Diamond Sawblades Mfrs.' Coal. v. United States

Slip Op. 18 - 146

UNITED STATES COURT OF INTERNATIONAL TRADE

: THE DIAMOND SAWBLADES : MANUFACTURERS’ COALITION, : : Plaintiff, : : v. : Before: R. Kenton Musgrave, Senior Judge : Court No. 17-00167 UNITED STATES, : : Defendant, : : and : : BOSUN TOOLS CO., LTD., : : Defendant-Intervenor. : :

OPINION and ORDER

[Remanding 2014-15 administrative review of antidumping duty order on diamond sawblades and parts thereof from the People’s Republic of China.]

Dated: October 23, 2018

Daniel B. Pickard, Maureen E. Thorson, and Stephanie M. Bell, Wiley, Rein & Fielding, LLP, of Washington, DC, for the plaintiff.

John J. Todor, Senior Trial Counsel, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, of Washington, DC, for the defendant. With him on the brief were Chad A. Readler, Acting Assistant Attorney General, Jeanne E. Davidson, Director, and Franklin E. White, Jr., Assistant Director. Of Counsel on the brief was Paul K. Keith, Attorney, Office of the Chief Counsel for Trade Enforcement and Compliance, U.S. Department of Commerce, of Washington, DC.

Gregory S. Menegaz, J. Kevin Horgan, and Alexandra H. Salzman, deKeiffer & Horgan, PLLC, of Washington, DC, for the defendant-intervenor.

Court No. 17-00167 Page 2

Musgrave, Senior Judge: The plaintiff Diamond Sawblades Manufacturers’ Coalition (“DSMC”) challenges three aspects of the 2014-15 administrative review of the antidumping (“AD”) duty order on diamond sawblades (“DSBs”) and parts thereof from the People’s Republic of China (“PRC”)1, to wit: (1) Commerce’s decision not to apply adverse facts available (“AFA”) to an aspect of Bosun’s record-keeping; (2) Commerce’s selection for the copper powder and copper iron “clab”2 factor of production of Thai import statistics as the “best available information” for Bosun; and (3), to the extent remand of either of the forgoing issues impacts the final determination on the margin for the separate respondents, DSMC also seeks recalculation thereof. The defendant-intervenor, Bosun Tools Co., Ltd. (“Bosun”), an exporter and/or producer of subject merchandise and one of the two mandatory respondents during the review, joins the defendant in support of the administrative record and determinations thereon by the International Trade Administration, U.S. Department of Commerce (“Commerce” or “Department”). For the following reasons, the plaintiff’s motion for judgment on the agency record persuades that the case requires remand.

I. Jurisdiction and Standard of Review

Jurisdiction is pursuant to 28 U.S.C. §1581(c). The standard of judicial review on an action invoking 19 U.S.C. §1516a(a)(2)(A) and (B)(iii) is to decide whether the administrative

1 DSBs and Parts Thereof From the PRC, 82 Fed. Reg. 26912 (June 12, 2017) (“Final Results”), Public Record Document (“PDoc”) 404, and accompanying issues and decision memorandum, PDoc 389 (June 12, 2017) (“IDM”); see also DSBs and Parts Thereof From the PRC, 81 Fed. Reg. 89046 (Dec. 9, 2016) (prelim. results of 2014-2015 antidumping duty admin. rev.), PDoc 360. 2 The court defers to the parties’ apparent and mutual understanding of the term as advanced in this matter, although no such word appears to exist in the English language; the nearest similarity would seem to be “clabber,” meaning “mud” or “curdled milk.” Cf., e.g., Webster’s New International Dictionary of the English Language, Unabridged, p. 493 (2nd ed. 1956).

Court No. 17-00167 Page 3 determination is “unsupported by substantial evidence on the record, or otherwise not in accordance with law”. 19 U.S.C. §1516a(b)(1)(B)(i).

II. Discussion

A. Bosun’s Record-Keeping

DSMC’s first challenge is to Commerce’s decision not to apply AFA to an aspect of Bosun’s record-keeping. See 19 U.S.C. §1677e(b).

1. Background

In general, the AD statute expects that the margin for subject merchandise from a non-market economy (“NME”) such as the PRC shall be determined by comparing its U.S. price with a “normal value” determined by reference to data covering the factors involved in production of subject merchandise (“FOPs”) plus general and other expenses and profit obtained from one or more surrogate market economies at a level of economic development comparable to the NME country at issue that is/are also significant producer(s) of comparable merchandise. See 19 U.S.C. §1677b(c)(1)&(4).3

In order to reach its determinations, Commerce is required to rely on “facts otherwise available” on the record if “necessary” information is missing from the record. 19 U.S.C. §1677e(a) (i.e., “shall”). If such information is missing due to a party’s failure to act to the “best of its ability,” Commerce “may” use inferences adverse to the non-cooperating party in selecting from among the facts otherwise available, also known as “AFA.” Id. §1677e(b). Resort to AFA is not mandatory,

3 See also Letter from Minoo Hatten, Program Manager, AD/CVD Operations, Off. I, to All Interested Parties, re: Diamond Sawblades and Parts Thereof from the People’s Republic of China: Request for Surrogate Country and Surrogate Value Comments and Information (Feb. 24, 2016) at 1-2.

Court No. 17-00167 Page 4 see, e.g., Nippon Steel Corp. v. United States, 337 F.3d 1373, 1382 (Fed. Cir. 2003), and the AFA statute has been held to confer administrative discretion in its application and to require judicial deference thereto. E.g., Shangdong Huarong Machinery Co., Ltd. v. United States, 30 CIT 1269, 1297, 435 F. Supp. 2d 1261, 1285-86 (2006); AK Steel Corp. v. United States, 28 CIT 1408, 1416, 346 F. Supp. 2d 1348, 1355 (2004). Further, in the exercise of that discretion AFA is intended to be remedial: “The purpose of the adverse facts statute is ‘to provide respondents with an incentive to cooperate’ with Commerce’s investigation, not to impose punitive damages.” Essar Steel Ltd. v. United States, 678 F.3d 1268, 1276 (Fed. Cir. 2012), quoting F.lli De Cecco Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir. 2000).

