” under § 1673d(c)(5)(A). Transcom, Inc. v. United States, 294 F.3d 1371, 1374 (Fed. Cir. 2002); Amanda Foods (Vietnam) Ltd. v. United States, 647 F. Supp. 2d 1368, 1379 (Ct. Int’l Trade 2009). As such, Commerce will typically use the weighted average of all mandatory respondents’ rates, excluding any de minimis and AFA rates. Appellee 6 YANGZHOU BESTPAK GIFTS v. US Berwick’s Br. at 15. If all dumping margins established are only either de minimis or AFA rates, Commerce accordingly applies the exception found in § 1673d(c)(5)(B). II. Bestpak is a Chinese exporter of narrow woven rib- bons. These are woven ribbons with a width that is less than or equal to 12 centimeters and have finished edges that keep the fabric from unraveling or fraying. On August 6, 2009, Commerce initiated an antidumping investigation of narrow woven ribbons from China and Taiwan for the period spanning January 1 to June 30, 2009. See Narrow Woven Ribbons with Woven Selvedge from the People’s Republic of China and Taiwan, 74 Fed. Reg. 39,291, 39,292 (Aug. 6, 2009). Commerce began its investigation by issuing quantity and value data (Q&V) questionnaires to all known Chi- nese exporters of this product for the purpose of selecting mandatory respondents to review. Id. at 39,296. Nine- teen companies responded, including Bestpak, and Com- merce determined that it was impractical to individually review all of them. Resp’t Selection Mem. (Sept. 11, 2009), Pub. Doc. 94 at 4 (“After careful consideration . . . we find that we can reasonably examine no more than two producers and/or exporters.”). Notably, Bestpak fully cooperated. On September 11, 2009, Commerce selected the two largest exporters as mandatory respondents for individual investigation: Ningbo Jintian Import & Export Co., Ltd. (Jintian) and Yama Ribbons & Bows Co., Ltd. (Yama). Pub. Doc. 94 at 3. Yama exported 135,000 kg of subject merchandise to the United States during the period of investigation, while Jintian exported 100,000 kg. No other Chinese respondent exported more than 48,000 kg, and Bestpak exported 21,000 kg, ranking seventh of the nineteen exporters submitting Q&V information. YANGZHOU BESTPAK GIFTS v. US 7 Between September and November of 2009, Com- merce received responses from Yama, but nothing from Jintian. Commerce also received separate rate applica- tions from twelve exporters, including Bestpak. As part of those applications, each applicant provided a photocopy of an invoice of a U.S. commercial transaction during the period of investigation. No exporter requested voluntary investigation, and Commerce did not choose a replace- ment mandatory respondent for Jintian even though Commerce, in the past, has chosen a replacement re- spondent when it was clear that a mandatory company would not participate. See, e.g., Prestressed Concrete Steel Wire Strand from the People’s Republic of China, 74 Fed. Reg. 61,104 (Nov. 23, 2009). In sum, Commerce’s investi- gation was left with one participant after Jintian’s with- drawal. In July 2010, Commerce issued its antidumping duty order and accompanying decision memo. See Narrow Woven Ribbons with Woven Selvedge from the People’s Republic of China, 75 Fed. Reg. 41,808 (July 19, 2010) (Final Determination), as amended Narrow Woven Rib- bons with Woven Selvedge from the People’s Republic of China, 75 Fed. Reg. 51,979 (Aug. 24, 2010); Issues and Decision Memorandum for the Final Determination in the Antidumping Duty Investigation of Narrow Woven Rib- bons with Woven Selvedge from the People’s Republic of China, (Deputy Assistant Sec’y July 12, 2010) (Issues and Decision Mem.), http://ia.ita.doc.gov/frn/summary/PRC/2010-17568-1.pdf. Commerce imposed dumping margins ranging from 0% to 247.65% against Chinese exporters of narrow woven ribbons with woven selvedge. With respect to the two mandatory respondents, Ya- ma was assigned a de minimis dumping margin whereas Jintian received the AFA China-wide rate of 247.65%— which Jintian was assigned because of its refusal to cooperate in the investigation. Commerce corroborated 8 YANGZHOU BESTPAK GIFTS v. US this 247.65% AFA China-wide rate by comparing it to the margin found for Yama and finding it to be within the range of margins for ten models of subject ribbons sold by Yama. Final Corroboration Mem. (July 12, 2010), Pub. Doc. 376 at 2, Conf. Doc. 375 at 2. Commerce further found all twelve exporters that submitted separate applications, including Bestpak, eligible for separate rates. In calculating a separate rate for these twelve companies, Commerce took a simple average of the de minimis rate of Yama and the AFA China-wide rate assigned to Jintian, yielding a 123.83% margin. Final Determination, 