Rebar Trade Action Coal. v. United States

2015 CIT 130

• Court: United States Court of International Trade
• Decided: November 23, 2015
• Docket: 14-00268
• Status: Published

Opinion

Slip Op 15 - 130

UNITED STATES COURT OF INTERNATIONAL TRADE

: REBAR TRADE ACTION COALITION, et al., : : Plaintiffs, : : v. : Before: R. Kenton Musgrave, Senior Judge : UNITED STATES, : Court No. 14-00268 : Defendant, : : and : : ICDAS CELIK ENERJI TERSANE VE ULASIM : SANAYI, A.S., and HABAS SINAI VE TIBBI : GAZLAR ISTIHSAL ENDUSTRISI A.S., : : Defendant-Intervenors. : :

OPINION AND ORDER

[Remanding negative less-than-fair-value determination on rebar from Turkey.]

Dated: November 23, 2015

Alan H. Price, John R. Shane, Maureen E. Thorson, and Jeffrey O. Frank, Wiley Rein LLP, of Washington, DC, for plaintiffs.

Richard P. Schroeder, Trial Attorney, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, of Washington, DC, for the defendant. With him on the brief were Benjamin C. Mizer, Principal Deputy Assistant Attorney General, Jeanne E. Davidson, Director, and Reginald T. Blades, Jr., Assistant Director. Of Counsel on the brief was David W. Richardson, Attorney, Office of the Chief Counsel for Trade Enforcement and Compliance, U.S. Department of Commerce, of Washington, DC.

Matthew M. Nolan, Nancy A. Noonan, and Diana Dimitriuc Quaia, Arent Fox LLP, of Washington, DC, for defendant-intervenor Icdas Celik Enerji Tersane ve Ulasim, A.S. Court No. 14-00268 Page 2

David J. Simon, Law Office of David L. Simon, of Washington, DC, for defendant-intervenor Habas Sinai ve Tibbi Gazlar Istihsal Endustrisi A.S.

Musgrave, Senior Judge: Domestic industry representatives Rebar Trade Action

Coalition and its individual members (plaintiffs or “RTAC”) challenge a number of aspects on the

record of Steel Concrete Reinforcing Bar From Turkey: Final Negative Determination of Sales at

Less Than Fair Value and Final Determination of Critical Circumstances, 79 Fed. Reg. 21986 (Sep.

15, 2014) (“Final Results”), and accompanying issues and decision memorandum (“IDM”), as

compiled by the U.S. Department of Commerce, International Trade Administration (“Commerce”).

The period of investigation is July 2012, through June 2013.

ICDAS Celik Enerji Tersane ve Ulasim, A.S. (“Icdas”) and Habas Sinai ve Tibbi

Gazlar Istihsal Endustrisi A.S. (“Habas”), respondents at the administrative proceeding, have

intervened in defense of Commerce’s determination. The plaintiffs’ USCIT Rule 56.2 motion for

judgment coalesce their nine-count complaint into four broad issues: (1) calculation of duty

drawback adjustments for each respondent, specifically the fact that the cost side of the adjustments

is on a different basis than the sales side, as well as grant of the adjustments in the first instance; (2)

use of invoice date as the U.S. date of sale in the antidumping duty margin calculation for Icdas; (3)

acceptance of potentially misreported yield strength information for rebar produced and sold by

Icdas; and (4) failure to collect alloy usage cost information from Icdas and declining to adjust

Icdas’s costs to reflect alloy useage differentials.

Commerce has requested voluntary remand of the issue concerning duty drawback

calculation, which has also prompted the plaintiffs to file a motion to expedite reconsideration of that Court No. 14-00268 Page 3

single issue. The defendant and Icdas oppose that motion, arguing that bifurcation resulting in

multiple remands is disfavored and that the plaintiffs have not shown good cause therefor, while

Habas did not file a position thereon. In view of the quality of the briefing, this opinion moots the

motion for oral argument as well as the motion for bifurcation and expedition of only one of the

issues, and the matter as a whole will be remanded in accordance with the following.

II. Jurisdiction and Standard of Review

Jurisdiction is here pursuant to 28 U.S.C. §1581(c). In this type of proceeding, the

court holds unlawful any determination, finding, or conclusion found “unsupported by substantial

evidence on the record, or otherwise not in accordance with law”. 19 U.S.C. §1516a(b)(l)(B)(i).

III. Discussion

A. Procedural History

The plaintiffs filed their antidumping duty petition regarding rebar from Turkey on

September 4, 2013. See IDM at 2. On October 2, 2 013, Commerce initiated its investigation thereof

and issued an affirmative preliminary determination in April 2014. See Steel Concrete Reinforcing

Bar From Turkey, 79 Fed. Reg. 22804 (Apr. 24. 2014) (“Preliminary Determination”) and

accompanying issues and decision memorandum (“Pre-IDM”), PDoc 212. Commerce calculated

preliminary margins of 0.00 percent for Habas, 2.64 percent for Icdas, and 2.64 percent as the “all

others” rate. Preliminary Determination, 79 Fed. Reg. at 22805. After verification of the

respondents and review of case and rebuttal briefs, on September 15, 2014, Commerce published

its final determination as negative, i.e., that it did not find sales of rebar from Turkey to the United Court No. 14-00268 Page 4

States to have been sold at less than fair value, and terminated the investigation. 79 Fed. Reg. 54965.

See IDM at 1. This appeal followed.

B. Duty Drawback Adjustment Issues

The plaintiffs’ first challenges are to Commerce’s decision to grant respondents a duty

drawback adjustment with respect to a particular tax imposed by Turkey and the calculation thereof.

1. Statutory and Regulatory Framework

Pursuant to the Tariff Act of 1930, as amended, Commerce will upwardly adjust

export prices by “the amount of any import duties imposed by the country of exportation which have

been rebated, or which have not been collected, by reason of the exportation of the subject

merchandise to the United States.” 19 U.S.C. 1677a(c)(1)(B). The United States Court of Appeals

for the Federal Circuit has explained that “if a foreign country would normally impose an import

duty on an input used to manufacture the subject merchandise, but offers a rebate or exemption from

the duty if the input is exported to the United States, then Commerce will increase [the export price]

to account for the rebated or unpaid import duty (the ‘duty drawback’).” Saha Thai Steel Pipe

(Public) Company Ltd. v. United States, 635 F.3d 1335, 1338 (Fed. Cir. 2011) (“Saha Thai”).

