ArcelorMittal USA LLC v. United States , 337 F. Supp. 3d 1285 ( 2018 )


Menu:
  •                                         Slip Op. 18-121
    UNITED STATES COURT OF INTERNATIONAL TRADE
    ARCELORMITTAL USA LLC,
    Plaintiff,
    and
    AK STEEL CORPORATION,
    NUCOR CORPORATION, AND
    UNITED STATES STEEL CORPORATION,
    Plaintiff-Intervenors,
    NOVOLIPETSK STEEL PUBLIC JOINT
    STOCK COMPANY,                                  Before: Gary S. Katzmann, Judge
    Consol. Court No. 16-00168
    Consolidated Plaintiff,
    v.
    UNITED STATES,
    Defendant,
    and
    PAO SEVERSTAL AND
    SEVERSTAL EXPORT GMBH,
    Defendant-Intervenors.
    OPINION
    [Commerce’s Final Determination is remanded for further explanation as appropriate.]
    Dated: September 19, 2018
    John M. Herrmann, II, and Brooke Ringel, Kelly Drye & Warren, LLP, of Washington, DC, argued
    for plaintiff. With them on the joint brief were Alan J. Price, Timothy C. Brightbill, and
    Christopher B. Weld, Wiley Rein LLP, of Washington, DC, for plaintiff-intervenor, Nucor
    Corporation; Thomas M. Beline, Cassidy Levy Kent, LLP, of Washington, DC, for plaintiff-
    Court No. 16-00168                                                                       Page 2
    intervenor, United States Steel Corporation, and on the brief were Jeffrey D. Gerrish and Luke A.
    Meisner, of Skadden, Arps, Slate, Meagher & Flom LLP, of Washington, DC; and Daniel L.
    Schneiderman and Stephen A. Jones, King & Spalding LLP, of Washington, DC, for plaintiff-
    intervenor, AK Steel Corporation.
    Matthew P. McCullough, Marat S. Umerov, and Tung A. Nguyen, Curtis, Mallet-Prevost, Colt &
    Mosle LLP, of Washington, DC, argued for consolidated plaintiff. With them on the brief were
    William H. Barringer and Valerie S. Ellis.
    Patricia M. McCarthy, Assistant Director, Commercial Litigation Branch, Civil Division, U.S.
    Department of Justice, of Washington, DC, argued for defendant. With her on the brief were Chad
    A. Readler, Principal Deputy Assistant Attorney General, Jeanne E. Davidson, Director, and Renee
    A. Burbank, Trial Attorney. Of counsel on the brief was Michael T. Gagain, Senior Attorney,
    Office of the Chief Counsel for Trade Enforcement & Compliance, U.S. Department of
    Commerce, of Washington, DC.
    Daniel J. Cannistra, Crowell & Moring LLP, of Washington, DC, argued for defendant-
    intervenors.
    Katzmann, Judge: The complex litigation that unfolds in the United States Court of
    International Trade is typically populated by multiple parties and agencies, shifting alignments,
    and intersecting claims and issues. Its resolution often calls for diagramming on a blackboard
    with multi-colored chalk. The case now before this Court may be viewed in such context.
    Plaintiffs in two separate cases -- ArcelorMittal USA LLC (“ArcelorMittal”) in one case,
    and Novolipetsk Steel Public Joint Stock Company (“NLMK”) in the other -- challenged different
    elements of the United States Department of Commerce’s (“Commerce”) final affirmative
    determination of its Countervailing Duty Investigation of Certain Cold-Rolled Steel Flat Products
    From the Russian Federation, 81 Fed. Reg. 49,935 (Dep’t Commerce July 29, 2016) (“Final
    Determination”) and the accompanying July 20, 2016 Issues and Decision Memorandum, C–821–
    823 (“IDM”). The United States (“the Government”), on behalf of Commerce, was the defendant
    in both cases. Domestic steel manufacturers AK Steel Corporation, Nucor Corporation, and
    United States Steel Corporation joined ArcelorMittal’s case as plaintiff-intervenors and supported
    its arguments; all four companies joined NLMK’s case as defendant-intervenors and supported the
    Court No. 16-00168                                                                      Page 3
    Government’s position with respect to the elements of the Final Determination that NLMK
    contested. In addition, PAO Severstal and its wholly-owned affiliate, Severstal Export GMBH
    (collectively, “Severstal”) joined ArcelorMittal’s case as defendant-intervenor and supported the
    Government’s position with respect to the elements of the Final Determination that ArcelorMittal
    contested. The Court, upon motion by the parties, consolidated the two cases into the case
    currently before the Court.
    In essence, this case requires the Court to assess the countervailing duty (“CVD”) rates
    selected, as well as the invocation and application of adverse facts available (“AFA”), by
    Commerce in its Final Determination. Commerce applied AFA in determining CVD rates for both
    company respondents to the CVD investigation, Severstal and NLMK. Plaintiff ArcelorMittal
    contests the CVD rate Commerce applied to respondent and defendant-intervenor Severstal, while
    consolidated plaintiff NLMK contests the CVD rate Commerce applied to it.
    The primary question posed with respect to Severstal’s CVD rate is whether Commerce
    erred in its selection of a program-specific AFA rate for Severstal’s unreported use of the tax
    deduction program for mining-exploration expenses.
    The following questions are posed with respect to NLMK’s CVD rate: (1) Was
    Commerce’s determination -- that the Government of Russia’s (“the GOR”) alleged provision of
    natural gas for less than adequate remuneration (“LTAR”) was de facto specific to steel
    manufacturing -- supported by substantial evidence and in accordance with law? (2) Were
    Commerce’s benefit and benchmark determinations supported by substantial evidence and in
    accordance with law?
    Court No. 16-00168                                                                         Page 4
    BACKGROUND
    A.     Countervailable Subsidies Generally.
    If Commerce determines that the government of a country is providing, directly or
    indirectly, a countervailable subsidy with respect to the manufacture, production, or export of a
    class or kind of merchandise imported, or sold, or likely to be sold for import, into the United
    States, and the International Trade Commission determines that an industry in the United States is
    materially injured or threatened with material injury thereby, then Commerce shall impose a
    countervailing duty upon such merchandise equal to the amount of the net countervailable subsidy.
    See Section 701 of the Tariff Act of 1930, as amended, 19 U.S.C. § 1671(a) (2012). 1 An
    investigation of countervailable subsidies shall commence whenever an interested party files a
    petition with Commerce, on behalf of an industry, 2 which alleges the elements necessary for the
    imposition of the duty, and which is accompanied by information reasonably available to the
    petitioner supporting those allegations. 19 U.S.C. § 1671a(b)(1), (c)(2).
    Generally, a subsidy is countervailable if it consists of a foreign government’s financial
    contribution to a recipient, which is specific, and also confers a benefit upon the recipient, as
    defined under 19 U.S.C. § 1677(5). A benefit is conferred when, in the case where goods or
    1
    Further citations to the Tariff Act of 1930, as amended, are to the relevant provision of Title 19
    of the U.S. Code, 2012 edition. Citations to 19 U.S.C. § 1677e, however, are not to the U.S. Code
    2012 edition, but to the unofficial U.S. Code Annotated 2018 edition. The current U.S.C.A.
    reflects the amendments made to 19 U.S.C. § 1677e (2012) by the Trade Preferences Extension
    Act of 2015, Pub. L. No. 114–27, § 502, 129 Stat. 362, 383–84 (2015). The TPEA amendments
    are applicable to all determinations made on or after August 6, 2015, and therefore, are applicable
    to this proceeding. See Dates of Application of Amendments to the Antidumping and
    Countervailing Duty Laws Made by the Trade Preferences Extension Act of 2015, 80 Fed. Reg.
    46,793, 46,794 (Dep’t Commerce Aug. 6, 2015).
    2
    “The term ‘industry’ means the producers as a whole of a domestic like product, or those
    producers whose collective output of a domestic like product constitutes a major proportion of the
    total domestic production of the product.” 19 U.S.C. § 1677(4)(A).
    Court No. 16-00168                                                                       Page 5
    services are provided, such goods or services are provided for less than adequate remuneration.
    19 U.S.C. § 1677(5)(E)(iv). Furthermore, the statute states that:
    [T]he adequacy of remuneration shall be determined in relation to prevailing market
    conditions for the good or service being provided or the goods being purchased in
    the country which is subject to the investigation or review. Prevailing market
    conditions include price, quality, availability, marketability, transportation, and
    other conditions of purchase or sale.
    
    Id. The regulation
    on “adequate remuneration” provides a multi-tiered analysis under which
    Commerce may determine a suitable benchmark for the purposes of determining the existence and
    amount of a benefit conferred. Under Tier One, 19 C.F.R. § 351.511(a)(2)(i), Commerce assesses
    market prices from actual transactions within the country under investigation; under Tier Two, 
    id. § 351.511(a)(2)(ii),
    Commerce assesses world market prices that would be available to purchasers
    in the country under investigation; and under Tier Three, 
    id. § 351.511(a)(2)(iii),
    Commerce
    assesses whether the government price is consistent with market principles. See IDM at 16.
    The subsidy must also be “specific,” either in law or fact, as defined under 19 U.S.C. §
    1677(5A). As relevant to this case, a subsidy may be de facto specific when “[a]n enterprise or
    industry is a predominant user of the subsidy.” 19 U.S.C. § 1677(5A)(D)(iii)(II).
