CIT 将韩国碳素钢板 CVD 行政复审发回 Commerce 重审（Slip Op. 26-90）

Slip Op. 26-90

UNITED STATES COURT OF INTERNATIONAL TRADE

HYUNDAI STEEL COMPANY,

Plaintiff,

and

DONGKUK STEEL MILL CO., LTD.,

Consolidated Plaintiff,

and

GOVERNMENT OF THE REPUBLIC OF KOREA,

Plaintiff-Intervenor,

v.

UNITED STATES,

Defendant,

and

NUCOR CORPORATION,

Defendant-Intervenor.

Before: Claire R. Kelly, Judge

Consol. Court No. 24-00190

PUBLIC VERSION

OPINION AND ORDER

[Granting Plaintiffs’ motions for judgment on the agency record and remanding Commerce’s determination.]

Dated: August 11, 2026 PUBLIC VERSION

Brady Warfield Mills, Donald B. Cameron, Jr., Edward J. Thomas, III, Eugene Degnan, Jordan L. Fleischer, Julie Clark Mendoza, Mary Shannon Hodgins, Nicholas C. Duffey, Jr., and Rudi Will Planert, Morris, Manning & Martin, LLP, of Washington D.C., for Plaintiff Hyundai Steel Company.

Jarrod Mark Goldfeder and Kenneth Neal Hammer, Trade Pacific PLLC, of Washington D.C., for Consolidated Plaintiff Dongkuk Steel Mill Co., Ltd.

Yujin Kim McNamara, Daniel Martin Witkowski, Devin Scott Sikes, and Sung Un K. Kim, Akin, Gump, Strauss, Hauer & Feld, LLP, of Washington D.C., for Plaintiff- Intervenor Government of the Republic of Korea.

Augustus J. Golden, National Courts Section, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, of Washington D.C., for Defendant United States. Also on the brief were Brett A. Shumate, Assistant Attorney General, Patricia M. McCarthy, Director, and Tara K. Hogan, Assistant Director. Of counsel was Jack Dunkelman, Attorney, U.S. Department of Commerce.

Alan Hayden Price, Adam Milan Teslik, Christopher Bright Weld, Derick G. Holt, Enbar Toledano, Maureen Elizabeth Thorson, Paul A. Devamithran, and Theodore Paul Brackemyre, Wiley Rein, LLP, of Washington D.C., for Defendant-Intervenor Nucor Corporation.

Kelly, Judge: Before the Court are motions for judgment on the agency record

pursuant to United States Court of International Trade Rule (“USCIT R.”) 56.2, filed

by Plaintiff Hyundai Steel Company (“Hyundai”), Consolidated Plaintiff Dongkuk

Steel Mill Co., Ltd. (“DSM”) (collectively, “Plaintiffs”), and Plaintiff-Intervenor

Government of the Republic of Korea (“Gov’t of Korea” or “Plaintiff-Intervenor”). See

generally [Hyundai] Mot. J. Agency Rec., Apr. 24, 2025, ECF Nos. 31, 32 (“Hyundai

Mot.”); [DSM] Mot. J. Agency Rec., Apr. 24, 2025, ECF No. 33 (“DSM Mot.”); [Gov’t of

Korea] Mot. J. Agency Rec., May 29, 2025, ECF Nos. 38, 39 (“Gov’t of Korea Mot.”);

see also USCIT R. 56.2; see also Certain Cut-to-Length Carbon-Quality Steel Plate PUBLIC VERSION

From the Republic of Korea: Final Results of Countervailing Duty Administrative

Review; 2022, 89 Fed. Reg. 15,825 (Sept. 6, 2024) (“Final Results”), PD 318, bar code

4626807-03 and accompanying Issues & Decision Mem. (“Final Decision Memo.”), PD

319, bar code 4626807-02 (Sept. 6, 2024). Plaintiffs and Plaintiff-Intervenor argue

that the U.S. Department of Commerce’s (“Commerce”) determination that KEPCO’s

provision of electricity (the “Electricity Program”) for less than adequate

remuneration (“LTAR”) is de facto specific under Section 771(5A)(D)(iii) of the Tariff

Act of 1930,1 as amended, 19 U.S.C. § 1677(5A)(D)(iii), is unsupported by substantial

evidence and otherwise not in accordance with law. Hyundai Mot. at 11–19; DSM

Mot. at 7–11; Gov’t of Korea Mot. at 4–14. Hyundai and the Gov’t of Korea argue

Commerce’s decision not to rely on the 2022 Korea Electric Power Corporation

(“KEPCO”) cost data submitted by the Gov’t of Korea in response to a supplemental

questionnaire—rather than the 2021 KEPCO data originally submitted by the Gov’t

of Korea in its initial questionnaire response—is unsupported by substantial evidence

and is otherwise not in accordance with law. Hyundai Mot. at 19–22; Gov’t of Korea

Mot. at 14–16. DSM argues that Commerce’s determination that KEPCO’s

Electricity Program was provided at LTAR is not supported by substantial evidence

and is contrary to law. DSM Mot. at 7–8, 11–14. Hyundai and the Gov’t of Korea

argue that Commerce’s calculation of the benefit from the provision of electricity for

1 Further citations to the Tariff Act of 1930, as amended, are to the relevant provisions of Title 19 of the U.S. Code, 2024 edition.

PUBLIC VERSION

LTAR under 19 U.S.C. § 1677(5)(E), using KEPCO’s cost data carried forward from,

and already used in, the countervailing duty analysis for the prior period of review

(“POR”) is unsupported by substantial evidence and not in accordance with law.

Hyundai Mot. at 22–26; Gov’t of Korea Mot. at 16–19. For the following reasons,

Plaintiffs’ and Plaintiff-Intervenor’s motions for judgment on the agency record are

granted, and Commerce’s determination is remanded for further explanation or

reconsideration.

BACKGROUND

On April 11, 2023, Commerce initiated its administrative review of the

countervailing duty (“CVD”) order on Certain Cut-to-Length Carbon-Quality Steel

Plate for the 2022 POR. See Initiation of Antidumping and Countervailing Duty

Administrative Reviews, 88 Fed. Reg. 21,609 (Dep’t Commerce Apr. 11, 2023).

Commerce evaluated many subsidy programs, including the Electricity Program. See

generally Certain Cut-to-Length Carbon-Quality Steel Plate From the Republic of

Korea: Preliminary Results and Partial Rescission of Countervailing Duty

Administrative Review; 2022, 89 Fed. Reg. 15,825 (Mar. 5, 2024) (“Preliminary

Results”), PD 248, bar code 4517008-03 and accompanying Prelim. Issues & Decision

Mem. (“Prelim. Decision Memo.”), PD 243, bar code 4517008-02 (Mar. 5, 2024). On

June 14, 2023, Commerce selected DSM and Hyundai as mandatory respondents and

issued an initial questionnaire to the Gov’t of Korea. See Letter from [Commerce] to PUBLIC VERSION

[Gov’t of Korea] Pertaining to Initial Questionnaire, CD 28, bar code 4389302-01

(June 14, 2023) (“Initial Questionnaire”).