“The purpose of the duty drawback adjustment is to account for the fact that the

producers remain subject to the import duty when they sell the subject merchandise domestically,

which increases home market sales prices and thereby increases [normal value].” Id. “A duty

drawback adjustment is meant to prevent dumping margins that arise because the exporting country

rebates import duties and taxes that it had imposed on raw material used to produce merchandise that Court No. 14-00268 Page 5

is subsequently exported.” Allied Tube & Conduit Corp. v. United States, 29 CIT 502, 506, 374 F.

Supp. 2d 1257, 1261 (2005) (citations omitted).

In order to determine whether a respondent is eligible for a duty drawback adjustment,

Commerce employs a two-prong test, pursuant to which the respondent must establish (1) that the

rebate and import duties are dependent upon one another, or in the context of an exemption from

import duties that the exemption is linked to the exportation of the subject merchandise, and (2) that

there are sufficient imports of the raw material to account for the duty drawback on the exports of

the subject merchandise. Saha Thai, 635 F.3d at 1340 (citation omitted).

2. Further Background

Commerce requested Icdas and Habas to “[r]eport the unit amount of any duty

drawback received upon exportation of the subject merchandise to the United States . . . [and]

[e]xplain how the amount of the duty drawback received is calculated”. Commerce Sections B-D

Questionnaire at C-24, PDoc 61. Both Icdas and Habas reported that they had participated in the

Turkish “Inward Processing Regime” (“IPR”), which permitted them to import into Turkey “raw

materials free of import duties, the resource utilization fund (KKDF) and value added tax if such

inputs are intended for producing final goods for export.” See Icdas Section C Questionnaire Resp.

at C-33--C-34, PDoc 118; Habas Section C Questionnaire Resp. at C-34, PDoc 110.

The IPR defines its “tax” as “[a]ll financial obligation such as taxes, duties, fees, fund

payments etc. which are stipulated for collection during import and export goods.” See Habas

Section C Questionnaire Resp., Ex. C-16 at 3. Icdas further explained that the “resource utilization

fund”, or “KKDF tax”, is “a tax imposed on foreign currency loans”, and that the IPR made such Court No. 14-00268 Page 6

foreign currency loans exempt from the KKDF tax if the loans were “used to finance imported inputs

that are used to manufacture goods for export.” Icdas Supplemental Questionnaire Resp. at SC-13--

SC-14, PDoc 167. Both Icdas and Habas explained that the IPR is similar to the United States’

system of “substitution” duty drawback, meaning that the company does not have to “track whether

particular imports are physically incorporated into particular exports; it is enough that the imports

were actually imported, that the exports were actually exported, and that the products that were

exported were made from inputs of the type of imports.” Icdas Supplemental Questionnaire Resp.

at SC-12--SC-13, PDoc 167; Habas Supplemental Questionnaire Resp. at 28, PDoc 161 (highlighting

omitted). That is, the respondents claimed that due to the fungible nature of the imported steel input,

tracing the specific molecules of the steel input from the import to the export was unnecessary as

long as the export of the finished product (i.e., rebar) contained sufficient amounts of the steel to

account for the quantity of the imports. Id.

To benefit from duty drawback, a firm must “obtain a Domestic Processing

Certificate/Authorization” (“IPC”). See Habas Section C Questionnaire Resp., Ex. C-16 at 7, Art.

9, PDoc 110. Once the export obligation under an IPC has been met, the party must disclose the

specific export commitment by submitting evidence that the processed products were exported. See

id. at 11, Art. 19. Icdas and Habas provided an explanation thereof and exhibits, including IPCs,

letters, and other documentation pertinent thereto. See Icdas 2nd Supplemental Questionnaire Resp.

at S2C-1--S2C-3, PDoc 198; Habas 2nd Supplemental Questionnaire Resp. at 1-2, PDoc 187.

According to Commerce, these submissions provided for the record evidence that (1) the products

were sold for export, (2) that they were produced from imported scrap under an IPC, (3) that they Court No. 14-00268 Page 7

actually were exported, (4) that the necessary approval therefor was obtained from the Turkish

government, and (5) that Icdas and Habas received the completion report from the Turkish

government. Def’s Resp. at 9 (citations omitted).

In the Preliminary Determination, Commerce granted duty drawback adjustments to

both Icdas and Habas. Pre-IDM at 17-18. Commerce explains that before reaching that

determination it had compared a list of imports during the period of investigation with the IPCs

submitted by the respondents, and it also reviewed the documentation that they claimed to have

provided to the Turkish government “when they closed out an IPC.” Def’s Resp. at 10, quoting id.

at 18. Commerce observed that the documentation the respondents submitted provided a tally of

imports and exports. Pre-IDM at 17-18. Applying its two-pronged test for granting a duty drawback

adjustment, Commerce preliminarily determined that (1) “respondents established sufficient linkage

between their respective inputs and the exports of subject merchandise during the [period of

investigation]” and (2) respondents had sufficient imports to account for the duty drawback

received.” Id.

At verification, Commerce officials requested and reviewed the relevant Turkish

customs regulations and law. See Icdas Verification Report at 21-22, PDoc 272. Commerce officials

were provided access to Icdas’s Turkish Customs online account and found it to be consistent with

the data submitted to Commerce. Id. at 22. Commerce confirmed that the IPCs used during the

period of investigation were closed and that quantities for the period of investigation matched the

online database. Id. Commerce also reviewed the relevant customs duties and the KKDF tax rates

for inputs from several countries, and it tracked the applicable duty rates to the rates that Icdas used Court No. 14-00268 Page 8

for its drawback calculations. Id. For two sales, Commerce officials tied the reported quantity by

vessel and the relevant IPC to the United States sales listing. Id. Commerce officials also traced a

single import and a single export through Icdas’s financial records. Id. Finally, Commerce officials

confirmed that the percentage of scrap reported was based on imported scrap. Id.

Commerce found that all the information presented on this issue was consistent with

Icdas’s questionnaire response. Id. It also made similar inquiries at the Habas verification and, other

than minor corrections reported by Habas at the beginning of the verification, Commerce found no

discrepancies in Habas’s questionnaire responses. See Habas Verification Report, at 23-24, PDoc

271.