During the administrative review Bosun maintained production facilities in both Thailand and the PRC. Letter from deKieffer & Horgan, PLLC to Sec’y Commerce, Sections C&D Questionnaire Response (July 1, 2016) (“QR”) at C-1, CDocs 132-143, PDocs 207-210. In the United States, Bosun’s sales affiliates were Bosun Tools, Inc. (“Bosun USA”) and Pioneer Tools, Inc. (“Pioneer”). Id. Both U.S. sales affiliates sold to U.S. customers diamond sawblades produced by both Thai and PRC facilities during the POR. Id. at C-2.

These products, both subject and non-subject merchandise, were maintained in containers purportedly indicating their country of origin. Memorandum to File, re: Verification of the U.S. Sales Response of Bosun Tools Co., Ltd. (May 17, 2017) (“VR”), PDoc 383, CDoc 365, at 4. Bosun’s U.S. sales affiliates did not, however, record this information when selling them to U.S. customers. Id. Lacking direct documentation during the review on which sales were of subject merchandise, Bosun reconstructed country of origin for those sales by: (1) identifying the models of sawblades that Pioneer and Bosun USA purchased through “unique product codes” assigned to Court No. 17-00167 Page 5 each affiliate; (2) identifying the country of origin by matching those codes to unit purchase prices; and (3) applying a first-in, first-out (“FIFO”) methodology to assign a country of origin to each sale. QR at C-2 to C-3.

In light of such reporting, DSMC argued that AFA should be applied to Bosun because Bosun could not demonstrate that it reported a complete and accurate universe of U.S. sales. Letter from Wiley Rein LLP to Sec’y Commerce (Jan. 17, 2017) (“DSMC Case Brief”), PDoc 373, CDoc 344, at 2-11. Specifically, DSMC argued that Bosun had conceded that it was unable to definitively identify its U.S. sales of subject merchandise made during the POR, see id., and that Bosun’s reported methodology for identifying U.S. sales was insufficiently supported and not demonstrated to be accurate. Id. at 6-11. In rebuttal, Bosun asserted that it had fully complied with Commerce’s information requests and had demonstrated the steps it undertook to identify sales of subject merchandise. See Letter from deKieffer & Horgan, PLLC to Sec’y Commerce (Jan. 24, 2017) (“Bosun Case Brief”), PDoc 367, at 6-8.

Due to this conflict, Commerce decided to verify Bosun’s U.S. sales responses. See Memorandum from Deputy Assistant Sec’y for Antidumping and Countervailing Duty Operations to Acting Assistant Sec’y for Enf’t and Compliance, re: Issues and Decision Memorandum for the Administrative Review of the Antidumping Duty Order on DSBs and Parts Thereof from the PRC (June 6, 2017); see also IDM at 22. Verification confirmed to Commerce that Bosun USA’s and Pioneer’s inventories of sawblades were maintained in boxes indicating the blades’ origin and that this information was not recorded during the sales process. VR, CDoc 365, PDoc 383, at 4. According to Bosun, as stated during verification, its affiliates did not record country of origin Court No. 17-00167 Page 6 because it “was not necessary information which needed to be recorded in its computer system for purposes of its business operations.” Id. As to its reporting and methodology, Bosun contended

that the petitioner’s argument omits the second stage and jumps from the first stage to the FIFO stage. Bosun argues that they were logical steps assigning the country of origin to the vast majority of products, not assumptions. Bosun explains that the Department extensively replicated and tested these stages in the database of all sales for Bosun USA and Pioneer USA and found no discrepancies.

IDM at 25.

Commerce reviewed Bosun’s first-in-first-out (“FIFO”) method, constructed after the fact for the review, for identifying merchandise of PRC or Thai origin. VR at 9-11. Commerce found that Bosun could not replicate the reported results of the FIFO step of its methodology for one of the three or four pre-selected sales for which this step was reviewed at verification. Id. at 10-11. Bosun “was unable to explain this discrepancy.” Id. at 11 (citation omitted).

Commerce also learned of errors in Bosun’s control number (“CONNUM”) reporting of multiple physical characteristics for certain of the examined sales. These errors were manifest for three out of the sixteen transactions examined, or nearly 19% of those sales reviewed at verification. Id. at 2, 7; IDM at 31. Specifically, Commerce discovered that Bosun had reported the incorrect code for cutting edge, segment height, segment length, and core thickness for two of those sales, VR at 7, and in preparing for verification Bosun also discovered that it had reported the incorrect diamond grade for an additional sale. Id. at 2; IDM at 31.

Commerce then permitted the parties to submit additional case briefs based on its verification of Bosun. See IDM at 22. DSMC continued to argue to Commerce that AFA should be applied to Bosun, highlighting that Bosun had failed to maintain country of origin information that would allow it to accurately identify sales of merchandise made in the PRC despite having the Court No. 17-00167 Page 7 ability to do so. Letter from Wiley Rein LLP to Sec’y Commerce, re: Results of Bosun’s U.S. Sales Verification (May 23, 2017) (“DSMC Verification Case Brief”), PDoc 385, CDoc 366, at 12-14. DSMC also argued that Commerce’s verification demonstrated that Bosun’s identification of sales of subject merchandise was unreliable because it could not replicate the results of its multi-step process for determining origin. Id. at 15-16. Finally, DSMC noted that verification revealed errors with other aspects of Bosun’s reporting, specifically the physical characteristics used to develop control numbers. Id. at 19-21.