In its initial questionnaire for this review, Commerce requested that the Gov’t

of Korea “provide the amount of electricity provided to the steel industry during the

POR and the largest 10 industries consuming electricity during the POR” as well as

the “percentage for the steel industry and largest 10 industries consuming electricity

for the POR based on total electricity consumed and the total industrial classification

consumed.” Initial Questionnaire at 20. The Gov’t of Korea submitted its initial

questionnaire response on August 11, 2023, which included usage data for the steel

industry and ten largest electricity-consuming industries. See Gov’t of Korea

Response to Initial Questionnaire, CD 105–142, PD 104–125, bar codes 4416305-01–

38; 4416377-01–22 (Aug. 11, 2023) (“Gov’t of Korea IQR”); Hyundai Mot. at 4. After

the Gov’t of Korea had submitted its initial questionnaire response, and although

Commerce did not request it, the Gov’t of Korea submitted a second, revised version

of usage data for the steel industry and ten largest electricity-consuming industries.

Hyundai Mot. at 4. The Gov’t of Korea explained that the first dataset was

“inaccurate in the sense that each category covers industries that do not match their

names; for example, the ‘Steel’ category covers not only steel but also copper and non-

ferrous metal, aluminum, etc.” Gov’t of Korea Second Supplemental Questionnaire

Response at 2, PD 207, CD 285, bar code 445588601 (Nov. 1, 2023) (“Gov’t of Korea

2SQR”). The Gov’t of Korea also explained that the revised data complied with the PUBLIC VERSION

official industry classification scheme in Korea. Id. at 2. Commerce accepted this

revised data, discussed it in the Preliminary Results, and then verified it at the Gov’t

of Korea’s verification, but it did not verify the original dataset. Final Decision Memo.

at 22; Hyundai Mot. at 14–15; Gov’t of Korea Mot. at 6–8. Despite accepting and

verifying the revised dataset provided in the supplemental questionnaire response,

Commerce continued to rely on the initial dataset in making its specificity

determination. Def. Resp. at 21 (citing Final Decision Memo. at 21–22).

On March 5, 2024, Commerce published its preliminary results of the review,

in which it determined the Electricity Program provided a financial contribution and

benefit to Hyundai and DSM, and that the provision of electricity was de facto specific

because the steel industry “received a disproportionately large amount of the

subsidy.” Prelim. Decision Memo. at 36. Thus, Commerce assigned Hyundai a

countervailing duty rate of 1.47 percent ad valorem and DSM a rate of 1.61 percent

ad valorem. Id. at 35–36.

The interested parties submitted comments on the Electricity Program that

Commerce deemed countervailable. See Final Decision Memo. at 8, cmts. 1–4. The

Electricity Program includes a cost pass-through tariff system for electricity which

isolates fuel and environmental cost elements in KEPCO’s tariffs. Final Decision

Memo. at 23 (citing Gov’t of Korea IQR at Exhibit E-1). Under this system, KEPCO

applies a “fuel cost adjustment charge” (“FCAC”) that is revised quarterly to reflect

changes in fuel prices, subject to certain regulatory limits. Final Decision Memo. at PUBLIC VERSION

22–23 (citing Hyundai Case Br. at 33–34). The FCAC appears as a distinct line item

on customers’ electricity bills and is part of the total tariff that industrial users pay.

Id. at 24. When fuel costs rise and government regulations do not permit full recovery

through the quarterly FCAC, KEPCO does not immediately write off the unrecovered

portion. Id. at 24–25 (citing Gov’t of Korea IQR at Exhibit E-2). Instead, KEPCO

explains that it accumulates unrecovered fuel costs and rolls them into its “Total

Comprehensive Cost” for the following term. Id. at 24–25 (citing Gov’t of Korea IQR

at Exhibit E-2). The Total Comprehensive Cost figure is then used to set the base

and usage charges for future tariffs, with the stated objective of allowing KEPCO to

recover its operating costs and a reasonable return over time. Id. at 24–25 (citing

Gov’t of Korea IQR at Exhibit E-2).

On July 18, 2024, Hyundai and the Gov’t of Korea submitted administrative

case briefs. See Letter from Yoon & Yang LLC to Sec’y Commerce, “Case Brief of the

Gov’t of the Republic of Korea,” CD 374, PD 307, bar code 4598797-01 (July 18, 2024)

(“Gov’t of Korea Case Br.”); Letter from Morris, Manning & Martin LLP to Sec’y of

Commerce, “Certain Cut-to-Length Carbon-Quality Steel Plate From the Republic of

Korea,” CD 375, PD 308, bar code 4598797-01 (July 18, 2024) (“Hyundai Case Br.”).

PUBLIC VERSION

On September 11, 2024, Commerce published the Final Results. See generally

Final Results; Final Decision Memo.2 In the Final Results, Commerce relies upon

the dataset submitted in the Gov’t of Korea’s initial questionnaire response and does

not consider the second, revised dataset submitted in the Gov’t of Korea’s

supplemental questionnaire response. See Final Decision Memo. at 21–22.

Commerce verified the revised electricity consumption data, but it determined that

the initial questionnaire response data was “more appropriate” to rely upon for the

review. See id. at 22.

On April 24, 2025, Hyundai and DSM filed motions for judgment on the agency

record, arguing that Commerce’s final determinations are unsupported by

substantial evidence and otherwise not in accordance with law. See Hyundai Mot. at

1; DSM Mot. at 1. On May 29, 2025, the Gov’t of Korea filed a brief in support of

Hyundai’s motion for judgment on the agency record, arguing the same. See Gov’t of

Korea Mot. at 1. On October 1, 2025, Defendant filed its response to the motions for

judgment on the agency record, arguing Commerce’s determination is supported by

substantial evidence and should be sustained. See generally Def. Resp. [Hyundai],

[DSM] and [Gov’t of Korea] Mot., Oct. 1, 2025, ECF Nos. 43, 44 (“Def. Resp.”). On

October 24, 2025, Defendant-Intervenor Nucor Corporation filed a response

2 On December 20, 2024, Defendant filed indices to the public and confidential administrative records underlying Commerce’s final determination. See ECF Nos. 25, 26.

PUBLIC VERSION

supporting Commerce’s determination. See generally [Nucor’s] Resp. [Hyundai],

[DSM] and [Gov’t of Korea] Mot., Oct. 24, 2025, ECF Nos. 45, 46 (“Nucor Resp.”). On

November 20, 2025, Hyundai filed its reply, see generally [Hyundai] Reply Br. Supp’n

[Hyundai Mot.], Nov. 20, 2025, ECF Nos. 48, 49 (“Hyundai Reply”), DSM filed its

reply, see generally [DSM’s] Reply Br. Supp’n [DSM Mot.], Nov. 20, 2025, ECF No.

