Jiangsu Zhongji Lamination Materials Co., (HK) Ltd. v. United States

2023 CIT 84

• Court: United States Court of International Trade
• Decided: June 7, 2023
• Docket: 21-00138
• Status: Published

Opinion

Slip Op. 23-84

UNITED STATES COURT OF INTERNATIONAL TRADE

Court No. 21-00138

JIANGSU ZHONGJI LAMINATION MATERIALS CO., (HK) LTD., et al., Plaintiffs, v. UNITED STATES, Defendant, and ALUMINUM ASSOCIATION TRADE ENFORCEMENT WORKING GROUP AND ITS INDIVIDUAL MEMBERS, et al., Defendant-Intervenors.

Before: M. Miller Baker, Judge

OPINION

[The court denies Plaintiffs’ motion for judgment on the agency record, sustains the Department of Com- merce’s determination, and grants judgment on the agency record to Defendant and Defendant-Interve- nors.]

Dated: June 7, 2023 Ct. No. 21-00138 Page 2

Jeffrey S. Grimson, Sarah M. Wyss, Bryan P. Cenko, and Wenhui “Flora” Ji, Mowry & Grimson, PLLC, of Washington, DC, on the papers for Plaintiffs.

Brian M. Boynton, Acting Assistant Attorney General; Patricia M. McCarthy, Director; Reginald T. Blades, Jr., Assistant Director; and Catharine M. Parnell, Trial Attorney, Civil Division, U.S. Department of Jus- tice of Washington, DC, on the papers for Defendant. Of counsel for Defendant was JonZachary Forbes, Staff Attorney, Office of the Chief Counsel for Trade Enforcement and Compliance, U.S. Department of Commerce of Washington, DC.

John M. Herrmann, Paul C. Rosenthal, Joshua R. Mo- rey, and Grace W. Kim, Kelley Drye & Warren LLP of Washington, DC, on the papers for Defendant-Interve- nors.

Baker, Judge: A Chinese aluminum foil producer challenges the Department of Commerce’s imposition of antidumping duties in an administrative review based on the Department’s calculation of surrogate values, denial of a double remedies adjustment, and liquidation instructions. Finding it supported by sub- stantial evidence, the court sustains the determina- tion. Ct. No. 21-00138 Page 3

I

A

An antidumping duty represents the amount by which the “normal value” of subject merchandise ex- ceeds its “export price.” 19 U.S.C. § 1673. If an inves- tigation involves a non-market economy such as China, then Commerce determines normal value using surrogate values for “the factors of production utilized in producing the merchandise,” along with “an amount for general expenses and profit plus the cost of contain- ers, coverings, and other expenses.” Id. § 1677b(c)(1).

“In selecting surrogate values, Commerce ‘attempts to construct a hypothetical market value of [the sub- ject merchandise] in the [nonmarket economy].’ ” Changzhou Trina Solar Energy Co. v. United States, 975 F.3d 1318, 1330 (Fed. Cir. 2020) (first alteration in original) (quoting Downhole Pipe & Equip., L.P. v. United States, 776 F.3d 1369, 1375 (Fed. Cir. 2015)). The Department values factors of production, “to the extent possible,” using data from surrogate countries that have market economies and that are (A) “at a level of economic development comparable to that of the nonmarket economy country,” and (B) “significant producers of comparable merchandise.” 19 U.S.C. § 1677b(c)(4)(A)–(B).

By statute, the Department must value factors of production using the “best available information re- garding the values of such factors in a market economy Ct. No. 21-00138 Page 4

country or countries considered” appropriate. Id. § 1677b(c)(1); see also Seah Steel VINA Corp. v. United States, 950 F.3d 833, 842 (Fed. Cir. 2020); Dorbest Ltd. v. United States, 462 F. Supp. 2d 1262, 1268 (CIT 2006). Because the statute is silent about what consti- tutes the “best available information,” Commerce has “broad discretion” in deciding what record evidence meets the criteria. Zhejiang DunAn Hetian Metal Co. v. United States, 652 F.3d 1333, 1341 (Fed. Cir. 2011); Nation Ford Chem. Co. v. United States, 166 F.3d 1373, 1377 (Fed. Cir. 1999). “Commerce’s analysis when selecting the ‘best available information’ on the record inherently involves a comparison of the compet- ing data sources to identify what available information is ‘best’ to value factors of production . . . .” Weishan Hongda Aquatic Food Co. v. United States, 917 F.3d 1353, 1367 (Fed. Cir. 2019) (citing 19 U.S.C. § 1677b(c)(1)); see also Ass’n of Am. Sch. Paper Suppli- ers v. United States, 716 F. Supp. 2d 1329, 1334 (CIT 2010).

In practice, the Department selects surrogate val- ues that are product-specific, representative of a broad market average, publicly available, contemporaneous with the review period, and exclusive of tax and duty. See Import Administration Policy Bulletin 04.1, Non- Market Surrogate Country Selection Process at 4 Ct. No. 21-00138 Page 5

(Mar. 1, 2004) (Policy Bulletin); 1 Jiaxing Brother Fas- tener Co. v. United States, 822 F.3d 1289, 1294 (Fed. Cir. 2016) (citing 19 C.F.R. § 351.408(c)(2)); see also Changzhou Trina Solar, 975 F.3d at 1331.

Commerce’s data need not be perfect. Jiaxing, 822 F.3d at 1301. And the Department need not duplicate a manufacturer’s precise experience. See Nation Ford, 166 F.3d at 1377. Instead, it seeks information that “most accurately represents the fair market value.” Id. at 1377.

B

The antidumping statute requires the Department to avoid imposing a double remedy when it simultane- ously imposes countervailing duties and antidumping duties based on its non-market economy calculation methodology. See 19 U.S.C. § 1677f-1(f). This issue arises in non-market economy cases because the use of surrogate values—that is, values from countries other than the non-market economy country at issue— means that a countervailable subsidy “is not embed- ded in the price used as normal value. Consequently, the subsidy could potentially be remedied both by the [countervailing duty] and by the [antidumping duty].”

1 http://enforcement.trade.gov/policy/bull04-1.html. The pincite above is to a .PDF printout of the Policy Bulletin webpage. Ct. No. 21-00138 Page 6

Vicentin S.A.I.C. v. United States, 404 F. Supp. 3d 1323, 1339 n.26 (CIT 2019).

In applying § 1677f-1(f), the Department examines (1) whether a countervailable subsidy has been pro- vided; (2) whether that subsidy has been shown to have reduced the average price of imports during the relevant period; and (3) whether Commerce can rea- sonably estimate the extent to which that countervail- able subsidy, in combination with the use of normal value determined under 19 U.S.C. § 1677b(c), has in- creased the weighted-average dumping margin for the class or kind of merchandise. Appx02468 (citing 19 U.S.C. § 1677f-1(f)(1)(A)–(C)). For a subsidy meet- ing these criteria, the statute requires the Department to reduce the antidumping rate by the estimated amount of the increase in the weighted-average dump- ing margin, subject to a specified cap. Id. (citing 19 U.S.C. § 1677f-1(f)(1)–(2)).