    B.     Adverse Facts Available.
    During the course of its CVD proceeding, Commerce requires information from both the
    producer respondent and the foreign government alleged to have provided the subsidy. See Fine
    Furniture (Shanghai) Ltd. v. United States, 
    748 F.3d 1365
    , 1369–70 (Fed. Cir. 2014). Information
    submitted to Commerce during an investigation is subject to verification. 19 U.S.C. § 1677m(i)(1).
    When a respondent: (1) withholds information that has been requested by Commerce, (2)
    fails to provide such information by Commerce’s deadlines for submission of the information or
    in the form and manner requested, (3) significantly impedes an antidumping proceeding, or (4)
    Court No. 16-00168                                                                          Page 6
    provides information that cannot be verified, then Commerce shall “use the facts otherwise
    available in reaching the applicable determination.” 19 U.S.C. § 1677e(a)(2). This subsection
    thus asks whether necessary or requested information is missing from the administrative record,
    and provides Commerce with a methodology to fill the resultant informational gaps. See Nippon
    Steel Corp. v. United States, 
    337 F.3d 1373
    , 1381 (Fed. Cir. 2003).
    Under certain circumstances, in an investigation, Commerce may assign an AFA rate to an
    investigated respondent as to a given subsidy program, instead of the countervailable subsidy rate
    that the respondent might receive for that program under normal circumstances. Specifically,
    Commerce “may use an inference that is adverse to the interests of that party in selecting from
    among the facts otherwise available” if it “finds that an interested party has failed to cooperate by
    not acting to the best of its ability to comply with a request for information[.]” 19 U.S.C. §
    1677e(b)(1)(A). A respondent’s failure to cooperate to “the best of its ability” is “determined by
    assessing whether [it] has put forth its maximum effort to provide Commerce with full and
    complete answers to all inquiries.” Nippon 
    Steel, 337 F.3d at 1382
    ; see Özdemir Boru San. ve
    Tic. Ltd. Sti. v. United States, 41 CIT ___, ___, 
    273 F. Supp. 3d 1225
    , 1231, 1241–42 (2017).
    When applying an adverse inference, Commerce may rely on information from the petition,
    a final determination in the investigation, a previous administrative review, or any other
    information placed on the record. 19 U.S.C. § 1677e(b)(2); 19 C.F.R. § 351.308(c). Notably, if
    Commerce uses an adverse inference under § 1677e(b)(1)(A) in selecting among facts otherwise
    available, Commerce is not required to demonstrate that the dumping margin used “reflects an
    alleged commercial reality of the interested party.” 19 U.S.C. § 1677e(d)(3); see Hyundai Steel
    Co. v. United States, 41 CIT ___, ___, 
    279 F. Supp. 3d 1349
    , 1355–56 (2017).
    Court No. 16-00168                                                                           Page 7
    Typically, an AFA rate is higher than the normally calculable subsidy rate for an
    investigated program, and thus ultimately results in a higher CVD rate. See 19 U.S.C. § 1677e.
    Commerce maintains that its practice ensures “that the party does not obtain a more favorable
    result by failing to cooperate than if it had cooperated fully.” 
    Özdemir, 273 F. Supp. 3d at 1233
    (quoting Statement of Administrative Action, accompanying the Uruguay Round Agreements Act,
    H.R. No. 103–316, vol. 1, at 870 (1994), reprinted in 1994 U.S.C.C.A.N. at 4199 (“SAA”)). 3
    Commerce’s practice when selecting an adverse rate from among the possible sources of
    information is also intended to ensure that the applied rate is sufficiently adverse to the respondent
    so as to deter future noncompliance. 4 See 
    id. at 1245.
    Commerce has developed a hierarchy when selecting the subsidy rate to be applied
    pursuant to AFA. IDM at 14–15, 126. Specifically, in the first step of the CVD AFA hierarchy -
    - the step applied, and primarily at issue, in this case -- Commerce examines whether, “in the
    context of the instant investigation, there is a calculated program subsidy rate for the identical
    program at issue. If so, [Commerce] will use the calculated program rate for that particular
    program as the basis of the AFA rate.” 
    Id. at 15.
    If there is no identical program match within the
    3
    See 19 U.S.C. § 3512(d), which states in relevant part that “[t]he statement of administrative
    action . . . shall be regarded as an authoritative expression by the United States concerning the
    interpretation and application of the Uruguay Round Agreements and this Act in any judicial
    proceeding in which a question arises concerning such interpretation or application.” See also
    RHP Bearings Ltd. v. United States, 
    288 F.3d 1334
    , 1346 n.7 (Fed. Cir. 2002).
    4
    The relevant statutory provision covering application of adverse inferences addresses both
    countervailing duty cases and antidumping cases in tandem. See generally 19 U.S.C. § 1677e.
    Federal Circuit case precedent too demonstrates that the statute’s underlying purposes are equally
    applicable to countervailing duty cases. See Fine 
    Furniture, 748 F.3d at 1372
    –73 (applying
    statutory purpose of ensuring uncooperative party does not benefit from its non-cooperation in
    countervailing duty case); KYD, Inc. v. United States, 
    607 F.3d 760
    , 768 (Fed. Cir. 2010)
    (applying statutory purpose of deterrence in countervailing duty case).
    Court No. 16-00168                                                                         Page 8
    investigation, or if the rate is zero, then Commerce in the second step of the hierarchy applies the
    highest non-de minimis rate calculated for the same program in another CVD investigation
    involving the same country. If there is none, then in the third step Commerce uses the highest non-
    de minimis rate calculated for a similar program in another CVD investigation involving the same
    country. If that step also does not produce an applicable AFA rate, then Commerce in the fourth
    step uses the highest calculated subsidy rate for any program otherwise identified in a CVD case
    involving the same country that could conceivably be used by the non-cooperating companies.
    See 
    Özdemir, 273 F. Supp. 3d at 1232
    (discussing Commerce’s AFA hierarchy); Essar Steel. Ltd.
    v. United States, 
    753 F.3d 1368
    , 1371–74 (Fed. Cir. 2014); see, e.g., Welded Line Pipe from The
    Republic of Turkey, 80 Fed. Reg. 61,371 (Dep’t of Commerce Oct. 13, 2015) (final affirm. CVD
    determ.), and accompanying Issues & Decision Memorandum at 4–7.
    C.     Factual and Procedural Background.
    On August 17, 2015, in response to a petition from petitioners, 5 Commerce initiated a CVD
    investigation concerning various subsidy programs that allegedly benefitted Russian cold-rolled
    steel producers. Certain Cold-Rolled Steel Flat Products From Brazil, India, the People’s Republic
    of China, the Republic of Korea, and Russia, 80 Fed. Reg. 51,206, 51,210 (Dep’t Commerce Aug.
    24, 2015) (initiation notice), P.R. 49. These programs included the deduction of research and
    development (“R&D”) and exploration costs from taxable income, and the reduction in extraction
    taxes. See Commerce’s Initiation Checklist at 11–12, P.R. 44 (Aug. 19, 2015). Commerce also
    investigated whether the cold-rolled steel producers were provided natural gas for less than
    adequate remuneration. 
    Id. at 12–14.
    5
    Petitioners include plaintiff ArcelorMittal and plaintiff-intervenors AK Steel Corporation, Nucor
    Corporation, and United States Steel Corporation.
    Court No. 16-00168                                                                        Page 9
    On September 14, 2015, Commerce selected Severstal Export GMBH and Novex Trading
    (Swiss) SA (Novex) as mandatory respondents. 6 Respondent Selection Memorandum at 5, P.R.
    71 (Sept. 14, 2015); 19 U.S.C. § 1677f-1(e)(2)(A)(i); 19 C.F.R. 351.204(c)(2). Commerce then
    issued initial and supplemental questionnaires to the GOR and both respondents. See Commerce’s
    Letter to Ministry of Economic Development of the Russian Federation Re: Countervailing Duty
    Investigation of Certain Cold-Rolled Steel Flat Products from the Russian Federation:
    Countervailing Duty Questionnaire, P.R. 72, (Sept. 14, 2015) (“GOR CVD Questionnaire”).
    NLMK provided responses on behalf of its wholly-owned affiliate, Novex, as well as several other
    companies that Commerce determined were cross-owned under 19 C.F.R. § 351.525(b)(6). 7 IDM
    at 9–10. Severstal provided responses on behalf of its wholly-owned affiliate, Severstal Export
    GMBH, and other companies which Commerce treated as cross-owned. IDM at 10.
    6
    In countervailing duty investigations, Commerce may select mandatory respondents pursuant to
    19 U.S.C. § 1677f-1(e)(2), which provides:
    If [Commerce] determines that it is not practicable to determine individual
    countervailable subsidy rates [in investigations or administrative reviews] because
    of the large number of exporters or producers involved in the investigation or
    review, [Commerce] may--
    (A) determine individual countervailable subsidy rates for a reasonable
    number of exporters or producers by limiting its examination to--
    (i) a sample of exporters or producers that the administering
    authority determines is statistically valid based on the information
    available to the administering authority at the time of selection[.]
    7
    19 C.F.R. § 351.525(b)(6)(vi) states, in relevant part, “[c]ross-ownership exists between two or
    more corporations where one corporation can use or direct the individual assets of the other
    corporation(s) in essentially the same ways it can use its own assets. Normally, this standard will
    be met where there is a majority voting ownership interest between two corporations or through
    common ownership of two (or more) corporations.”