For the final determination Commerce, determined not to apply AFA to Bosun. IDM at 25-31. In doing so, Commerce accepted Bosun’s FIFO methodology for identifying the country of origin for U.S. sales by Pioneer and Bosun USA. IDM at 25-29. See id. at 28; see also VR at 9. Commerce found that Bosun “fully complied” with Commerce’s requests for additional information regarding this issue. See IDM at 28. Commerce ultimately concluded that the errors identified in the verification report did not justify the application of AFA because they were “isolated” to the transactions identified at verification. Id. Accordingly, Commerce found that Bosun had satisfied the “best of its ability” standard, i.e., “Bosun was [not] inattentive, careless, or inadequate in keeping the country of origin record enough to warrant AFA.” Id. (italics added). Commerce therefore based the margin for Bosun on the information Bosun provided. Id. at 25.

2. Analysis

a. Compliance with the “Best of Its Ability” Standard

DSMC contends that Bosun reasonably should have anticipated country of origin data as information necessary to this administrative proceeding, as covered by precedent, and it asks how Bosun could have possibly met the “best of its ability” standard and thereby avoided application of Court No. 17-00167 Page 8 AFA? DSMC argues Commerce has yet to provide an adequate explanation of the problem that would satisfy the standard of Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mutual Auto. Ins. Co., 463 U.S. 29, 43 (1983) (“the agency must examine the relevant data and articulate a satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made’”), quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168 (1962).

Responding, Bosun contends that DSMC’s restatement of what transpired is “myopic”; that Commerce can only resort to “facts available” if information is “missing”; and that at the end of the day, because Bosun presented a logical and verified hierarchical test using available records, Commerce concluded the information was not missing. Bosun Resp. at 3-4. More precisely, the defense supports the adequacy of Commerce’s decision on five grounds: (1) Commerce reasonably explained how Bosun’s behavior complied with the “best of its ability” standard because Bosun was “able to segregate the sales of subject merchandise using its sales identification methodology”; (2) Commerce found that Bosun was “not ‘inattentive, careless, or inadequate’” in maintaining origin information; (3) Commerce reasonably exercised its discretion not to apply adverse inferences to Bosun; (4) Commerce explained that application of adverse inferences here would be inconsistent with 19 U.S.C. §1677m(d); and (5) Commerce reasonably explained that Bosun’s origin-identification methodology met the requirements of 19 U.S.C. §1677m(e), such that application of AFA would be inappropriate. Def.’s Br. at 12-16; Bosun’s Br. at 4-6.

In holding that the “best of its ability” standard of 19 U.S.C. 1677e(b) inherently “requires that [interested parties] . . . take reasonable steps to keep and maintain full and complete records documenting the information that a reasonable [party] should anticipate being called upon Court No. 17-00167 Page 9 to produce”, Nippon Steel Corp., supra, 337 F.3d at 1382, appellate precedent has explicitly interpreted that statute to require potential interested parties to a future administrative proceeding to anticipate the information that would be necessary thereto and conduct their information acquisition and maintenance accordingly. But in this instance, it is unclear whether Commerce’s conclusion in the IDM that Bosun had met the “best of its ability” standard is on the basis of Bosun’s responses to Commerce’s requests for information, or whether Commerce’s conclusion was with respect to Bosun’s overall conduct prior to and during the review. Either instance would appear to be at odds with appellate precedent on the meaning of the “best of its ability” standard and the facts on the record at bar, and therefore remand at least for clarification, or reconsideration if Commerce deems that appropriate, is necessary.

Generally speaking, Commerce recognized that while the “best of its ability” standard does not require “perfection,” it does not condone inattentiveness, carelessness, or inadequate record keeping. IDM at 27 (referencing Nippon Steel, 337 F.3d at 1382 & 19 U.S.C. §1677m(d) & (e)). Nonetheless, Commerce in this instance appears to have elided over the fact that Bosun is not a naif in these proceedings, having been individually reviewed in the original investigation and two prior annual reviews. See IDM at 27. Commerce also appears to have elided over Bosun’s awareness of the necessity of country of origin information as an aspect of these proceedings. Bosun’s seemingly cavalier statement to Commerce, that country of origin data were “not necessary information which needed to be recorded in its computer system for purposes of its business operations” is odd, given that accurate identification of the country of origin is unquestionably necessary to distinguishing U.S. sales of subject merchandise and to determining accurate duty margins. See 19 U.S.C. § 1675(a); see, e.g., Kyocera Solar, Inc. v. United States, 41 CIT ___, ___, 253 F. Supp. 3d 1294, 1312 (2017). Court No. 17-00167 Page 10 Notwithstanding that Bosun claims to have attempted to provide that information by other means, the “best of its ability” standard “requires the respondent to do the maximum it is able to do,” inclusive of “maintain[ing] full and complete records” of relevant data. Peer Bearing Co.-Changshan v. United States, 766 F.3d 1396, 1400 (Fed. Cir. 2014); Nippon Steel, 337 F.3d at 1382. Obviously, Bosun’s origin identification methodology would not have been necessary had it “maintain[ed] full and complete records” of the origin of the sawblades sold in the United States in the first place. See id.

On the first of the defense’s points, as mentioned Commerce apparently concluded that because Bosun was “able to segregate the sales of subject merchandise using its sales identification methodology”, this meant that Bosun had cooperated to the best of its ability. Def.’s Br. at 12; IDM at 27-28. DSMC’s counter is that this explanation fails to recognize that Bosun could not fully replicate the results of its methodology for determining origin at verification. DSMC Reply at 6, referencing VR at 10-11. While this might seem a weak reed for DSMC’s argument, as there was “only” one error identified which related to the reporting of an incorrect quantity and DSMC does not address whether that is error impugning Bosun’s methodology as a whole or whether the error resulted from misreporting, either instance would seem to necessitate further clarification.