II

In 2019, Commerce opened the first administrative review of duties on aluminum foil from China covering November 2, 2017, through March 31, 2019. See Initi- ation of Antidumping and Countervailing Duty Ad- ministrative Reviews, 84 Fed. Reg. 27,587, 27,589 (Dep’t Commerce June 13, 2019), Appx25964, Appx25966. The Department selected two mandatory respondents: (1) Jiangsu Zhongji Lamination Ct. No. 21-00138 Page 7

Materials Co., (HK) LTD (Zhongji); and (2) Xiamen Xiashun Aluminum Foil Co. See Appx02464.

Zhongji submitted surrogate value comments rec- ommending that the aluminum dross/ash produced as a byproduct in its aluminum foil production process should be classified under Harmonized Tariff Sched- ule (HTS) heading 7602.00.19. Appx28870. It also rec- ommended classifying rolling oil and rolling oil addi- tive it uses in its production of aluminum foil under HTS 2710.12.21. Appx28868. Finally, the company suggested using information from Xeneta AS, a Nor- wegian shipping company, or the Descartes Group, a Canadian logistics company, to calculate international freight costs. Appx01026; Appx02483.

As to double remedies, Zhongji argued that its ma- terial inputs are subsidized according to a countervail- ing duty determination by the Department, and that those subsidies lower its costs. Appx02489–02490. In connection with the second part of the statutory double remedy test, the company explained that the price of all its aluminum materials—inputs bought and foil sold—is directly related to the London Metal Ex- change ingot price, which the company contends fur- ther showed that its aluminum prices declined during the review period. See Appx02490.

Zhongji then asked Commerce to modify its liquida- tion instructions. See generally Appx02488–02489. The company explained that some of its customers re- invoiced sales prior to import and thus the importer of Ct. No. 21-00138 Page 8

record could differ from the final customer. Id. Conse- quently, it worried that U.S. Customs and Border Pro- tection might liquidate its imports at the China-wide rate. Id. To avoid this, Zhongji requested that the De- partment insert the words “resold or imported” into the instructions. Appx02488.

Commerce published the preliminary results in June 2020 and calculated a dumping margin of zero percent for Zhongji. See Appx02465.

In the final determination, however, that rate changed to 23.62 percent. See Certain Aluminum Foil from the People’s Republic of China: Final Results of Antidumping Duty Administrative Review; Final De- termination of No Shipments; 2017–2019, 86 Fed. Reg. 11,499, 11,500 (Dep’t Commerce Feb. 25, 2021), Appx3640. The final determination measured Zhongji’s aluminum ash byproduct using the four-digit HTS code 2620. See Appx02485–02486. Commerce re- jected the company’s proposed HTS code 7602.00.19, concluding that at the four digit-level HTS 7602 cor- rectly covers scrap, cuttings, and other byproducts, but not ash, and emphasized that because Zhongji re- ported its byproduct as “dross/ash,” HTS 7602 was not the most specific category available. Appx02486.

The Department used HTS 3403.99 and HTS 3811.90 to value Zhongji’s rolling oil and rolling oil ad- ditive, finding that the company’s “description of these Ct. No. 21-00138 Page 9

inputs available on the record is not sufficiently de- tailed to support selection of the eight-digit HTS cate- gories” Zhongji recommended using instead. Appx02487; see also Appx01241, Appx01039. Com- merce used Maersk data for international freight ra- ther than data from Xeneta 2 or Descartes. Appx02487.

The Department also found that evidence submit- ted by Zhongji in its double remedies response was in- sufficient to establish a subsidy-to-cost link or a cost- to-price link. Appx02490. Commerce therefore did not make a double remedies adjustment to the company’s purchases of primary aluminum, aluminum plate, and electricity. Appx02490–02492. Finally, the Depart- ment rejected Zhongji’s request to modify its liquida- tion instructions, finding it unnecessary. Appx02489.

The company then commenced this litigation seek- ing relief under 19 U.S.C. §§ 1516a(a)(2)(A)(i)(I) and 1516a(a)(2)(B)(iii). ECF 8, ¶ 2 (complaint). 3 The Alu- minum Association Trade Enforcement Working Group intervened to support the government. ECF 20. Zhongji then moved for judgment on the agency record. See USCIT R. 56.2; see also ECF 25 (motion), ECF 47

2Commerce rejected the Xeneta data mainly because it de- termined that the source is not publicly available. See Appx02487. 3Three days later, Commerce published an amended final determination to correct “ministerial errors” relating to other companies. Zhongji’s margin remained the same. See Appx04450–04452. Ct. No. 21-00138 Page 10

(public brief), ECF 49 (confidential brief). The govern- ment (ECF 30) and the Association (ECF 31, confiden- tial; ECF 32, public) opposed and the company replied (ECF 48, public; ECF 50, confidential). Zhongji later requested, with the consent of the other parties, that the court decide the case without oral argument, and the court obliges that request. ECF 51.

III

The court has subject-matter jurisdiction under 28 U.S.C. § 1581(c).

In 19 U.S.C. § 1516a(a)(2) actions such as this, “[t]he court shall hold unlawful any determination, finding, or conclusion found . . . to be unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i). That is, the question is not whether the court would have reached the same decision on the same record— rather, it is whether the administrative record, taken as a whole, permits the Department’s conclusion.

Substantial evidence has been defined as more than a mere scintilla, as such relevant evidence as a reasonable mind might accept as adequate to support a conclusion. To determine if substan- tial evidence exists, we review the record as a whole, including evidence that supports as well as evidence that fairly detracts from the sub- stantiality of the evidence. Ct. No. 21-00138 Page 11

Nippon Steel Corp. v. United States, 337 F.3d 1373, 1379 (Fed. Cir. 2003) (cleaned up).

Specifically for surrogate value calculations, the “court’s duty is ‘not to evaluate whether the infor- mation Commerce used was [actually] the best availa- ble, but rather whether a reasonable mind could con- clude that Commerce chose the best available infor- mation.’ ” Zhejiang, 652 F.3d at 1341 (quoting Gold- link Indus. Co. v. United States, 431 F. Supp. 2d 1323, 1327 (CIT 2006)). Affirming the Department’s deter- mination “requires a reasoned explanation from Com- merce that is supported by the administrative record.” Dorbest, 462 F. Supp. 2d at 1269–70; see also Longkou Haimeng Mach. Co. v. United States, 617 F. Supp. 2d 1363, 1369 (CIT 2009).

In addition, Commerce’s exercise of discretion in § 1516a(a)(2) cases is subject to the default standard of the Administrative Procedure Act, which authorizes a reviewing court to “set aside agency action, findings, and conclusions found to be . . . arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A); see Solar World Amer- icas, Inc. v. United States, 962 F.3d 1351, 1359 n.2 (Fed. Cir. 2020) (explaining that in § 1516a cases, i.e., cases brought under section 516A of the Tariff Act of 1930, APA “section 706 review applies since no law provides otherwise”) (citing 28 U.S.C. § 2640(b)). “[I]t is well-established that an agency action is arbitrary when the agency offers insufficient reasons for Ct. No. 21-00138 Page 12

treating similar situations differently.” See SKF USA Inc. v. United States, 263 F.3d 1369, 1382 (Fed. Cir. 2001) (cleaned up).