    Court No. 16-00168                                                                         Page 10
    The GOR CVD Questionnaire requested information regarding whether the two mandatory
    respondents, NLMK and Severstal, received countervailable subsides in the form of the provision
    of natural gas for LTAR by Public Joint Stock Company Gazprom (“Gazprom”), a Russian
    government authority, during the period of investigation (“POI”). See GOR CVD Questionnaire
    at Section II at 4–7. The questionnaire also requested that the GOR “[e]xplain in detail how natural
    gas rates are set in Russia and provide copies of all laws, regulations and pricing guidelines that
    govern the setting of the rates.” 
    Id. at 6.
    The GOR responded that it does not maintain statistics
    on industries that purchase natural gas or the amount of natural gas purchased by the metallurgical
    industry, of which both mandatory respondents are a part. GOR’s Initial Questionnaire Resp. at
    29, P.R. 133–41, C.R. 138–46, (Oct. 27, 2015).
    In its supplemental questionnaire to the GOR, Commerce asked the GOR to suggest an
    alternate source of data that could be used to evaluate natural gas purchases in the domestic market
    during the POI. Commerce’s Letter to Ministry of Economic Development of the Russian
    Federation Re: Countervailing Duty Investigation of Certain Cold-Rolled Steel Flat Products from
    the Russian Federation: Countervailing Duty Supplemental Questionnaire at 4–7, P.R. 186, (Nov.
    12, 2015). In its response, the GOR identified Gazprom’s annual reports, which had been placed
    on the record as part of the GOR’s initial questionnaire response, as a potential source of the
    requested information. GOR’s Suppl. Questionnaire Resp. at 7, P.R. 239–62, C.R. 220–43, (Nov.
    19, 2015).
    In Severstal’s response to Commerce’s Initial Questionnaire, it addressed Commerce’s
    questions regarding the tax deduction of mining-related R&D and exploration costs by providing
    information concerning the tax deduction of R&D expenses, but responded that “Severstal did not
    receive any benefits under the tax deduction for exploration costs.”             Severstal’s Initial
    Court No. 16-00168                                                                          Page 11
    Questionnaire Resp. at 17–21, 23–26, P.R. 145–52, C.R. 147–75, (Oct. 27, 2015) (“Severstal’s
    IQR”). Commerce subsequently issued a supplemental questionnaire to Severstal requesting
    additional information and clarification regarding Severstal’s response to the Initial Questionnaire.
    See Letter to Crowell & Moring LLP Re: Countervailing Duty Investigation of Cold-Rolled Steel
    Flat Products from the Russian Federation, P.R. 225, C.R. 214, (Nov. 17, 2015) (“Supplemental
    Questionnaire”). In its response to the Supplemental Questionnaire, Severstal did not respond to
    Commerce’s questions about the deduction of mining-related exploration expenses.                 See
    Severstal’s Suppl. Questionnaire Resp. at S-11-5-13, P.R. 269–309, C.R. 253–300, (Nov. 25,
    2015) (“Severstal’s Suppl. QR”).
    On December 22, 2015, Commerce issued its preliminary determination. See Certain
    Cold-Rolled Steel Flat Products From the Russian Federation, 80 Fed. Reg. 79,564 (Dep’t
    Commerce Dec. 22, 2015) (“Preliminary Determination”), P.R. 369, and accompanying
    Preliminary Decision Memorandum, P.R. 364, (Dec. 16, 2015) (“PDM”). Commerce found, with
    respect to the provision of natural gas for LTAR, that: (1) Gazprom is a government authority
    providing a financial contribution in the form of the provision of natural gas; (2) the provision of
    natural gas is specific to cold-rolled steel producers in Russia; (3) the Russian natural gas market
    is distorted by Gazprom’s presence; and (4) “Tier Two” prices from certain European and Asian
    export markets were appropriate benchmarks for measuring the benefit of the program to the
    mandatory respondents. See PDM at 13–19; see also 19 C.F.R. § 351.511(a)(2)(ii). Furthermore,
    Commerce determined that the GOR tax deduction program for exploration expenses (1) provided
    a financial contribution to companies by allowing them to forego tax payments that would
    otherwise be due to the government, pursuant to 19 U.S.C. § 1677(5)(D)(ii), PDM at 21; and (2)
    was de facto specific under 19 U.S.C. § 1677(5A)(D)(iii)(I). Commerce preliminarily calculated
    Court No. 16-00168                                                                        Page 12
    overall subsidy rates of 6.33 percent for NLMK and a de minimis 0.01 for Severstal. Preliminary
    Determination, 70 Fed. Reg. at 79,565. However, in its Final Determination, Commerce explained
    that it had inadvertently relied on the incorrect data for Severstal in performing this calculation
    and instead clarified its understanding that Severstal had claimed non-use of the program. See
    IDM at 122–23; Ministerial Error Memorandum at 4, P.R. 534, (Aug. 16, 2016).
    Following its Preliminary Determination, Commerce verified the facts placed on the record
    by mandatory respondents and the GOR. Commerce provided the GOR with its outline for
    verification on February 29, 2016. See Verification of the Questionnaire Responses Submitted
    by the GOR for the Provision of Natural Gas for LTAR (“GOR Verification Outline”), P.R. 39
    (Feb. 29, 2016). Both in its verification instructions to the GOR and at verification, Commerce
    officials “asked the Gazprom representatives to provide data to support the composition of
    domestic sales reported in the company’s annual reports for 2012, 2013, and 2014.” GOR
    Verification Report at 7, P.R. 471, C.R. 426 (May 16, 2016); see GOR Verification Outline at
    5. Specifically, Commerce requested that the GOR be prepared to provide “supporting records
    (such as, print-outs from Gazprom’s database or sales reports) which were used to build-up the
    annual sales data and to compute the percentages reported” in Gazprom’s 2014 annual report.
    GOR Verification Report at 4. Gazprom did not allow Commerce personnel to review the
    specific data contained in the company’s quarterly sales reports for 2012, 2013, and 2014,
    claiming that “the forms are confidential and cannot be examined.” See 
    id. at 7.
    Instead, Gazprom
    provided Commerce with, inter alia, a blank copy of an internal form that the sales departments
    within Gazprom purportedly complete on a quarterly basis. See 
    id. During verification
    of Severstal, Commerce found that there were “previously unreported
    deductions” due to Severstal’s use of the exploration expense tax deduction program. IDM at 124.
    Court No. 16-00168                                                                         Page 13
    Commerce learned during the GOR’s verification -- which preceded Severstal’s verification -- that
    “line 040” of a Russian income tax return is where companies report their “indirect expenses,”
    which include “expenses for exploration activities and R&D.”           IDM at 123 (citing GOR
    Verification Report at 8) (emphasis in original). Based on this information, Commerce requested
    Severstal provide a breakout and sub-breakout of line 040 of its tax return, which revealed that the
    sub-breakout contained exploration related accounts, the amount of which did not link to any
    deduction amounts that Severstal had previously reported to Commerce. 
    Id. Commerce declined
    to collect the specific line 040 breakout and sub-breakout amounts “because they would have
    constituted untimely new factual information.” 
    Id. at 123–24.
    Consequently, the specific amounts
    within line 040 of Severstal’s tax returns corresponding to exploration expenses are not on the
    record.
    On July 29, 2016 Commerce issued its Final Determination. Commerce continued to find
    that Gazprom’s provision of natural gas to NLMK Companies was de facto specific under 19
    U.S.C. § 1677(5A)(D)(iii)(II), on the basis that the metallurgy sector is the predominant user of
    natural gas provided by Gazprom for LTAR.           Commerce made this de facto specificity
    determination by applying AFA due to the GOR’s refusal to place on the record the specific
    information -- supporting records such as sales reports -- requested by agency officials. See IDM
    at 48–50.
    In regards to the appropriate benchmark to be used in calculating benefits under the
    provision of natural gas for LTAR program, Commerce continued to find that it could not use a
    Tier One benchmark as a result of the domestic Russian natural gas market being distorted. See
    
    id. at 52–56;
    see also 19 C.F.R. § 351.51l (a)(2)(ii). However, diverging from its Preliminary
    Determination, Commerce determined that it would rely on a Tier Three benchmark consisting
    Court No. 16-00168                                                                         Page 14
    of world market prices -- specifically, regional European prices - - to measure the adequacy of
    remuneration for the natural gas that Gazprom sold to the NLMK Companies during the POI.
    See IDM at 66–67. Commerce found that Gazprom’s prices were not market based, and that as
    a result Commerce could not “conclude that the government natural gas prices are reflective of
    market principles.” 
    Id. at 69.
    Because the government natural gas prices in Russia are not set in
    accordance with market principles, pursuant to the agency’s regulations and practice, Commerce
    looked for “an approximate proxy to determine a market-based natural gas benchmark.” 
    Id. Commerce concluded
    that regional European natural gas pricing was the appropriate Tier Three
    benchmark price, as Russia is part of the European gas market, the two markets are interconnected,
    and “regional European prices are market-determined in the regional market to which Russia
    belongs.”   
    Id. at 70.
