AG der Dillinger Hüttenwerke v. United States

Slip Op. 23 - 94

UNITED STATES COURT OF INTERNATIONAL TRADE

AG DER DILLINGER HÜTTENWERKE,

Plaintiff,

and

ILSENBURGER GROBBLECH GMBH,
SALZGITTER MANNESMANN GROBBLECH
GMBH, SALZGITTER FLACHSTAHL GMBH,
SALZGITTER MANNESMANN INTERNATIONAL
GMBH, and FRIEDR. LOHMANN GMBH,

Consolidated Plaintiffs,
and
Before: Leo M. Gordon, Judge
THYSSENKRUPP STEEL EUROPE AG, Consol. Court No. 17-00158
Plaintiff-Intervenor,

v.

UNITED STATES,

Defendant,

and

NUCOR CORPORATION and
SSAB ENTERPRISES LLC,

Defendant-Intervenors.

OPINION and ORDER

[Commerce’s application of facts otherwise available to Dillinger and partial
adverse facts available to Salzgitter sustained; Commerce’s application of its
model-match methodology remanded.]

Dated: June 23, 2023
Consol. Court No. 17-00158 Page 2

Marc E. Montalbine, deKieffer & Horgan, PLLC, of Washington, D.C., argued for
Plaintiff AG der Dillinger Hüttenwerke. With him on the brief were Gregory S. Menegaz,
Alexandra H. Salzman, and Merisa A. Horgan.

Ron Kendler and Allison Kepkay, White & Case LLP, of Washington, D.C., argued
for Consolidated Plaintiffs Ilsenburger Grobblech GmbH, Salzgitter Mannesmann
Grobblech GmbH, Salzgitter Flachstahl GmbH, and Saltzgitter Mannesmann
International GmbH. With them on the brief was David E. Bond.

Kara M. Westercamp, Trial Attorney, Commercial Litigation Branch, Civil Division,
U.S. Department of Justice of Washington, D.C., argued for Defendant United States.
On the brief were Brian M. Boynton, Principal Deputy Assistant Attorney General,
Patricia M. McCarthy, Director, and Tara K. Hogan, Assistant Director. Of counsel was
Ayat Mujais, Attorney, U.S. Department of Commerce, Office of Chief Counsel for Trade
Enforcement and Compliance of Washington, D.C.

Jeffrey Gerrish, Schagrin Associates, of Washington, D.C., argued for
Defendant-Intervenor SSAB Enterprises LLC. With him on the brief were Roger B.
Schagrin, Luke A. Meisner, and Nicholas J. Birch.

Stephanie M. Bell, Wiley Rein LLP, of Washington, D.C., argued for
Defendant-Intervenor Nucor Corporation. With her on the brief were Alan H. Price and
Christopher B. Weld.

Gordon, Judge: This consolidated action involves challenges to the final

determination in the antidumping (“AD”) investigation conducted by the U.S. Department

of Commerce (“Commerce”) of certain carbon and alloy steel cut-to-length plate

(“CTL plate”) from the Federal Republic of Germany. See Certain Carbon and Alloy Steel

Cut-to-Length Plate from the Federal Republic of Germany, 82 Fed. Reg. 16,360 (Dep’t of

Commerce Apr. 4, 2017) (“Final Determination”), and accompanying Issues and Decision

Memorandum, A-428-844 (Mar. 29, 2017),

http://enforcement.trade.gov/frn/summary/germany/2017-06628-1.pdf (last visited this

date) (“Decision Memorandum”).
Consol. Court No. 17-00158 Page 3

Before the court are Commerce’s Final Results of Redetermination Pursuant to

Court Remand, ECF No. 153 (“Third Remand Results”) filed pursuant to the court’s

remand order in AG der Dillinger Huttenwerke v. United States, 46 CIT ___, 592 F.

Supp. 3d 1344 (2022) (“Dillinger II”). Plaintiff AG der Dillinger Hüttenwerke (“Dillinger”)

challenges Commerce’s determination to use “likely selling price” for the cost of

production for non-prime plate as facts otherwise available when it was missing

necessary actual cost information, as well as Commerce’s rejection of Dillinger’s

proposed change to the agency’s model-match methodology to include a proposed

additional quality code for “sour transport plate.”1 See Pl. Dillinger’s Comments in Opp’n

to Final Results of Redetermination, ECF No. 162 (“Dillinger Comments”); see also Def.’s

Resp. to Comments on Remand Redetermination, ECF No. 168 (“Def.’s Resp.”);

Pl. Dillinger Mem. in Supp. of Rule 56.2 Mot. for J. on the Agency R., ECF No. 40

(“Dillinger MSJ”); Def.’s Mem. Opp. Pls.’ Rule 56.2 Mots. for J. on the Admin. R., ECF

No. 55 (“Def.’s MSJ Resp.”); Reply Br. of Pl. Dillinger, ECF No. 62 (“Dillinger MSJ Reply”).

Separately, Consolidated Plaintiffs Ilsenburger Grobblech GMBH, Salzgitter

Mannesmann Grobblech GMBH (“SMSD”), Salzgitter Flachstahl GMBH, and Salzgitter

Mannesmann International GMBH (collectively, “Salzgitter”) challenge Commerce’s

determination from the results of the previous remand to use partial AFA for certain home

1 The parties refer to the products covered by proposed quality code 771 with different
terms including “Sour Service Petroleum Transport Plate” and “Sour Service Line Pipe
Steel.” See Decision Memorandum at 77 (“Dillinger first proposed a distinct quality
reporting code for sour service petroleum transport plate in its Dillinger Model Match
Comments.”); Dillinger Br. at 11 (describing “sour service petroleum transport or line pipe
steel (code 771)”). The court will continue to use the shorthand term “sour transport plate”
for consistency.
Consol. Court No. 17-00158 Page 4

market CTL plate sales made by their respective affiliates when Salzgitter failed to submit

manufacturing information. See Salzgitter Consol. Pls.’ Comments on Remand

Redetermination, ECF No. 135 (“Salzgitter Comments”); Commerce’s Final Results of

Redetermination Pursuant to Court Remand, ECF No. 129 (“Second Remand Results”);

see also Def.’s Resp. to Comments on Remand Redetermination, ECF No. 141 (“Def.’s

2RR Resp.”); Def.-Int. SSAB’s Comments on Remand Redetermination, ECF No. 139;

Def.-Int. Nucor Corporation’s Comments on Remand Redetermination, ECF No. 146.