IV

Zhongji’s challenges to the final determination fall into three broad categories: (a) challenges to the calcu- lation of surrogate values, (b) a challenge to the denial of a double remedies adjustment, and (c) a challenge to the liquidation instructions. The court considers these in turn.

Zhongji argues that Commerce’s selection of surro- gate values for its aluminum ash, rolling oil and roll- ing oil additive, and international freight is unsup- ported by substantial evidence because the selections did not reflect the best available information on the record. ECF 47, at 2.

Specifically, the company argues that the Depart- ment selected an overly broad HTS code that was not specific to the aluminum ash it used; that Commerce failed to provide a reasoned explanation for selecting data to value Zhongji’s rolling oil and rolling oil addi- tive that differed from evidence the company submit- ted; and that the Department failed to provide a rea- soned explanation for valuing the company’s interna- tional freight using Maersk data rather than the Des- cartes or Xeneta datasets. Id. at 2–3. Ct. No. 21-00138 Page 13

In all three surrogate value calculations Zhongji challenges, imperfect data forced the Department to make compromises. For aluminum ash, it reasonably selected a second-best category that emphasized ash rather than aluminum. As to rolling oil, it reasonably chose a broader code rather than a narrower code be- cause the company’s data did not compel a narrower choice. Finally, the Department reasonably selected the Maersk database for shipping cost comparisons, because on at least some of the criteria outlined in Commerce’s Policy Bulletin the Maersk dataset was superior. All these decisions met the substantial evi- dence standard for the reasons stated below. There- fore, this court sustains the agency’s surrogate value calculations for rolling oil, rolling oil additive, and commercial shipping.

In its final determination, Commerce relied on Bul- garian data from the Global Trade Atlas for HTS 2620 to value Zhongji’s aluminum ash byproduct. See Appx02505; see also Appx01248. The company argues that data it submitted show that the Department should instead have used the more specific HTS code 7602.00.19. See Appx28870, Appx28983.

The tariff schedule applicable to Bulgaria includes an HTS code (subheading 2620.40) that provides for aluminum ash—the exact material Zhongji reported as a byproduct. See Appx28287, Appx28290, Appx28327–28332; Appx28992–28994. During the ap- plicable period of review, however, there were no Ct. No. 21-00138 Page 14

Bulgarian imports classified under HTS subheading 2620.40. See Appx28330.

So the Department was faced with a choice between two imperfect options: (1) Bulgarian import statistics for merchandise classified under HTS 2620, which pro- vides for “ash and residues (other than from the man- ufacture of iron or steel), containing metals, arsenic or their compounds,” including aluminum ash, but also ash comprised of other metals; or (2) Bulgarian import statistics for merchandise classified under HTS sub- heading 7602.00.19, which provides for aluminum waste and scrap other than “turnings, shavings, chips, milling waste, sawdust and filings; waste of coloured, coated or bonded sheets and foil, of a thickness (ex- cluding any backing) not exceeding 0.2 mm,” but does not provide for aluminum ash (or any ash for that mat- ter). See ECF 47, at 21; Appx28290, Appx28327– 28332; Appx28992–28994, Appx29116–29121.

Zhongji argues that HTS code 2620 was unrepre- sentative of its aluminum ash because it is not specific to aluminum and instead includes slag, ash, and other residues. ECF 47, at 13. In contrast, it argues that HTS 7602.00.19 matched its product very well. Id. at 25–26. The Global Trade Atlas data covering HTS 7602.00.19 are, according to the company, specific to its aluminum waste. Id. (citing Appx28867–28875, Appx28983, Appx29116–29118).

In response, the government argues that Com- merce reasonably determined that the tariff heading Ct. No. 21-00138 Page 15

associated with ash, although not specific to alumi- num, was more appropriate than the tariff subheading associated with other non-ash aluminum waste prod- ucts. ECF 30, at 6. The Department objected to HTS code 7602.00.19 because “the broader HTS category 7602 under which it falls describes scrap, cuttings, and other such by-products, not ash,” Appx02486, 4 and ex- plained that “metallurgic content is not the sole factor in determining the value of a by-product,” id. The gov- ernment also contends that “[a] tariff subcategory that includes merchandise such as scrap coils of flat-rolled aluminum products is inconsistent with the physical characteristics of ash, which . . . should be mostly com- prised of the chemical residues from fluxing agents that remove impurities from the melt.” ECF 30, at 16– 17.

Commerce regularly rejects overly broad datasets. See Polyethylene Retail Carrier Bag Comm. v. United States, 29 CIT 1418, 1444 (2005). For instance, in Jiangsu Jiasheng Photovoltaic Technology Co. v. United States, 28 F. Supp. 3d 1317 (CIT 2014), the

4 Zhongji argues that this is a “misstatement of the HTS schedule as 7602.00.90 includes scrap,” ECF 47, at 25, and asserts that “7602.00.11 includes other types of aluminum waste such as ‘shavings, chips, milling, sawdust and fil- ings,’ ” id. at 26 (citing Appx29116–29118). “By selecting [a] more detailed eight-digit HTS code—here, 7602.00.19”—the company sought “to exclude the type of materials that Commerce determined were not representa- tive of its input.” Id. Ct. No. 21-00138 Page 16

court upheld the Department’s decision to value Trina Solar’s aluminum frames in its antidumping duty in- vestigation using an eleven-digit HTS code instead of a six-digit HTS code, HTS 7616.99, in part because “7616.99 is a catch-all category that covers many di- verse aluminum products . . . whose value is not rea- sonably comparable to that of respondent’s aluminum solar panel frames.” Id. at 1338. 5

However, if reasonable minds could differ, that suf- fices to sustain Commerce’s conclusion. See Pastificio Lucio Garofalo, S.p.A. v. United States, 783 F. Supp. 2d 1230, 1233 (CIT 2011) (“[I]f a reasonable mind could accept the connection presented between the facts found and the conclusion reached, an alter- native judgment may not be substituted for that of the agency.”) (citing FCC v. Fox Television Stations, Inc., 556 U.S. 502, 513 (2009)).

Zhongji concedes that the HTS code the Depart- ment used describes ash. ECF 47, at 22–23. The com- pany argues, however, that its recommended code is also specific to ash, see id. at 25, and is therefore strictly better, since the government concedes that

5 Zhongji argues that here, unlike in Jiangsu Jiasheng, Commerce selected a broad four-digit HTS code that was not specific to the aluminum ash and therefore cannot rep- resent the best available information. ECF 47, at 23. Ct. No. 21-00138 Page 17

Zhongji’s code is specific to aluminum, see ECF 30, at 15–16.