50 (“DSM Reply”), and Gov’t of Korea filed its reply, see generally [Gov’t of Korea]

Reply Br. Supp’n [Gov’t of Korea Mot.], Dec. 19, 2025, ECF Nos. 53, 54 (“Gov’t of

Korea Reply”).

JURISDICTION AND STANDARD OF REVIEW

This Court has jurisdiction over this action, which contests Commerce’s final

results in an administrative review of an antidumping duty order, under 19 U.S.C.

§ 1516a(a)(2)(B)(iii) and 28 U.S.C. § 1581(c). The Court will sustain Commerce’s

determination unless it is unsupported by substantial evidence on the administrative

record or is otherwise not in accordance with law. 19 U.S.C. § 1516a(b)(1)(B)(i).

Substantial evidence is “such relevant evidence as a reasonable mind might accept

as adequate to support a conclusion.” Huaiyin Foreign Trade Corp. (30) v. United

States, 322 F.3d 1369, 1374 (Fed. Cir. 2003) (quoting Consol. Edison Co. v. NLRB,

305 U.S. 197, 229 (1938)). An abuse of discretion occurs when Commerce bases its

determination on an erroneous interpretation of the law, on factual findings that are

not supported by substantial evidence, or on an unreasonable judgment in weighing PUBLIC VERSION

relevant factors. Consol. Bearings Co. v. United States, 412 F.3d 1266, 1269 (Fed.

Cir. 2005).

DISCUSSION

Plaintiffs and Plaintiff-Intervenor challenge Commerce’s specificity analysis

as contrary to law and unsupported by substantial evidence. Specifically, Plaintiffs

and Plaintiff-Intervenor argue that the Electricity Program is not specific because it

is broadly available throughout the economy, Commerce’s grouping of industries it

uses to demonstrate that some industries receive a disproportionate amount of a

subsidy is arbitrary, and Commerce’s interpretation and understanding of

“disproportionate” is contrary to law. Hyundai Mot. at 11–19; Gov’t of Korea Mot. at

4–12. Plaintiff Hyundai and Plaintiff-Intervenor also argue Commerce abused its

discretion by failing to use accurate and verified data. Hyundai Mot. at 19–22; Gov’t

of Korea Mot. at 14–15. Plaintiff DSM argues that Commerce acted contrary to law

in conducting its Tier III analysis and determining that electricity was provided for

LTAR. DSM Mot. at 11–16. Finally, Plaintiff Hyundai and Plaintiff-Intervenor

challenge Commerce’s calculation of benefit, arguing that Commerce included

amounts irrelevant to the cost of supplying electricity in the POR. Hyundai Mot. at

22; Gov’t of Korea Mot. at 16. For the following reasons, Commerce’s determinations

are remanded for reconsideration or further explanation.

PUBLIC VERSION

I. Specificity

Plaintiffs and Plaintiff-Intervenor argue that Commerce’s determination is

unsupported by substantial evidence and not in accordance with law because (1) the

steel industry did not receive a disproportionately large amount of the subsidy alone,

or when grouped together with two other industries; and (2) Commerce’s decision to

group the steel industry with two other large industries for its specificity

determination is not reasonable. Hyundai Mot. at 2–3, 16–17. Defendant contends

that Commerce’s final determination is supported by substantial evidence and in

accordance with law because Commerce reasonably determined that a group of three

industries, including the steel industry, received a disproportionately large amount

of the Electricity Program subsidy, and accordingly, Plaintiff received a de facto

specific countervailable subsidy. Def. Resp. at 8; Final Decision Memo. at 17–18, 19.

For the following reasons, Commerce’s determination is remanded for further

explanation or reconsideration.

A “countervailable subsidy” is a financial contribution by a government

authority that confers a benefit on a recipient. 19 U.S.C. § 1677(5)(B). To be

countervailable, the subsidy must be “specific;” i.e., the subsidy must be (i) an import

substitution subsidy; (ii) an export subsidy; or (iii) a domestic subsidy that is specific

to an enterprise or industry within the jurisdiction of the authority providing the

subsidy. Id. § 1677(5A)(A)–(D). A domestic subsidy may be specific as a matter of PUBLIC VERSION

law (“de jure”) or as a matter of fact (“de facto”).3 Id. § 1677(5A)(D). A domestic

subsidy is de facto specific where:

(I) The actual recipients of the subsidy, whether considered on an enterprise or industry basis, are limited in number. (II) An enterprise or industry is a predominant user of the subsidy. (III) An enterprise or industry receives a disproportionately large amount of the subsidy. (IV) The manner in which the authority providing the subsidy has exercised discretion in the decision to grant the subsidy indicates that an enterprise or industry is favored over others. In evaluating the factors set forth in subclauses (I), (II), (III), and (IV), the administering authority shall take into account the extent of diversification of economic activities within the jurisdiction of the authority providing the subsidy, and the length of time during which the subsidy program has been in operation.

Id. § 1677(5A)(D)(iii). Thus, even where the number of industries receiving a subsidy

is not limited, and the authority does not exercise discretion in administering the

program, a subsidy may still be de facto specific if an industry is a predominant user

or receives a disproportionately large amount of the subsidy. See id.

§ 1677(5A)(D)(iii)(II)–(III).

Commerce determines whether an entity receives a disproportionately large

amount of a subsidy on a case-by-case basis by assessing the amount received “not in

3 A subsidy may be de jure specific under 19 U.S.C. § 1677(5A)(D)(i) where an authority or legislation expressly limits access to a subsidy to a sufficiently small number of enterprises, industries, or groups. Id.

PUBLIC VERSION

relation to the benefits of others, but in relation to some other comparator depending

on the circumstances.” Hyundai Steel Co. v. United States, 745 F. Supp. 3d 1345 at

1352–53 (Ct. Int’l Trade 2024) (“Hyundai I”) (citing AK Steel Corp. v. United States,

192 F.3d 1367, 1385 (Fed. Cir. 1999); Royal Thai Gov’t v. United States, 436 F.3d

1330, 1336 (Fed. Cir. 2006)). The phrase “disproportionately large” requires “an

enterprise or industry [be] favored in some way (i.e., it receives more than its fair

share [of the subsidy]).” Hyundai I, 745 F. Supp. 3d at 1352–53. Thus, the fact that

an industry or group of industries receives more—even much more—of an absolute

dollar amount from the program than others receive does not, by itself, establish

disproportionality. See Hyundai Steel Co. v. United States, 798 F. Supp. 3d 1283,

1288 (Ct. Int’l Trade 2025) (“Hyundai II”) (interpreting 19 U.S.C.