The court has jurisdiction pursuant to Section 516A(a)(2)(B)(iii) of the Tariff Act of 1930,

as amended, 19 U.S.C. § 1516a(a)(2)(B)(iii), 2 and 28 U.S.C. § 1581(c) (2018).

For the reasons set forth below, the court sustains: (1) Commerce’s determination

to assign the “likely selling price” as the cost of production for non-prime plate recorded

in Dillinger’s books and records as “the best available information on the record” for

evaluating and adjusting the cost of production under 19 U.S.C. § 1677b(f); and

(2) Commerce’s application of partial AFA to Salzgitter. The court remands the issue of

Commerce’s application of its model-match methodology to Dillinger for further

explanation, or if appropriate, reconsideration.

I. Standard of Review

The court sustains Commerce’s “determinations, findings, or conclusions” unless

they are “unsupported by substantial evidence on the record, or otherwise not in

accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i). More specifically, when reviewing

2Further citations to the Tariff Act of 1930, as amended, are to the relevant provisions of
Title 19 of the U.S. Code, 2018 edition.
Consol. Court No. 17-00158 Page 5

agency determinations, findings, or conclusions for substantial evidence, the court

assesses whether the agency action is reasonable given the record as a whole.

Nippon Steel Corp. v. United States, 458 F.3d 1345, 1350–51 (Fed. Cir. 2006);

see also Universal Camera Corp. v. NLRB, 340 U.S. 474, 488 (1951) (“The substantiality

of evidence must take into account whatever in the record fairly detracts from its weight.”).

Substantial evidence has been described as “such relevant evidence as a reasonable

mind might accept as adequate to support a conclusion.” DuPont Teijin Films USA v.

United States, 407 F.3d 1211, 1215 (Fed. Cir. 2005) (quoting Consol. Edison Co. v.

NLRB, 305 U.S. 197, 229 (1938)). Substantial evidence has also been described as

“something less than the weight of the evidence, and the possibility of drawing two

inconsistent conclusions from the evidence does not prevent an administrative agency’s

finding from being supported by substantial evidence.” Consolo v. Fed. Mar. Comm’n,

383 U.S. 607, 620 (1966). Fundamentally, though, “substantial evidence” is best

understood as a word formula connoting reasonableness review. 3 Charles H. Koch, Jr.

& Richard Murphy, Administrative Law and Practice § 9.24[1] (3d ed. 2023). Therefore,

when addressing a substantial evidence issue raised by a party, the court analyzes

whether the challenged agency action “was reasonable given the circumstances

presented by the whole record.” 8A West’s Fed. Forms, National Courts § 3.6 (5th ed.

2023).
Consol. Court No. 17-00158 Page 6

II. Discussion

A. Use of “Likely Selling Price” to Calculate Cost of Production under § 1677b

The court presumes familiarity with its prior decisions regarding Commerce’s

calculation of the cost of production of Dillinger’s non-prime products under 19 U.S.C.

§ 1677b. In its most recent opinion, the court held that “[b]ecause Dillinger has failed to

place information on the record demonstrating the actual cost of production of its

non-prime products, Commerce may reasonably rely on facts otherwise available

pursuant to § 1677e(a)(1).” Dillinger II, 46 CIT at ___, 592 F. Supp. 3d at 1349. However,

the court remanded the determination of facts otherwise available for Commerce to

“explain how its reliance on information indicating the ‘likely selling price’ of non-prime

products accords with its obligation to ensure that the reported costs of production

reasonably reflect the cost of producing the merchandise under consideration.” Id.

On remand, Commerce explained “how the information recorded for non-prime products

in Dillinger’s normal books and records is not only the best available information on the

record, but also ensures that the reported costs reasonably reflect the cost of producing

both prime and non-prime products.” Third Remand Results at 5; see also id. at 4 (noting

that Commerce continues “to rely on [‘the likely selling price’ information from] Dillinger’s

normal books and records,” which Commerce maintains is “the only reasonable approach

for determining the allocation of total costs between prime and non-prime products, and

the per-unit costs of non-prime products.”).

Dillinger continues to challenge the reasonableness of Commerce’s finding that

Dillinger values the cost of producing non-prime merchandise at the “likely selling price”
Consol. Court No. 17-00158 Page 7

in its normal books and records. Dillinger contends that the application of facts otherwise

available, i.e., Commerce’s reliance on the “likely selling price” of the non-prime

merchandise recorded in Dillinger’s books and records, was unreasonable given the

totality of the record as well as the guidance from the U.S. Court of Appeals for the Federal

Circuit (“Court of Appeals”) in Dillinger France S.A. v. United States, 981 F.3d 1318, 1321

(Fed. Cir. 2020). See Dillinger Comments at 1. Dillinger further maintains that Commerce

misread the record by finding that Dillinger uses the likely selling price of non-prime

products to value costs in its audited financial statements. Id. at 4. Dillinger also argues

that “[b]y using the likely selling price of non-prime plate rather than the actual cost of

production allocated to non-prime plate in Dillinger’s verified cost calculation, Commerce

has imposed an impermissible adverse inference.” Id. at 14. As explained below,

because Dillinger has failed to demonstrate that Commerce’s application of facts

otherwise available was unreasonable given the limited information in the record,

the court is unpersuaded by Dillinger’s arguments and sustains Commerce’s

determination on this issue.