But a review of the codes and the record under- mines that argument. The plain language of the HTS states that subheading 2620.40 provides for aluminum ash, and that subheading 7602.00.19 does not. See Appx28992–28994, Appx29116–29121. Thus, ulti- mately, this dispute boils down to a technical judg- ment by Commerce over whether to prioritize “alumi- num” or “ash” in selecting between two imperfect cat- egories.

This sort of decision is a technical judgment for the Department to make. Even if reasonable minds might differ, the substantial evidence standard is met. See Jacobi Carbons AB v. United States, 313 F. Supp. 3d 1344, 1352 (CIT 2018) (“The court may not ‘reweigh the evidence or . . . reconsider questions of fact anew.’ ”) (ellipsis in original) (quoting Downhole Pipe, 776 F.3d at 1377).

The court therefore concludes that the Depart- ment’s choice was reasonable and thus supported by substantial evidence.

In its final determination, Commerce valued Zhongji’s rolling oil using Bulgarian Global Trade At- las data under HTS 3403.99 and rolling oil additive under 3811.90. See Appx02505; Appx02487; see also Ct. No. 21-00138 Page 18

Appx01248. 6 The company instead urged the Depart- ment to use code 2710.12.21 to value its rolling oil and rolling oil additive. 7 See Appx28868, Appx28882– 28884. Zhongji argues that Commerce’s determina- tions to value its rolling oil under HTS 3403.99 and its rolling oil additive under HTS 3811.90 were unsup- ported by substantial evidence and otherwise not in accordance with law. ECF 47, at 13, 27.

Zhongji argues that HTS 2710.12.21 is more spe- cific to the rolling oil and rolling oil additive it used because its oil is made from petroleum and has a flash point that meets the specification in the EU tariff

6 Chapter 34 covers “Soap, organic surface-active agents, washing preparations . . .” and Chapter 38 covers “miscel- laneous chemical products.” See Appx29053–29064; Appx12791–12794. 7 Chapter 27 covers “mineral fuels, mineral oils and prod- ucts of their distillation . . . .” Appx28995–29004. The spe- cific HTS code Zhongji submitted, HTS 2710.12.21, covers: Petroleum oils and oils obtained from bituminous minerals (other than crude) and preparations not elsewhere specified or included, containing by weight 70% or more of petroleum oils or of oils ob- tained from bituminous minerals, these oils being the basic constituents of the preparations, other than those containing biodiesel and other than waste oils: Light oils and preparations: For other purposes: Special spirits: White spirit. Appx28999. Ct. No. 21-00138 Page 19

schedule notes covering white spirits. ECF 47, at 29. The company acknowledges that “the law does not re- quire Commerce to build the record on the plaintiffs’ behalf,” id. (quoting Linyi City Kangfa Foodstuff Drinkable Co. v. United States, Ct. No. 15-00184, Slip Op. 16-89, at 8–9, 2016 WL 5122648, at *3 (CIT Sept. 21, 2016)), but argues that a specification sheet for its rolling oil and rolling oil additive was enough to compel the use of Chapter 27, id. (citing Appx12804– 12814). 8

Zhongji therefore argues that Commerce failed to provide a “reasoned explanation” for rejecting its

8 The company argues that the Department failed to ad- dress this record evidence and “merely found ‘unpersuasive Zhongji’s argument that the eight-digit HTS classifications that it has offered to value refining oils and additives . . . are more specific to Zhongji’s inputs used in its production process than the four- or six-digit HTS classifications that Commerce selected . . . .’ ” Id. at 30 (quoting Appx02487 and citing Appx01241). Thus, according to the company, the Department failed to consider information that “fairly detracts from [the] weight” of its conclusion. Id. at 30–31 (brackets in original) (quoting Nippon Steel Corp. v. United States, 458 F.3d 1345, 1351 (Fed. Cir. 2006)). But Com- merce did consider this evidence—it found that the evi- dence in the specification sheet did not sufficiently show why HTS 2710.12.21 would be the most appropriate sub- heading to value these products. Appx02487. Specifically, it determined that Zhongji’s “description of [its additives] available on the record is not sufficiently detailed to sup- port selection of the eight-digit HTS categories for these in- puts in this case.” Id. Ct. No. 21-00138 Page 20

proposal. ECF 47, at 31 (citing Dorbest, 462 F. Supp. 2d at 1269). The company seeks either a reversal of that decision or a remand for the Department to exam- ine the specification sheet Zhongji submitted and pro- vide a reasoned explanation for why it does not estab- lish that the inputs fall under HTS 2710.12.21. Id.

As noted, see above note 8, Commerce found that the company did not sufficiently describe the relevant inputs on the record to support selection of the eight- digit HTS categories rather than the six-digit catego- ries that the Department selected. Appx02487. In choosing the six-digit category, Commerce cited the Association’s rebuttal brief and the Department’s orig- inal determination as supporting its conclusion. Id. & nn.50, 52.

The Association asserted that Zhongji’s infor- mation confirms that the rolling oil consumed in its operations is not a raw oil (i.e., a product typically clas- sified under Chapter 27 of the tariff schedule). See Appx25849. 9 Thus, the Association contends, although the rolling oil Zhongji used is made of white spirit, it has undergone sufficient refining to become a purified oil. Id. Commerce used the value under HTS 3403.99

9The Association also argued that Zhongji submitted infor- mation confirming that the rolling oil used in its operations must meet FDA, Kosher, and Halal requirements followed by the mill, and is thus purified, not raw. Id. Ct. No. 21-00138 Page 21

covering “lubricating preparations not containing pe- troleum oils or oils obtained from bituminous miner- als,” finding that it “best matched [the company’s] de- scription of the input.” Appx01241 (title case re- moved). The government argues that “HTS 3811.90 specifically references additives” like Zhongji’s rolling oil additive but the HTS code the company used, 2710.12.21, “does not reference additives.” ECF 30, at 21–22. The government also notes that, “in the orig- inal investigation, the mandatory respondent identi- fied HTS subheading 3811.90 as the proper provision for rolling oil additive.” Id. at 22 (citing Appx31361). Because of these characteristics, and because other mandatory respondents identified Commerce’s sub- heading as proper, the government argues that the De- partment’s selection of HTS 3811.90 was based on the best information available and therefore meets the substantial evidence burden. Id. at 22–23.

The government has the better of this argument. Zhongji raises legitimate points about why its HTS code was superior. But a reasonable mind could easily conclude that the agency was right about the technical details and right to privilege the original HTS code. Despite the company’s objections, the record shows that Commerce did consider its evidence—the Depart- ment just rejected it. The court therefore finds that Commerce’s classification of the company’s rolling oil and rolling oil additive products was based on substan- tial evidence. Ct. No. 21-00138 Page 22

Zhongji also argues that the Department failed to provide a reasonable explanation for rejecting the Des- cartes shipping data the company submitted. ECF 47, at 13–14. It argues that Commerce’s decision to in- stead use Maersk data was unsupported by substan- tial evidence. Id. at 14, 32.