75 Fed. Reg. at 41,811. Commerce explained that normally it would determine the separate rate through a weighted average of dumping margins, excluding zero and de minimis margins or margins based entirely upon AFA, pursuant to § 1673d(c)(5)(A). In this case, however, the only two dumping margins on the administrative record were Yama’s de minimis rate and Jintian’s AFA China-wide rate. Therefore, Commerce applied the exception found in § 1673d(c)(5)(B) and took a simple average of the two rates. Issues and Decision Mem., at 18–19. In effect, the resulting separate rate equaled one half of the AFA China-wide rate. Commerce also determined it was not feasible, given statutory time constraints, to identify, investigate, and analyze another respondent. Id. at 21–22. Commerce noted that neither Bestpak, nor any other separate rate respondent requested to be a voluntary respondent. Further, Commerce determined that while Bestpak later challenged the selection of such a small number of man- datory respondents, Commerce did not have “sufficient time” to “obtain, evaluate, and employ additional factual information from the separate rate companies” because the challenge was filed with less than three months remaining in the statutory period to complete the investi- gation. Id. at 20–22. YANGZHOU BESTPAK GIFTS v. US 9 Thereafter, Bestpak appealed the separate rate de- termination to the Court of International Trade. Yang- zhou Bestpak Gifts & Crafts Co. v. United States, 783 F. Supp. 2d 1343 (Ct. Int’l Trade 2011) (Bestpak I). Bestpak argued the use of a simple average methodology was contrary to law and not supported by substantial evidence on the record. Bestpak further argued that the most reasonable calculation was to assign Bestpak the de minimis margin calculated for the only individually investigated respondent, Yama. The Court of International Trade sustained-in-part and remanded-in-part, holding that although Commerce was permitted by law to use its chosen methodology, Commerce did not support with substantial evidence the reasonableness of its determination of Bestpak’s potential dumping margins. Bestpak I, 783 F. Supp. 2d at 1350–51. The Court of International Trade explained that merely because the SAA “allow[ed] for use of a simple average of an [AFA] rate and zero or de minimis rates [did] not absolve the agency from ensuring that a separate rate reasonably reflect[ed] potential dumping margins . . . or from rationally connecting the record evidence with the final conclusions.” Id. at 1351. The Court of Internation- al Trade also recognized that “the calculated separate rate [was] exceptionally larger than the rate calculated for the lone cooperative mandatory respondent.” Id. The Court of International Trade remanded the case to Commerce to provide “a more rigorous explanation” since Commerce had “not provided information suggesting that Bestpak dumps its sales at such levels.” Id. In accordance with the Court of International Trade’s order, Commerce reviewed the administrative record for information that would support or detract from the mar- gin applied to Bestpak. Commerce compared the estimat- ed average unit values (AUVs) calculated from available information provided by Jintian, Yama, and Bestpak in their Q&V questionnaire responses. Draft Results of 10 YANGZHOU BESTPAK GIFTS v. US Redetermination Pursuant to Court Order (Sept. 7, 2011) (Draft Remand Results). Commerce explained that an estimated AUV “is a ratio calculated by dividing a re- spondent’s total value of sales by its total quantity of sales, which provides a rough, estimated snapshot of a respondent’s pricing practices.” Id. at 6. Commerce explained that “all things being equal,” a low estimated AUV could indicate the existence of a larger dumping margin, while a high estimated AUV could “indicate the reverse to be true.” Id. In its Draft Remand Results, Commerce determined that Yama had a higher estimated AUV, Jintian had a lower estimated AUV, while Bestpak’s estimated AUV fell between the two mandatory respondents’ AUVs. Com- merce found that a “simple average of Jintian and Yama’s estimated AUVs” equaled a rate which was “very close” to Bestpak’s “estimated AUV.” Id. at 7. Commerce there- fore determined that using a simple average of Jintian’s and Yama’s margins “reasonably reflects . . . Bestpak’s potential dumping margin.” Id. In response, Bestpak contested the correlation be- tween Jintian’s, Yama’s, and Bestpak’s comparative AUVs and dumping margins. See Yangzhou Bestpak Comments on Draft Redetermination (Sept. 15, 2011). Commerce acknowledged the flaws in its reliance on AUVs as support for its determination, but stated that the record was