DSMC argues that Commerce appears to have applied a more “relaxed” standard to Bosun on this issue and in this particular instance, i.e., “d[id] not find that Bosun was inattentive, careless, or inadequate in keeping the country of origin record enough to warrant AFA.” See IDM at 27 (italics added). The court agrees that the standard Commerce applied in this instance does not appear to be consistent with the standard Commerce has applied in other proceedings. Lacking here is explanation of how Commerce’s conclusion comports with judicial articulations of the “best of Court No. 17-00167 Page 11 its ability” standard, or alternately how it can lawfully substitute for that standard a “looser” one that does not require a respondent to “maintain full and complete records” of relevant data. See IDM at 27. Agency action is arbitrary when insufficient reasons are offered for treating similar situations differently. E.g., Motor Vehicle Mfrs. Ass’n, 463 U.S at 57 (1983); SKF USA, Inc. v. United States, 263 F.3d 1369, 1382 (Fed. Cir. 2011).

Bosun argues that Commerce reasonably concluded that Bosun’s behavior met the requirements of Nippon Steel, because Bosun supplied all the records in its possession and developed an alternative strategy for determining origin that Commerce concluded was reliable. Bosun’s Br. at 4-6. Bosun further argues that the facts underlying Nippon Steel were distinct, involving a particularly recalcitrant respondent. Id. But again, the “best of its ability” standard requires a respondent to do the “maximum,” inclusive of “maintain[ing] full and complete records” of relevant data. Peer Bearing Co.-Changshan, 766 F.3d at 1400; Nippon Steel, 337 F.3d at 1382. Origin information is among the most basic data necessary for the calculation of a margin, and an experienced respondent would reasonably foresee the need to maintain it. Bosun apparently did not maintain the country of origin information that had been in its possession. While it may have been frank regarding its failure to do so, this court does not understand how such frankness transforms a recordkeeping insufficiency into compliance with the “best of its ability” standard. As the appellate court has indicated, the standard applies to not only malfeasant conduct but to “inattentiveness, carelessness, or inadequate recordkeeping.” Nippon Steel, 337 F.3d at 1382 (italics added).

As to the second line of defense, the defendant argues Commerce found that Bosun was not in fact “inattentive, careless, or inadequate” in its recordkeeping. Def.’s Br. at 12-13. But, the question DSMC has raised is not whether Commerce found that Bosun was appropriately Court No. 17-00167 Page 12 attentive, etc., but why it so found. To the extent Commerce found that Bosun did not act “inattentive[ly], careless[ly], or inadequate[ly]” by failing to track and maintain origin information, that finding is a conclusion. It does not amount to an explanation for a conclusion. Furthermore, while Commerce concluded that Bosun’s use of an after-the-fact, indirect methodology as a substitute for direct origin information met the “best of its ability” standard, it indicated that in any further reviews it would expect Bosun to record origin data at the time of sale, IDM at 27, which would seem to contradict finding that Bosun had, in fact, acted to the “best of its ability.”

The defense also claims, thirdly, that Commerce reasonably exercised its discretion not to apply adverse inferences. Def.’s Br. at 13-14. Within the substantial-evidence context of judicial review, once a party fails to meet the “best of its ability” standard the court would normally review an administrative decision on whether or not to apply AFA for abuse of discretion. See, e.g., POSCO v. United States, 42 CIT ___, ___, Slip Op. 18-117 at 28 (2018); Maverick Tube Corp. v. United States, 41 CIT ___, ___, 273 F. Supp. 3d 1293, 1312-13 (2017). Here, however, because Commerce found that Bosun met the “best of its ability” standard, IDM at 28-29, the exercise of discretion in whether or not to apply AFA is irrelevant, as DSMC argues. Cf. 19 U.S.C. §1677e(a) with 19 U.S.C. §1677e(b). As it stands, it is Commerce’s finding that Bosun had in fact acted to the best of its ability that DSMC contends has not been adequately explained. To the extent the defendant claims that Commerce would have acted reasonably in declining to apply adverse inferences had it been in a position to exercise its discretion (i.e., had Commerce found that Bosun did not act to the best of its ability), speculating as to what Commerce might have done in a different situation does not explain why it did what it did here. Court No. 17-00167 Page 13

The defense also argues, fourthly, that Commerce explained that it could not have applied adverse inferences consistently with 19 U.S.C. §1677m(d). Def.’s Br. at 14; Bosun’s Br. at 4. That statute requires Commerce, where it finds that a respondent has provided deficient data, to offer an opportunity for correction. 19 U.S.C. §1677m(d). If the respondent then fails to offer a “satisfactory” response, the statute specifies that Commerce may disregard the respondent’s information. Id. Conversely, where the respondent provides a “satisfactory” response to an opportunity to correct a deficiency, the statute indicates that the information should be considered. See id.

In its IDM, Commerce stated that it had questions regarding the origin methodology described in Bosun’s initial questionnaire responses and thus asked supplemental questions. IDM at 26, 28. Commerce stated that Bosun “fully complied” with these supplemental questions, indicating that it found Bosun’s overall origin data to be satisfactory. See id. But again, the question underlying DSMC’s appeal is that of why Commerce found the data “satisfactory,” i.e., why it found that Bosun acted to the “best of its ability” overall. That finding has yet to be explained in a manner that is reasonably grounded in the requirements of “best of its ability” standard, as articulated by the courts. See, e.g., Peer Bearing Co.-Changshan, 766 F.3d at 1400, quoting Nippon Steel, 337 F.3d at 1382 (indicating that the “best of its ability” standard requires respondents to “take reasonable steps to keep and maintain full and complete records documenting the information that a reasonable [respondent] should anticipate being called upon to produce”).