In response, the government argues that the De- partment selected the Maersk data because they are publicly available and do not incorporate non-market economy state-owned shipping data, as distinguished from Xeneta and Descartes. ECF 30, at 6–7. The gov- ernment argues that this decision was reasonable and supported by substantial evidence. Id.

As a matter of policy, Commerce tries to select sur- rogate values that are product-specific, representative of a broad market average, publicly available, contem- poraneous with the review period, and tax/duty exclu- sive. See Policy Bulletin at 4. Its practice is to use pub- licly available data. Id.

In this case, there were three datasets available on the record for the Department to consider. The first was Descartes, which had the advantage of being based on real rates and being publicly available—but had the disadvantage of including impermissible data Ct. No. 21-00138 Page 23

from Chinese shipping 10 and had other technical dis- advantages. The second option was Xeneta. That da- taset shared with Descartes the advantage of drawing from real rates and other technical advantages. But Commerce reasonably concluded that the Xeneta set was proprietary—a critical disadvantage based on De- partment policy. The Xeneta set, like Descartes, also included shipping data from China. The third option was Maersk. Maersk had two major advantages: it was publicly available, and it excluded rates from China. But it also had one key disadvantage: The Maersk data are based on freight quotes, not freight prices paid.

In its final determination, Commerce valued Zhongji’s international freight using Maersk data. See Appx02487, Appx02506. The Department considered the Xeneta data as a potential source to value the com- pany’s international freight, Appx02487, but found “Maersk to be a superior [surrogate value] source as compared to Xeneta” because of its “public availabil- ity” and because “the Xeneta data can reflect shipping data from China which is an impermissible source for [surrogate valuation].” Id. (citing Appx25863–25868).

10 The problem with using Chinese data is that Zhongji’s goods were themselves shipped from China, meaning a da- taset containing Chinese data could contain the very sub- sidy that Commerce is trying to identify through this entire investigation. Ct. No. 21-00138 Page 24

Zhongji argues that Commerce’s selection of the Maersk data was unsupported by substantial evidence because it failed to “compar[e] . . . the competing data sources to identify what available information is ‘best’ to value factors of production.” ECF 47, at 33 (altera- tions in original) (quoting Weishan, 917 F.3d at 1367). The company argues that the Maersk data “are funda- mentally flawed” because they are based on “mere quotes that have not been finalized.” See id. at 38 (cit- ing Appx28387, Appx28415–28738). These “estimates do not necessarily identify, or include[,] all relevant charges, and do not reflect consummated transac- tions.” Id. at 39. “By comparison, the Xeneta data con- sist of ‘several hundred thousand rates per month.’ ” Id. (quoting Appx12828). Zhongji argues that these shortcomings it contends mar the Maersk data are “even more egregious considering that the Court up- held Commerce’s selection of Descartes data to value [the company’s] international freight in the underlying investigation.” 11 ECF 47, at 34–35 (citing Jiangsu Zhongji Lamination Materials Co. v. United States, [3]96 F. Supp. 3d 1[3]34, 1353 (CIT 2019)).

The company argues in the alternative that Com- merce’s selection of the Maersk data to value its

11 The government points out that, in the underlying inves- tigation, “the only available surrogate values on the record were from Xeneta and Descartes, not Maersk.” ECF 30, at 27. Ct. No. 21-00138 Page 25

international freight was unsupported by substantial evidence and otherwise not in accordance with law be- cause the Xeneta data are clearly the best available information on the record. Id. at 35–36 (citing Appx12827–12846). 12

In response, the government argues that the De- partment reasonably determined that the Maersk data were the best available for two reasons. ECF 30, at 23. First, Commerce determined that the Xeneta data are proprietary, but the Maersk data are publicly available. Id. at 24 (citing Appx02487). Second, as dis- cussed above, the Department noted that the Xeneta data “can reflect shipping data from China, which is an impermissible source.” Id. at 26 (citing Appx02487). 13

12 Relatedly, and in reply to the final determination, Zhongji argues that the Department provided no support for its assertion that the Xeneta data reflect shipping data from China and instead merely referred to the Associa- tion’s rebuttal brief. ECF 47, at 38 (citing Appx02487). The company argues that in so doing Commerce failed to pro- vide a reasoned explanation for its finding. Id. (citing Dorbest, 462 F. Supp. 2d at 1269–70). 13 The Association argued that Xeneta’s and Descartes’s sampling of freight rates involving the movement of mer- chandise from China to the United States captures Chinese non-market economy ocean carriers, including state-owned entities. Appx25865. The Association also identified spe- cific evidence showing that certain rate sheets showed use of a state-owned shipping line, COSCO. Appx25866. Ct. No. 21-00138 Page 26

Zhongji disputes that the Xeneta data are not pub- licly available. It acknowledges that this court previ- ously held otherwise, ECF 47, at 36 (citing Jiangsu Zhongji, 396 F. Supp. 3d at 135[4]), 14 but argues that, in a separate recent countervailing duty investigation, Commerce found that data obtained from a pay-for- subscription service were public in nature because “any party [could] access the information if they pay for the service.” ECF 47, at 36–37 (alteration in origi- nal) (citing Certain Glass Containers from the People’s Republic of China: Final Affirmative Countervailing Duty Determination, 85 Fed. Reg. 31,141 (Dep’t Com- merce May 22, 2020) and accompanying I&D Memo at Cmt. 9, p. 44). 15 Zhongji further notes that while Glass Containers involved countervailing, rather than anti- dumping, duties, “Commerce interpreted the defini- tion of ‘publicly available’ under 19 CFR

14 “Commerce’s determination that the Xeneta data were not publicly available was . . . within its discretion and con- sistent with its past practice.” 396 F. Supp. 3d at 1353. 15Zhongji quotes Commerce as having said: “With regard to whether Xeneta data is proprietary, we disagree. Xeneta data is a pay for service subscription service, meaning any party can access the information if they pay for the service. In this respect Xeneta is not proprietary but public in na- ture.” Id. (quoting the Glass Containers I&D Memo at 44, which in turn cited 19 C.F.R. § 351.102(b)(21)(iii) for the definition of “publicly available”). Ct. No. 21-00138 Page 27

§ 351.102(b)(21)(iii), which is applicable to the dump- ing context.” Id. 16

The company also rightly notes that the alternative Descartes database has been used before. But the gov- ernment argues that this has generally occurred when Descartes was the only information available and con- tends that where the Department “has relied on the Descartes data, Commerce recognized the inclusion of rates from non-market economy carriers and ex- plained: ‘[f]or any rate that the Department deter- mined was from a non-market economy carrier, the Department has not included that rate in the period- average international freight calculation.’ ” ECF 30, at 27–28 (brackets in original) (quoting Certain Steel Wheels from the People’s Republic of China: Notice of Preliminary Determination of Sales at Less Than Fair Value, Partial Affirmative Preliminary Determination of Critical Circumstances, and Postponement of Final Determination, 76 Fed. Reg. 67,703, 67,713 (Dep’t Commerce Nov. 2, 2011)). 17