    Commerce calculated an overall subsidy rate of 6.95 percent ad valorem
    for NLMK Companies, based on countervailable subsidy program usage rates of 6.92 and 0.03
    percent ad valorem for the provision of natural gas for LTAR and tax deduction for exploration
    expenses subsidy programs, respectively.
    With respect to Severstal, Commerce applied AFA to determine the benefit the respondent
    received from the tax deduction for exploration expenses program. 
    Id. at 124.
    The program-
    specific AFA rate that Commerce selected for Severstal in connection with the tax deduction for
    exploration expenses was based on the subsidy rate that Commerce calculated for NLMK in
    connection with NLMK’s use of the same program. See 
    id. at 126.
    To determine this AFA rate,
    Commerce relied on the first step of its AFA hierarchy for investigations, 
    described supra
    ¸ in which
    it “examines whether, in the context of the instant investigation, there is a calculated program
    subsidy rate for the identical program at issue,” and, if so, uses “the calculated program rate for
    that particular program as the basis of the AFA rate.” 
    Id. at 14,
    126. Commerce calculated an
    Court No. 16-00168                                                                       Page 15
    overall subsidy rate for the Severstal Companies of 0.62 percent ad valorem, based on the sum of
    the Severstal Companies’ usage rates of the Reduction in Extraction Taxes and the Provision of
    Mining Rights for LTAR programs of 0.02 percent and 0.57 percent ad valorem, respectively. See
    
    id. at 23,
    31. Because 0.62 percent is de minimis, the Severstal entries would not be subject to a
    CVD order concerning cold-rolled steel from Russia. See Final Determination, 81 Fed. Reg. at
    49,936; see also 19 U.S.C. § 1671d(a)(3) (“In making a determination under this subsection,
    [Commerce] shall disregard any countervailable subsidy rate that is de minimis[.]”); 19 U.S.C. §
    1671b(b)(4).
    ArcelorMittal initiated this action challenging Commerce’s Final Determination on August
    25, 2016. Summ., ECF No. 1. ArcelorMittal filed its complaint on September 23, 2016. ECF No.
    8. Severstal filed a Consent Motion to Intervene as Defendant-Intervenor on October 3, 2016.
    ECF No. 10. The Court granted that motion on October 3, 2016. ECF No. 14. 8
    On October 14, 2016, Severstal filed a cross-claim against the United States. Case No. 16-
    00219, ECF No. 20. The Government filed a motion to sever and dismiss Severstal’s cross-claim
    on December 2, 2016, primarily arguing that Severstal had no standing to file a cross-claim. Case
    No. 16-00219, ECF No. 35. The Court granted Defendant’s motion to dismiss, and Defendant-
    Intervenor’s cross-claim was dismissed without prejudice on April 25, 2017. ArcelorMittal USA
    LLC v. United States, 41 CIT ___, 
    222 F. Supp. 3d 1293
    (2017).
    8
    AK Steel Corporation moved to intervene as plaintiff-intervenor on October 7, 2016. ECF No.
    22. The Court granted that motion on October 17, 2016. ECF No. 21. Nucor Corporation moved
    to intervene as plaintiff-intervenor on October 18, 2016. ECF No. 16. The Court granted that
    motion on October 19, 2016. ECF No. 26. U.S. Steel Corporation moved to intervene as plaintiff-
    intervenor on October 24, 2016. ECF No. 27. The Court granted that motion on October 27, 2016.
    ECF No. 31.
    Court No. 16-00168                                                                      Page 16
    NLMK initiated an action challenging Commerce’s Final Determination on October 17,
    2016. Case No. 16-00219, ECF No. 1; NLMK filed its complaint on November 17, 2016. Case
    No. 16-00219, ECF No. 10. 9
    The Court ordered Case Nos. 16-00168 and 16-00219 to be consolidated under the lead
    caption ArcelorMittal USA LLC v. United States, Consol. Case No. 16-00168, on May 10, 2017.
    ECF No. 62. Plaintiffs ArcelorMittal et al. and NLMK each filed motions for judgment on the
    agency record pursuant to USCIT Rule 56.2, as well as memoranda of points and authorities
    supporting these motions, on August 18, 2017. ArcelorMittal Br., ECF Nos. 65–66; NLMK Br.,
    ECF Nos. 67–68. ArcelorMittal filed a response in opposition to NLMK’s motion for judgment
    on the agency record as defendant-intervenor on November 17, 2017. ECF No. 74. Severstal also
    filed a response in opposition to ArcelorMittal’s motion for judgment on the agency record as
    defendant-intervenor on that same day. ECF Nos. 75–76. The Government filed its response in
    opposition to both ArcelorMittal’s and NLMK’s motions for judgment on the agency record on
    November 17, 2017. Government Br., ECF Nos. 77–78.
    The Government filed a motion to strike part of defendant-intervenor Severstal’s Rule 56.2
    response brief on December 1, 2017, arguing that much of the Severstal’s brief advances
    arguments that are not a response to ArcelorMittal’s motion, but in fact an improper attempt to
    advance the arguments contained in Severstal’s dismissed cross-claim. ECF No. 79. Severstal
    9
    AK Steel Corporation moved to intervene a plaintiff-intervenor on December 5, 2016. Case No.
    16-00219, ECF No. 12. The court granted that motion the next day. Case No. 16-00219, ECF No.
    16. Nucor Corporation moved to intervene as plaintiff-intervenor on December 13, 2016. Case
    No. 16-00219, ECF No. 17. ArcelorMittal moved to intervene as plaintiff-intervenor the next day
    Case No. 16-00219, ECF No. 21. The court granted both those motions on December 16, 2016.
    Case No. 16-00219, ECF Nos. 27–28. U.S. Steel Corporation also moved to intervene as plaintiff-
    intervenors on December 16, 2016. Case No. 16-00219, ECF No. 29. The Court granted that
    motion on December 20, 2016. Case No. 16-00219, ECF No. 33.
    Court No. 16-00168                                                                        Page 17
    then filed a response in opposition to the Government’s motion to strike on December, 20 2017.
    ECF No. 82. The next day, the Court issued an order denying without prejudice the motion to
    strike, and further ordered that the parties may move for reconsideration of the motion to strike
    following further discussion of these matters at oral argument. ECF No. 83. NLMK and
    ArcelorMittal et. al. filed reply briefs on January 16, 2018 and January 17, 2018 respectively. ECF
    Nos. 85–86.
    JURISDICTION AND STANDARD OF REVIEW
    The Court has jurisdiction over this action pursuant to 28 U.S.C. § 1581(c) and 19 U.S.C.
    § 1516a(a)(2)(A)(i)(I) and (a)(2)(B)(ii). The standard of review in this action is set forth in 19
    U.S.C. § 1516a(b)(l)(B)(i): “[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[.]”
    DISCUSSION
    I.          Commerce Did Not Sufficiently Explain Its Selection Of A Program-Specific AFA
    Rate For Severstal’s Unreported Use Of The Tax Deduction Program For Mining-
    Exploration Expenses.
    As detailed above, in its Final Determination, Commerce applied NLMK’s rate for the
    program to Severstal as its AFA rate under step one of Commerce’s AFA hierarchy. See IDM at
    126. ArcelorMittal argues that Commerce’s decision was not in accordance with law because
    Commerce did not adequately explain why it relied on step one of its AFA hierarchy and did not
    use other alternatives in establishing Severstal’s AFA rate. ArcelorMittal Br. at 26. Specifically,
    Commerce did not explain its reason for resorting to the AFA hierarchy, or evaluate the remaining
    steps in the hierarchy beyond step one. 
    Id. at 27.
    In doing so, ArcelorMittal argues, Commerce
    deprived the Court of the material needed to “conduct meaningful judicial review” of Commerce’s
    Court No. 16-00168                                                                           Page 18
    decision as is required under the law. 
    Id. at 26;
    see Altx, Inc. v. United States, 
    25 CIT 1100
    , 1104,
    
    167 F. Supp. 2d 1353
    , 1361 (2001) (“[I]n order to ascertain whether [agency] action is arbitrary,
    or otherwise not in accordance with law, reasons for the choices made among various acceptable
    alternatives usually need to be explained.” (quoting Bando Chem. Indus., Ltd. v. United States, 
    16 CIT 133
    , 136, 
    787 F. Supp. 224
    , 227 (1992), aff’d, 
    26 F.3d 139
    (Fed. Cir. 1994))).
    ArcelorMittal also contends that Commerce’s selected rate was not in accordance with law
    because the rate was not sufficiently adverse to Severstal to effectuate the statute’s dual purposes:
    (1) ensuring that the respondent does not benefit from its non-cooperation by receiving a more
    favorable rate than it otherwise would have and (2) deterring future non-cooperation by
    respondents. ArcelorMittal Br. at 32, 34–35; see SAA, 1994 U.S.C.C.A.N. at 4199; Nan Ya
    Plastics Corp. v. United States, 
    810 F.3d 1333
    , 1347–48 (Fed. Cir. 2016). ArcelorMittal asserts
    that Commerce may not use its discretion to select an AFA rate which is insufficiently adverse to
    achieve the statutory objectives. ArcelorMittal Reply Br. at 8; see IPSCO, Inc. v. United States
    
    899 F.2d 1192
    , 1195 (Fed. Cir. 1990). Additionally, the statute provides Commerce with
    discretion to rely on record information (but not necessarily the hierarchy) to apply the highest rate
    possible to effectuate the purposes of the statute. ArcelorMittal Br. at 31; see 19 U.S.C. §
    1677e(b)(2), (d)(2). ArcelorMittal also cites to prior Commerce determinations to demonstrate
    that it regularly deviates from its AFA hierarchy to ensure that its selected AFA rate is sufficiently
    adverse to the non-cooperative party, 10 and contends that Commerce had a “legal obligation” to
    do so here. See ArcelorMittal Br. at 30–32.