§ 1677(5A)(D)(iii)(III)). Therefore, Commerce must explain “how the combined

industries it identifies benefit more than would be expected, based on their usage

given that the subsidy in question is designed to confer benefits on usage levels, or in

relation to some other comparator.” Hyundai I, 745 F. Supp. 3d at 1352–53; see also

Samsung Electronics Co. Ltd. v. United States, 973 F. Supp. 2d 1321, 1326 (Ct. Int’l

Trade 2014) (“[i]n focusing solely on Samsung’s relative share of the total benefit,

Commerce failed to consider . . . that [the Gov’t of Korea] confers [] tax credits based

on usage . . . ”).

Commerce may identify “a group of . . . enterprises or industries” to evaluate

whether those enterprises or industries receives a disproportionately large amount PUBLIC VERSION

of the subsidy. 19 U.S.C. § 1677(5A)(D)(iii)(III); see, e.g., POSCO v. United States,

794 F. Supp. 3d 1402, 1408 (Ct. Int’l Trade 2025). Commerce has “reasonable

flexibility” in grouping enterprises or industries. See Mosaic Company v. United

States, 160 F.4th 1340, 1347–48 (Fed. Cir. 2025) (applying Commerce’s “reasonable

flexibility in carrying out the de facto specificity inquiry” to its determinations of

comparator groups for purposes of 19 U.S.C. § 1677(5A)(D)(iii)(II)) (citing AK Steel

Corp. v. United States, 192 F.3d 1367, 1385 (Fed. Cir. 1999)); 19 C.F.R. § 351.502(b)

(Commerce need not consider “whether there are shared characteristics among the

enterprises or industries” when grouping); POSCO, 794 F. Supp. 3d at 1408

(Commerce must reasonably exercise the discretion it is granted under 19 U.S.C.

§ 1677(5A)(D)(iii)(III) to make grouping determinations). Commerce’s flexibility is

“not unbounded.” Mosaic, 160 F.4th at 1348.

Here, Commerce determines that the steel industry and two other industries

received a disproportionately large amount of the Electricity Program subsidy

because the group of industries “consumed a majority of the industrial class

electricity during the POR.” Final Decision Memo. at 19. Commerce’s determination

identifies the raw amount of electricity consumed and the commensurate amount of

Electricity Program contribution received as a result of that consumption. Id. at 19.

Commerce explains the Gov’t of Korea has a “wide diversification of economic

activities . . . including 19 industry groupings with a broad range of distinctly

different types of economic activities within the groupings,” yet the steel industry and PUBLIC VERSION

two other industries,4 received a large amount of the subsidy. Id. at 18–19 (citing

Placement of Korea Economic Diversification Memo., PD 30, barcode 438930801

(June 14, 2023)). Commerce also quantifies the steel industry’s particular

consumption of electricity among industrial consumers.5 Id. at 19. However, contrary

to Defendant’s and Defendant-Intervenor’s arguments,6 the use of a larger amount of

electricity compared to other consumers does not indicate that the steel industry

receives a disproportionately large amount of the subsidy. See Def. Resp. at 19; Def-

Int. Resp. at 20 (citing AK Steel Corp. v. United States, 192 F.3d 1367, 1384 (Fed.

4 The other two industries are [[ ]], and [[ ]] and together, the three industries used [[ ]] percent of industrial class electricity. Def. Resp. at 13 (citing Gov’t of Korea IQR at 42–43). 5 Steel consumed [[ ]] percent of the total electricity consumed by industrial consumers in Korea during the POR. Def. Resp. at 13 (citing Gov’t of Korea IQR at 42–43). 6 Defendant’s and Defendant-Intervenor’s reliance upon AK Steel is misplaced. In AK Steel, Commerce determined that the steel industry, which received revaluation of assets under a Korean revaluation program, did not receive a disproportionately large amount of the benefit. AK Steel Corp. v. United States, 192 F.3d 1367, 1383 (Fed. Cir. 1999). Commerce found that the program was available to a large number of companies across a wide variety of industries, and the company’s assets were revalued upwards at the same average rate of increase as all other companies that revalued assets under the program. Id. at 1384–85. Domestic producers argued Commerce unreasonably analyzed disproportionality because it should have looked at the “percentage of the total benefit of a subsidy program” the industry is receiving. Id. at 1385. The Court of Appeals held that Commerce’s determination was reasonable because it considered the relative percentage benefit rather than the absolute benefit conferred. Id. at 1385. Unlike Commerce’s determination in AK Steel, in which Commerce reasonably compared the proportion of benefit each industry received, here, Commerce compares only the nominal amount of the subsidy each industry received, without taking into account whether that nominal amount is proportionate to the industries’ respective use of electricity. See id. at 1384–85; Final Decision Memo. at 19.

PUBLIC VERSION

Cir. 1999)); Hyundai I, 745 F. Supp. 3d at 1352–53; Hyundai II, 798 F. Supp. 3d at

1289. Commerce’s interpretation suggests that each industry should receive the

same portion of the subsidy regardless of the amount of electricity use. Such an

interpretation of “disproportionately large” under 19 U.S.C. § 1677(5A)(D)(iii)(III)

renders the “predominant user” factor under 19 U.S.C. § 1677(5A)(D)(iii)(II)

meaningless.

Defendants cannot rely on Mosaic to support Commerce’s finding here. See

Def. Notice of Supp. Authority at 2–3, Jan. 8, 2026, ECF No. 58 (citing Mosaic, 160

F.4th at 1346–47). In Mosaic, the Court of Appeals rejected the argument that

Commerce was limited in its choice of comparator groups for assessing whether a

recipient of a subsidy was a predominant user of the subsidy. See Mosaic, 160 F.4th

at 1346–47 (Fed. Cir. 2025). Commerce compared the agrochemical industry’s

subsidy use to other industrial users rather than to all subsidy users. Id. at 1346–

47. The Court of Appeals held that Commerce has “flexibility” to identify a

comparator group. Id. at 1346–47. Here, the question is not whether Commerce has

discretion in identifying a comparator for measuring disproportionality, but whether

Commerce has identified a comparator at all. See Hyundai II, 798 F. Supp. 3d at

1289 (“Rather, here Commerce fails to identify a comparator that would measure

disproportionality.”); Mosaic, 160 F.4th at 1346 (Commerce’s discretion is “not

unbounded, [and does not] provide Commerce with a loophole for rendering all

subsidies de facto specific. Rather, Commerce must engage in reasoned decision PUBLIC VERSION

making”) (internal citations omitted). Thus, absent additional explanation or

evidence that the steel industry received “more than its fair share” of the subsidy,

Commerce’s determination that the subsidy is de facto specific based on

disproportionality is not in accordance with law or supported by substantial evidence.

Commerce fails to point to record evidence that indicates the steel industry receives

a larger percentage of its subsidized electricity costs than other consumers receive

and therefore its determination is remanded for reconsideration.