In the Final Results, Commerce continued to grant both Icdas and Habas the duty

drawback adjustment. IDM at 13-14. Commerce again applied its two-prong test, id., and with

respect to the first prong of that test, Commerce expressly noted that it looks for a “reasonable link

between the duties traced and rebated or exempted” and that it does not require “the imported input

be traced directly from importation through exportation.” Id.

In finding that the respondents had met the first prong of the duty drawback test,

Commerce relied on the Turkish laws and regulations, as well as the paperwork that is required to

be filed to qualify for the KKDF tax exemption submitted with questionnaire responses and

examined at verification. IDM at 14-15. Commerce determined, based on this record information,

that “[e]ach respondent had demonstrated that[,] although the KKDF [tax] is related to the type of

financing used, the tax is import dependant and export contingent.” Id. at 14. Court No. 14-00268 Page 9

In finding that the respondents had satisfied the second prong of the duty drawback

test -- that is, that both respondents had sufficient imports to account for the exported merchandise --

Commerce relied, among other things, on respondents’ lists of imports during the period of

investigation, documentation matching the imports to the exports, and Commerce’s verification

findings, including an examination of the Turkish government’s on-line customs database. Id. at 15.

Commerce also found that a comparison of the Turkish and United States duty

drawback systems did not undermine the legitimacy of the Turkish system. Id. at 16. It further

addressed a number of cases on which the plaintiffs had relied to support their arguments, finding

that they did not require Commerce to reach a different result. Id.

In sum, after considering the evidence on the record, as well as the parties’ arguments,

IDM at 9-13, Commerce concluded that both Icdas and Habas were entitled to a duty drawback

adjustment.

3. Calculation of Duty Drawback Adjustment

The plaintiffs argue that the “sales side” duty drawback adjustments for each

respondent involved dividing total drawback duties by total exports, in contrast to the “cost side”

adjustments that relied on total production cost for both domestic and export sales; this, the plaintiffs

contend, results in distorted adjustments. See Compl. ¶¶ 41-42; Pls’ Br. at 14-21.

The defendant acknowledges the inconsistency (“because, although the adjustment

to the costs averaged the total duties over total domestic and foreign sourced input costs, the

sales-side adjustment spread the entire duty drawback amount on the just exported sales”) and

requests voluntary remand in order to reconsider pursuant to SKF USA Inc. v. United States, 254 Court No. 14-00268 Page 10

F.3d 1022, 1029 (Fed. Cir. 2001) (court has discretion to issue a remand if the government so

requests even without admission of error).1

Icdas defends this aspect of the Final Results, contending that the petitioners seek to

overturn established precedent on arguments that are inconsistent with the statute and based on a

faulty premise. Icdas Resp. at 12-17. Icdas cites a number of cases, including Saha Thai Steel Pipe

(Public) Co. v. United States, 635 F.3d 1335 (Fed. Cir. 2011) (“Saha Thai”), indicating that

Commerce normally calculates (and must calculate) a cost side duty drawback adjustment wherever

it calculates a sales side adjustment. Icdas Resp. at 12-14. The plaintiffs reply that none of the cited

cases deal with the allocation basis for that adjustment, and that even Saha Thai appears to have been

concerned solely with the question of whether a cost side duty drawback adjustment should be made,

and not with “how” the “corresponding increase” to the cost of production is to be calculated. Pls’

Reply at 2, referencing Saha Thai, 635 F.3d at 1342-43.

Icdas also argues that the Tariff Act of 1930 defines “cost of production” such that

there can only be one, single cost of production determined, covering both home-market and export

sales. Icdas Resp. at 14-15. The plaintiffs reply that while the statute specifies the costs that make

up the cost of production, it does not mandate the methodology by which those costs must be

calculated. See 19 U.S.C. § 1677b(b)(3); see also SeAH Steel Corp. v. United States, 34 CIT 605,

614, 704 F. Supp. 2d 1353, 1363 (2010) (the statute “does not dictate the method by which

Commerce may calculate costs of production”). Nor, they argue, is it true that Commerce never

1 Def’s Resp. at 16-17, referencing IDM at 17-18 & Pls’ Br. at 14-21. The defendant’s request prompted the plaintiffs’ motion to expedite remand of this single issue. See ECF Nos. 67-69 (Sep. 18-21, 2015). As indicated above, in light of this opinion that motion, as well as the plaintiffs’ motion for oral argument, ECF No. 66 (Sep. 18, 2015), must be, and hereby are, denied as moot. Court No. 14-00268 Page 11

calculates multiple costs of production; rather, it routinely calculates separate costs of production

for each model of merchandise at issue. Pls’ Reply at 3, referencing id. at 1358. Moreover, they

point out, Commerce regularly adjusts for differences in costs that are incidental to selling to specific

markets, such as freight. Id.

Finally, in arguing that the plaintiffs’ claims are based on a faulty premise, Icdas

states that “export sales simply did not incur the raw material duty costs as prescribed by the duty

drawback.” Icdas Resp. at 16. The plaintiffs reply that the sales-side adjustment requires that

exempted duties be added to export price, although export sales did not incur the related costs, thus

putting the export prices on the same basis as home-market prices and avoiding phantom dumping

margins. Likewise, they contend, on the cost side of the equation imputed duties should be allocated

so as to avoid phantom “cushions” against dumping.

For its part, Habas also defends the Final Results as they stand, arguing that it

reported its costs in accordance with Commerce’s instructions, and that the plaintiffs’ arguments are

inconsistent with the “matching principle.” Habas Resp. at 8-12. The plaintiffs reply they are not

arguing any failure to respond to the agency’s questionnaires, they are arguing Commerce has not

adequately explained its rejection of their cost side allocation claim and that the allocation distorts

the margin. Regarding the matching principle of accounting, the plaintiffs contend that the sales side

drawback adjustment requires adding imputed duty values to U.S. prices while home-market prices

require no adjustment because they already reflect the value of any import duties; thus, they contend,

in order to match the sales side adjustment on the cost side, per Saha Thai, the costs for exported

material must be adjusted to account for duties not actually paid, while no adjustment is required to Court No. 14-00268 Page 12

domestically-sold product costs.2 Likewise, the plaintiffs further contend, under the matching

principle it would be illogical to impute the costs of the import duties that have been added to the

export price sales to home market production -- which should already reflect any import duty costs --

but, as the Final Results currently stand, Commerce “did exactly that”. Id. (confidential portion of

argument omitted).