    10
    Issues and Decision Memorandum for the Final Determination in the Countervailing Duty
    Investigation of Truck and Bus Tires from the People’s Republic of China; and Final Affirmative
    Determination of Critical Circumstances, in Part at 14 (Dep’t Commerce Jan. 19 2017); Decision
    Memorandum for Final Results of Countervailing Duty Administrative Review; Certain Oil
    Court No. 16-00168                                                                         Page 19
    With respect to the first statutory purpose of ensuring the respondent does not benefit,
    ArcelorMittal asserts, based on purported relative size of their respective mining operations, that
    Severstal’s mining-exploration expense deductions must have been much higher than NLMK’s.
    
    Id. at 32–34.
    Consequently, ArcelorMittal argues, applying NLMK’s rate is more beneficial to
    Severstal than the rate that would have been applied if Severstal had cooperated, in contravention
    of one of the statutory purposes. 
    Id. With respect
    to the second statutory purpose of deterrence,
    in ArcelorMittal’s view, the selected AFA rate was too low to serve as a deterrent against future
    non-cooperation, in part because it produced a perverse outcome: Severstal (the uncooperative
    party) received a de minimis overall CVD rate for, while NLMK (which was cooperative) received
    an affirmative CVD rate. See 
    id. at 32.
    Altogether, according to ArcelorMittal, Commerce’s
    decision was contrary to law because it improperly used its discretion to select an AFA rate for
    Severstal that was insufficiently adverse to satisfy either of the statute’s purposes.
    The Court first considers whether Commerce adequately explained its decision to rely on
    step one of its AFA hierarchy, and whether the subsequent steps were “potentially acceptable
    alternatives.” See 
    Altx, 167 F. Supp. 2d at 1361
    ; Asociacion Colombiana de Exportadores de
    Flores v. United States, 
    22 CIT 1174
    , 1176, 
    704 F. Supp. 1068
    , 1071 (1998). “While Commerce
    must explain the bases for its decision, ‘its explanations do not have to be perfect.’” Taian Ziyang
    Food Co., Ltd. v. United States, 37 CIT ___, ___, 
    918 F. Supp. 2d 1345
    , 1354 (2013) (quoting
    NMB Singapore Ltd. v. United States, 
    557 F.3d 1316
    , 1319 (Fed. Cir. 2009)). Still, “‘the path of
    Commerce’s decision must be reasonably discernable’ to support judicial review.” 
    Id. at 1354
    (quoting NMB 
    Singapore, 557 F.3d at 1319
    ).             ArcelorMittal cites no authority requiring
    Country Tubular Goods from the People’s Republic of China at 23–24 (Dep’t Commerce Aug. 25,
    2014).
    Court No. 16-00168                                                                        Page 20
    Commerce to provide an explanation for not relying on the hierarchy’s subsequent steps even
    though the first step was applicable. Therefore, Commerce’s explanation, limited to the first step
    of the AFA hierarchy on which it based its selected rate, was sufficient.
    The Court next considers whether Commerce’s application of NLMK’s program-specific
    rate to Severstal was sufficiently adverse to ensure Severstal did not benefit from its non-
    cooperation, and to deter future non-cooperation. The SAA expressly sets forth that “one factor
    [Commerce] will consider is the extent to which a party may benefit from its own lack of
    cooperation.” SAA, 1994 U.S.C.C.A.N. at 4199. As the Federal Circuit has recognized, the
    statute’s “expectation” is that Commerce’s selected AFA rate will have a deterrent effect. 11 See
    Nan Ya 
    Plastics, 810 F.3d at 1348
    . Here, ArcelorMittal’s assertion that Severstal benefitted from
    Commerce’s application of NLMK’s program-specific rate, because it produced a lower rate than
    would have been applied had Severstal cooperated, is speculative. While facts on the record do
    establish the disparity between the size of Severstal’s and NLMK’s respective mining operations,
    ArcelorMittal fails to show that necessarily means Severstal’s mining-exploration deductions
    under the program during the period of investigation were resultantly larger. Under the statute,
    Commerce is not required to infer that there was necessarily a connection between the relative size
    of Severstal’s mining operations and the larger size of its deductions under the program. See 19
    U.S.C. § 1677e(b)(1)(B), (d)(3).
    11
    This view is also supported by Commerce’s own prior determinations. See Issues and Decision
    Memorandum for Final Results of Antidumping Duty Administrative Review: Certain Circular
    Welded Non-Alloy Steel Pipe from Mexico at cmt 4 (Dep’t Commerce June 13, 2011), ref’d in 76
    Fed. Reg. 36,086 (Dep’t Commerce June 21, 2011) (final results) (“The Department has a duty to
    both ensure that uncooperative parties do not benefit from their lack of cooperation and to
    encourage their future compliance.” (citing Final Determination of Sales at Less than Fair Value:
    Static Random Access Memory Semiconductors from Taiwan, 63 Fed. Reg. 8,909, 8932 (Dep’t
    Commerce Feb. 23, 1998)) (citation omitted)).
    Court No. 16-00168                                                                        Page 21
    The Court determines, however, that Commerce’s justification does not adequately explain
    why the application of NLMK’s program-specific rate was sufficiently adverse to deter future non-
    cooperation. First, the Government argues that the selected AFA rate is higher than the rate that
    would have been applied pursuant to Severstal’s initial claim of non-usage. See Government Br.
    at 23. However, this justification is essentially an assertion that Commerce’s selected rate is
    sufficiently adverse because it is above 0% -- the rate that is applied when a respondent truly has
    not utilized the program at issue. Moreover, Severstal’s claim of non-usage was proven to be false
    at verification, which is the reason Commerce is resorting to AFA. It was unreasonable for
    Commerce to use Severstal’s demonstrably false non-usage claim as a basis for what rate is
    sufficiently adverse to Severstal. Commerce’s contention that “a difference between a finding of
    use and non-use is adverse” does not adequately remedy this deficiency.
    Second, the Government notes that Commerce’s selected rate is significantly higher than
    the rate Commerce assigned Severstal for the program in its Preliminary Determination. See
    Government Br. at 23. However, Commerce acknowledged in its Final Determination that it had
    inadvertently relied on the incorrect data in performing this calculation for its Preliminary
    Determination. See IDM at 122–23. Using the value set in its Preliminary Determination as a
    benchmark for measuring the adversity of the selected rate to Severstal was arbitrary, given that
    the value was admittedly incorrect based on an administrative error. 
    Id. Because both
    of the
    benchmarks Commerce used to justify the adversity of its selected AFA rate to Severstal were
    unreasonable, the Court remands this issue so that Commerce may provide a more satisfactory
    explanation as to why NLMK’s program-specific rate is sufficiently adverse to deter future non-
    cooperation.
    Court No. 16-00168                                                                            Page 22
    Finally, the Court considers ArcelorMittal’s contention that Commerce was legally
    obligated to deviate from its hierarchy in this case in order to fulfill the purpose of the statute. The
    statute’s provision on adverse inferences consistently uses permissive language to describe how
    Commerce may go about applying AFA. See 19 U.S.C. § 1677e(b), (d). Commerce thus has wide
    latitude in its selection of an appropriate AFA rate. However, Commerce’s discretion is not
    without limit. The Federal Circuit in 
    IPSCO, 899 F.2d at 1195
    , noted that a Court cannot uphold
    an agency’s “exercise of administrative discretion if it contravenes statutory objectives.” In other
    words, Commerce has discretion to select an AFA rate by any means permissible under the statute,
    so long as the rate it produces does not contravene the purposes of the statute. Furthermore,
    ArcelorMittal identifies no authority that requires Commerce to consider the aggregate effect on a
    company’s overall subsidy rate in selecting AFA for a particular program-specific rate.
    Consequently, whether or not Severstal’s overall rate is affirmative or de minimis does not
    necessarily affect whether the program-specific rate it selects is sufficiently adverse. Therefore,
    on remand Commerce is not obligated to deviate from its AFA hierarchy or produce a program-
    specific rate that necessarily results in an affirmative overall rate, but it must provide adequate
    explanation as to why the program-specific rate it selected was sufficiently adverse to satisfy the
    underlying statutory purposes. If it cannot do so, it must select another rate that can be justified
    under the statute’s purposes. 12
    12
    In order to provide Commerce with further clarity as it performs the remand, the Court addresses
    ArcelorMittal’s argument that Commerce’s rejection of line 040 as the basis for Severstal’s use of
    the mining-exploration expenses tax deduction program (“the program”) was not supported by
    substantial evidence. See ArcelorMittal Br. at 23. ArcelorMittal criticizes as speculative
    Commerce’s finding that using line 040 would “overstate[] the benefit” to Severstal under the
    program, asserting that Commerce properly declined to accept information about the sub-
    breakouts of line 040 at verification and thus lacked any record evidence to its overstatement
    conclusion. ArcelorMittal Br. at 21–24; see, e.g., Marsan Gida Sanayi ve Ticaret A.S. v. United
    States, 37 CIT ___, ___, 
    931 F. Supp. 2d 1258
    , 1280 (2013) (agreeing that the “purpose of
    Court No. 16-00168                                                                        Page 23
    II.    Commerce’s Determination That The GOR’s Alleged Provision Of Natural Gas For
    Less Than Adequate Remuneration Is Specific To Steel Manufacturing Is Not
    Supported By Substantial Evidence And In Accordance With Law.
    a. The GOR Did Not Verify The Information Necessary For Commerce To Make Its
    Specificity Determination.