Separately, Plaintiffs and Plaintiff-Intervenor also argue that Commerce’s

decision to group the steel industry with two other large industries was arbitrary and

capricious because Commerce did so without basis or explanation. Hyundai Mot. at

16–17; DSM Mot. at 8; Gov’t of Korea Mot. at 10. In the 2021 review, this Court

remanded Commerce’s decision to use the four “top industrial users of electricity in

Korea” as a group because Commerce failed to provide a rational explanation for the

grouping.7 Hyundai I, 745 F. Supp. 3d at 1353. Here, Commerce again finds that

“the steel industry and two other industries” consumed a majority of the industrial

class electricity during the POR. Final Decision Memo. at 19. Commerce explains

that the remaining seven industries in the top ten users of industrial class electricity

7 On remand, Commerce “offer[ed] the same basis for its decision” to group the four industries, and the Court remanded again. Hyundai II, 798 F. Supp. 3d at 1291. On second remand, Commerce abandoned its prior grouping and instead, reconsidered the basis of its determination and abandoned its prior grouping of certain industries. Hyundai Steel Co. v. United States, No. 23-00211, 2026 WL 1558544, at *1 n.1 (Ct. Int’l Trade May 18, 2026).

PUBLIC VERSION

“accounted for a significantly smaller share” of electricity consumed. Final Decision

Memo. at 19 (citing Certain Cut-To-Length Carbon-Quality Steel Plate from the

Republic of Korea; 2022: Final Specificity Analysis PR 320, CD 378–379 barcode

4626810-02 (Sept. 5, 2024)). As Commerce notes, under 19 C.F.R. § 351.502(b),

Commerce is not required to determine whether there are shared characteristics

among the grouped enterprises or industries. 19 C.F.R. § 351.502(b); Final Decision

Memo. at 19–20. Defendant and Defendant-Intervenor argue that because the

statute and regulations do not require Commerce’s grouping determinations to

“depend on ‘whether there are shared characteristics among the enterprises or

industries,’” Commerce’s decision to group the three largest users of electricity is

reasonable. Def. Int. Resp. at 12 (citing 19 C.F.R. § 351.502(b)); Def. Resp. at 14

(citing 19 C.F.R. § 351.502(b)). Though the regulations do not require Commerce to

determine grouping based on shared characteristics, Commerce is required to support

its formulation of a “group” with reasonably discernable explanation for its decision.

See Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto Ins. Co., 463 U.S. 29, 43

(1983). Commerce provides no explanation for its grouping decision beyond

emphasizing its discretion to define groups under 19 U.S.C. § 1677(5A)(D)(iii).8

8 Commerce reasons that because three industries consume such a large percentage of the electricity, Commerce is not required to “compare those unique industries within the group to others and their individual energy consumption or relative size.” Final Decision Memo. at 20. At oral argument, Defendant again argued that

(footnote continued)

PUBLIC VERSION

Without reasoned explanation, Commerce’s grouping determination is unsupported

by substantial evidence and therefore remanded for further explanation or

reconsideration.9

II. Commerce’s Decision Not to Rely Upon Revised Dataset

Plaintiff and Plaintiff-Intervenor challenge Commerce’s reliance on the Gov’t

of Korea’s first industrial electricity consumption dataset submission, rather than on

the second, revised submission, as unsupported by substantial evidence, contrary to

law, and arbitrary. See Hyundai Mot. at 19–22; Gov’t of Korea Mot. at 14–16.

Defendant argues that data verification is not mandatory, and Commerce has

discretion to choose the data upon which it relies to make its determinations. Def.

Commerce need not use an “external metric” to find disproportionality because of the substantially large use of the three industries and explained that the respondent could rebut that conclusion by pointing to an external metric. Oral Argument at 23:20–24:00. Defendant’s argument again conflates predominance with disproportionality. To find that an industry received a disproportionately large amount of a subsidy, Commerce must explain to what the receipt of the benefit is disproportionate and explain why the industry’s receipt of the benefit was disproportionate to that comparator; a raw comparison of energy consumption alone is insufficient. See Hyundai I, 745 F. Supp. 3d at 1352–53 (citing AK Steel, 192 F.3d at 1385); Hyundai II, 798 F. Supp. 3d at 1288. Further, Defendant cites no authority for the proposition that, just because an industry or group of industries consumes a large amount of an underlying resource, the respondent bears the burden of proving that its use is proportionate. Rather, Commerce must consider the record as a whole, including information that detracts from a finding that the industry received a disproportionate amount of the subsidy. See Timken U.S. Corp. v. United States, 421 F.3d 1350, 1355–56 (Fed. Cir. 2005). 9 Should Commerce, on remand, find that the Electricity Program is not de facto specific, the remaining issues regarding Commerce’s selection of data, determination of LTAR and calculation of benefits are moot. However, in the interest of judicial economy, the Court addresses each issue.

PUBLIC VERSION

Resp. at 23–24. For the following reasons, Commerce’s determination regarding the

choice of dataset is remanded for further explanation.

Commerce must make its determinations as “accurate as possible” when

administering the trade remedies laws. Rhone Poulenc, Inc. v. United States, 899

F.2d 1185, 1191 (Fed. Cir. 1990). To meet that obligation, Commerce must consider

“the entire record,” including evidence that detracts from its final conclusion.

Siemens Energy, Inc. v. United States, 806 F.3d 1367, 1369 (Fed. Cir. 2015) (citing

Universal Camera Corp. v. NLRB, 340 U.S. 474, 488 (1951)). Consistent with the

goal of accuracy, parties may submit factual information “to rebut, clarify or correct”

previously submitted information within fourteen days after the initial questionnaire

response. 19 C.F.R. § 351.301(b)(2), (c)(1)(v). However, Commerce may reject factual

information as untimely.10 Id. Nevertheless, acceptance incorporates information

into the administrative record, and Commerce must address that information in its

analysis. See Siemens, 806 F.3d at 1369. Although Commerce has discretion to

10 Even where Commerce’s rejects information as untimely, the Court of Appeals has explained that Commerce must, in some instances, accept untimely submissions where a party seeks to correct an error. Timken U.S. Corp. v. United States, 434 F.3d 1345, 1353–54 (Fed. Cir. 1993) (finding Commerce’s refusal to consider corrective data submitted before the final results an abuse of discretion); see also NTN Bearing Corp. v. United States, 74 F.3d 1204, 1208 (Fed. Cir. 1995) (holding that Commerce abused its discretion by failing to consider “untimely” corrective information submitted before the final determination); Fine Furniture (Shanghai) Ltd. v. United States, 865 F. Supp. 2d 1254, 1267 (Ct. Int’l Trade 2012) (finding an abuse of discretion where Commerce refused to correct an error with accurate, though untimely, information).

PUBLIC VERSION

choose among sets of information on the record, its discretion is not unlimited, and

its choice must be reasonable and adequately explained. The Mosaic Co. v. United

States, 589 F. Supp. 3d 1298, 1314 (Ct. Int’l Trade 2022) (citing Timken, 16 C.I.T. at

147).