16 In response, the government points out that Maersk in- formation was not on the record in that case for comparison with Xeneta and Descartes data. ECF 30, at 25. “Further, other factual concerns presented in this review with re- spect to the Xeneta and Descartes data . . . were not raised in Certain Glass Containers.” Id. (citing Certain Glass Con- tainers I&D Memo at 44–46). 17The government also explains that no such adjustment was possible here, because the information Zhongji submit- ted did not permit the identification of non-market Ct. No. 21-00138 Page 28

This court recently addressed the issue here at some length. In two cases, we remanded Commerce’s reliance on Maersk data where the Department failed to explain why it found it reasonable “to choose price quotes over broad data sets.” Changzhou Trina Solar Energy Co. v. United States, 492 F. Supp. 3d 1322, 1330 (CIT 2021) (Trina I); see also Changzhou Trina Solar Energy Co. v. United States, 532 F. Supp. 3d 1333, 1336–37 (CIT 2021) (Trina II). Zhongji argues that the reasoning in Trina I and Trina II is persua- sive because the facts are analogous. 18 ECF 47, at 39. But we also stated that “it is reasonable for Commerce

economy carrier rates included in the Descartes data. Id. at 28 (citing Appx25867). 18 In response, the government seeks to distinguish this case by arguing that the facts presented in Trina II differ in meaningful ways. ECF 30, at 26. The government notes that the parties to the administrative review in Trina II never raised the question of whether the freight data Com- merce used were publicly available. Id. (citing Solar Cells from China I&D Memo at 27–32). The government also notes that in Trina I, the parties never raised whether the freight data included Chinese non-market carrier infor- mation—the case was about specificity. Id. (citing Trina I, 492 F. Supp. 3d at 1327–32). The government argues that the court should not rely on Trina I here because the facts underlying that matter “differ substantially from those present in this case.” Id. Ct. No. 21-00138 Page 29

to choose price quotes over broad data sets in some cir- cumstances.” Trina I, 492 F. Supp. 3d at 1330.

Similarly, in Trina II we upheld Commerce’s deter- mination upon remand to rely on Xeneta data, instead of Maersk data, to calculate its surrogate value for in- ternational freight. 532 F. Supp. 3d at 1337. But we did so under the substantial evidence standard. Under that forgiving standard, it seems possible that either deciding the Xeneta evidence was admissible or decid- ing it was not would both be supported by substantial evidence. 19

The Maersk data were not dominated by either the Xeneta or the Descartes. On the criteria identified by the Policy Bulletin (completeness, quality of data, pub- lic availability, and permissibility of data sources) the Maersk database beat both the Xeneta source (on pub- lic availability and permissibility) and the Descartes source (on permissibility). Because weighting each as- pect of the Policy Bulletin is within Commerce’s dis- cretion, the Department met the burden required by the substantial evidence standard.

Zhongji argues that Commerce’s determination not to grant it a double remedies adjustment was unsup- ported by substantial evidence because it established

19 Zhongji does not appear to have considered this possibil- ity. Ct. No. 21-00138 Page 30

that (1) there is a link between the countervailable subsidies it received and its cost of manufacturing and (2) there is a link between its cost of manufacturing and its selling price of aluminum foil. ECF 47, at 3, 41–43. The company also argues that Commerce could reasonably estimate the extent to which the assigned dumping margin increased as the result of subsidies. Id. at 15.

In response, the government argues that the De- partment’s denial was based on substantial evidence because Zhongji failed to demonstrate (1) a “subsidy- to-cost link” and (2) a “cost-to-price link” as required by Commerce’s interpretation of 19 U.S.C. § 1677f-1(f)(1)(B). ECF 30, at 7.

The first statutory requirement is whether “a coun- tervailable subsidy (other than an export subsidy . . .) has been provided with respect to the class or kind of merchandise.” 19 U.S.C. § 1677f-1(f)(1)(A). Zhongji re- ported that certain inputs were simultaneously sub- ject to countervailable subsidies in the form of primary aluminum, aluminum plate, and strip/electricity pro- vided for less than adequate remuneration. See Appx16029; see also ECF 47, at 46 (arguing that the company satisfied the first statutory element). The government does not dispute that the company estab- lished this point. See ECF 30, at 30–36 (addressing other statutory elements but not the first one). Ct. No. 21-00138 Page 31

The second consideration is whether the subsidies identified have “been demonstrated to have reduced the average price of imports of the class or kind of mer- chandise during the relevant period.” 19 U.S.C. § 1677f-1(f)(1)(B). Commerce looks for a subsidies-to- cost link and a cost-to-price link. See Appx02490– 02492. Failure to show either of these elements allows the Department to deny the double-remedy adjust- ment. See Vicentin S.A.I.C. v. United States, 503 F. Supp. 3d 1255, 1263 (CIT 2021) (discussing Com- merce’s interpretation that § 1677f-1(f)(1) requires the producer/exporter to show both a “subsidies-to-cost link” and a “cost-to-price link”).

In its final determination, Commerce found that Zhongji “failed to establish either a ‘subsidy-to-cost link’ or a ‘cost-to-price link.’ ” Appx02490. The com- pany disputes that finding and argues that it demon- strated how the pricing of each direct input was incor- porated directly into its cost-of-manufacturing ledgers, and further showed that its input prices and its cost of manufacturing both trended down during the period of review. See ECF 47, at 47–48 (citing Appx16025, Appx16049–16050).

As for subsidy-to-cost, in this case, the majority of the cost of manufacturing comes from foil stock. See Appx16025. Zhongji told the Department that the price of foil stock is lowered by subsidies on other Ct. No. 21-00138 Page 32

aluminum materials. See Appx16029. 20 Since alumi- num ingot, strip, and stock constitute much of the cost of manufacturing the subject merchandise here, the company argues that any decrease in the cost of inputs necessarily decreased the cost of manufacturing. ECF 47, at 50. Zhongji asserts that it submitted subledgers showing how its electricity expense ties back to its overall cost of manufacturing. Id. (citing Appx16063–16066). The company says it provided ex- pense records rather than an accounting of subsidies because the aluminum and electricity subsidies are not recognized as subsidies in China and are therefore not reflected in the company’s financial statements.21 Id. at 51.

20 Two of the company’s foil stock suppliers, Jiangsu Huafeng Aluminum Industry and Anhui Maximum Alumi- num Industries, both produce foil stock, subject to the sub- sidy on primary aluminum and aluminum strip. Both Huafeng’s and Maximum’s aluminum inputs account for most of their production costs. Id. 21Zhongji argues that, in other recent and similar cases, the Department has not required a line-item accounting. For instance, in a recent investigation of quartz surface products from China, Commerce found basic expense infor- mation sufficient. Id. at 52–53 (citing Certain Quartz I&D Memo at 12 and Certain Corrosion Inhibitors from the Peo- ple’s Republic of China: Final Affirmative Determination of Sales at Less Than Fair Value, 86 Fed. Reg. 7,532 (Dep’t Commerce Jan. 29, 2021) and accompanying Mem. on Fi- nal Double Remedy Calc. at 2). Ct. No. 21-00138 Page 33

Zhongji argues that the facts related to subsidy-to- cost are substantively identical to other recent cases in which Commerce has ruled that a respondent met the requirements of 19 U.S.C. § 1677f-1(f)(1)(B). Id. at 52– 54. 22 The company argues that, because the Depart- ment provided no reasoning permitting parties to dis- cern why it treated this case differently from a case like Certain Quartz, its decision was not based on sub- stantial evidence. 23 ECF 47, at 52–55.