limited and AUVs were the “only basis” for comparison between these respondents. See Final Results of Redetermination Pursuant to Court Order, at 16 (Sept. 26, 2011) (Final Remand Results). In view of that, Com- merce concluded that the calculated separate rate margin was reasonably reflective of Bestpak’s potential dumping margin based on the support of correlating estimated AUVs. Id. Bestpak appealed again to the Court of International Trade. Bestpak argued that Commerce’s AUV analysis YANGZHOU BESTPAK GIFTS v. US 11 was flawed and further pointed out that in addition to Bestpak’s AUV information, the record contained a sales invoice relating to its commercial activity during this period of investigation. Bestpak II, 825 F. Supp. 2d at 1352 n.1. Bestpak asked the Court of International Trade to find that the record, including this invoice, supported a more reasonable separate rate of zero percent. In re- sponse, Commerce asked the Court of International Trade to strike Bestpak’s invoice argument since it was not raised before the agency. In its opinion, the Court of International Trade first acknowledged that “the real problem [was] the absence of enough sales data.” Id. at 1351–52. It found that “the record contain[ed] little information as to what Bestpak’s (or the other separate rate respondents’) potential dump- ing margin might be, or whether it [was] closer to 0% or 247.65%. . . . [Commerce] simply [could not] know on this administrative record whether the separate rate ‘reason- ably reflects’ commercial reality.” Id. at 1352. Nonethe- less, while the Court of International Trade found that Commerce’s separate rate determination “may [have been] unfortunate and even frustrating, . . . it [was] not unreasonable on this limited administrative record.” Id. at 1353. With respect to the invoice, while the Court of Inter- national Trade did not grant the motion to strike, it declined to entertain Bestpak’s argument because Bestpak did not raise this argument with Commerce. Id. at 1352 n.1. The Court of International Trade further noted that it was “not persuaded that one sales invoice [was] sufficient to demonstrate that the separate rate should be 0%.” Id. Ultimately, the Court of International Trade sustained Commerce’s Final Remand Results and this appeal followed. Bestpak appeals three issues: (1) whether Commerce employed a reasonable methodology to calculate the 12 YANGZHOU BESTPAK GIFTS v. US separate rate; (2) whether Commerce’s calculated sepa- rate rate for Bestpak was supported with substantial evidence showing a reasonable correlation to Bestpak’s potential dumping margins; and (3) whether the Court of International Trade abused its discretion when it deter- mined that Bestpak failed to exhaust its administrative remedies in raising the invoice argument that should first have been raised with Commerce during the remand. This court has jurisdiction under 28 U.S.C. § 1295(a)(5). III. This court reviews decisions of the Court of Interna- tional Trade concerning Commerce’s antidumping deter- minations by applying the same standard of review used by the trade court. Gallant Ocean (Thailand) Co. v. United States, 602 F.3d 1319, 1323 (Fed. Cir. 2010). Commerce’s determination will be sustained unless it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” § 1516a(b)(1)(B)(i). The first issue on appeal concerns whether Com- merce’s methodology accords with law. This court turns to the two-part test articulated in Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S 837 (1984). First, the court must determine whether Congress directly spoke to the precise question at issue and clearly ex- pressed its purpose and intent in the governing statute. Id. at 842–43. If the statute does not clearly answer the relevant question, then the court must turn to the second step and decide whether the agency’s interpretation amounts to a reasonable construction of the statute. Id. at 843. To survive judicial scrutiny, Commerce’s interpre- tation need not be “the only reasonable interpretation or even the most reasonable interpretation.” Koyo Seiko Co. v. United States, 36 F.3d 1565, 1570 (Fed. Cir. 1994). Importantly, the court will defer to a reasonable interpre- YANGZHOU BESTPAK GIFTS v. US 13 tation even where the court might have adopted a differ- ent interpretation. Chevron, 467 U.S. at 843 n.11. Here, the relevant statutory text does not directly ad- dress the precise question at issue: whether it is permis- sible to use a simple average to calculate separate rates in a nonmarket economy investigation where one respondent receives an AFA rate and the other receives a de mini- mis rate. § 1673d(c)(5)(B). Therefore, the court