Here, though Commerce did not request the Gov’t of Korea’s revised

submission and the Gov’t of Korea submitted the dataset beyond the regulation’s

fourteen-day window for correction, it was submitted on November 1, 2023, before

Commerce published the Preliminary Results in February 2024, before Commerce

began verification in June 2024, and ten months before Commerce issued the Final

Results. See Preliminary Results; Final Decision Memo. at 21; Hyundai Mot. at 7.

Commerce may reject untimely corrective submissions, but here, it chose to accept

and verify the second corrective submission. See Final Decision Memo. at 22; 19

C.F.R. § 351.301(c)(1)(v). Therefore, when Commerce accepted the revised dataset,

it was obligated to consider the data in its analysis. See Siemens, 806 F.3d at 1369.

Defendant argues that by submitting revised data in its supplemental

questionnaire response, the Gov’t of Korea was “non-responsive” to the questionnaire,

which asked the Gov’t of Korea to “provide KEPCO’s unconsolidated financial

statements for 2022 . . . .” Oral Argument at 41:30–42:30, July 9, 2026, ECF No. 77

(“Oral Argument”); Gov’t of Korea 2SQR at 1. The Gov’t of Korea responded by

pointing Commerce to an attached exhibit providing the requested financial

statements and stating “[t]he [Gov’t of Korea] would like to use this opportunity to PUBLIC VERSION

provide the Department with a more detailed and accurate response regarding the

amount of electricity consumed by the 10 largest industries during the POR.” Gov’t

of Korea 2SQR at 1. Commerce explains that the first dataset is responsive and

therefore “more appropriate,” than the revised dataset, however, Commerce fails to

explain why it is “more appropriate” or why it relied on a “more appropriate” dataset

that appears to be less accurate. See Final Decision Memo. at 22.

Commerce likewise fails to explain why the second, revised dataset, which it

accepted as part of the record, is “not usable.” See id. The Gov’t of Korea, in its second

questionnaire response, reasoned that the revised dataset is more “specific” and thus

more accurately represented electricity usage as categorized by industry.11 See Gov’t

of Korea 2SQR at 2. Commerce concludes that the data is unusable but fails to

explain why more “specific” data would not be usable to calculate accurately. Final

Decision Memo. at 22. Although Commerce asserts that the revised dataset reports

electricity consumption at “a lower level of industrial classification” than requested,12

11 Indeed, one of the corrections identified in the Gov’t of Korea’s revised dataset is a revision of the steel industry’s usage category, as previously represented, corrected to reflect usage by exclusively the “steel industry.” See Gov’t of Korea 2SQR at 2. Specifically, the Gov’t of Korea indicated that the initial dataset had erroneously included usage from the steel, copper, non-ferrous metal and aluminum industries under steel, see Gov’t of Korea 2SQR at 1–3, which led to an overinflated calculation of the electricity consumed by the steel industry. Hyundai Mot. at 21–22. Hyundai explains that the second, revised dataset provided more detailed and therefore more accurate representation of industry use to determine whether the steel industry received a greater benefit from KEPCO. Id. at 21–22. 12 The initial questionnaire requested the Gov’t of Korea “provide the amount of electricity provided to the steel industry during the POR and to the largest 10 PUBLIC VERSION

and uses a different format from the initial dataset, 13 it does not explain why the

revised dataset lacks sufficient detail, why its alternative format renders it

unsuitable for consideration, or why the initial dataset is both “usable” and more

accurate than the revised data. Id. at 21–22. Commerce also argues that the

Plaintiffs submitted the revised dataset, claiming it is more accurate, but failed to

provide “supporting documentation” for that claim. Id. at 21–22. However, neither

the regulations nor the questionnaire requires parties to submit supporting

documentation for submissions of factual information. See Gov’t of Korea IQR. The

regulations define “factual information” as “[e]vidence, including statements of fact,

documents and data submitted either in response to initial and supplemental

questionnaires . . . .” 19 C.F.R. § 351.102(b)(21)(i). Just as the initial dataset

constitutes factual information submitted without supporting documentation, the

revised dataset is also factual information, which Commerce accepted into the record

and is therefore required to provide reasonable explanation as to why it was not

industries consuming electricity during the POR.” Initial Questionnaire at 20. It did not request that data be provided in accordance with specific standards or “levels of industrial classification.” See id. at 20. 13 Specifically, Defendant explains that the revised dataset conflicts with Commerce’s requirements because it provides data for “manufacture of basic iron and steel”, which was not an industry listed in the Gov’t of Korea’s initial questionnaire response, and because the Gov’t of Korea did not explain why the steel category named in the first data submission also encompassed “copper and non-ferrous metal, aluminum, etc.” Def. Resp. at 22 (citing Gov’t of Korea 2SQR), 23.

PUBLIC VERSION

“usable.” Commerce’s decision not to rely on the corrective dataset provided by the

Gov’t of Korea is therefore remanded for further consideration or explanation.

III. Determination of LTAR

Plaintiff DSM argues that Commerce’s finding of LTAR is unsupported by

substantial evidence and contrary to law. DSM Mot. at 7–8. DSM argues where an

entity sets prices in accordance with market principles, neither Commerce’s

regulation nor practice requires the entity earn a profit every year. Id. at 7–8.

Further, DSM argues Commerce’s practice is to ensure that an entity’s cost recovery

is sufficient to ensure future operations, not to require a specific rate of return. Id.

at 11–14.14 Defendant explains that Commerce examined both whether KEPCO’s

tariff system was set with market principles, and whether it is applied using market

principles. Def. Resp. at 28 (citing Final Decision Memo. at 13–14). Defendant also

explains that despite KEPCO’s price setting methodology being based upon market

principles, including “processes to fully recover costs and a rate of return,” the data

showed the “implementation of this pricing methodology ‘conferred a benefit to the

purchasers of industrial categories of electricity.’” Id. As a result, in analyzing

KEPCO’s tariff system, Commerce determined that KEPCO did not recover its costs

14 Plaintiff-Intervenor Gov’t of Korea makes a related argument when it argues that Commerce (1) failed to properly consider KEPCO’s FCAC and pass-through system when determining that KEPCO did not recover costs within the POR and therefore, that the system is not based on market principles, and at the same time; (2) included KEPCO’s FCAC in its calculation of benefits. See Gov’t of Korea Mot. at 16–19.

PUBLIC VERSION

(inclusive of return) from industrial users during the POR, and therefore, KEPCO

provided electricity at LTAR when the tariff system was applied. Final Decision

Memo. at 14. For the following reasons, Commerce’s determination is remanded for

further consideration or explanation.