In response, the government argues that “Zhongji did not offer a sufficient explanation of how quoted London Metal Exchange prices for primary aluminum ingot establish a monthly decline in the prices of alu- minum jumbo rolls, a distinct product and not of the same class or kind of merchandise as called for under

22 As in Certain Quartz, Zhongji asserts that it reported here that it consumes countervailed inputs that flow into its cost of manufacturing, see id. at 53 (citing Appx16029), and that the subsidy programs affected its cost of manufac- turing, see id. (citing Appx16025). Indeed, it claims to have further reinforced the subsidies-to-cost link by showing that its aluminum inputs constitute a significant portion of the merchandise cost of manufacturing. See id. (citing Appx16049–16050). 23 Zhongji adds that Commerce’s decision is arbitrary be- cause it treated a similar situation differently “without ad- equate explanation and factual support on the record.” ECF 47, at 55 (quoting Thai Plastic Bags Indus. Co. v. United States, 949 F. Supp. 2d 1298, 1302 (CIT 2013), and citing Transactive Corp. v. United States, 91 F.3d 232, 237 (D.C. Cir. 1996)). Ct. No. 21-00138 Page 34

19 U.S.C. § 1677f-1(f)(1)(B).” ECF 30, at 32 (citing Appx02491). Commerce’s final determination gave that same explanation. 24 Appx02491.

The Department found unpersuasive Zhongji’s claim that London Metal Exchange prices establish the existence of a “subsidy-to-cost link” for its alumi- num inputs. Id. 25 The government states that the Lon- don Market Exchange is “a published price index that nearly all aluminum contracts throughout the world incorporate in pricing.” ECF 30, at 33. Commerce de- termined that the company failed to sufficiently show how incorporating this global index element into its purchases established a “subsidy-to-cost link” for its aluminum inputs. Appx02491.

The government argues that the Department rea- sonably determined that Zhongji failed to establish a subsidy-to-cost link because it “did not provide addi- tional documents to demonstrate a connection be- tween subsidies received and cost of manufacture

24 The government also notes that, when determining whether there has been a monthly decline in prices, “Com- merce typically looks to average unit value data rather than to specific proprietary price data” as provided by a re- spondent. ECF 30, at 32. 25Commerce based its analysis on the information that Zhongji provided; there is no dispute here of material fact. Appx02491. Ct. No. 21-00138 Page 35

beyond showing how the cost of materials and electric- ity are accounted for in its records.” ECF 30, at 34.

Commerce’s second requirement as it interprets 19 U.S.C. § 1677f-1(f)(1)(B) is cost-to-price. According to Zhongji, “[a]luminum pricing generally consists of two components: (1) a conversion premium, negotiated and agreed by the parties and (2) the price of the metal component that is based on the [London Metal Ex- change] ingot price.” ECF 47, at 59 (citing Appx16024– 16025). The company asserts that under this pricing structure, price fluctuation “is automatically trans- lated to aluminum strip, foil stock, and foil,” such that all changes in the export prices of subject merchandise “are largely driven by fluctuations of the aluminum in- put cost.” 26 Id. (citing Appx16025–16026).

To show that those statements are accurate here, Zhongji reported that the head of the sales department “monitors the [London Metal Exchange] ingot price and checks with the purchase manager and financial manager regarding the cost of the domestic purchase prices of foil stock, aluminum strip, aluminum ingot,

26 Zhongji’s submission to Commerce stated that “[w]hen setting and changing the prices of exports to the United States, the primary factor . . . is the overall cost of manu- facturing, including the costs of the main inputs such as foil stock, aluminum strip, aluminum ingot, and the cost of electricity, labor, overheads etc.” Appx16025. Ct. No. 21-00138 Page 36

electricity and other elements of the cost of manufac- turing periodically to consider the prices quoted to the customers.” Appx16026. The company argues that this evidence directly responds to Commerce’s requests for information about internal processes, showing how price changes are realized and how cost of manufac- turing is incorporated directly into pricing. 27 ECF 47, at 58 (citing Appx16024–16027).

In reply, the government notes that, for the “cost- to-price link,” “a company must demonstrate a connec- tion between subsidies received and cost of manufac- ture, and how a change in the cost of manufacture is transferred to the price of the subject merchandise.” ECF 30, at 35 (citing Appx02491–02492). According to the government, Zhongji’s accounting records estab- lished only how it tracks its usage of these three in- puts. Id. (citing Appx02492). As a result, the govern- ment argues, this information failed to establish a link between the cost of manufacture and the price of the merchandise. Id.

The company also spends some time trying to show that (f)(1)(C) is met. ECF 47, at 63–66. The third ele- ment of the double remedies test is whether the

27Zhongji analyzes two cases at length to support the claim that in prior cases the record would suffice to establish a cost-to-price link and argues that Commerce varied from a recent case with similar facts when it should not have (Cer- tain Quartz) and instead analogized this case to another investigation with dissimilar facts. ECF 47, at 61–62. Ct. No. 21-00138 Page 37

Department “can reasonably estimate the extent to which the countervailable subsidy . . . , in combination with the use of [normal value] determined pursuant to [19 U.S.C. § 1677b(c)], has increased the weighted av- erage dumping margin for the class or kind of mer- chandise.” 19 U.S.C. § 1677f-1(f)(1)(C). By again apply- ing and analyzing Certain Quartz, Zhongji argues that Commerce could have used similar methodology to properly estimate the appropriate reduction here. ECF 47, at 65–66.

Finally, the company argues that if the Department found its information deficient, then the agency should have given it a chance to remedy any deficiency, but did not. ECF 47, at 57 (citing 19 U.S.C. § 1677m(d)). But “[t]he interested party that is in possession of the relevant information has the burden of establishing . . . the amount and nature of a particular adjustment.” 19 C.F.R. § 351.401(b)(1); see also NTN Corp. v. United States, 306 F. Supp. 2d 1319, 1328 (CIT 2004) (same).

The question is whether Zhongji “demonstrated” to the Department that the Chinese countervailable sub- sidy “reduced the average price of imports of the class or kind of merchandise during the relevant period.” 19 U.S.C. § 1677f-1(f)(1)(B). The company argues that it did; the government argues it did not.