Lastly on this point, the defense argues Commerce explained that Bosun’s information complied with the terms of 19 U.S.C. §1677m(e), making adverse inferences inappropriate. Def.’s Br. at 15-16; Bosun’s Br. at 4; see also IDM at 28-29. Pursuant to 19 U.S.C. Court No. 17-00167 Page 14 §1677m(e), where a respondent avails itself of the opportunity to correct a deficiency under 19 U.S.C. §1677m(d), but provides information that “does not meet all applicable requirements,” Commerce shall not decline to rely on the data so long as certain pre-requisites are met. 19 U.S.C. §1677m(d) & (e). During the review, Commerce found that although it had questions about aspects of Bosun’s reporting (most notably its origin determination methodology), Bosun complied with Commerce’s requests for further information, and the methodology was reviewed at verification and found acceptable. IDM at 28. Commerce also noted that Bosun’s inability to recreate the reported results of its application of the FIFO step of the methodology affected only one of the three or four sales for which this step was reviewed at verification. Id.; see also VR at 10-11. Accordingly, Commerce found that the pre-requisites of 19 U.S.C. §1677m(e) were met, such that it would be inappropriate for it to decline to rely on Bosun’s reported origin information. IDM at 28.

Commerce’s explanation, however, does not acknowledge that among 19 U.S.C. §1677m(e)’s pre-requisites is that the respondent demonstrate that it “acted to the best of its ability.” 19 U.S.C. §1677m(e)(4). Thus, any finding that Bosun’s origin reporting methodology complied with 19 U.S.C. §1677m(e) does not itself adequately explain why Commerce determined that Bosun “acted to the best of the ability” for purposes of 19 U.S.C. §1677e(a)-(b), despite failing to maintain direct origin data that was, or had been, within its possession. At best, it means that Commerce has failed to explain its determination that Bosun “acted to the best of its ability” both in applying 19 U.S.C. §1677e(a)-(b) and in applying 19 U.S.C. §1677m(e). And at this point, DSMC’s fundamental argument -- that Commerce has not adequately explained how Bosun’s efforts to compensate for a lack of recordkeeping satisfies a standard that is fundamentally concerned with a respondent’s duty to “keep and maintain full and complete records”, Nippon Steel, 337 F.3d at 1382 -- is persuasive. Court No. 17-00167 Page 15

2. Commerce’s Conclusion on the Reliability of Bosun’s Data

DSMC also challenges Commerce’s finding that errors in Bosun’s reported information did not indicate a failure to comply with the “best of its ability” standard. DSMC Br. at 17-20. As mentioned, at verification Commerce found that Bosun could not replicate the results of the FIFO step of its origin-identification methodology for one of the three or four pre-selected sales reviewed. VR at 10-11. Commerce also learned that Bosun had misidentified the physical characteristics of the goods included in certain sales. Id. at 2, 7. These errors affected three out of sixteen transactions examined at verification, and involved multiple product characteristics and more than one model of merchandise. Id.; see also IDM at 31. Commerce nonetheless found that Bosun had acted to the best of its ability in reporting its data, concluding that the errors identified at verification were isolated and did not implicate the overall reliability of Bosun’s reported data. IDM at 29-31.

Commerce’s explanation for this decision is here defended as adequate for two reasons. Def.’s Br. at 15-17; Bosun’s Br. at 6-9. The defendant argues that Commerce explained that Bosun’s data met the requirements of 19 U.S.C. §1677m(e). Def.’s Br. at 15-16. The defense also argues that Commerce adequately explained its conclusion that Bosun’s errors were isolated and not indicative of unreliability in the reported data as a whole. Id. at 16-17; Bosun’s Br. at 6-9.

First argued is that Commerce’s explanation that Bosun’s information complied with 19 U.S.C. §1677m(e) and regardless of any errors identified at verification. Def.’s Br. at 15-16; see also IDM at 28-29. However, as discussed above, 19 U.S.C. §1677m(e) premises the use of nonstandard data on a respondent’s compliance with the “best of its ability” standard. Accordingly, to conclud ethat a respondent acted to the best of its ability because its data complies with 19 U.S.C. Court No. 17-00167 Page 16 §1677m(e) is akin to concluding that it acted to the best of its ability because it acted to the best of its ability. Such a tautology does not rise to the level of “a satisfactory explanation for [agency] action including a ‘rational connection between the facts found and the choice made.’” Motor Vehicle Mfrs., 463 U.S. at 43, quoting Burlington Truck Lines, 371 U.S. at 168. See, e.g., Husteel Co. v. United States, 40 CIT ___, ___, 180 F. Supp. 3d 1330, 1348 (2016), aff’d, 710 F. App’x 890 (Fed. Cir. 2018).

Next argued is that Commerce adequately explained its view that the errors identified at verification were “isolated,” and otherwise not indicative of problems in Bosun’s reporting. Def.’s Br. at 16-17; Bosun’s Br. at 6-9. The defense notes that Commerce explained that the error identified in Bosun’s application of the FIFO step of its origin-identification methodology affected only one of the three or four sales for which this step was reviewed at verification, while the errors identified in Bosun’s physical characteristics reporting affected only two of the product codes examined at verification. Def.’s Br. at 16-17. For its part, Bosun argues that the agency acted diligently at verification, and that discovery of errors is common and does not necessarily implicate the overall reliability of a respondent’s prior-reported information. Bosun’s Br. at 6-9. However, verification is meant to spot-check the accuracy of a respondent’s prior-reported data. See, e.g., Hyundai Steel Co. v. United States, 42 CIT ___, ___, 282 F. Supp. 3d 1332, 1350 (2018), referencing Micron Tech., Inc. v. United States, 117 F.3d 1386, 1396 (Fed. Cir. 1997). Here, in conducting that spot check, Commerce identified a mistake in Bosun’s application of its FIFO methodology, which affected one out of the three or four sales traces for which Commerce reviewed the FIFO methodology. Id.; see also VR at 10. Given a maximum sample size of four sales traces, Court No. 17-00167 Page 17 Commerce’s conclusion that the error was “isolated” and did not affect other sales is not sufficiently explained. IDM at 29-30.