must determine whether Commerce reasonably interpreted the statute under the second step of Chevron. Chevron, 467 U.S. at 843. The statutory text allows Commerce to use any rea- sonable method to establish the estimated separate rate. See § 1673d(c)(5)(B) (“[Commerce] may use any reasona- ble method . . . .”); see also SAA (“Commerce may use other reasonable methods.”). Hence, this court must determine whether Commerce used “reasonable methods” by taking the simple average of a de minimis rate and total AFA rate assigned to the two mandatory respond- ents to calculate the separate rate. In accord with the Court of International Trade’s hold- ing in Bestpak I, this court finds that the methodology used by Commerce—although somewhat questionable— meets the statute’s lenient standard of “any reasonable method.” Bestpak contends that Commerce is not permit- ted to include an AFA rate in the calculation of a separate rate. However, § 1673d(c)(5)(B) and the SAA explicitly allow Commerce to factor both de minimis and AFA rates into the calculation methodology. Although again ques- tionable in terms of economic reality, this court detects no legal error in Commerce’s use of a simple average rather than a weighted average. The statute also specifically allows for an averaging or any other reasonable method. Because the agency employed a methodology similarly derived from the relevant statutory language, this court affords the appropriate deference due to Commerce. 14 YANGZHOU BESTPAK GIFTS v. US Nevertheless, “[w]hile various methodologies are permitted by the statute, it is possible for the application of a particular methodology to be unreasonable in a given case.” Thai Pineapple Canning Indus. Corp. v. United States, 273 F.3d 1077, 1085 (Fed. Cir. 2001). “[F]orm should be disregarded for substance and the emphasis should be on economic reality.” United States v. Eurodif S.A., 555 U.S. 305, 317–18 (2009). This court finds that this case presents that situation. Although Commerce may be permitted to use a simple average methodology to calculate the separate rate, the circumstances of this case renders a simple average of a de minimis and AFA China- wide rate unreasonable as applied. Similarly, a review of the administrative record reveals a lack of substantial evidence showing that such a determination reflects economic reality. This court may review whether substantial evidence supports Commerce’s determination by asking whether there was “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Uni- versal Camera Corp. v. NLRB, 340 U.S. 474, 477 (1951). This standard requires Commerce to examine the record and articulate a satisfactory explanation for its action. Amanda Foods, 647 F. Supp. 2d at 1379. Commerce may not base that determination “on the basis of mere conjec- ture or supposition.” § 1677(7)(F)(ii). Nor may Commerce explain the absence of evidence by invoking procedural difficulties that were at least in part a creature of its own making. As justification, Commerce points to only one eviden- tiary finding in its Final Remand Results as substantial evidence to support the calculated margin as being a reasonable reflection of Bestpak’s potential dumping margin—the AUV analysis. Commerce concluded that the simple average margin assigned to Bestpak was reasonably reflective of commercial reality based on an analysis showing that a simple average of the estimated YANGZHOU BESTPAK GIFTS v. US 15 AUVs of Yama and Jintian was very close to Bestpak’s estimated AUV. Final Remand Results, at 6, 15–16. This court does not find Commerce’s late attempt to backfill with these AUV estimates, untethered to the three respondents’ actual dumping margins, as amount- ing to substantial evidence. Commerce only received detailed information from one mandatory respondent: Yama. Jintian’s estimated AUV, with little connection to its calculated dumping margin, is not enough. In other words, Commerce really only had one substantiated and calculated basis for dumping margins and that infor- mation came from Yama. Commerce cannot base a de- termination of economic reality on such slim findings. There must be more. An overriding purpose of Commerce’s administration of antidumping laws is to calculate dumping margins as accurately as possible. Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed. Cir. 1990). When there is only one benchmark, Commerce’s comparison of the potential dumping margins with the estimated AUVs based on scant information available here is not reasona- ble. Even the Court of International Trade noted that “Commerce put itself in a precarious situation when it selected only two mandatory respondents.” Bestpak I, 783 F. Supp. 2d at 1351 n.4. This record simply does not supply enough data for Commerce to calculate its sepa- rate rate determination based on only one individually investigated respondent. In fact, Jintian’s actual dumping margin is unknown. The 247.65% margin was imposed on Jintian because of its lack of participation. Although this 247.65% rate was corroborated, even that corroboration was based on Ya- ma’s commercial activity. More importantly, Commerce acknowledged that the link between Jintian’s AUV and its AFA China-wide dumping margin was created by Commerce when it assigned Jintian the AFA margin. 16 YANGZHOU BESTPAK GIFTS v. US While Bestpak’s estimated AUV aligned with a simple average of Jintian’s and Yama’s estimated AUVs, Com- merce’s inference that their dumping margins paralleled that same correlation is speculative. As such, using the AUV analysis as evidence that Bestpak’s dumping margin is likewise in line with a simple average of Jintian’s and Yama’s dumping margins finds no credible economic support in the record. The record here is so thin that Commerce could nei- ther have reached a valid decision nor reasonably have found evidence to support the determination that Bestpak deserves a margin that more than doubles the import’s sales price. The 123.83% rate assigned to Bestpak is far in excess of the de minimis rate assigned to the only cooperating, non-government controlled, and mandatory respondent: Yama. It is worth noting that similar to Yama, Bestpak successfully proved that it was independ- ent of government control. However, Commerce ultimate- ly assigned Bestpak a margin that was exactly half of the China-wide rate—a rate for those presumed to be under foreign government control. Assigning a non-mandatory, separate rate respondent a margin equal to over 120% of the only fully investigated respondent with no other information is unjustifiably high and may amount to being punitive, which is not permitted by the statute. See F.Lii de Cecco di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir. 2000) (“[T]he pur- pose of section 1677e(b) is to provide respondents with an incentive to cooperate, not to impose punitive, aberration- al, or uncorroborated margins.”). The reasoning of Gallant Ocean is applicable here. In Gallant Ocean, this court found that the high rate deter- mined for the Thai respondent—a rate that was more than ten times higher than the dumping margin for cooperating respondents—was unsupported by substan- tial evidence because there was nothing in the record that tied that rate to the respondent’s actual dumping margin. YANGZHOU BESTPAK GIFTS v. US 17 Gallant Ocean, 602 F.3d at 1325. Notwithstanding that the Thai respondent was unresponsive and application of adverse facts was warranted, the AFA rate assigned was required to reflect commercial reality and thus, to be “a reasonably accurate estimate” of actual dumping rates. Id. at 1324. Even with determinations of an AFA-rate, Commerce may not select unreasonably high rates having no relationship to the respondent’s actual dumping mar- gin. Id. at 1323. Likewise, rate determinations for non- mandatory, cooperating separate rate respondents must also bear some relationship to their actual dumping margins. The record here, however, does not contain any infor- mation—save the AUV estimate—that indicates what Bestpak’s individually calculated margin might be. There is no basis in the record to tie this 123.83% rate to Bestpak’s commercial activity. What could have been a coincidental correlation of the three data points is not enough to be substantial supporting evidence of economic reality. Commerce points to no other evidence in the record to substantially support the 123.83% margin derived from the simple average methodology applied here. The result is not only limited and frustrating, as the Court of International Trade described it, but is also unreasonable. This case is peculiar in that Commerce identified only two significant exporters/producers, yet one was assigned a de minimis dumping margin while the other was as- signed the highest possible AFA China-wide margin. This situation undercut the reasonableness of the evidence relied upon to set the margin. Moreover, Commerce applied that simple average in determining the rate for the twelve respondents that qualified for a separate rate. Commerce was certainly aware of this situation when Jintian was being unresponsive early in the investigation and could have gathered more information. Although Commerce argues in its Final Remand Results that it was 18 YANGZHOU BESTPAK GIFTS v. US the best it could do because of the limited record, this court finds no support in this court’s precedents