Summarizing, the plaintiffs argue the cost side of the duty drawback allocation was

not well-explained in this instance, resulted in distorted margins, and delivered results that appear

contrary to the agency’s intent. Whether Commerce would agree with that characterization, the court

finds that Commerce’s request for remand expresses a substantial and legitimate concern for

reconsidering its duty drawback adjustment calculation, and the matter will be remanded therefor.

4. Sufficiency of the Agency’s Explanation for Including the KKDF Tax as Part of the Duty Drawback Adjustment Calculation

The plaintiffs also contend that Commerce failed to explain sufficiently the

determination to grant a duty drawback adjustment with respect to the KKDF tax in the first place.

The IDM’s discussion of the KKDF tax begins by stating that “[i]n order for Turkish

companies to qualify for exemptions from paying customs duties and KKDF [tax] on imported

inputs for rebar exports under the IPR, each respondent demonstrated that it applied for[,] or

‘opened,’ and the GOT maintained[,] an IPC[,] which is the official mechanism under the IPR by

which companies justify, and the GOT affirms, entitlement to such exemptions.” IDM at 14. The

plaintiffs contend that the sum of the IDM’s explanation and analysis of the KKDF tax, including

2 “As Saha Thai explained, ‘it would be illogical to increase EP to account for import duties that are purportedly reflected in NV, while simultaneously calculating NV based on a COP and CV that do not reflect those import duties.’” Pls’ Reply at 4, quoting Saha Thai, 635 F.3d at 1342. Court No. 14-00268 Page 13

citation to the verification reports, is that the respondents had demonstrated that “although the KKDF

is related to the type of financing used, the tax is import-dependent and export contingent.” Id.,

citing Habas Sales Verification Report at 23-25 and SVE-21; and Icdas Sales Verification Report

at 21-22 and SVE-29.

The plaintiffs argue that this is inadequate explanation for determining that the KKDF

tax qualified as a statutory “import duty” under 19 U.S.C. §1677a(c)(l)(B) or, for that matter, that

the tax was “import-dependant and export contingent.” The KKDF tax is not imposed on imports

as such, the plaintiffs argue, but on commercial loans that are financed in certain ways, and

regardless of whether those loans are used to support imports or not. Their argument is that the tax

amount is not based on the value of goods secured, nor is it paid through the mechanism of the loan

itself, but is rather based on the amount of money borrowed. See, e.g., Pls’ Reply at 5, referencing

Icdas Sections A and C Supplemental Response at Ex. SC-14, PDoc 166. As such, they contend they

pointed out to Commerce that the KKDF tax did not qualify as an “import duty” within the meaning

of 19 U.S.C. § 1677a(c)(l)(B) because the KKDF tax can be avoided altogether, even with respect

to loans to support imports, simply by avoiding certain types of financing options such as acceptance

loans or loans denominated in foreign currencies. See PDoc 283, CDoc 504, at 46-49. They also

argued that the record did not show that the loans by which respondents financed their import

purchases were of a type that would incur the KKDF tax, such that the tax could be rebated or

exempted by reason of their exports of rebar to the United States. Pls’ Br. at 9, referencing PDocs

166, 168. Elaborating here, the plaintiffs acknowledge that financing, if documented to support the

importation of goods under the IPR, is “exempt” from KKDF taxes, but they argue that this fact does Court No. 14-00268 Page 14

not, by itself, explain the basis for finding that the KKDF tax is itself an “import duty” contemplated

by the statute. For example, they contend, the fact that the IPR can exempt income taxes does not

transform income taxes into “import duties”, and they further complain that Commerce never

explained how export contingency could signify that a tax is an import duty. Id. at 10.

Commerce responds that it specifically reviewed the Turkish import system, including

the customs regulations specific to the IPR and the relevant IPCs, and explained that the applicable

Turkish government bylaws concerning the KKDF tax provides, in relevant part, that a six percent

rate shall be applied on “imports made with acceptance loan, deferred letter of credit and in the form

of cash on delivery.” Def’s Resp. referencing PDoc 166 at Ex. SC-14 (“Communique On Resource

Utilization Fund Regarding Bylaw”), Art. 2, Sec. (7)D, PDoc 168. After also quoting Article 2,

Section 8 of those resource utilization fund bylaws, which describes an applicable KKDF tax rate

of a zero percent rate on loans for financing exports, Commerce emphasizes that the bylaws

specifically reference the “inward processing license” (IPC) with which its determination was

concerned. Def’s Resp. at 12-13. Commerce states that because the explicit exemption in the

bylaws for the KKDF tax on imports under the Turkish duty drawback system is contingent on

import and export commitments, it reasonably found the KKDF tax to be eligible for adjustment in

accordance with the Turkish duty drawback scheme.

Although the relevant translation of the Turkish decree might speak for itself, and

although Commerce’s response borders on impermissible post hoc explanation,3 Commerce’s

3 The plaintiffs also contend the verification reports do not appear to furnish the required explanations. See id. They point out that the sales verification report for Habas, for example, notes that the agency collected a copy of the KKDF decree (which had previously been provided in the (continued...) Court No. 14-00268 Page 15

response does not, in any event, completely address the plaintiffs’ points insofar as it assumes that

KKDF tax was in fact owing on the respondents’ methods of import financing. The plaintiffs further

plead for address of their argument that even if the KKDF did constitute an “import duty” it was not

one that was rebated or which went uncollected by reason of exports as required by 19 U.S.C.