Commerce determines whether a financial contribution confers a benefit on a

recipient by evaluating whether goods and services are provided for less than

adequate remuneration. 19 U.S.C. § 1677(5)(E)(iv); see also Nucor Corp. v. United

States, 927 F.3d 1243, 1251 (Fed. Cir. 2019). Commerce assesses adequacy of

remuneration in relation to prevailing market conditions in the country under

investigation or review. 19 U.S.C. § 1677(5)(E)(iv). It evaluates “price, quality,

availability, marketability, transportation, and other conditions of purchase or sale.”

Id. Commerce implements this standard through a three-tier framework to

determine whether a recipient is receiving a good or service below fair-market value.

See 19 C.F.R. § 351.511(a)(2); Nucor Corp., 927 F.3d at 1246.

Relevant here, the regulation provides:

(iii) World market price unavailable. If there is no world market price

available to purchasers in the country in question, the Secretary will

normally measure the adequacy of remuneration by assessing whether

the government price is consistent with market principles. In making

an assessment of whether a government price is consistent with market

principles under this provision, the Secretary may assess such factors

as costs (including rates of return sufficient to ensure future operations),

the government's price setting methodology, possible price

discrimination, or a government price derived from actual sales from

competitively run government auctions . . .

PUBLIC VERSION

19 C.F.R. § 351.511(a)(2)(iii).15 If neither a market-determined nor a world market

price is available, Commerce evaluates whether the government price was set in

accordance with “market principles.” 19 C.F.R. § 351.511(a)(2)(iii); see Nucor Corp.

v. United States, 927 F.3d 1243, 1254 (Fed. Cir. 2019) (citing Countervailing Duties,

63 Fed. Reg. 65,348, 65,378 (Nov. 25, 1998)).

At Tier III, the regulations allow Commerce to assess whether a company

benefits from a subsidy by looking at long term cost recovery. 19 C.F.R.

§ 351.511(a)(2)(iii). Thus, Commerce examines whether the government price is

15 Tiers I and II provide:

(i) In general. The Secretary will normally seek to measure the adequacy of remuneration by comparing the government price to a market-determined price for the good or service resulting from actual transactions in the country in question. Such a price could include prices stemming from actual transactions between private parties or actual imports. In choosing such transactions or sales, the Secretary will consider product similarity; quantities sold or imported; and other factors affecting comparability.

(ii) Actual market-determined price unavailable. If there is no useable market-determined price with which to make the comparison under paragraph (a)(2)(i) of this section, the Secretary will seek to measure the adequacy of remuneration by comparing the government price to a world market price where it is reasonable to conclude that such price would be available to purchasers in the country in question. Where there is more than one commercially available world market price, the Secretary will average such prices to the extent practicable, making due allowance for factors affecting comparability.

19 C.F.R. § 351.511(a)(2)(i)–(ii).

PUBLIC VERSION

consistent with market principles by considering “such factors as” the government

supplier’s price-setting philosophy, its costs (including rates of return sufficient to

ensure future operations), and possible price discrimination. See Countervailing

Duties, 63 Fed. Reg. 65,348, 65,378 (Nov. 25, 1998) (final rule adopting Commerce’s

countervailing duty regulations under the Uruguay Round Agreements Act,

including the adequacy of remuneration benchmark regulation codified at 19 C.F.R.

§ 351.511). The rule further explains that the factors are non-hierarchical, and

Commerce may consider one or more of the factors as “circumstances of each case

vary widely.” Id.

Here, Commerce appears to conclude that KEPCO operates within a long-term

cost recovery framework. When Commerce examined whether KEPCO’s prices were

set in accordance with market principles, it analyzed both “KEPCO’s price-setting

philosophy and costs (including rates of return sufficient to ensure future

operations).” Final Decision Memo. at 12; Prelim. Decision Memo. at 34 (examining

KEPCO’s pricing mechanism that recovers costs and separately examining individual

tariff rate cost recovery). Commerce reviewed the 2021 updates to the Gov’t of Korea’s

electricity tariff system and found that the 2021 revisions were adopted through the

same process and methodology used for the 2013 tariff revisions, which Commerce

had previously determined were consistent with market principles. Final Decision

Memo. at 12 (citing Gov’t of Korea IQR at 14–18, 43–47, Exs. E-7, E-8, E-8.1, E-8.2,

E-8.3, E-10, E-15). Commerce also examined the FCAC added in 2021 and concludes PUBLIC VERSION

that—by design and under its price-setting methodology—the charge itself is likewise

consistent with market principles. Final Decision Memo. at 12–13.

Nonetheless, Commerce examined KEPCO’s cost recovery rates during the

POR and explains:

Finally, we examined KEPCO’s cost recovery rates by tariff

classification during the POR, which are submitted to the MOTIE each

year. As noted in the Preliminary Results, these data demonstrate that

no industrial class tariff rate recovered costs (inclusive of a rate of

return) during the POR. This finding did not negate our analysis that

KEPCO’s pricing methodology is based upon “laws, regulations, and

processes to fully recover costs and a rate of return;” it merely highlights

the complexity of evaluating KEPCO’s electricity pricing and reflects the

fact that an examination of the actual application of the aforementioned

laws and regulations, particularly with regard to cost recovery, is

necessary to confirm that the Korean electricity market was based on

market principles in both law and application of the law. Commerce

must consider not only the mechanism, but also its implementation, and

in this regard, we find that KEPCO conferred a benefit to the purchasers

of industrial categories of electricity.

Id. at 13.16 Thus, although Commerce finds that KEPCO’s pricing methodology was

in accordance with market principles, it concludes that the cost-recovery analysis

16 In the Final Results, Commerce either misconstrues the regulatory requirements under 19 C.F.R. § 351.511(a)(2)(iii) or attempts to create a hybrid methodology not provided by the regulations to conform the facts to its conclusion. The regulations do not instruct Commerce to first analyze the “mechanism” to find prices were set in accordance with market principles and then separately, analyze the “implementation,” as Commerce did here. See Final Decision Memo. at 13. Rather, the regulations instruct Commerce to consider cost recovery, both in the mechanism

(footnote continued) and in the implementation of government prices, as a part of its determination of whether the government price was set in accordance with market principles, and PUBLIC VERSION

showed that no industrial electricity tariff rate recovered costs during the POR, and

therefore, the system, as applied, is not in accordance with market principles. Final

Decision Memo. at 13. Despite its design based on “laws, regulations, and processes

to fully recover costs and a rate of return,” the tariff system, in operation, failed to

produce cost recovery during the POR; therefore, Commerce concludes that KEPCO

nevertheless conferred a benefit on Plaintiffs. Id. at 13.