or the statute's plain text for the proposition that limited re- sources or statutory time constraints can override fairness or accuracy. See SNR Roulements v. United States, 402 F.3d 1358, 1363 (Fed. Cir. 2005) (“Antidumping laws intend to calculate antidumping duties on a fair and equitable basis.”). As Bestpak contends, “[i]f the record before the agency does not support the agency action . . . the proper course, except in rare circumstances, is to remand to the agency for additional investigation or explanation.” Appellant’s Br. at 40 (quoting Fla. Power & Light Co. v. Lorion, 470 U.S. 729, 744 (1985)). This court therefore vacates and remands the Court of International Trade’s decision so that it may remand the case back to Commerce. The final issue on appeal centers on Bestpak’s invoice from its separate rate application. In its second appeal to the Court of International Trade, Bestpak argued that Commerce should have considered information derived from this sample invoice in determining the separate rate. Commerce required each applicant to submit a photocopy of a commercial invoice from the period of investigation in the separate rate application, but Bestpak made no reference to this sample invoice until the Final Remand Results were being reviewed by the Court of International Trade. Bestpak, at this late time, argued that the sample invoice was evidence of its commercial behavior and strongly supported a determination that Bestpak was entitled to a zero dumping rate. The Court of Interna- tional Trade refused to entertain Bestpak’s arguments since Bestpak failed to raise this argument before Com- merce, “depriving Commerce of the opportunity to address it.” Bestpak II, 825 F. Supp. 2d at 1352 n.1. The court “shall, where appropriate, require the ex- haustion of administrative remedies.” 28 U.S.C. § 267(d). The doctrine of exhaustion provides “that no one is enti- YANGZHOU BESTPAK GIFTS v. US 19 tled to judicial relief for a supposed or threatened injury until the prescribed administrative remedy has been exhausted.” Sandvik Steel Co. v. United States, 164 F.3d 596, 599 (Fed. Cir. 1998). “[T]he CIT generally takes a ‘strict view’ of the requirement that parties exhaust their administrative remedies . . . .” Corus Staal BV v. United States, 502 F.3d 1370, 1379 (Fed. Cir. 2007); SeAH Steel Corp. v. United States, 764 F. Supp. 2d 1322, 1325 (Ct. Int’l Trade 2011). Certain exceptions to the exhaustion requirement apply, such as where exhaustion would be “a useless formality,” or where the party “had no opportuni- ty” to raise the issue before the agency. Jiaxing Brother Fastener Co. v. United States, 751 F. Supp. 2d 1345, 1355–56 (Ct. Int’l Trade 2010). Even “where Congress has not clearly required exhaustion, sound judi- cial discretion governs.” McCarthy v. Madigan, 503 U.S. 140, 144 (1992). This court thus reviews the Court of International Trade’s decision for an abuse of discretion. See Corus Staal, 502 F.3d at 1381. Bestpak first raised this invoice issue in its second appeal to the Court of International Trade. Bestpak, however, had other opportunities to raise this issue before Commerce. For instance, Bestpak appealed the same claim in Bestpak I and argued that it deserved Yama’s de minimis rate, but did not mention the invoice. Bestpak I, 783 F. Supp. 2d at 1350. In addition, the issue of whether the separate rate reasonably reflects Bestpak’s dumping margin had been squarely in play throughout the proceed- ings, as that was the Court of International Trade’s exact reason for remand in Bestpak I. 783 F. Supp. 2d at 1353 (“The court remands the issue to Commerce so that the agency may more thoroughly explain whether the sepa- rate rate reasonably reflects Bestpak’s potential dumping margins.”). However, Bestpak did not address the invoice in its submissions to Commerce during the remand rede- termination proceeding. Bestpak was further aware of Commerce’s sole reliance on the comparative AUVs after 20 YANGZHOU BESTPAK GIFTS v. US issuance of the Draft Remand Results, yet Bestpak still failed to highlight the invoice at issue. If Bestpak had raised this argument before Com- merce, it would not have been futile, but rather would have permitted Commerce the opportunity to address the argument in the first instance. Therefore, this court holds that the Court of International Trade did not abuse its discretion by finding that Bestpak failed to exhaust its administration remedies with respect to the invoice. IV. Accordingly, this court vacates and remands the Court of International Trade’s decision so that it may remand the case to Commerce for further proceedings consistent with this opinion. VACATE and REMAND