§1677a(c)(1)(B). Cf. RTAC Case Br. at 46-49, with IDM at 14. The plaintiffs contend that in their

case brief they argued that it was the respondents’ responsibility to clearly establish that their import

3 (...continued) questionnaire responses), see Mem. to The File from George McMahon & Jolanta Lawska, Senior International Trade Analysts, Office III, Antidumping and Countervailing Duty Operations, re: Verification of the Sales Response of Habas in the 2012-13 Antidumping Duty Investigation of Concrete Reinforcing Bar from Turkey (June 23, 2014) at 23 (“Habas SV Report”), see also Icdas Section A and C Supplemental Response at Ex. SC-14, but, they argue, the verification report offers no analysis of the decree, and accordingly, it does not explain the agency’s reasons for finding the KKDF tax to be “import-dependent and export contingent.” Further, the plaintiffs contend the only other apparent reference to the KKDF tax in Habas’ sales verification report indicates that the agency verifiers looked up KKDF rates on a website and matched these rates with the rates that Habas used in the drawback calculations it had presented to the agency. Pls’ Br. at 11, referencing Habas SV Report at 24. As the plaintiffs point out, however, the fact that standard KKDF rates obtained from websites tied to Habas’s own calculation of its duty drawback adjustment does not establish that the KKDF tax itself is “import dependent.” Here, the plaintiffs would again call attention to the fact that the decree itself indicates that the KKDF is not a tax on imports as such but rather is a tax on certain types of loans, apparently regardless of the reasons such loans are taken out, and the plaintiffs complain that neither the IDM nor the verification reports engage with this fact, or explain the agency’s rationale in discounting RTAC’s arguments based on it. They further argue that the discussion of the KKDF tax in the Icdas verification report is even more cursory, consisting only of a statement that the agency verifiers looked up KKDF rates on a website and matched these rates with the rates that Icdas used in the drawback calculations it had presented to the agency. See Mem. to File from Jolanta Lawska and George McMahon, Senior International Trade Analysts, Office III, Antidumping and Countervailing Duty Operations, re: Verification of the Sales Response of Icdas Celik Enerji Tersane ve Ulasim Sanayi A.S. (Icdas) in the 2012-13 Investigation of Steel Concrete Reinforcing Bar from Turkey (June 27, 2014) at 21 (“Icdas SV Report”) & reference to Ex. SVE-29, pp. 21-40. As such, the plaintiffs contend, there is still no explanation of the agency’s decision to treat the KKDF as “import-dependent and export contingent,” or otherwise as an “import duty.” Court No. 14-00268 Page 16

financing transactions would normally have been subject to the KKDF tax,4 and they again note that

the tax does not apply to imports per se but only to certain types of financing that may or may not

be used in conjunction with imports. In other words, they argue that the mere fact that imports

occurred does not suffice for finding that the KKDF tax was applicable, and thus it would be

improper to simply assume that KKDF taxes were rebated or uncollected on respondents’ import

financing, as the respondents were first required to show that the financing was of a type to incur the

tax in the first place.

Most importantly, the plaintiffs aver, nothing in the record shows that respondents’

import financing was such as to incur the tax. See Pls’ Reply at 6-8 (discussing lack of confirmation

in the record that respondents’ imports were financed in taxable ways). “If no tax was ever owed,

then it could not have either been rebated or foregone by reason of exports to the United States”, Pls’

Br. at 13, referencing RTAC Case Br. at 48-49, and they maintain that in the IDM, Commerce

neither addressed their arguments nor pointed to any factual information on the record indicating that

respondents in fact owed the KKDF tax on any financing associated with imports claimed under the

IPR as relating to U.S. sales of rebar, id. referencing IDM at 14.

The foregoing persuades that the IDM is lacking clear reasons for rejecting the points

the plaintiffs raise, and without such an explanation the basis for the agency’s determination to

include the KKDF tax as part of the respondents’ duty drawback adjustments cannot be sustained.

See, e.g., Altx, Inc. v. United States, 25 CIT 1100, 1103 (2001), citing United States v. Nova Scotia

4 See, e.g., Allied Tube v. United States, 25 CIT 23, 29 (2001) (“[a]s with all favorable adjustments to normal value or export price, respondent bears the burden of establishing both prongs of the test, and therefore, its entitlement to a duty drawback adjustment”). Court No. 14-00268 Page 17

Food Prods., 568 F.2d 240, 252 (2d Cir. 1977) (“[i]t is not in keeping with the rational [agency]

process to leave vital questions, raised by comments which are of cogent materiality, completely

unanswered”). On remand, Commerce is requested to either provide reasons therefor, or reconsider

inclusion of “foregone” KKDF taxes in any adjustment for duty drawback.

C. Selection of Invoice Date as Icdas’s U.S. Date of Sale

In calculating antidumping duty margins, Commerce matches U.S. and home market

sales to one another on the basis, inter alia, of the respondent’s “date of sale” in each market. Under

Commerce’s regulations, the date of invoicing is the presumptive date of sale unless the record

demonstrates otherwise. 19 C.F.R. §351.401(i).

Considering the record, Commerce initially concluded that the material terms of

Icdas’s U.S. sales were set before invoicing, i.e., as of the date of the contract or purchase order

(“PO”) or the last-amended contract or PO. See, e.g., PDoc 212 at 16-17. For the Final Results,

Commerce verified that Icdas invoiced and shipped material in accordance with the terms of the last

amended contract or PO, but it nonetheless determined that the material terms were not set until

invoicing. PDoc 297 at 30-32.

The plaintiffs argue that Commerce did not adequately explain its reasons for

rejecting its preliminary conclusion or adequately support its final conclusion, and that Commerce’s

and Icdas’s post hoc explanations are insufficient to remedy the problem and are further inconsistent

with the record. They contend, as an initial matter, that Commerce identified no invoices that

reflected material terms distinct from those in the last amended contract or PO. Pls’ Br. at 25-26;

see also PDoc 297 at 30-31. Certain POs examined at verification were unsigned, PDoc 297 at 30, Court No. 14-00268 Page 18

and thus the plaintiffs argue that the actual practice of the respondents and their counterparties does

not indicate that they viewed signatures as a necessary indicator of a “meeting of the minds.” See

Pls’ Br. at 26; see also Icdas SV Report at Exs. SVE-9, SVE-34--SVE-38. Likewise, the plaintiffs

contend, it is unclear why Commerce concluded that the issuance of POs or contracts “within days

of the invoice date” meant that those documents did not reflect final terms. See Pls’ Br. at 26-27.

Finally, the plaintiffs argue that the sole precedent cited by the agency involved meaningfully

different facts, in the form of a documented instance of an invoice that reflected different material

terms than the related contract, while the agency’s reasons for discounting the precedents offered by

the plaintiffs are unclear. Id. at 27- 28; PDoc 297 at 30-31.

Commerce defends its determination not by pointing to any instances in which

invoices reflected different terms than the last-amended contracts or POs, but by arguing that the

plaintiffs “are second-guessing Commerce’s exercise of its judgment and expertise on this issue.”