The parties’ dispute centers on Commerce’s finding that “[t]he information

recorded in Dillinger’s normal books and records, including the likely selling price of

non-prime products, to allocate costs for the [period of investigation (“POI”)] is the most

reasonable information on the record to fill in the informational gap caused by Dillinger’s

failure to provide either the actual cost of producing non-prime products and their physical

characteristics, or other information from its production records.” Third Remand Results

at 9. Commerce emphasizes that Dillinger “could have provided Commerce with the
Consol. Court No. 17-00158 Page 8

information needed to ascertain the non-prime product’s actual costs and to comply with

the Federal Circuit’s directive [in Dillinger France] to determine the actual costs of prime

and non-prime products.” Id. Commerce highlights the fact that it had previously

re-opened the record to allow Dillinger to provide such critical actual cost information for

Commerce’s calculations, but Dillinger’s failure to provide such information resulted in

Commerce resorting to using facts otherwise available under 19 U.S.C. § 1677e(a). Id.

at 3, 9–10.

Dillinger maintains that Commerce should have used Dillinger’s proffered

information regarding the average actual total cost of manufacture for all of its plate sold

during the POI. See Dillinger Comments at 7. While Dillinger acknowledges that its

proposal would require Commerce to accept data from an “average,” Dillinger maintains

that its preferred calculation nonetheless represents the “most reasonable calculation of

the actual production costs” because Dillinger’s proffered information “is based upon

actual costs.” Id. In rejecting Dillinger’s proposed alternative, Commerce explained that:

Dillinger’s normal books and records are more reasonable to
use as facts otherwise available because they recognize that
the lost value of the non-prime products, which is an inevitable
result of Dillinger’s production of prime products, is
appropriately considered to be a cost of producing the prime
products. Consequently, Dillinger’s proposal to assign the
overall average cost of all prime products is unreasonable
because it would distort the disparity in cost across prime CTL
plate products, as well as the disparity in “size, specification,
and grade” among non-prime products. Thus, although both
Dillinger’s proposal and Dillinger’s normal books and records
are flawed because Dillinger chooses not to track the actual
costs of producing non-prime products, we find that the use of
the amounts recorded in Dillinger’s normal books and records
is reasonable for use as facts otherwise available.
Consol. Court No. 17-00158 Page 9

Third Remand Results at 5–6.

Dillinger responds by emphasizing Commerce’s obligation under 19 U.S.C.

§ 1677b(b)(3) to calculate Dillinger’s actual cost of production of non-prime products.

Dillinger contends that Commerce may resort to facts otherwise available under

§ 1677e(a) only to fill an “informational gap” in the record, and that Commerce’s reliance

on the likely sales price for non-prime merchandise as a substitute for the actual cost of

production is an unreasonable application of facts otherwise available as the estimated

sales values of non-prime merchandise “has absolutely nothing to do with the costs of

production.” Dillinger Comments at 2–4. Dillinger maintains that Commerce

unreasonably relied on this selling price information because this information was not how

Dillinger actually valued the cost of production for non-prime products in its audited

financial statements. See id. at 1–2 (arguing that Commerce unreasonably conflated

record here with record in Dillinger France, which was subject to different generally

accepted accounting principles and practices).

Dillinger’s argument is unpersuasive. Dillinger placed this likely selling price

information on the record as part of its response in the Supplemental Section D

Questionnaire regarding “the ‘quantity and value of non-prime, defective, and low quality

plates sold during the POI.’” See id. at 3 (citing Dillinger’s Supplemental

Section D Questionnaire Response). Commerce has previously explained the

importance of reviewing information as to the “physical characteristics of the non-prime

products produced and the actual cost of producing the non-prime products,” and even
Consol. Court No. 17-00158 Page 10

re-opened the record to allow Dillinger to place actual cost information on the record.

See Third Remand Results at 3. When Dillinger failed to provide this actual cost

information, Commerce determined it was necessary to resort to facts otherwise available

under § 1677e(a) and to use the best available information on the record to fill this gap,

a determination already sustained in Dillinger II. See id. at 9–10. Commerce found that

this likely selling price information submitted by Dillinger is the “best available information”

on the record to value the cost of producing non-prime products in the absence of

accurate, actual cost of production data. See id. at 4, 5.

Despite maintaining that Commerce’s reliance on Dillinger’s likely selling price

information was unreasonable as that information was unrelated to the cost of production,

Dillinger fails to demonstrate that Commerce acted unreasonably in finding that “Dillinger

values non-prime products at their likely selling price, rather than full cost.” See id. at 10.

In light of this finding based on Dillinger’s questionnaire response, coupled with Dillinger’s

failure to put data corresponding to the actual cost of production of non-prime products

on the record, the court sustains Commerce’s use of the likely selling price information to

value the cost of production of non-prime products as a reasonable application of facts

otherwise available under § 1677e(a).

Dillinger lastly contends that “[b]y using the likely selling price of non-prime plate

rather than the actual cost of production allocated to non-prime plate in Dillinger’s verified

cost calculation, Commerce has imposed an impermissible adverse inference.” Dillinger

Comments at 14. Dillinger maintains that “[u]nder the statute, Commerce may only

impose an adverse inference when it ‘finds that an interested party has failed to cooperate
Consol. Court No. 17-00158 Page 11

by not acting to the best of its ability to comply with the request for information.’” Id.

(quoting 19 U.S.C. § 1677e(b)). Dillinger argues that “[b]y rejecting all of these other cost

of production figures and applying an unreasonably low cost of production to non-prime

plate based upon resale value, Commerce is applying an adverse inference that

impermissibly shifts costs from non-prime plate to prime plate and thereby increases the

dumping margin.” Id. at 20.