    NLMK acknowledges that the Gazprom annual reports were “necessary information” for
    Commerce’s specificity determination, because they provided the natural gas consumption
    percentages by various industries in the Russia. See NLMK Br. at 13–14. The only contested
    issue is whether the reports are verifiable. 
    Id. NLMK asserts
    that the GOR sufficiently verified
    verification is not to collect new information”). ArcelorMittal further notes this Court’s holding
    that, in selecting an AFA rate, there must be a “built in increase” over a respondent’s actual rate
    in order to deter uncooperative behavior. ArcelorMittal Reply Br. at 15 (citing China Steel Corp.
    v. United States, 
    28 CIT 38
    , 61, 
    306 F. Supp. 2d 1291
    , 1311 (2004)). Thus, according to
    ArcelorMittal, Commerce’s decision to reject line 040 as AFA is contrary to the law’s requirement
    to include a “built in increase” as a response to Severstal’s uncooperative behavior. 
    Id. The Court
    determines that Commerce’s finding that line 040 would overstate the benefit
    of the program is supported by substantial evidence. Substantial evidence is “more than a mere
    scintilla,” but “less than the weight of the evidence.” 
    Altx, 370 F.3d at 1116
    . “A finding is
    supported by substantial evidence if a reasonable mind might accept the evidence as sufficient to
    support the finding.” Maverick Tube Corp. v. United States, 
    857 F.3d 1353
    , 1359 (Fed. Cir. 2017)
    (citing Consol. Edison Co. of N.Y. v. NLRB, 
    305 U.S. 197
    , 229 (1938)). Despite the fact that the
    values of each category of expense under line 040 are not on the record, a reasonable mind could
    conclude that using the line 040 value -- which aggregates the values of each sub-breakout expense
    category -- would overstate the amount Severstal deducted under the program, given that most of
    the sub-breakout expense categories would not qualify for deduction under the program. See
    Severstal Verification Report at 6–7 (listing 11 entries under the breakout for line 040). No
    evidence in the record suggests that the values of the non-deductible expense categories under line
    040 equal zero. Commerce’s decision to reject line 040 as inappropriate for AFA was also
    reasonable and in accordance with law. The statute at 19 U.S.C. § 1677e(b)(1) and (d)(3) expressly
    discusses adjustments and information that Commerce is not required to consider in selecting
    AFA, but does not mandate that Commerce consider any particular information in its selection.
    Moreover, prior decisions of this Court and the Federal Circuit demonstrate that Commerce has
    significant discretion in its selection of appropriate AFA. See PAM S.p.A v. United States, 
    582 F.3d 1336
    , 1340 (Fed. Cir. 2009); SolarWorld Ams., Inc. v. United States, 41 CIT ___, ___, 
    229 F. Supp. 3d 1362
    , 1366 (2017). Commerce thus reasonably concluded that using line 040 would
    “overstate the benefit” of the program to Severstal, and would be contrary to Commerce’s
    responsibility to select AFA that does not unreasonably overestimate the actual usage rate.
    Court No. 16-00168                                                                           Page 24
    the annual reports by (1) presenting original copies of the reports (2) making available Gazprom
    officials who could attest to the authenticity of the original reports and (3) providing a blank
    authentic quarterly sales report form to demonstrate the statistical categories used to build the
    annual reports were the same as those presented in the annual reports. 
    Id. at 14.
    According to
    NLMK, these steps were sufficient because there was no more granular set of statistical data than
    that presented in the annual reports, so they adequately verify the statistical categories used in the
    data, which NLMK claims is the focus of Commerce’s specificity analysis. 13                  See 
    id. Consequently, the
    completed quarterly sales reports could offer no further insights into these
    statistical categories, and so Commerce need not have looked at them. See NLMK Br. at 14.
    Additionally, NLMK argues that it was inappropriate for Commerce to find the annual reports
    untrustworthy for its specificity finding, but rely on them for its market distortion and
    benefit/benchmark finding. See NLMK Br. at 16. Therefore, NLMK contends that Commerce’s
    decision to find the necessary information unverifiable, and consequently apply AFA, was not
    supported by substantial evidence because, for the reasons stated above, the records were
    verifiable. 
    Id. The Government
    argues that NLMK misunderstands the issue at hand. According to the
    Government, the real issue is not whether the statistical categories match between the quarterly
    sales forms and the annual reports, but whether the GOR provided the underlying data necessary
    to verify that the consumption percentages between them match. See Government Br. at 32–33.
    13
    NLMK attempts to support this proposition by citing to a prior determination it claims shows
    that Commerce’s key consideration in such investigations is to confirm that the statistical
    categories produced for verification are those used in everyday business. NLMK Br. at 15; Live
    Swine from Canada: Final Negative Countervailing Duty Determination, 70 Fed. Reg. 12,186
    (Dep’t Commerce Mar. 11, 2005) and accompanying Issues and Decision Memorandum at 11–14.
    Court No. 16-00168                                                                          Page 25
    The very purpose of verification is to “verify the accuracy and completeness of submitted factual
    information.” 19 C.F.R. § 351.307(d); see Government Br. at 33. Commerce has wide latitude in
    selecting its verification procedures. See Micron Tech., Inc. v. United States, 
    117 F.3d 1386
    , 1396
    (Fed. Cir. 1997). Therefore, the Government argues, Commerce cannot take for granted that just
    because the names of the general statistical categories match, the percentages of sales across the
    categories in the annual reports were also verified. 
    Id. at 32.
    Essentially, because the GOR did
    not provide the underlying data necessary -- in the form of the completed quarterly sales reports -
    - Commerce could not verify the consumption percentages in the annual reports, and was thus
    justified in applying AFA to come to its de facto specificity finding. 
    Id. at 33.
    Additionally, the Government asserts that Commerce was justified in finding the Gazprom
    annual report unreliable for its specificity finding but still using evidence from the report for its
    benefit/benchmark finding. See Government Br. at 37. The Government contends that the fact
    that Commerce could not verify the consumption percentages in one discrete section of the annual
    report does not make the entire report unreliable, since there was nothing that else that put the
    validity of the rest of the report into question. 
    Id. As a
    result, Commerce was not precluded from
    relying on other parts of the annual report for its benefit/benchmark determination. 
    Id. The Court
    concludes that because the GOR did not provide information at verification that
    would allow Commerce to verify the underlying data used to produce the values of the natural gas
    consumption percentages in the Gazprom annual reports, Commerce was justified in applying
    AFA for its de facto specificity finding. The presentation of authentic original versions of the
    annual reports, along with Gazprom officials to verify their authenticity, does not prove the
    consumption percentages were indeed accurate. These materials simply demonstrate that the
    consumption percentages found in the copies of the annual reports sent to Commerce in the GOR’s
    Court No. 16-00168                                                                        Page 26
    questionnaire responses had not been altered from the values listed in the original annual report.
    However, this does nothing to verify the accuracy of the consumption percentage values in the
    original annual reports. In addition, the GOR’s provision of blank quarterly sales forms does
    nothing to verify the accuracy of the values found in the annual reports, because this blank form,
    by definition, contains no underlying data. Thus, because the GOR refused to allow Commerce
    verification officials to view the completed quarterly sales form -- the only known documents that
    could verify the accuracy of the annual report consumption percentages -- on the grounds that they
    were “confidential,” the Consumption percentages were not verified. See GOR Verification
    Report at 4.
    b. Commerce’s Finding That The GOR Failed To Cooperate By Not Acting To The Best
    Of Its Ability Was Supported By Substantial Evidence And In Accordance With Law.
    NLMK argues that Commerce’s guidance in its verification outline did not clearly identify
    the types of documents the GOR needed to produce at verification, so the GOR reasonably believed
    that it was acting to the “best of its ability” by producing the materials that it did. NLMK Br. at
    18, 20. NLMK claims that “nowhere in the outline are actual sales data expressly required.” 
    Id. at 20.
    NLMK advances two arguments for why the GOR was reasonable in believing that it was
    in compliance with Commerce’s requests to verify the accuracy and completeness of the
    consumption percentages in the Gazprom annual reports. 
    Id. at 20–23.
    First, the character of the
    information at issue is “self-authenticating” because the Gazprom annual reports are published by
    “a publicly traded company subject to securities laws, which were not prepared for the purposes
    of the CVD proceeding.” 
    Id. at 21.
    Therefore, it was reasonable for the GOR to expect that the
    veracity of this document would not be challenged beyond the four corners of the report, consistent
    with prior Commerce determinations which effectively presume the accuracy of such documents.
    Court No. 16-00168                                                                       Page 27
    See 
    id. at 21–22.
    14 Second, the GOR could reasonably rely on Commerce’s past practice in de
    facto specificity determinations to expect that the materials it produced were adequate since they
    demonstrated the statistical categories in the annual reports were those “used in the everyday
    course of business.” 