Bosun also argues that DSMC has mischaracterized the importance of the FIFO step of its origin-identification methodology. Bosun’s Br. at 6-7. Bosun describes this step as applying to a “small number of sales.” Id. at 7, quoting VR at 10. The verification report is not ideally clear with respect to the quantum of Bosun’s sales for which origin was determined pursuant to the FIFO step, but it does indicate that the agency reviewed the FIFO step’s application to three or four transactions, and found an error in one of the transactions reviewed, i.e., an error effecting 25%-33% of reviewed transactions. As such, even if the FIFO step was applied only in the last resort, Commerce has yet to explain its conclusion that the error discovered at verification was not replicated in other sales, which were not reviewed at verification, to which the FIFO step applied. Further, because Bosun was unable to explain how the error was made, the particularities of the error do not themselves suggest that the error was limited to this single sales trace. VR at 10-11. Nor did Commerce identify anything unique about the affected sales trace that indicates that the mistake did not affect other, unreviewed sales. IDM at 29-30.

Commerce’s explanation for concluding that errors in reporting physical characteristics were not widespread is similarly insufficient. Id. at 31. The errors affected three out of sixteen transactions identified at verification, and related to multiple product characteristics. Id.; see also VR at 2, 7. Nor were the errors limited to Bosun’s reporting of a single model of merchandise. IDM at 30-31; VR at 2, 7. Similarly, Commerce’s conclusion that errors identified in Bosun’s reporting at verification did not indicate Bosun’s failure to act to the best of its ability is insufficiently explained. While Commerce found that Bosun’s data met the standards of 19 U.S.C. Court No. 17-00167 Page 18 §1677m(e), that statute incorporates the “best of its ability” standard, such that mere reference to the statute does not logically or adequately support Commerce’s conclusion. Further, while Commerce downplayed the errors identified at verification, those errors affect a substantial percentage of the sample transactions reviewed at verification, and Commerce has not pointed to anything unique about the reviewed samples that would suggest such errors were limited to those samples.

As previously noted, the standard of review requires Commerce to “articulate a satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made[,]’” Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43, quoting Burlington Truck Lines, Inc., 371 U.S. at 168, and the defense here underscores the lack of adequate explanation for Commerce’s finding and its tension with judicial articulations of the “best of its ability” standard. Remand of Commerce’s conclusion that Bosun acted to the best of its ability is thus necessary for further clarification and/or reconsideration.

B. Selection of Surrogate Values

DSMC’s second challenge is to Commerce’s selection of Thai import statistics for the copper powder and copper iron clab factor-of-production (“FOP”) as the “best available information” for Bosun. See 19 U.S.C. §1677b(c)(1).

1. Background

The submission of potential surrogate country data from interested parties for the record obligates Commerce to determine the evidence that represents the “best available information.” See id. Commerce’s publically announced policy in that regard is that it “normally will value all factors in a single surrogate country.” 19 C.F.R. §351.408(c)(2). Further, because the AD statute “does not mandate that Commerce use any particular data source” in reaching a particular Court No. 17-00167 Page 19 determination, Guangdong Chemicals Import & Export Corp. v. United States, 30 CIT 1412, 1416, 460 F. Supp. 2d 1365, 1368-69 (2006), Commerce is conferred “wide” discretion to determine what constitutes the best available FOP information. See Nation Ford Chemical Co. v. United States, 166 F.3d 1373, 1377 (Fed. Cir. 1999).

Each party placed on the record data from Thailand to serve as the surrogate values (“SVs”) for most FOPs. See DSMC SV Comments at Ex. 1; Bosun SV Comments at Ex. SV-1; Fengtai SV Comments at Ex. 1. To value copper powder and copper iron clab in particular, DSMC also provided South African import statistics under Harmonized Tariff Schedule (“HTS”) 7406.10, and DSMC placed on the record import statistics under HTS 7406.10 for each additional country, as well as the United States, identified on Commerce’s “OP List” as economically comparable to China (Bulgaria, Ecuador, Mexico, and Romania), as well as the United States, in support of it selection of South African data to value copper powder and copper iron clab. Letter from Wiley Rein LLP to Sec’y Commerce, re: Diamond Sawblades and Parts Thereof from the People’s Republic of China: DSMC’s Rebuttal Surrogate Value Comments (May 31, 2016) at Ex. 1C, PDoc 195-196. Those data are as follows:

Country AUV (USD/kg) Thailand 1.32 Mexico 4.92 Romania 8.35 United States 9.68 South Africa 12.14 Bulgaria 15.50 Ecuador 59.24

DSMC Case Brief at 15-16. Bosun and Fengtai also proposed Thai import statistics under HTS 7406.10 to value these inputs. Bosun SV Comments at Ex. SV-1; Fengtai SV Comments at Ex. 1. Court No. 17-00167 Page 20

DSMC argued in its comments prior to Commerce’s preliminary determination that the South African import statistics should be used to value the copper powder and copper iron clab. Letter from Wiley Rein LLP to Sec’y Commerce, re: Diamond Sawblades and Parts Thereof from the People’s Republic of China: DSMC’s Pre-Preliminary Determination Comments Regarding Surrogate Values (Nov. 15, 2016) at 2-4, PDoc 340. DSMC argued that the record demonstrated that the Thai import statistics for HTS 7406.10 were aberrational, as the average unit value (“AUV”) calculated from these data was substantially below the AUVs based on the import data from all other countries on the record. Id. at 3.