Dillinger’s argument is unsupported by the record. Dillinger’s naked assertion that

Commerce is applying an adverse inference lacks any basis beyond the fact that

Commerce’s selection of facts otherwise available ultimately resulted in an increase in

Dillinger’s calculated dumping margin. As Commerce explained:

Dillinger is quite simply mistaken that Commerce’s reliance on
its books and records to fill an informational gap created by
Dillinger’s decision is an impermissible adverse inference
because Commerce’s reliance on the information recorded in
Dillinger’s normal books and records accords with its own
recognition that the information recorded in its normal books
and records results in the total direct and indirect costs
reasonably attributable to the production of prime products
being allocated to prime products.

Third Remand Results at 16. Defendant further highlights that Commerce did not make

a finding that an adverse inference was warranted pursuant to 19 U.S.C. § 1677e(b)—

a prerequisite for applying an adverse inference when selecting from among the facts

otherwise available on the record. See Def.’s Resp. at 9 (citing Third Remand Results).

Dillinger’s dissatisfaction with its resulting dumping margin, without more, does not

demonstrate that Commerce’s selection of facts otherwise available was made with an

impermissible adverse inference. Dillinger’s remaining arguments and cited case law are
Consol. Court No. 17-00158 Page 12

without merit as they are predicated on the unfounded assumption that Commerce

applied an adverse inference here. Accordingly, the court sustains Commerce’s reliance

on Dillinger’s normal books and records as a reasonable application of facts otherwise

available.

B. Application of Partial AFA to Salzgitter

In a previous remand redetermination, Commerce explained that it used different

AFA methodologies to calculate Dillinger and Salzgitter’s margins, resulting in totals of

4.98% and 22.9% respectively, because the scope of Salzgitter’s non-disclosures was

significantly larger than Dillinger’s non-disclosures. See Second Remand Results at 27,

ECF No. 129. Salzgitter challenged the reasonableness of this determination, and the

court reserved decision on this issue in Dillinger II. See Dillinger II, 46 CIT at ___, 592 F.

Supp. 3d at 1347; see also Salzgitter Comments; Def.’s 2RR Resp. In calculating

Salzgitter’s margin, Commerce applied AFA to incentivize Salzgitter’s future cooperation.

Second Remand Results at 27. Specifically, Commerce explained that:

[T]he application of the Dillinger France I[3] partial AFA
methodology to Salzgitter deprives Commerce of the ability to
3 In Dillinger France S.A. v. United States, 42 CIT ___, 350 F. Supp. 3d 1349 (2018)
(“Dillinger France I”), the court remanded Commerce’s application of partial AFA to
Dillinger France, concluding that the decision “to utilize the highest non-aberrational net
price among Dillinger’s downstream home market sales” was unreasonable because
“the reliability of the reported sales prices has not been called into question and there is
no informational gap in the sale prices for Commerce to fill.” See id. at ___, 350 F.
Supp. 3d at 1364. On remand, Commerce followed the court’s guidance and determined
that it would “treat[] these downstream home market sales transactions as Dillinger
France-produced plate, rather than treating these transactions as sales of plate produced
by an unrelated manufacturer; and 2) rel[y] on the sale prices as reported.” See Dillinger
France S.A. v. United States, 43 CIT ___, ___ 393 F. Supp. 3d 1225, 1228 (2019) (quoting
Commerce’s remand results adopting Dillinger France I methodology), rev’d in part on
other grounds, 981 F.3d 1318 (Fed. Cir. 2020); see also AG der Dillinger Huttenwerke v.
Consol. Court No. 17-00158 Page 13

apply [19 U.S.C. § 1677c] meaningfully in this proceeding. It
is well established that Congress intended Commerce to use
AFA as a means to induce cooperation in its proceedings and
address evasion concerns. The purpose of AFA is to provide
respondents with an incentive to cooperate in Commerce’s
investigations and reviews and ensure that necessary
information is placed on the record to enable Commerce to
reach a reasonable determination. However, the change in
the AFA methodology prescribed by the Court in Dillinger
France I and applied to Salzgitter in the Dillinger I Remand
Redetermination frustrates Commerce’s goal of inducing
cooperation by ensuring that a non-cooperating respondent
does not receive a more favorable AFA rate than it would have
received it would have fully cooperated.

Id. at 29.

Dillinger reported manufacturer information for more than 99 percent of its

downstream sales in this matter and, while the “number of sales with missing

manufacturer information was not on the record” in Dillinger France, Commerce reported

that “it was only a small number of Dillinger France’s downstream sales.” Id. at 27.

Commerce could therefore approximate what Dillinger and Dillinger-France’s margins

would have been had they disclosed manufacturer information for all their downstream

sales and could be sure that the Dillinger France I methodology would not materially

impact either margin calculation. Id. at 27–28.

In contrast, Salzgitter did not report manufacturer information for approximately

28,000 downstream sales of CTL plate, representing a not-insignificant percentage of

United States, 43 CIT ___, ___, 399 F. Supp. 3d 1247, 1256–57 (2019) (“Dillinger I”)
(explaining that in Dillinger France, Commerce initially applied highest net-aberrational
price to all sales without manufacturer information, but ultimately accepting the sales
prices as reported, classifying all sales without manufacturer information as Dillinger
produced sales—Dillinger France I methodology).
Consol. Court No. 17-00158 Page 14

home market sales used in Commerce’s analysis. Id. at 27. Thus, Commerce maintains

that it “could not determine what Salzgitter’s margin would have been if Salzgitter had

fully cooperated with [its] requests for information and properly reported the manufacturer

of the downstream sales at issue,” so Salzgitter “may well receive a more favorable

margin [using the Dillinger France I methodology] than it would have received if [it] had

fully cooperated.” Id. at 29–31. As a result, Commerce applied the highest

non-aberrational net price for all of Salzgitter’s sales without manufacturer information to

insure it did not receive a lower margin than it otherwise would have. Id. at 30.

Salzgitter maintains that this approach is unreasonable because Commerce

compared the scope of each exporter’s non-disclosures inconsistently. First, Salzgitter

argues that “substantial evidence does not support Commerce’s conclusion that the sales

at issue for Dillinger France were smaller than the sales at issue for Salzgitter.” Salzgitter

Comments at 6. Further, Salzgitter notes that even if the scope of its non-disclosure was

larger, very few of those sales would be necessary to calculate its antidumping margin.