    Id. at 23;
    see Live Swine from Canada: Final Negative Countervailing Duty
    Determination, 70 Fed. Reg. 12,186 (Dep’t Commerce Mar. 11, 2005) and accompanying Issues
    and Decision Memorandum at 11–14; Live Cattle from Canada: Preliminary Negative
    Countervailing Duty Determination, 64 Fed. Reg. 25,277, 25,279 (Dep’t Commerce May 11,
    1999). Thus, providing the completed quarterly forms would have done nothing more to prove
    that the statistical categories in the annual reports were the ones used in the course of everyday
    business. 
    See supra
    at pp. 24–26 (discussing materials provided by GOR that allegedly verified
    statistical categories used in annual report); NLMK Br. at 25. For the reasons stated above, NLMK
    contends that Commerce unreasonably found that the verification outline clearly reflected
    Commerce’s intention to examine actual Gazprom sales data outside the context of that presented
    in the annual reports. Therefore, in NLMK’s view, the GOR acted to “the best of its ability” to
    comply with the requests made in Commerce’s verification outline by providing the materials it
    did in line with its reasonable interpretation of the instructions.
    The Court first considers whether the GOR acted to “the best of its ability” in responding
    to Commerce’s verification instructions. As noted above, “[c]ompliance with the ‘best of its
    ability’ standard is determined by assessing whether respondent has put forth its maximum effort
    to provide Commerce with full and complete answers to all inquiries in an investigation.” Nippon
    14
    See, e.g., Countervailing Duty Investigation of Certain Amorphous Silica Fabric From the
    People’s Republic of China: Preliminary Determination of the Countervailing Duty Investigation
    and Alignment of Final Determination With Final Antidumping Duty Investigation, 81 Fed. Reg.
    43,579 (Dep’t Commerce July 5, 2016) and accompanying Issues and Decision Memorandum at
    21–25.
    Court No. 16-00168                                                                        Page 28
    
    Steel, 337 F.3d at 1382
    . “While the standard does not require perfection and recognizes that
    mistakes sometimes occur, it does not condone inattentiveness, carelessness, or inadequate record-
    keeping.” 
    Id. Whether the
    GOR acted “to the best of its ability” rests on two distinct but related
    questions. First, the Court must determine whether Commerce’s verification outline instructions
    were clear enough that it was unreasonable for the GOR to believe the materials it produced at
    verification were adequate. Commerce’s verification instructions explicitly state: “Have available
    the supporting records (such as, print-outs from Gazprom’s database or sales reports) which were
    used to build-up the annual sales data and to compute the percentages reported” on the 2012–2014
    Gazprom annual reports. See GOR Verification Outline at 4 (emphasis added); IDM at 49. The
    Oxford English Dictionary defines “compute” as “to calculate.” Compute, OXFORD ENGLISH
    DICTIONARY (3d ed. 2008). This definition implies that there must be numerical values included
    in any materials used to “compute” the reported percentages. Thus, materials such as the blank
    quarterly sales forms, by definition, cannot reasonably be said to be used to compute the reported
    percentages because they do not include any numbers from which to do so. These instructions, in
    combination with additional instructions in the verification outline were clear enough that the GOR
    should have known it needed to provide the completed quarterly sales reports, to which the GOR
    expressly denied Commerce verifiers access. See 
    id. at 39;
    see also IDM at 49; GOR Verification
    Report at 7. Given the word choice and specificity of the instructions, NLMK cannot reasonably
    assert that the GOR “reasonably interpreted” these instructions as not requiring the provision of
    the completed quarterly sales reports. Altogether, the GOR did not act to the “best of its ability”
    in this regard.
    Court No. 16-00168                                                                        Page 29
    Second, the Court considers whether the Gazprom annual reports were self-authenticating.
    According to NLMK, if the Gazprom annual reports were self-authenticating by their very nature
    as audited financial documents, then the GOR was reasonable in believing that it need not provide
    any underlying data, in the form of completed quarterly sales reports, in response to Commerce’s
    verification instructions. See NLMK Br. at 21. NLMK argues that it is contrary to Commerce’s
    practice to look behind data in audited financial statements because Commerce regards them as
    “the touchstone for accuracy.” 
    Id. at 22.
    NLMK attempts to support this proposition by citing to
    Geum Poong v. United States, 
    26 CIT 322
    , 325, 
    193 F. Supp. 2d 1363
    , 1367 (2002), which it
    asserts establishes that Commerce considers audited financial statements to be “self-verifying.” In
    that case, the Court quoted a Commerce request for additional information from the respondent,
    in which Commerce “indicated that ‘any information submitted must be credible and self-
    verifying, e.g. audited financial statements . . . .’” 
    Id. However, in
    the very next sentence, the
    Court pointed out that Commerce “could have requested still more information if it considered
    the submission somehow incomplete.” 
    Id. Despite NLMK’s
    assertion to the contrary, this
    language demonstrates that Geum Poong stands for the proposition that Commerce is broadly
    entitled to ask for additional information from respondents if it feels that the materials provided
    were not sufficient to verify the facts placed on the record, regardless of Commerce’s
    characterization of the facial reliability of such documents. This is precisely what Commerce did
    in the instant case, and the GOR expressly refused to provide the additional materials requested by
    Commerce (i.e., the completed quarterly sales forms) on the grounds that they were “confidential.”
    Therefore, Gazprom’s annual reports were not “self-authenticating,” and as such it was
    unreasonable for the GOR to believe it did not need to produce the underlying quarterly sales data
    in response to Commerce’s instructions.
    Court No. 16-00168                                                                         Page 30
    NLMK further argues that the GOR’s response to the Commerce verification team’s
    request for information was merely deficient, requiring Commerce to give notice and an
    opportunity for remedy which Commerce did not provide. NLMK Br. at 26–27. The statute, 19
    U.S.C. § 1677m(d), requires Commerce to give a cooperating party an opportunity to remedy or
    explain deficiencies in its submissions before it may use AFA. NLMK Br. at 26; see 19 U.S.C. §
    1677m(d). According to NLMK, if Commerce wanted the GOR to produce more materials than
    what the GOR could have reasonably expected from the verification instructions, and was prepared
    to invoke AFA if the GOR did not comply, the GOR was entitled to some notice and opportunity
    to remedy the submissions that Commerce found inadequate. NLMK Br. at 27.
    “Nothing in the statute compels Commerce to treat intentionally incomplete data as a
    ‘deficiency’ and then to give a party that has intentionally submitted incomplete data an
    opportunity to ‘remedy’ as well as to ‘explain.’” Papierfabrik August Koehler SE v. United States,
    
    843 F.3d 1373
    , 1384 (Fed. Cir. 2016). In the instant case, Commerce specifically requested the
    quarterly sales forms in its verification outline instructions. It is unreasonable for NLMK to argue
    that the completed quarterly sales forms were not “sales reports” used to “compute the annual
    consumption percentages” per Commerce’s instructions.          
    See supra
    pp. 28.      Furthermore,
    Commerce again specifically asked for the quarterly sales reports at verification, and the GOR
    informed Commerce in no uncertain terms that it would not be able to provide those materials
    because they were “confidential.” NLMK’s arguments related to notice and opportunity to remedy
    under 19 U.S.C. § 1677m(d) are unpersuasive, since it cannot be argued that the GOR’s response
    was “merely deficient” if the instructions were unambiguous as to what was expected of it.
    Court No. 16-00168                                                                         Page 31
    c. Commerce’s Affirmative Specificity Finding Was Not Supported By Substantial
    Evidence.
    NLMK argues that Commerce’s de facto specificity determination was not based on any
    facts on the record. See NLMK Br. at 28. NLMK points out that Commerce does not cite to
    anything in the record in its de facto specificity finding. See id.; IDM at 16, 50. NLMK asserts
    that when Commerce “simply reaches an adverse conclusion without any resort to facts on the
    record, that conclusion cannot be said to be supported by substantial evidence.” NLMK Br. at 28;
    See Changzhou Trina Solar Energy Co., Ltd. et al. v. United States, 40 CIT ___, ___, 
    195 F. Supp. 3d
    1334, 1350 (2016).
    The Government argues that Commerce’s determination that the GOR’s provision of
    natural gas was de facto specific was supported by substantial evidence, even though Commerce
    did not cite any record facts to support that determination. First, the Government contends that
    Changzhou Trina’s holding is not binding on this Court and outlines several reasons why it
    disagrees with the decision. See Government Br. at 42; Algoma Steel Corp. v. United States, 
    865 F.2d 240
    , 243 (Fed. Cir. 1989). The Government then argues that regardless of Changzhou Trina,
    Commerce’s de facto specificity finding was based on “facts” on the record. Government Br. at
    43. Per the Government, although Commerce’s finding was not accompanied by a specific record
    citation, the “procedural history, and the inferences flowing therefrom, are themselves ‘facts’
    supporting this determination.” 
    Id. at 43–44.
    Moreover, the Government asserts that Commerce’s
    finding is consistent with “the culmination of Commerce’s analysis of the Gazprom 2014 annual
    report in the preliminary determination,” and with Commerce’s decision to initiate the
    investigation. 
    Id. at 44.