In its preliminary determination, Commerce relied on Thai import statistics under HTS 7406.10 to value copper powder and copper iron clab. Memorandum from Yang Jin Chun, Senior Int’l Trade Compliance Analyst, AD/CVD Operations, Off. I, Through Minoo Hatten, Program Manager, AD/CVD Operations, Off. I, to The File, re: Diamond Sawblades and Parts Thereof from the PRC: Surrogate Values for the Preliminary Results of Review (Dec. 5, 2016) at Ex. 2, PDoc 343. In making this determination, Commerce did not address DSMC’s arguments. See id. at 3-8; Memorandum from Christian Marsh, Deputy Assistant Sec’y for Antidumping and Countervailing Duty Operations, to Paul Piquado, Assistant Sec’y for Enf’t and Compliance, re: Diamond Sawblades and Parts Thereof from the PRC: Decision Memorandum for Preliminary Results of Antidumping Duty Administrative Review; 2014-2015 (Dec. 5, 2016), PDoc 345.

DSMC then argued again in its case brief that South African, not Thai, import statistics should be used to value copper powder and copper iron clab. DSMC Case Brief at 13-16. DSMC highlighted that Commerce is required to rely on the best available information in valuing factors of production and argued that the AUV based on Thai import statistics was substantially Court No. 17-00167 Page 21 below the AUV based on import statistics from every other country identified as economically comparable to the PRC. Id. at 13-15. DSMC also noted that the AUV based on South African import statistics was well within the range of AUVs calculated for other economically comparable countries and was consistent with the AUV based on U.S. import statistics. Id. at 15-16.

In rejecting the argument that South African and not Thai import statistics constituted the best available information, in its final determination Commerce found that DSMC did not provide evidence to support the claim that the Thai surrogate value was aberrational. IDM at 47-48. In doing so, Commerce pointed to its preference for relying on surrogate values from a single source and stated that DSMC had not “substantiated its claim” that the Thai AUV was aberrational, because the “mere appearance of an AUV on the high or low end of a range of AUVs is not generally sufficient to determine that an AUV at one end is aberrational” and the record did not contain historical data that “would permit us to evaluate whether this Thai AUV is aberrational.” Id. at 48. Accordingly, for the final results, Commerce continued to use Thai import statistics to value copper powder and copper iron clab.

2. Analysis

The parties agree that import data for HTS 7406.10 is appropriate for valuing the copper powder and copper iron clab inputs. See IDM at 47-48. DSMC acknowledges Commerce’s preference to value all FOPs with costs from a single country, if possible. See 19 C.F.R. §351.408(c)(2). DSMC argues that Commerce failed to rely on the best available information for the provision as required by the statute insofar as Commerce has not provided any rationale for its decision to rely on the Thai AUV for HTS 7406.10. Court No. 17-00167 Page 22

Commerce’s usual practice for determining whether data is aberrational is to require a quantitative analysis, comparing data from economically comparable countries and/or historical data from the country at issue, in order to determine if the data is unreliable or an outlier.4 Here the defendant, echoed by Bosun, reiterates Commerce’s usual preference for valuing inputs using data from the primary surrogate country and supports Commerce’s conclusion that DSMC did not place “sufficient data”to undermine finding the Thai data aberrant by contending that Commerce’s explanation for that decision (i.e., “the mere appearance of an AUV on the high or low end of a range of AUVs is not generally sufficient to determine that an AUV at one end is aberrational”) is sufficient by reference to various rulings that it believes are analogous. Def’s Resp. at 19, citing IDM at 48 (citing Polyethylene Terephthalate Film, Sheet, and Strip from the PRC, 80 Fed. Reg. 33241 (June 11, 2015) (final results admin. review), and accompanying issues and decision

4 See, e.g., Tri Union Frozen Products, Inc. v. United States, 41 CIT ___, ___, 254 F. Supp. 3d 1290, 1296 (2017), appeal filed, No. 17-2523 (Fed. Cir. Sep. 8. 2017); Ad Hoc Shrimp Trade Action Committee v. United States, 41 CIT ___, ___, 234 F. Supp. 3d 1315, 1319 (2017); Jinxiang Chengda Import & Export Co. v. United States, 37 CIT ___, ___, Slip Op. 13-40 at 10 (2013) (“clear from the record” that transfer price was not only high in comparison to Customs AUV but an outlier vis-a-vis the data thereof), aff’d, 553 Fed. App’x 1005 (Fed. Cir. 2014); Nan Ya Plastics Corp. v. United States, 37 CIT ___, ___, 906 F. Supp. 2d 1348, 1353-54 (2013) (on claim that “interquartile range comparison (applied here) is the average dumping margin for the 25% of sales with the lowest dumping margin compared to the average dumping margin for the 25% of sales with the highest dumping margin” plaintiff “presented what appear to be good and compelling statistical arguments that test the reasonableness of Commerce’s total AFA rate”, which Commerce needed to address), aff’d on other grounds, 810 F.3d 1333 (Fed. Cir. 2016); Tianjin Tiancheng Pharm. Co. v. United States, 29 CIT 256, 261 n.4 (2005) (“a simple modal analysis would have allowed Commerce to show its reasons for disregarding that month’s value with greater precision”), referencing Laurence C. Hamilton, Data Analysis for Social Scientists 78-82 (Duxbury Press, 1996). Cf. Ozdemir Boru San. ve Tic. Ltd. v United States, 42 CIT ___, ___, 282 F. Supp. 3d 1352, 1352 (2018) (“there is a reasonable basis for treating the Istanbul and Yalova land parcels as outliers”).

Court No. 17-00167 Page 23 memorandum5 at 17), and Camau Frozen Seafood Processing Import Export Corp. v. United States, 37 CIT ___, ___, 929 F. Supp. 2d 1352, 1356 n.9 (2013) (“Camau Frozen Seafood”) (“[o]n this record, the Bangladeshi data is not aberrational, it is merely the lowest price in a range of prices”). More broadly, the defendant and Bosun contend that Commerce examines the record for import values for other potential surrogate countries and also historical data and determines if such values or data support concluding that a particular value is aberrational. Both emphasize that the record here lacked such historical information, albeit amplifying certain of the statements therein. See Def’s Resp. at 19-20; Bosun’s Resp. at 10.