Id. at 4. Specifically, Salzgitter notes that:

Commerce claimed that the universe of sales considered with
respect to Salzgitter was larger than the universe of sales
considered with respect to Dillinger France, Commerce did
not similarly consider the linkage between the number of
Salzgitter sales affected and Salzgitter’s dumping margin.
Indeed, were Commerce to apply the analysis used for
Dillinger France to Salzgitter, it is clear that only a very small
fraction of SMSD’s sales for which manufacturer information
was unknown were use “as a basis for normal value” and were
“actually compared to U.S sales prices.”
Consol. Court No. 17-00158 Page 15

Id. (quoting Dillinger France S.A. v. United States, 43 CIT ___, ___, 393 F. Supp. 3d 1225,

1228 (2019)). Salzgitter maintains that, if Commerce only considered sales that were

necessary for its home market comparison, there would be little difference between the

scope of Salzgitter and Dillinger’s non-disclosures. Id.

Salzgitter further contends that Commerce’s application of § 1677e is

unreasonable because there is no indication that Salzgitter benefitted from not fully

disclosing all requested information, and there is no evidence that Salzgitter intentionally

obscured any information for this purpose. First, Salzgitter notes that it did not maliciously

or dishonestly omit information, but rather its information systems were not equipped to

record all of the information Commerce requested. Id. at 6. Second, Salzgitter maintains

that it does not benefit from these omissions because its antidumping margin would likely

have been zero percent even if it had disclosed all requested information. Id. at 8.

Commerce disagrees. First, Commerce notes that there is a factual difference

between the overall number of sales that Dillinger and Dillinger-France reported without

manufacturer information and the number of sales that Salzgitter reported without

manufacturer information, and not just a difference in how many sales are relevant to

each exporter’s margin calculation. Second Remand Results at 28. Specifically,

Commerce notes that the AFA methodology applied to Dillinger’s sales without

manufacturer information in this matter, as well as Dillinger-France’s sales without

manufacturer information, did not impact the margin calculation for Dillinger in either

proceeding. Id. at 27; see also First Remand Results at 2, ECF No. 85 (finding that

applying partial AFA methodology of Dillinger France I to Dillinger did not impact
Consol. Court No. 17-00158 Page 16

Dillinger’s margin calculation). Salzgitter’s margin, however, would have been reduced

from 22.90 percent, when Commerce applied the highest net-aberrational price,

to zero percent under the Dillinger France I methodology. Second Remand Results

at 28, 54.

Commerce further explained that “these differences affected Commerce’s goals in

using partial AFA as a means to induce cooperation because the margin result for

Salzgitter under the Dillinger France I methodology provides no incentive for Salzgitter to

cooperate by providing requested information to Commerce.” Id. at 54. Commerce

rejected Salzgitter’s suggested view of the record, stating that “Salzgitter would have

Commerce establish a new test of materiality to determine whether AFA is warranted –

a test that would allow a respondent, not Commerce, to determine what information is

relevant for Commerce’s analysis.” Id. at 55. Commerce maintains that Salzgitter’s

margin must reflect the full extent of its non-disclosure, and determined that using the

Dillinger France I methodology to assign Salzgitter a zero percent margin would not

incentivize future cooperation. Id. at 54–57.

Since a zero percent margin cannot, by definition, be higher than what Salzgitter’s

margin would otherwise have been if it had disclosed all its manufacturer information,

Commerce reasonably found that applying AFA to Salzgitter using the Dillinger France I

methodology would be inconsistent with the intent of § 1677e. For the same reason,

Commerce’s refusal to adopt one of Salzgitter’s three proposed alternative methods for

calculating normal value is also reasonable, as all three of Salzgitter’s proposed

alternatives would have left Salzgitter with a de minimis dumping margin. See Salzgitter
Consol. Court No. 17-00158 Page 17

Comments at 8–9 (explaining Salzgitter’s proposed alternatives that Commerce calculate

its margin by (1) treating none of the sales as Salzgitter-manufactured plate; (2) treating

all sales as Salzgitter-manufactured plate; or (3) treating a percentage of each sale as

Salzgitter-manufactured plate based on SMSD’s purchases from each supplier”);

see also Second Remand Results at 55–56 (noting that “[u]nder the Dillinger France I

partial AFA methodology, Salzgitter would receive a zero rate and, consequently, would

be excluded from the AD order. Because of Salzgitter’s failure to provide requested

information, Commerce cannot determine what the resulting margin would have been

if Salzgitter had complied fully with Commerce’s requests to report the manufacturer

information for all of its home market sales. Thus, it is reasonable to assume that

Salzgitter would receive a more favorable result under the Dillinger France I methodology

as a result of withholding information than by providing the requested information and

allowing Commerce to properly analyze the sales in question”).

19 U.S.C. § 1677e provides Commerce with discretion in applying AFA

methodologies. See, e.g., Nippon Steel Corp. v. United States, 337 F.3d 1373,1383

(Fed. Cir. 2003) (noting that 19 U.S.C. § 1677e does not require Commerce to find

“evidence of nefarious intentions” to apply AFA against the importer); F.lli de Cecco

di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir. 2000)

(stating that 19 U.S.C. § 1677c gives Commerce “broad discretion” in calculating

antidumping margins for “uncooperative respondents”). As the court observed in

Dillinger I, Salzgitter has failed to demonstrate that its proposed alternative methods

provide a reliable measure or approximation of what its margin would be if it fully disclosed
Consol. Court No. 17-00158 Page 18

all relevant information. See Dillinger I, 43 CIT at ___, 399 F. Supp. 3d at 1255–56. Since

Salzgitter did not provide any additional information to show that one of these alternative

methodologies constituted the only reasonable path forward on this record, the court

again concludes that Commerce acted reasonably in rejecting those proposed

alternatives.