    Instead of Commerce’s de facto specificity finding being based on verified
    evidence in the final determination, which is not required for an AFA determination, Commerce
    Court No. 16-00168                                                                         Page 32
    declined to reward the GOR for its non-cooperation and found that the “facts” supported a finding
    of de facto specificity. 
    Id. The Court
    determines that Commerce did not fulfil its obligation to support its de facto
    specificity finding with any record evidence. The AFA statute lists four potential sources of
    information on which Commerce may rely when making adverse inferences: (1) the petition, (2) a
    final determination in the investigation under this title, (3) any previous review under 19 U.S.C. §
    1675 or determination under § 1675b, or (4) any other information placed on the record. See 19
    U.S.C. § 1677e(b)(2). Although this list provides Commerce wide latitude in selecting facts upon
    which to base its adverse inference, it is clear it must be based upon record evidence.
    Changzhou Trina is an analogous case reinforcing this proposition. In that case, Commerce
    resorted to AFA after the respondent and the Government of China were both uncooperative in
    helping Commerce identify countervailable programs used by the respondent. See Changzhou
    Trina, 
    195 F. Supp. 3d
    at 1343–45. As a result, Commerce found the government grants it
    discovered the respondent had used to be de facto specific and therefore countervailable. See 
    id. at 1347.
    Commerce’s Final Determination in that case did not cite to any facts that it relied upon
    in finding that the programs at issue were de facto specific. 
    Id. at 1347–48.
    Instead, Commerce’s
    finding was “a sweeping legal conclusion lacking any factual foundation.” 
    Id. at 1349.
    The Court
    held that because Commerce “improperly reached legal conclusions without the support of
    requisite factual findings, the agency’s determination . . . must be remanded for reconsideration.”
    
    Id. at 1350.
    In the instant case, Commerce provided no citation to any facts whatsoever -- on the record
    or otherwise -- in its finding that the GOR’s provision of natural gas was de facto specific. See
    IDM at 16, 50. Commerce essentially rested its specificity finding on the proposition that because
    Court No. 16-00168                                                                          Page 33
    it may use an adverse inference, it therefore may find the GOR’s provision of natural gas was de
    facto specific. 
    Id. Such a
    statement is the type of “sweeping legal conclusion” the Changzhou
    Trina Court held to be inadequate under the statute. As noted, the Government attempts to justify
    Commerce’s lack of citation to any record facts by arguing that the “procedural history, and the
    inferences flowing therefrom, are themselves ‘facts’ supporting this determination.” Government
    Br. at 43–44. However, the government provides no authority to support this proposition, or
    explain why it satisfies the statutory standard. Because both the statute’s language and prior
    decisions of this Court are clear that Commerce must rely on some actual record fact in applying
    adverse inferences, the Court remands Commerce’s determination so that the agency may identify
    record facts on which it bases its de facto specificity finding. 15
    III.    Severstal Can Raise Arguments Related To Commerce’s Application Of AFA In Its
    Defendant-Intervenor Brief.
    As noted above, the Government filed a motion to strike arguments advanced by Severstal
    in its defendant-intervenor brief that pertain to whether Commerce was correct in resorting to AFA
    in determining Severstal’s CVD rate. The Government asserts Severstal is now improperly raising
    arguments that support its cross-claim, which the Court previously dismissed for lack of standing.
    See 
    id. at 3–5;
    see also ArcelorMittal, 
    222 F. Supp. 3d 1293
    . Severstal argues that it is appropriate
    to raise such arguments in its defendant-intervenor brief, because they are merely rebutting “the
    various arguments and factual assertions in Plaintiff’s opening brief, which claims that Severstal
    failed to report certain tax benefits and that the Department therefore properly relied on [AFA].”
    15
    The issues raised by the parties pertaining to the benchmark Commerce selected to measure the
    adequacy of NLMK’s remunerations for the GOR’s provision of natural gas are predicated on
    Commerce appropriately finding that the GOR’s provision of natural gas for LTAR was de facto
    specific. Therefore, until Commerce can sufficiently justify its de facto specificity finding, the
    Court will not address those issues.
    Court No. 16-00168                                                                           Page 34
    Resp. to Mot. to Strike at 3, Dec. 20, 2017, ECF No. 82. Furthermore, Severstal contends that it
    can bring such arguments in its brief without running afoul of the Court’s prior rulings because
    those pertained to whether Severstal had standing for a cross-claim, which this is not. 
    Id. at 4–5.
    Under the Court’s rules, it “may strike from a pleading an insufficient defense or any
    redundant, immaterial, impertinent, or scandalous matter.” USCIT R. 12(f). However, motions
    to strike constitute extraordinary remedies, and “should be granted only in cases where there has
    been a flagrant disregard of the rules of court.” Jimlar Corp. v. United States, 
    10 CIT 671
    , 673,
    
    647 F. Supp. 932
    , 934 (1986) (citing Application of Harrington, 55 CCPA 1459, 1462, 
    392 F.2d 653
    , 655 (1968)). The party’s brief must demonstrate “a lack of good faith, or that the court would
    be prejudiced or misled by the inclusion in the brief of the improper material.” Jimlar, 647 F.
    Supp. at 934; see also Fla. Tomato Exch. v. United States, 38 CIT ___, ___, 
    973 F. Supp. 2d 1334
    ,
    1338 (2014).
    In its order granting the Government’s motion to dismiss, the Court expressly ruled that
    Severstal’s argument that Commerce should not have resorted to AFA did not impermissibly
    expand the issues in dispute because it is encompassed within the primary issue in this case:
    “whether the AFA rate assigned to Severstal is supported by substantial evidence and in
    accordance with law.” Given this characterization of the issue, Severstal’s inclusion of this
    argument in its brief does not meet the high standard of “flagrant disregard” of the court’s rules to
    justify striking it from the brief.
    IV.     Commerce’s Application Of AFA To Severstal Was Appropriate.
    Severstal argues that Commerce failed to meet the legal requirements of the statute’s AFA
    and related provisions, and was wrong, as a factual matter, that Severstal failed to report its use of
    the program. Severstal Br. at 12–22. In essence, Severstal contends that it was inappropriate for
    Court No. 16-00168                                                                            Page 35
    Commerce to resort to AFA on the grounds that Severstal did not disclose its use of the program,
    because Severstal did disclose its use of the program. See 
    id. at 12–19.
    As such, Severstal claims
    that Commerce’s finding that it “failed to act to the best of its ability” was not supported by
    substantial evidence.    See 
    id. Accordingly, Severstal
    asserts that Commerce should have
    calculated a program rate based on the same methodology that it used in its Preliminary
    Determination. See 
    id. at 10,
    22, 28–29.
    Severstal’s claims that it disclosed its use of the program are incorrect. In Severstal’s initial
    questionnaire response, it unequivocally claimed that “it did not receive any benefits under the tax
    deduction for exploration costs.” Severstal’s IQR at 23. Furthermore, Severstal’s supplemental
    questionnaire response directed Commerce to line 054 of Severstal’s tax return, which did not
    include any amount for exploration-related deductions, only R&D expenses deductions -- a
    separate program. See Severstal’s Pre-Preliminary Comments at 10; IDM at 123–24. Moreover,
    Severstal did not respond to Commerce’s direct question regarding exploration expense deduction
    usage, only its extraction tax deduction. See 
    id. at S-11,
    S-13. 16 Commerce only discovered the
    actual expenses related to Severstal’s use of the program at verification, when it observed the
    relevant values under line 040 of Severstal’s tax return that “did not trace to any of the deduction
    amounts previously reported by the Severstal Companies.” IDM at 123. Based on the standard
    
    described supra
    , these facts demonstrate that Severstal did not comply with Commerce’s requests
    about the program to “the best of its ability,” and thus Commerce was justified in resorting to AFA
    to determine Severstal’s subsidy rate under the program.
    16
    Commerce’s question was “Did the company deduct any exploration expenses or take a
    reduction of its extraction taxes on the company’s 2013 tax returns?”
    Court No. 16-00168                                                                           Page 36
    CONCLUSION
    Commerce’s selected AFA rate for Severstal for the program was contrary to law because
    it did not adequately explain why the rate it selected was sufficient to deter future non-cooperation,
    as the statute and binding case precedent require. The Court thus remands the Final Determination
    so that Commerce may either provide a satisfactory explanation as to why its selected AFA rate
    was sufficiently adverse to satisfy the statute’s purpose, or select another rate which does comport
    with the statute’s purpose of deterrence in accordance with the guidance laid out in this opinion.
    Regarding NLMK, the consumption percentages in the Gazprom annual reports were not
    verified by the materials produced by the GOR at verification, and thus the GOR did not act to
    “the best of its ability.” Commerce was therefore warranted in applying AFA. However, even
    though Commerce’s decision to apply AFA was appropriate, its de facto specificity finding was
    still improper because it did not provide any specific factual basis for its conclusion.
    Consequently, the Court remands the Final Determination so that Commerce may identify the
    record facts it relied upon in making its de facto specificity determination.
    Commerce shall file with the Court and provide to the parties the results of its
    redetermination on remand within 90 days of the date of this order; thereafter, the parties shall
    have 30 days to submit briefs addressing the redetermination to the Court and the parties shall have
    15 days thereafter to file reply briefs with the Court.
    SO ORDERED.
    /s/    Gary S. Katzmann
    Gary S. Katzmann, Judge
    Dated: September 19, 2018
    New York, New York