DSMC responds that while historical data could provide additional information on whether a particular data set is aberrational, Commerce failed to explain why the lack of such data “necessarily” means that the Thai data it selected are not aberrational and/or are the best available information. DSMC Reply at 18 (Commerce “has not explained why contemporaneous data are insufficient, or conversely, why historical data is required”), referencing IDM at 47-48. The court, however, does not interpret Commerce’s statement regarding historical data to be a requirement for the record; Commerce’s comment on the lack thereof on this record simply meant that the record lacked an aspect that would otherwise have assisted in determining aberrancy or normality. Accord, e.g., Certain Activated Carbon from the PRC, 80 Fed. Reg. 61172 (Oct. 9, 2015) (final results of antidumping duty admin. rev.; 2013-2014), and accompanying I&D Memo at 31 (“the record does not contain historical data . . . which would permit us to evaluate whether the[ ] data are aberrational”).

5 Non-IDM memoranda hereinafter “I&D Memo”.

Court No. 17-00167 Page 24

On the other hand, DSMC did provide a type of quantitative analysis. Commerce’s preference for valuing all FOPs in a single country may be reasonable where there is no reason to suspect that data from the primary surrogate country are inferior, Jacobi Carbons AB v. United States, 38 CIT ___. ___, 992 F. Supp. 2d 1360, 1376 (2014), but DSMC pointed out that the AUV of the Thai import data is almost four times lower than the next-higher value and vastly below all other AUVs on the record, and that excluding the AUVs for Thailand and Ecuador (the latter of which being nearly eight times greater than the Thai AUV)6, the average “inner” AUV is $10.12. DSMC’s fundamental argument is that the Thai AUV is an outlier.

On this issue, there is weakness in the analyses of both sides. Commerce did not adequately address DSMC’s argument that the Thai AUV is an outlier, but at the same time DSMC did not further advance its argument on aberrancy, arguing only as to the degree to which the Thai AUV figure differed from the other data points rather than, for example, putting on the record further evidence or any recognized statistical argument on outlier data.7 Nonetheless, given that Commerce did not directly address the substance of the DSMC’s argument, the standard of review of substantial evidence compels remand in this instance, because one cannot discern the “leap” Commerce has

6 DSMC contends the 59.24 AUV for Ecuador is aberrant, and statistical analysis bears that out. See infra, note 9. 7 E.g., Chauvenet’s criterion; Grubbs’ test; Dixon’s Q test; et cetera. See, e.g., Vic Barnett and Toby Lewis, Outliers in Statistical Data (1994 Wiley, 3rd ed); Douglas M. Hawkins, Identification of Outliers (1980 Chapman and Hall). Applying John Tukey’s “fences,” for example: the median of all the data points is 9.68; the first and third quartile values are 6.635 and 13.82, respectively; the inter-quartile range is 7.185; and outliers would be points less than -4.1425 or greater than 24.5975.

Court No. 17-00167 Page 25 made in the IDM as to the reliability of the Thai AUV in the face of the unaddressed DSMC’s arguments on the”aberrancy” of the Thai AUV on the record.

The regulatory preference for valuing inputs for an NME respondent’s production using data from a single “primary” surrogate country has been held insufficient to explain decisions to reject data from a non-primary surrogate country if questions remain unanswered as to the suitability of the primary surrogate country data. See Calgon Carbon Corp.v. United States, 40 CIT ___, ___, 145 F. Supp. 3d 1312, 1327-28 (2016) (“Commerce, by relying on its single surrogate country preference and nothing more, improperly rejected other SVs for anthracite coal derived from POR6-contemporaneous data from other countries. This ‘preference’. . . carries the day only when it is used to ‘support a choice of data as the best available information where the other available data ‘upon fair comparison, are otherwise seen to be fairly equal’’”) (quoting, inter alia, Peer Bearing Co.-Changshan v. United States, 35 CIT 1626, 1643, 804 F. Supp. 2d 1337, 1353 (2011)); Camau Frozen Seafood Processing Import Export Corp. v. United States, 37 CIT ___.___, 929 F. Supp. 2d 1352, 1355 (2013) (“it is not sufficient for Commerce to cite the policy of using a single surrogate country where, as here, there is reason to believe that the primary surrogate country may not provide the best available information for a particular FOP”); Peer Bearing Co.-Changshan v. United States, 35 CIT 103, 124, 752 F. Supp. 2d 1353, 1373 (2011) (“because the statute requires Commerce to compare the chosen data set with other data sets on the record and thereby determine what is the best available information, the regulatory preference cannot suffice as adequate reasoning if it is the only factor that Commerce considers”), vacated on other grounds, 766 F.3d 1396 (Fed. Cir. 2014).

In sum, Commerce has not adequately explained its determination to reject South African surrogate values for Bosun’s copper powder and copper iron clab or to rely on Thai Court No. 17-00167 Page 26 surrogates for these inputs. Rather, Commerce appears to have relied solely on regulatory preferences that have previously been deemed insufficient in the face of unaddressed arguments, and/or otherwise provided conclusory rationales for its choice, without engaging in a reasoned analysis of the record data. Commerce’s selection of the surrogate values for copper powder and copper iron clab will therefore be, and hereby are, remanded for further consideration.

C. Separate Rate

In view of the foregoing, remand of the margin calculation for the separate respondents is also necessary.

III. Conclusion

For the above reasons, the case must be, and hereby is, remanded to the International Trade Administration, U.S. Department of Commerce for further proceedings consistent with this opinion. The results of remand shall be due February 20, 2019. By the fifth business day after the filing thereof with the court, the parties shall confer and file a joint status report as to a proposed scheduling of comments, if any, on those results.

So ordered.

/s/ R. Kenton Musgrave R. Kenton Musgrave, Senior Judge

Dated: October 23, 2018 New York, New York