Salzgitter contends that, even if Commerce acted reasonably in applying a

different AFA methodology than was applied to Dillinger, Commerce still unreasonably

ignored information that Salzgitter had already placed on the record in calculating its

margin. Salzgitter Comments at 10–11. Specifically, Salzgitter maintains that under

19 U.S.C. §1677m(e) “Commerce was not permitted on remand to disregard [its] verified

sales prices for the sales at issue as a result of the missing manufacturer [information].”

Id. at 10. Salzgitter maintains that it has demonstrated that “it would not receive a more

favorable AFA rate using the methodology applied to Dillinger France than it would have

received if it reported the manufacturer for all sales.” Id. at 8. Nevertheless, Salzgitter

admits that this conclusion requires Commerce to “not unjustifiably ‘ignore record

information that is not in dispute,’ namely the prices and other information for the SMSD

sales, which Commerce verified.” Id. Although the court in Dillinger France raised

concerns about Commerce’s refusal to consider the submitted sales price data in applying

AFA, this Court refused to reach the same conclusion in Dillinger I, observing that

“Commerce has clear statutory authority pursuant to 19 U.S.C. § 1677m(d) to ‘disregard

all or part of the original and subsequent responses’ in an adverse inference scenario.”

Dillinger I, 43 CIT at ___, 399 F. Supp. 3d at 1256; see also Second Remand Results
Consol. Court No. 17-00158 Page 19

at 55 (highlighting that “the Court acknowledged Commerce’s statutory authority under

section 782(d) of the Act to ‘disregard all or part of the original and subsequent responses’

when relying on AFA”).

Salzgitter responds that Commerce may only exercise this authority subject to

§ 1677m(e) and contends that Salzgitter’s pricing information could not be disregarded

by Commerce because Salzgitter’s submission of information met all of the criteria under

this provision. Salzgitter Comments at 10–11. The court previously addressed and

rejected this same argument. See Dillinger I, 43 CIT at ___, 399 F. Supp. 3d at 1253

(explaining that “the ‘information’ to which § 1677m(e) refers, in the context of this

proceeding, is the missing manufacturer information, not the remainder of ‘the

information’ that Plaintiffs submitted. Plaintiffs acknowledge that the identity of the CTL

plate manufacturers is relevant to whether home market transactions should or should

not be included in margin calculations, and that they did not identify all of them. Plaintiffs

thus cannot escape the conclusion that they failed to satisfy § 1677m(e) with respect to

that information.”). Because Salzgitter has failed to demonstrate any error in the court’s

prior analysis of this issue, the court again concludes that “Plaintiffs’ reliance upon

§ 1677(m)(e) is misplaced.” Id. As a result, the court cannot agree that Commerce’s

selected methodology for applying partial AFA to Salzgitter was unreasonable.

Lastly, Salzgitter contends that “Commerce’s selection of the highest

non-aberrational net price as AFA is inappropriate.” Salzgitter Comments at 12.

Salzgitter maintains that “that the sale from which this price was derived would not even

be used as a basis for normal value in Salzgitter’s margin calculation” because the
Consol. Court No. 17-00158 Page 20

product at issue in that sale “was so dissimilar to the products sold to the United States

that it was not compared to a single U.S. sale.” Id. Consequently, Salzgitter argues that

“[i]t is unreasonable and punitive for Commerce to extrapolate the price of a wholly

dissimilar product, and use that price as the basis for normal value for all of Salzgitter’s

home-market sales for which it could not identify the manufacturer.” Id. at 13. Commerce

stated that “[t]o determine the highest non-aberrational net price [] to be assigned to the

downstream sales with missing manufacturer information, Commerce sorted all of

SMSD’s net prices for these sales in descending order and selected the transaction at

the beginning of a smooth continuum of net prices.” Second Remand Results at 30

(confidential information omitted). Commerce further explained that “[b]ecause Salzgitter

failed to report the manufacturer of these sales, we cannot determine if the net prices

correlated to the manufacturer of the CTL plate. Commerce cannot rule out the possibility

that the sales with the highest prices were entirely or primarily of CTL plate manufactured

by Salzgitter, and Salzgitter’s failure to report the manufacturer information was an

attempt to obscure this fact, thereby distorting the margin.” Id. at 30–31.

Beyond generally decrying the unreasonableness of Commerce’s selected AFA

sale price, Salzgitter fails to suggest an alternative basis for an AFA sale price that would

instead be the one and only reasonable option on the record. While Salzgitter

emphasizes the fact that Commerce does not have “unlimited authority” in applying AFA,

Salzgitter does not identify how Commerce exceeded the bounds of reasonableness

here, or what alternative AFA price Commerce should have selected in order to meet the

purpose of § 1677e(b). See Salzgitter Comments at 13 (noting that “[t]he purpose of
Consol. Court No. 17-00158 Page 21

section 1677e(b) is to provide respondents with an incentive to cooperate, not to impose

punitive, aberrational, or uncorroborated margins.” (quoting F.Lii de Cecco di Filippo

Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed. Cir. 2000)). There is

no dispute that Commerce has the discretion, where appropriate, to select the highest

non-aberrational net price in applying AFA. See BMW of N. Am. v. United States,

926 F.3d 1291, 1300 (Fed. Cir. 2019) (noting that court has “previously held that

Commerce has wide discretion to assign the ‘highest calculated rate’ to uncooperative

parties,” but warning that “use of the highest rate is not automatic, however, and

‘will depend upon the facts of a particular case.’” (internal citations omitted)). Here,

Commerce has considered the totality of the record and explained the factors that led to

its differing application of AFA to Salzgitter as compared to Dillinger. See Second

Remand Results at 57 (highlighting differences in “(1) the number of sales lacking the

requested manufacturer information; (2) the net prices among the sales with the missing

data; and (3) the impact on the margin caused by the respondents’ failure to provide the

requested information.”). While Salzgitter contends that Commerce’s selected AFA price

(and resulting margin of 22.9%) is “punitive,” Salzgitter fails to explain how Commerce’s

selection was unreasonable given the totality of the circumstances on the record.