Thus, Commerce appears to either modify the Tier III framework it purports

to apply, or it unartfully explains its analysis and application of the Tier III

framework. Commerce finds KEPCO’s price-setting methodology to be “consistent

with market principles.” Id. at 13. If KEPCO’s price setting is consistent with market

principles, then Commerce has no basis to find that the program provided a good or

service for LTAR under 19 C.F.R. § 351.511(a)(2)(iii). Alternatively, it may be that

Commerce does not conclude that the system is consistent with market principles,

because at the same time Commerce finds the price-setting methodology suitable to

recover costs, and therefore consistent with market principles, it appears to also

doubt whether there will in fact be long-term costs recovery. Final Decision Memo.

at 13. The doubt appears to stem from KEPCO’s failure to recover costs during this

POR. Id. at 14; Oral Argument at 1:18–1:24. Yet, Commerce has explicitly rejected

using a single year of charges to customers to undermine the Tier III cost recovery

thus, whether a benefit was conferred. See Countervailing Duties, 63 Fed. Reg. 65,348, 65,378 (Nov. 25, 1998) (interpreting 19 C.F.R. § 351.511(a)(2)(iii)).

PUBLIC VERSION

analysis. See Large Diameter Welded Pipe from the Republic of Korea: Final Results

of Countervailing Duty Administrative Review; 2018-2019, 87 Fed. Reg. 5780 (Feb.

2, 2022) and accompanying Issues & Decisions Memo. at cmt. 1 (Feb. 2, 2022)

(explaining that Commerce’s Tier III analysis consistently evaluates an entity’s

overall “financial performance . . . rather than the prices actually paid by the

respondent” during the POR.) Thus, while Commerce says that KEPCO’s prices are

consistent with market principles, it seems to ultimately conclude that they are not.

See Final Decision Memo. at 13. On remand, Commerce must reconsider its Tier III

analysis or further explain how the regulations provide for a finding that prices set

in accordance with market principles can also fail to recover costs and therefore confer

a benefit.

IV. Benefit Calculation

Once Commerce determined that KEPCO provided electricity conferring a

countervailable benefit under Tier III, Commerce calculated the extent to which a

benefit was conferred. See Final Decision Memo. at 13; 19 U.S.C. § 1677(5)(E)(iv).

Plaintiffs contend that Commerce improperly included the FCAC and carried-forward

2021 costs in its benefit calculation, arguing the inclusion of those costs is an

irrelevant and unreasonable representation of KEPCO’s actual 2022 tariff and cost-

recovery methodology and results in “double counting.” Hyundai Mot. at 24–26; Gov’t

of Korea Mot. at 16–18. Defendant argues that if the FCAC is built into the tariff

customers actually pay, then Commerce properly determined the FCAC is part of the PUBLIC VERSION

revenue stream that the Gov’t of Korea uses to determine whether KEPCO recovered

electricity costs, and therefore, it is necessarily included in the benefit calculation.

Def. Resp. at 30–31; Final Decision Memo. at 23–24. For the following reasons,

Commerce’s calculation of benefit is remanded.

Where Commerce applies the tier-three benchmark criteria and finds that the

government-charged price is inconsistent with market principles, it may treat a

benefit as conferred and calculate the amount of that benefit using the benchmark it

has selected. See 19 U.S.C. § 1677(5)(E)(iv); 19 C.F.R. § 351.511(a)(2)(iii); POSCO v.

United States, 977 F.3d 1369, 1372 (Fed. Cir. 2020). In applying the criteria,

Commerce must reasonably explain its selected benchmark and inputs it uses to

make the benefit calculation. See, e.g., ArcelorMittal USA LLC v. United States, 337

F. Supp. 3d 1285, 1291 (Ct. Int’l Trade 2018).

Whether Commerce employs a yearly cost of goods sold analysis or a long-term

cost recovery analysis, it cannot remedy the same benefit twice and record evidence

suggests that Commerce may have done so. In 2022, the 2021 fuel costs that were

not charged to customers, by way of the FCAC, were included as a line item that

KEPCO used to calculate its base charge. Hyundai Mot. at 23–24 (citing Gov’t of

Korea IQR at Exhibit E-9). Commerce included KEPCO’s carried-forward 2021 fuel

costs in its 2022 POR adequacy of remuneration analysis because KEPCO treated the

unrecovered fuel costs as part of its 2022 Total Comprehensive Cost and used that

figure to set the tariff rates in effect during the POR. Final Decision Memo. at 23– PUBLIC VERSION

24. Commerce reasons that because the carried-forward amounts were built into the

tariff framework governing 2022 electricity prices, those amounts were relevant to

determining whether KEPCO’s 2022 rates allowed recovery of costs plus a reasonable

return.17 Id. at 14. Commerce determined that excluding those amounts would

distort the Tier III benchmark inquiry by severing the analysis from KEPCO’s real-

world tariff-setting methodology. Id. at 14, 15. Further, Commerce determined that

the carried-through amounts should be included in the benefit calculation because

they were paid by consumers as a part of the tariff schedule. Id. at 24.

Yet, KEPCO’s financial statement submissions indicate that fuel costs were

countervailed in 2021.18 Hyundai Mot. at 23 (citing Gov’t of Korea IQR at Exhibit E-

9). Commerce fails to explain why it was reasonable to calculate the benefit incurred

in 2022 by including costs that were reported in KEPCO’s 2021 financial statements

and appeared to have been countervailed. Commerce, on remand, must provide

sufficient explanation and evidence to show that the same pass-through charges were

not countervailed twice, or reconsider its calculation of benefit to ensure any benefit

is not remedied twice.

CONCLUSION

17 The carried-forward fuel costs from 2021 were not included in the [[

]] for 2022, but as a [[ ]] amount in the Total Comprehensive Cost. Hyundai Mot. at 23–24 (citing Gov’t of Korea IQR at Ex. E-9). 18 KEPCO’s fuel costs were reflected in its [[ ]] in 2021. Gov’t of Korea IQR at Ex. E-9.

PUBLIC VERSION

Commerce’s determinations that: (1) the steel industry and two other

industries received a disproportionately large amount of the electricity subsidy and

therefore the program is specific; (2) the revised dataset provided by Gov’t of Korea

is unusable; (3) electricity was provided at LTAR; and (4) the fuel cost pass-through

charges should be included in the calculation of benefit are remanded for further

explanation or reconsideration consistent with this opinion. In light of the foregoing,

it is

ORDERED that the final results, see ECF No. 25-5, are remanded for further

explanation or reconsideration consistent with this opinion; and it is

ORDERED that Commerce shall file its remand redetermination with the

Court within 90 days of this date; and it is further

ORDERED that the parties shall have 30 days to file comments on the remand

redetermination; and it is further

ORDERED that the parties shall have 30 days to file their replies to the

comments on the remand redetermination; and it is further

ORDERED that the parties shall file the joint appendix, including the entire

confidential record, within 14 days after the filing of replies to the comments on the

remand redetermination; and it is further

ORDERED that Commerce shall file the administrative record within 14 days

of the date of filing its remand redetermination.

PUBLIC VERSION

/s/ Claire R. Kelly

Claire R. Kelly, Judge

Dated: August 11, 2026

New York, New York