Def’s Resp. at 20. In particular, Commerce argues that it reasonably treated unsigned POs or

contracts as not reflecting final terms, even though no changes occurred between the last-amended

PO/contract and invoicing. Commerce concedes that there were no such changes, id. at 21, but it

characterizes this fact as immaterial given that certain POs/contracts lacked signatures, certain

POs/contracts were issued “within days of the invoice date,” and Icdas had no uniform method of

amending POs or contracts across customers. Id. at 21-22. Commerce also argues that its case

citation was apposite, while those cited by the plaintiffs were not. Id. at 22-23.

The court concludes that further explanation from Commerce would aid the record.

Although Commerce apparently concedes that no invoices reflected material terms of sale different Court No. 14-00268 Page 19

from those in Icdas’s last-amended POs/contracts, id. a 21, the Final Results do not address this fact.

See PDoc 297 at 30-31. Similarly, Commerce’s elucidation of its case citation and rejection of the

plaintiffs’ cited cases go beyond the explanation provided for the Final Results. See Def’s Resp. at

22-23. Commerce offers a new reason, also not provided in the Final Results, for finding that Icdas’s

material terms were not set until invoicing, to wit, that Icdas did not have a uniform practice or

method of amending contracts or POs. Id. at 20-21.

The plaintiffs’ contention is that Commerce’s explanation runs counter not only to

the record but also precedent. They point out that Commerce appears to take the position that

material terms of sale are not established unless they are unchangeable; in other words, even where

no change occurs between a particular date on which material terms of sale are memorialized and

the invoice date, if the terms were “subject” to change in the abstract, then the terms should be

considered set only as of the invoice date. See Def’s Resp. at20-23. That position was rejected in

both Nucor Corp. v. United States, 33 CIT 207, 256-57, 612 F. Supp. 2d 1264, 1306-07 (2009) and

Habas Sinai ve Tibbi Gazlar Istihsal Endustrisi A.S. v. United States, 33 CIT 695, 735, 625 F. Supp.

2d 1337, 1373 (2009), wherein certain RTAC members were plaintiffs or intervened and argued

what appears to have been the same position that Commerce now appears to take -- that terms were

established only once they were no longer subject to even theoretical change. Not only did the Nucor

and Habas courts reject that approach, but it also appears that Commerce has repeatedly rejected it

in the past, selecting a date as the date of sale despite the fact that material terms actually changed

afterwards (and thus must have been “subject to change”.5

5 See, e.g., Issues and decision memorandum accompanying Circular Welded Carbon Steel (continued...) Court No. 14-00268 Page 20

In previous matters, the salient question has been held to be that of when the parties

to the transaction intended the terms to be final. E.g., Habas, 33 CIT at 735, 625 F. Supp. 2d at 1373

(“the focus of an agency date of sale analysis is to determine when the contracting parties reached

a ‘meeting of the minds’ on the material terms of sale”). The plaintiffs argue that from the record

of Icdas’s sales it does not appear that Icdas and its customers intended to -- or ever did -- continue

to negotiate as to material terms through invoice date. Pls’ Reply at 13, referencing Pls’ Br., Att. 2

(summarizing record sales traces). They further contend that the rationale offered by Commerce --

that the precedents they cited involved sales contract transactions rather than POs and therefore were

inapposite to the determination here -- “makes no sense” and is post hoc in any event. Pls’ Reply

at 13-14, referencing Def’s Resp. at 22-23. The gist of Commerce’s argument, they contend, appears

to be that documents termed “contracts” are uniformly treated as such by contracting parties and

represent an unchangeable “meeting of the minds” notwithstanding the agency’s prior experience

with changes to contracts. See id. The plaintiffs argue that the actual practice of the parties, as

indicated in the record, shows otherwise. Id. at 14 (citation omitted).

5 (...continued) Pipes and Tubes From Thailand, 78 Fed. Reg. 65272 (Oct. 31, 2013) (final rev. results) at cmt 6 (finding contract date or amended contract date, as applicable, to be the date of sale); Issues and decision memorandum accompanying Sulfanilic Acid from Portugal, 67 Fed. Reg. 60219 (Sep. 25, 2002) (final inves. results) at cmt 1 (finding contract date to be date of sale despite contract renegotiation after certain production quantities could not be met); Circular Welded Non-Alloy Steel Pipe from the Republic of Korea, 63 Fed. Reg. 32833, 32836 (June 16, 1998) (final rev. results) (choosing contract date as date of sale despite subsequent changes including changes to material terms of sale); Issues and decision memorandum accompanying Certain Welded Steel Carbon Pipes and Tubes from Thailand, 65 Fed. Reg. 60910 (Oct. 13, 2000) at cmt 1 (final rev. results) (selecting contract date as date of sale despite quantity changes). Court No. 14-00268 Page 21

Icdas’s defense of Commerce’s date of sale determination focuses on a line from

Icdas’s sales verification report, stating that Commerce officials “verified Icdas’[s] claim that for

these sales there were quantity, price, and/or product specification change [sic] between the purchase

order/sales contract date and the date Icdas issued an invoice.” Icdas Resp. at 18, quoting Sales

Verification Report at 30, PDoc 272, CDoc 500. The plaintiffs contend that Commerce did not rely

on this statement, either in its determination or as further explanation in its briefs, and they maintain

that the verification report is not inconsistent with their arguments, because they agree that changes

occurred between the initial PO/contract and invoicing, but Commerce preliminarily found no

evidence of changes between the last amended PO/contract and invoicing. In other words, whatever

the validity of Icdas’s summarization of the record evidence and the conclusions it draws therefrom,

it veers into post hoc territory that forms no basis for sustaining this determination as it stands.

In view of the foregoing and Commerce’s seeming departure from its own (and

judicially-confirmed) precedent, the matter also needs to be remanded at least for further explanation,

as argued by the plaintiffs in their briefs to the court, or for reconsideration, at Commerce’s

discretion, of the determination on the date of sale for Icdas’s U.S. sales.