Salzgitter also fails to suggest any alternative price from the record that Commerce could

have selected as a reasonable application of AFA. Based on the record as a whole, the

court cannot agree with Salzgitter’s contention that Commerce’s selection of the highest

non-aberrational net price on the record was “unreasonable and punitive.” See Salzgitter
Consol. Court No. 17-00158 Page 22

Comments at 13. Accordingly, the court sustains Commerce’s application of partial AFA

to Salzgitter.

C. Rejection of Dillinger’s Proposed Quality Code for Sour Transport Plate

In a previous memorandum and order addressing Dillinger’s challenge to

Commerce’s model-match methodology, the court sustained Commerce’s rejection of

Dillinger’s proposed quality code for sour vessel plate but stayed consideration of

“Dillinger’s challenge to Commerce’s rejection of Dillinger’s other proposed quality code

(sour transport plate), pending the outcome of the remanded issues.” See Memorandum

and Order, ECF No. 121 (Aug. 18, 2021); see also Dillinger MSJ; Def.’s MSJ Resp.;

Def.-Intervenor Nucor Corporation Resp. Br., ECF No. 58; Dillinger MSJ Reply. The court

assumes familiarity with that decision, which outlined the basic details as to how

Commerce applies its model-match methodology and how that methodology was applied

in this matter. The court remands this issue again to Commerce for further consideration,

and if appropriate, reconsideration.

Commerce rejected Dillinger’s proposed quality code 771 (for sour transport plate),

explaining:

In its Dillinger Model Match Comments, Dillinger identified two
examples of products contained in its proposed sour service
petroleum transport plate quality subcategory: NACE
TM0284/ISO 15156-2 and NACE MR0175/ISO 15156. We did
not adopt this suggested quality subcategory in the final
model match methodology issued to interested parties, and
instead we identified a single quality code for petroleum
transport plate.

Nonetheless, Dillinger reported sales in its questionnaire
responses using its proposed quality code subcategory for
Consol. Court No. 17-00158 Page 23

this product, and also changed the examples it provided for
the subcategory to be “steel grades L450MS-PSL2,
5L-X65MS-PSL2, etc.” without explanation. Dillinger did not
identify what standards it had provided, if any, to identify the
products to which it refers. The absence of any actual
standards, identifying the full range of properties for such
products, limits our ability to evaluate how such products
compare to other petroleum transport plate products.

Dillinger provided a “Presentation on Requirements for Steel
Plates in Sour Service” (Sour Service Presentation), which
appears to be a slide presentation containing information
about sour service. However, the Sour Service Presentation
does not provide a systematic or clear reference to the range
of properties of the products in question. Of the four example
products Dillinger listed in the Dillinger Model Match
Comments and its section B response, only one of them
(i.e., NACE FR0175/ISO 15156) appears to be clearly
identified in the Sour Service Presentation for use in the
corrosive hydrogen sulfide environments Dillinger indicates
require such plate, while the example products listed in
Dillinger’s section B response are not referenced at all.

Dillinger indicates that the sulfur content must be strictly
limited for sour service petroleum transport plate, and we note
that the Sour Service Presentation does appear to refer to a
maximum allowable percentage level, which it refers to as
“low.” However, it is not evident that such a requirement
applies to the two example “grades” (L450MS-PSL2,
5L-X65MS-PSL2) identified in Dillinger’s section B response.
Even assuming, arguendo, that those grades are within the
API 5L line pipe specification, as the petitioner states, that
would support the petitioner’s argument that Dillinger’s
petroleum transport plate products are covered under the
same specification as other petroleum plate products
identified by the quality code established by the Department.
The Sour Service Presentation also does not refer to the
content requirements of carbon or the “expensive alloys”
(i.e., copper and nickel) discussed in the Dillinger Model
Match Comments.

Furthermore, assuming these elements are pertinent to the
analysis, the Department’s model match methodology
Consol. Court No. 17-00158 Page 24

contains product characteristic fields that segregate products
based on minimum specified content of two of those three
elements (i.e., carbon and nickel). If the levels of these
chemical elements are important distinguishing factors for
sour service petroleum transport plate, as Dillinger indicates,
the separate product characteristic fields for those elements
would distinguish sour service petroleum transport plate
products from other plate products.

Similarly, the heat treatment product characteristic may also
distinguish these products from other petroleum transport
plate products. The Sour Service Presentation refers to the
use of “Q&T” (i.e., quenching and tempering) to effect the
desired end properties of the sour service petroleum transport
plate. Products that have been quenched and tempered will
be assigned a different heat treatment code than those which
have not undergone that treatment.

Therefore, we do not agree that a new quality reporting code
is required to distinguish sour service petroleum transport
plate from other products. We find that Dillinger did not
subsequently provide information that would justify either
allowing it to report revised quality codes for different
petroleum transport plate products or revisiting this issue once
parties had submitted their questionnaire responses. Instead,
we find that Dillinger has failed to both: 1) justify creating a
quality code subcategory for this product; and 2) clearly
identify the products that would be classified in its proposed
subcategory. Consequently, consistent with the Preliminary
Determination, we continued to reassign all products which
Dillinger reported with a quality code of 771 to have a quality
code 772, thereby assigning all petroleum transport plate
products the same quality code.

Decision Memorandum at 77–79 (footnotes omitted).

The model-match methodology, based on 19 U.S.C. § 1677(16)(A), determines

matches based on physical differences. Courts have noted that this is a consideration

apart from whether physical characteristics result in price and cost differences between

products. See Maverick Tube Corp. v. United States, 39 CIT ___, ___, 107 F. Supp. 3d
Consol. Court No. 17-00158 Page 25

1318, 1330 (2015) (“differences in costs do not constitute differences in products in and

of themselves”).