In its preliminary determination, Commerce ex- plained that (1) a “subsidy to cost link” exists where Ct. No. 21-00138 Page 38

there is a “subsidy effect” to the merchandise’s cost of manufacture and (2) a “cost to price link” exists where a change in the cost of manufacture results in a change to the price charged to customers. Appx02468 (citing Wooden Cabinets and Vanities and Components Thereof from the People’s Republic of China: Prelimi- nary Affirmative Determination of Sales at Less Than Fair Value, Postponement of Final Determination and Extension of Provisional Measures, 84 Fed. Reg. 54,106 (Dep’t Commerce Oct. 9, 2019), and accompanying pre- liminary memorandum at 48). The Department ex- plained that Zhongji had “failed to demonstrate that the programs discussed in [its] double remedies ques- tionnaire[ ] have led to a decrease to either input costs or” cost of manufacturing. Appx02469. Commerce also stated that the company had failed to provide suffi- cient responses to the questions about “cost and price changes during the relevant period.” Id. In the final determination, the Department reiterated its prelimi- nary findings and concluded that Zhongji had provided “no convincing explanation” and “failed to establish” the results it urged the Department to reach. Appx02491.

The company cites its questionnaire responses and argues that they show that Commerce’s decision was incorrect. See ECF 47, at 43–44 (citing Appx16029– 16030 and stating that they establish a link between input prices, cost of manufacture, and sales prices). The cited pages, however, state that (1) Zhongji “does not have a formal threshold for changes in the cost Ct. No. 21-00138 Page 39

item that would lead to adjustment of prices” because any price adjustments result from management dis- cussions and (2) “[t]he price charged to the customer changes based on an agreed reference to the LME in- got price, usually monthly.” Appx16030.

The two admissions cited above are enough to sup- port the Department’s conclusion. First, the company admits that it does not necessarily adjust prices when the inputs’ costs change. Second, the company admits that its pricing changes based on the London Market Exchange ingot price, which presumably has nothing to do with any countervailable subsidies provided by the Chinese government. In other words, whether “the average price of imports” to the United States was re- duced is apparently a matter of whether Zhongji’s management decides to adjust prices. The Depart- ment’s conclusion that the company failed to demon- strate eligibility for a double remedies adjustment is therefore supported by substantial evidence.

C

Finally, Zhongji argues that Commerce’s decision not to modify its liquidation instructions to include the phrase “resold or imported” where the company’s mer- chandise was re-invoiced before importation was un- supported by substantial evidence because these in- structions will lead to inaccurate and overly punitive liquidation. ECF 47, at 3–4. Zhongji also argues that the Department ignored its concerns about the inap- propriate application of the China-wide rate. Id. at 15. Ct. No. 21-00138 Page 40

The company argues that Commerce erred by not modifying its liquidation instructions because some U.S. customers re-invoiced the company’s sales to third parties before importation. Id. at 67 (citing Appx24809–24816). As a result of this sales arrange- ment sometimes a different importer of record appears on the entry documents than the initial customer. Id.

Zhongji thus argues that when the Department is- sued liquidation instructions, the sales subject to the re-sale arrangement would not be assessed duties commensurate with the company’s review-specific rate because the importer of record would differ from the initial customer as identified in the sales database. Id. at 69 (citing Appx09862–09870). According to the com- pany, this means that, when the re-seller does not ap- pear in the customer database, Commerce’s default language would cause these sales to liquidate at the China-wide rate instead of the company’s rate. Id. at 70. Zhongji therefore argues that the default liquida- tion language will result in an inaccurate, punitive as- sessment of duties. Id.

In response, the government argues that the De- partment did not modify its standard liquidation in- structions to Customs because Zhongji did not demon- strate how re-invoicing merchandise was unique and warranted a specific deviation in Commerce’s stand- ard practice. ECF 30, at 7.

Antidumping laws are “remedial[,] not punitive.” NTN Bearing Corp. v. United States, 74 F.3d 1204, Ct. No. 21-00138 Page 41

1208 (Fed. Cir. 1995) (citing Chaparral Steel Co. v. United States, 901 F.2d 1097, 1103–04 (Fed. Cir. 1990)). The function of antidumping law is to “reduc[e] or eliminat[e] discrepancies in pricing . . . .” U.S. Steel Grp. v. United States, 177 F. Supp. 2d 1325, 1330 (CIT 2001); see also C.J. Tower & Sons v. United States, 71 F.2d 438, 445 (CCPA 1934) (noting Congress’s ex- pressed purpose in the Anti-Dumping Act of 1921 was to impose “an amount of duty sufficient to equalize competitive conditions”). Also, “[t]he purpose of the an- tidumping law, as its name implies, is to discourage the practice of selling in the United States at [less than fair value] by the imposition of appropriately in- creased duties.” Melamine Chems., Inc. v. United States, 732 F.2d 924, 933 (Fed. Cir. 1984). Zhongji cites these cases to argue that Commerce was obliged to tai- lor its liquidation instructions to ensure that the du- ties imposed are “appropriately increased” solely to equalize market conditions. ECF 47, at 70–71. While the company tries to explain why that follows on the facts here, id. at 72–73, the Department said, “Because the importer of record is information that is available to [Customs], our practice is to issue liquidation in- structions on the basis of entered value rather than by U.S. customer.” Appx02489. Commerce said it could find no reason to depart from that practice here be- cause the importer of record was unknown to Zhongji. Id.

In this case, the Department made its calculation using the reported sales from Zhongji to its listed U.S. Ct. No. 21-00138 Page 42

customers. Neither Commerce nor the company knows the specifics of the sales information between those U.S. customers and third-party companies, and the Department’s margin calculations did not reflect such information. See 19 U.S.C. § 1677a(a) (“ ‘[E]xport price’ means the price at which the subject merchandise is first sold . . . before the date of importation by the pro- ducer or exporter . . . to an unaffiliated purchaser in the United States . . . .”) Because this is general policy and because Zhongji failed—in the agency’s eyes—to show why that general policy should not apply here, Commerce adhered to its standard instructions, a de- cision the government argues is supported by substan- tial evidence.

After reviewing the record, this court sustains the Department. It is not entirely clear what argument Zhongji is trying to make on this record. The company makes some very general points about the purpose of antidumping law that are correct. But it seems to vac- illate between arguing either (i) that some specific as- pect of its business makes Commerce’s standard liqui- dation instructions problematic here, or (ii) that, based on the general points the company raised, those liqui- dation instructions are unlawful as a general matter.

If Zhongji is trying to argue for (ii)—that, in gen- eral, the liquidation instructions are unlawfully puni- tive—it needs to produce far more evidence than a few pages of case analysis. If it is trying to argue instead for (i)—that special circumstances justify an Ct. No. 21-00138 Page 43

exception—then it needs to explain more thoroughly what those circumstances are and how they differ from other cases. By not clearly taking either path, the com- pany fails to provide a clear reason why the Depart- ment’s instructions ought to vary. The court therefore concludes that Commerce’s decision was based on sub- stantial evidence.

* * *

For the reasons provided above, the court denies Zhongji’s motion for judgment on the agency record, grants judgment on the agency record to the govern- ment and the Association, see USCIT R. 56.2(b), and sustains the Department of Commerce’s final determi- nation. A separate judgment will enter. See USCIT R. 58(a).

Dated: June 7, 2023 /s/ M. Miller Baker New York, NY M. Miller Baker, Judge