D. Icdas’s Yield Strength Coding

Yield strength is a physical characteristic of rebar attributable to carbon equivalency

and was among the factors that Commerce included in its model match criteria. Commerce’s

questionnaire asked the respondents to assign one of six numbered codes to each of their models of

rebar and also categorize according to yield strength based on psi; in the case of two of those codes,

by “maximum specified carbon equivalency [less than or equal to] 0.55%”. See Def’s Resp. at 24. Court No. 14-00268 Page 22

Rebar with a carbon equivalency over 0.55% cannot be welded, and the plaintiffs emphasize

repeatedly that Turkish standards require weldability, a point that Commerce apparently downplayed

or overlooked.6

Both Icdas and Habas initially reported home market sales of certain grades of rebar

as having a carbon equivalency over 0.55%.7 For the Preliminary Determination, Commerce

adjusted this reporting on the basis that rebar with such a carbon equivalency would not meet

weldability requirements. Commerce later collected mill test certificates (“MTCs”) from Habas8 but

not from Icdas. In the Final Results, Commerce reasoned that it lacked sufficient record evidence

to reject Icdas’s reporting and it accepted Icdas’s yield-strength coding as originally submitted. IDM

at 20.

The plaintiffs argue that Commerce’s determination was not adequately explained.

See Pls’ Br. at 29-34. They claim that Commerce’s response goes “well beyond” what was provided

at the administrative proceeding to essentially conclude that the lack of Icdas MTCs on the record

is determinative.9 The plaintiffs contend it is important to note here that the agency has not stated

that anything on the record suggests that Icdas’s rebar actually had carbon equivalencies greater than

0.55%, cf. Def’s Resp. at 23-39, and they note that while Commerce now claims that the relevant

6 See PDoc 217, CDoc 314, at 2-3; PDocs 136-37, CDocs 158-59, at 7-8 and Ex. 2; PDoc 33 at 5-8. 7 See PDoc 223, CDoc 331, at 4; PDoc 217, CDoc 314, at 2-3. 8 See PDoc 297 at 20; see also PDoc 300, CDoc 512, at 2-3. 9 Cf. Def’s Resp. at 23-39, with PDoc 297 at 20, CDoc 512, PDoc 300 at 2-3, CDoc 520, PDoc 303 at 2-3. Court No. 14-00268 Page 23

standards could potentially include such products, id. at 26, 28, Commerce has not stated whether

it is now finding that there is no requirement that rebar sold in Turkey be weldable, which would

seem to be the inference towards which Commerce’s determination would otherwise point. Further,

the plaintiffs argue it appears Commerce is only just now revealing that it accepted Habas’s

originally-reported yield-strength coding with respect to a particular grade of rebar, and they claim

this has taken them by surprise as an unexplained or unannounced change in the Final Results as

compared with the Preliminary Determination, and is an issue they claim they might have chosen

to appeal had they been apprised of it. See Pls’ Reply at 17 n.10.

Commerce admits that it addressed Habas’s data by collecting MTCs and did not

collect any MTCs from Icdas.10 Def’s Resp. at 26-27. The plaintiffs argue that the absence of Icdas

MTCs appears to have led Commerce to accept Icdas’s original reporting despite the agency’s

subsequent conclusion that the Turkish grade standards for certain products are ambiguous. See id.

at 27-28. Given that preliminary misgiving, the plaintiffs argue it is not clear that the agency acted

reasonably in declining to collect MTCs from Icdas.

For its part, Icdas argues that its reporting was consistent with Field 3.2 of the

agency’s Section C questionnaire and that the record does not show that weldability is a requirement

for rebar sold in the Turkish market. Icdas Resp. at 23-27. Whether those points are true, they go

beyond what the IDM articulates. Cf. id., with PDoc 297 at 20, CDoc 512, PDoc 300 at 2-3.

10 While Commerce depends on respondents’ cooperation to develop the record, the agency has its own obligation to attempt to ensure that the record allows for accurate calculations. For example, this court recently remanded Commerce’s use of a simple average in calculating an all-others rate, finding that the lack of public data necessary for weight-averaging was due to the agency’s failure to request it or enforce its regulatory requirements that respondents provide it. MacLean-Fogg Co. v. United States, 39 CIT ___, Slip Op. 15-85 (2015). Court No. 14-00268 Page 24

Regardless, the plaintiffs contend that Icdas’s points do not excuse or establish the reasonableness

of the agency’s decision not to collect MTCs from Icdas in the first place, either by following up on

its initial request or by requesting them at verification.

The agency’s determination on this issue, and the restatement(s) thereof (and

arguments thereon) among the papers here, are fairly muddled.11 Given Commerce’s realization, at

some point, of no theoretical maximum yield strength/carbon equivalency levels among the Turkish

standards for the particular products about which the plaintiffs complain, as reported by the

respondents, but which reporting by Habas Commerce determined to have been inaccurate in fact

(at least in part), based upon MTCs for the product verified with respect to Habas, the fact that the

record does not at present contain specific production data for Icdas by which to test Icdas’s reported

coding of the yield strength/carbon equivalency for the grades of rebar in question does not

reasonably lead to the inference that those grades were accurately coded (although they may well

have been), particularly given the seemingly unsettled question of Turkish weldability requirements,

either in general or (preferably) with respect to the grades of rebar in question. In view of the

foregoing, therefore, Commerce is requested on remand either to reopen the record and obtain

11 Just as an example, the defendant’s response brief states at page 25 that Habas originally reported one grade of rebar as a type with a maximum specified carbon equivalence of 0.55%, and then at page 26 states that Habas had originally reported the same grade of rebar as an “other” type, i.e., with no maximum specified carbon equivalency level; on page 25 again, the brief states that the record establishes that this same grade of rebar does not have a maximum specified carbon equivalency level based on the official Turkish specification standard, whereas in its preliminary analysis memorandum for Habas it stated that Habas had provided information in a supplemental response that “shows that [the particular grade] of specification TSE708 have a maximum carbon equivalency level of 0.50%.” See generally Def’s Resp. at 23-29; Mem. to The File from George McMahon, Case Analyst, AD/CVD Operations, Office III, re: Less-Than-Fair-Value Investigation of Steel Concrete Reinforcing Bar from Turkey: Preliminary Determination Analysis Memorandum for Habas Sinai ve Tibbi Gazlar Istihsal Endustrisi (Apr. 18, 2014), PDoc 217, CDoc 314, at 3. Court No. 14-00268 Page 25

relevant MTCs from Icdas as it did from Habas, or to reconsider and/or explain the reasonableness

of resorting to what appears to be an uncorroborated “facts available” determination under 19 U.S.C.