As noted above, Dillinger explains that its sour transport plate is used with “sour”

petroleum products containing high amounts of hydrogen sulfide, thus the sour transport

plate is made with “extremely low levels of phosphorus and sulfur” to withstand the

corrosion effects of the hydrogen sulfide. See Dillinger MSJ at 11. Dillinger thus

maintains that there are non-minor, commercially significant differences in physical

characteristics between sour transport plate and other petroleum transport products.

See id. at 11–15; Dillinger MSJ Reply at 4–7 (citing Pesquera Mares Australes Ltda. v.

United States, 266 F.3d 1372, 1384 (Fed. Cir. 2001) for proposition that merchandise can

only be treated as identical under 19 U.S.C. § 1677(16)(A) if it has either (1) no

differences in physical characteristics or (2) the differences are only minor and ‘not

commercially significant’”).

Dillinger highlights Bohler Bleche GMBH & Co. KG v. United States, 42 CIT ___,

324 F. Supp. 3d 1344 (2018) (“Bohler”), in which the court “struck down the model-match

methodology used in this investigation.” Dillinger MSJ Reply at 1. Relying on this

decision, Dillinger maintains that it should receive similar relief as the respondent in that

case. Id. at 2. In Bohler, the plaintiff-respondents challenged a final determination by

Commerce, which relied on the same model-match methodology that was used in the

underlying proceeding here, arguing that Commerce had failed to adequately account for

“the alloy content of Plaintiffs’ specialized high alloy steel products, thereby failing to

account for significant differences in physical characteristics, costs, and price.”
Consol. Court No. 17-00158 Page 26

See Bohler, 42 CIT at ___, 324 F. Supp. 3d at 1348. While Commerce there disagreed

“that the newly proposed methodologies would have the effect of creating closer matches

between exported merchandise and home market merchandise,” the court ultimately

agreed with the plaintiffs that the “methodology insufficiently accounts for alloy content in

Plaintiffs' products” and remanded the issue to Commerce for reconsideration. Id., 42 CIT

at ___, 324 F. Supp. 3d at 1348, 1354. On remand, Commerce changed course and

revised its methodology to better account for these alloy content differences. 4

Here, in a similar fashion, Commerce rejected Dillinger’s contention that the record

reflected that lower levels of phosphorus and sulfur in these steels distinguished them

from other petroleum transport plate. See Decision Memorandum at 77–79 (reviewing

record evidence cited by Dillinger in support of its position, and explaining findings that

“Dillinger has failed to both: 1) justify creating a quality code subcategory for this product;

and 2) clearly identify the products that would be classified in its proposed subcategory.”).

Thus, although Commerce acknowledged the record evidence supporting a finding that

Dillinger’s sour transport plate had different physical characteristics than other

comparable products (i.e., lower maximum sulfur content), Commerce ultimately did not

agree “that a new quality reporting code is required to distinguish sour service petroleum

4 While Commerce noted that it was changing its model-match methodology to meet the
respondent’s concerns in that matter “under protest,” the Government did not appeal the
court’s subsequent decision sustaining those remand results. See Bohler Bleche
Remand Results at 2, Court No. 17-00163, ECF No. 55 (explaining that Commerce would
adopt respondent’s proposed alternative model-match methodology under protest);
Bohler Bleche GMBH & Co. KG v. United States, 43 CIT ___, 362 F. Supp. 3d 1377
(2019) (sustaining as reasonable Commerce’s adoption on remand of plaintiffs’
alternative model-match methodology “as it fairly compares commercially significant
differences in physical characteristics”).
 Consol. Court No. 17-00158 Page 27

transport plate from other products.” Decision Memorandum at 79. Given Commerce’s

apparent recognition in Bohler that its model-match methodology insufficiently accounted

for variations in the alloy content of the products at issue in that proceeding, the court

concludes that Commerce should have the opportunity to explain why a similar outcome

is not warranted here.

Because Bohler and Commerce’s subsequent remand results in that action were

not published until after the submission of the Government’s response brief in this litigation,

Commerce has had no opportunity to address whether the circumstances in Bohler are

comparable to those here. At oral argument, the court noted its concern for the parties that

any response by the Government or Defendant-Intervenor to the circumstances of Bohler

might constitute post hoc rationalization that the court could not use to sustain the

decision-making of Commerce without potentially violating fundamental principles of

administrative law. See, e.g., Burlington Truck Lines Inc. v. United States, 371 U.S. 156,

168 (1962) (“courts may not accept appellate counsel’s post hoc rationalizations for agency

action’’); SEC v. Chenery Corp., 332 U.S. 194, 196 (1947) (warning that courts “must judge

the propriety of [agency] action solely by the grounds invoked by the agency”). As the

circumstances in Bohler appear analogous, the court reiterates its observation that

“[r]easoned decision-making requires a certain measure of consistency.” See Dillinger I,

43 CIT at ___, 399 F. Supp. 3d at 1257. Accordingly, the court remands this issue for

Commerce to further explain why its determination is reasonable in light of its approach in

Bohler, or if appropriate, reconsider its rejection of Dillinger’s proposed quality code for

sour transport plate.
Consol. Court No. 17-00158 Page 28

III. Conclusion

For the foregoing reasons, it is hereby

ORDERED that Commerce’s determinations as to the cost adjustments for

Dillinger’s non-prime plate, as well as the application of partial AFA to Salzgitter, are

sustained; it is further

ORDERED that Commerce’s determination to reject Dillinger’s proposed quality

code for sour transport plate is remanded for further explanation, and if appropriate,

reconsideration; it is further

ORDERED that Commerce shall file its remand results on or before September 7,

2023; and it is further

ORDERED that, if applicable, the parties shall file a proposed scheduling order

with page limits for comments on the remand results no later than seven days after

Commerce files its remand results with the court.

/s/ Leo M. Gordon
Judge Leo M. Gordon

Dated: June 23, 2023
New York, New York