§ 1677e(a)(1) in this instance.

E. Air-Cooled Rebar and Nonadjustment of Icdas’s Reported Costs

The characteristics of rebar are affected by the cooling process. E.g., PDoc 203,

CDoc 306, at 8-15; PDoc 188, CDoc 229, at 2 & Ex. S3D-1. Water cooling imparts a strength to

rebar that air-cooled rebar lacks, and thus additional alloying elements must be added to air-cooled

rebar to equilibrate strength. See id. While water-cooled rebar is accepted in Turkey, it is not

generally accepted in imports to the U.S. market; as such, additional alloys must be incorporated into

U.S. bound product. See, e.g., Icdas Br. at 27-28.

Commerce’s model-match criteria did not incorporate cooling method. See PDoc 212

at 9. Commerce did, however, seek information from respondents on the alloy cost differences

attributable to water- and air-cooled product. See Pls’ Br. at 35-36 (citations omitted). Habas

provided such information, PDoc 110, CDocs 54-55, at D-37, but Icdas stated that it did not track

such costs, and that while it might be able to construct them it was not worth doing so. PDocs 191-

97, CDocs 237-48, at 38-40 & Ex. SD-54; PDocs 252-57, CDocs 369-86, at 2SD-9--2SD-15. RTAC

then argued to Commerce that it should adjust Icdas’s cost reporting to reflect the greater alloy usage

in air-cooled rebar. PDoc 283, CDoc 504, at 17-21. Commerce declined, stating that Icdas’s cost

reporting was based on actual, product-specific costs, and thus already reflected all cost differences

for water- and air-cooled rebar. PDoc 297 at 35; CDocs 474-492, 502, PDoc 269, 276 (“Icdas CV

Report”) at 4, 6-7, 15-17 & Exs. CVE-6- CVD-9. Court No. 14-00268 Page 26

The plaintiffs here contend that Commerce’s explanation is inadequate, given that

Icdas itself had stated that it did not track or record alloy usage or costs, and that the verification

materials Commerce cited did not support the conclusion that Icdas tracked actual, product-specific,

alloy costs. Pls’ Br. at 37-39. They contend Commerce’s response indicates that at verification it

found Icdas’s repeated statements that it did not track product-specific alloy usage or cost essentially

not correct, see Def’s Resp. at 31-32, and therefore the plaintiffs argue that if this is true, it would

indicate that Icdas “repeatedly stymied” valid agency requests for information to which the company

had easy access. See Pls’ Br. at 37-39. The court is unclear as to whether the record merits such

construction, but in their reply the plaintiffs also argue it is not clear whether that is in fact true,

pointing out that while Commerce states that it verified that “the amount of alloy . . . is not averaged

over all products,” Def’s Resp. at 31, this finding is not reflected in the IDM, PDoc 297 at 35, nor

is it apparent from the verification report or relevant exhibits thereto. Pls’ Reply at 19,

referencing Icdas CV Report at 13-17 and Exs. CVE-6--CVE9, CVE-11.

The bottom line, the plaintiffs argue, is that Commerce is now glossing over the

report, with references to specific pages of lengthy exhibits that, in the report itself, are referenced

only in toto, see Def’s Resp. at 32, and that these specific pages are untranslated and largely illegible.

Pls’ Reply at 20, referencing Tab I to Conf. Appx to Def’s Resp. As such, they contend, the

defendant’s arguments are again post hoc and do not show that these pages “demonstrate the specific

quantity of each alloy used in the production of each specific billet used to produce the rebar

included in each of the selected CONNUMS”, Def’s Resp. at 32, in any event. Court No. 14-00268 Page 27

With regard to Commerce’s statement that Icdas requires only a certain discernible

modicum of additional alloys per ton of air-cooled rebar, Def’s Resp. at 34, the plaintiffs reply that

this is more post hoc explanation, as Commerce did not address their arguments regarding the value

of such costs during the administrative proceeding, and they further contend that the figure is only

an estimate that Icdas prepared with respect to just one of the main alloying elements. Cf. id. with

CDoc 237-248, PDoc 191-197 at 39 (stating that the figure is an estimate of what additional

manganese cost “could” be) & Ex. SD-54; see also Icdas Br. at 28 (stating that this figure is “not

actual”). They also note that to the extent Commerce defends its decision on the basis that it

correctly accepted Icdas’s yield strength coding, that issue is unsettled. See supra.

Icdas, for its part, argues that the plaintiffs conceded to the International Trade

Commission (“ITC”), in a sunset review involving rebar from countries other than Turkey, that cost

differences between air- and water-cooled rebar are not significant. Icdas Br. at 30-31 and Ex. 1. To

which the plaintiffs reply that the statements of a certain company’s official regarding the possibility

of both water- and air-cooled product conforming to ASTM standards does not support Icdas’s point,

that such statements are the views of one person, that such statements are not of record here, and that

such statements are not compelling in any event as against actual record data from a U.S. producer

demonstrating a more significant cost difference than the estimated modicum quoted by Commerce.

Pls’ Reply at 20, referencing PDoc 203, CDoc 306, at Ex. 3. It is for Commerce, however, to sort

out such matters on remand.

Icdas lastly argues that Commerce correctly found that it tracked costs with the

required level of specificity. The plaintiffs point out, however, that Icdas confirmed “allocated” Court No. 14-00268 Page 28

costs based on a testing of “each type of billet,” rather than individually measuring alloys in “each

specific billet used to produce the rebar” at issue here, as the defendant now concludes. Pls’ Reply

at 20, referencing Icdas Br. at 30, 31. The plaintiffs thus contend that “Icdas and Commerce still

cannot agree as to what Icdas’s accounting records contain, or how the company records costs.” Id.

In light of such confusion, with a view towards addressing all the concerns expressed

above, it is appropriate that this issue also be remanded, along with the others, for further explanation

or reconsideration of the support for Commerce’s determination in the first instance.

IV. Conclusion

In view of the foregoing, the case shall be, and hereby is, remanded for further

proceedings not inconsistent with this decision.

The results of remand shall be due February 22, 2016. After the results of remand

are filed with the court, the parties shall confer and file a joint status report within five days of that

filing to propose dates for filing comments and/or concerning any other matters needing the

assistance of the court.

So ordered.

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

Dated: November 23, 2015 New York, New York