American Tubular Products, LLC et al v. United States

Slip Op. 15-
UNITED STATES COURT OF INTERNATIONAL TRADE
OPINION
[The court sustains the remand results of an administrative review of an antidumping duty order
on oil country tubular goods from the People’s Republic of China.]
Dated:
Julie C. Mendoza, Donald B. Cameron, R. Will Planert, Brady W. Mills, Mary S.
Hodgins, and Sarah S. Sprinkle, Morris, Manning & Martin LLP, of Washington, DC, for
plaintiffs.
L. Misha Preheim, Senior Trial Counsel, Commercial Litigation Branch, Civil Division,
U.S. Department of Justice, of Washington, DC, for defendant. With him on the brief were
Jeanne E. Davidson, Director, and Claudia Burke, Assistant Director. Of counsel on the brief
was Whitney Rolig, Office of the Chief Counsel for Trade Enforcement & Compliance, U.S.
Department of Commerce, of Washington, DC.
Robert E. Lighthizer, Jeffrey D. Gerrish, and Luke A. Meisner, Skadden, Arps, Slate,
Meagher & Flom LLP, of Washington, DC, for defendant-intervenor United States Steel
Corporation.
AMERICAN TUBULAR PRODUCTS, LLC,
and JIANGSU CHENGDE STEEL TUBE
SHARE CO., LTD.,
Plaintiffs,
v.
UNITED STATES,
Defendant,
and
UNITED STATES STEEL CORPORATION,
TMK IPSCO, WHEATLAND TUBE
COMPANY, and V&M STAR L.P.,
Defendant-Intervenors.
Before: Richard W. Goldberg, Senior Judge
Court No. 13-00029
PUBLIC VERSION
98
August 28, 2015
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 1 of 21

Court No. 13-00029 	Page 2
Goldberg, Senior Judge: In its previous opinion, the court invalidated the final results of
WKHí11 review of the antidumping order on certain oil country tubular goods (“OCTG”)
from the People’s Republic of China (“PRC”). See Certain Oil Country Tubular Goods from the
People’s Republic of China, 77 Fed. Reg. 74,644 (Dep’t Commerce Dec. 17, 2012) (final admin.
review) (“Final Results”). The court remanded two decisions for the Department of Commerce
(“Commerce” or “the agency”) to reconsider: (1) the choice to value steel billet, an input used to
make OCTG, as alloy steel, and (2) the decision to use Indonesian surrogate values as stand-in
prices for carbon steel billet. See Am. Tubular Prods., LLC v. United States, Slip Op 14-116,
2014 WL 4977626, at *17 (CIT Sept. 26, 2014).
On remand, the agency revised the former decision, but not the latter. First, Commerce
found that it lacked the evidence necessary to value most of the billet as alloy steel. It opted
instead to value a majority of billet using a simple average of alloy and carbon steel prices. See
Final Results of Redetermination Pursuant to Remand í(&)1R-1 (“Remand Results”).
Second, the agency held that its original surrogate values for carbon steel billet were reasonable.
Id. DWí
Plaintiffs Jiangsu Chengde Steel Tube Share Co. (“Chengde”) and American Tubular
Products, LLC (“ATP”) (collectively “Plaintiffs”) now challenge both remand decisions as
unsubstantiated in evidence and contrary to law. Defendant-intervenor the United States Steel
Corporation (“U.S. Steel”) also lodges a challenge, but only regarding the decision to value the
billet using a simple average of carbon and alloy steel prices. Yet none of these objections have
traction. The court finds that Commerce’s determinations on remand were grounded in the
record and accorded with law. The court sustains the Remand Results in all their particulars.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 2 of 21

Court No. 13-00029 	Page 3
BACKGROUND
Here, as before, the discussion focuses on the process used to estimate normal value
(“NV”) for nonmarket economy merchandise. When Commerce calculates NV for goods from
China or other nonmarket countries, it starts by selecting artificial market prices or surrogate
values for each input consumed in production. See 19 U.S.C. § 1677b(c)(1) (2012). The statute
requires that surrogates be made of the “best available information” on the record. Id. And by
regulation, Commerce will normally value each input using data from a single market economy
at a level of development comparable to country under review. See 19 C.F.R. § 351.408(c)(2)
(2015). The agency then adds up the surrogate values for each input—along with amounts for
general expenses, profit, and other costs—to form NV.
Commerce was supposed to follow these rules as it created surrogates for Chengde’s steel
billet. The agency strayed from the path during the review below, but it made appropriate course
corrections on remand. The sections that follow outline the agency’s original decision, the
court’s critique of that decision, and the revised results after reconsideration.
I. 	The Administrative Proceeding
In its prior opinion, the court outlined relevant record data and explained how Commerce
chose surrogates for billet. The explanation was comprehensive and included two charts that
may be of use to those with access to the confidential docket. See Confidential Slip Op. 7, 10,
ECF No. 94 (“Conf. Slip Op.”); see also Am. Tubular Prods.:/DW íFor
the rest, the court repeats the key developments that shaped the agency’s surrogate value
decisions.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 3 of 21

Court No. 13-00029 	Page 4
A. The Agency Determines the Type of Billet Consumed
At the review’s outset, Commerce asked Chengde to identify the ingredients it used to
make OCTG. Chengde responded that it consumed steel billet, among other things, and it named
Harmonized Tariff Schedule (“HTS”) 7224.90.0075 as the proper tariff subheading for the input.
Chengde Resp. to Sections C&D Questionnaire (“C&D Resp.”) at Ex. D-&',9í 1RY
17, 2011). Subheading 7224.90 covers semifinished products of alloy steel, not carbon steel.
The agency later issued a supplemental questionnaire to gather more information about
Chengde’s billet. The appeal for data was meticulous in its detail. To begin, Commerce asked
“for a complete technical description” of each input, including “chemical specifications, purity,
grades/standards, and mineral/metal content.” First Suppl. Questionnaire (“First Suppl. Q.”) at 6,
CD IV 30 (Jan. 6, 2012). The agency then solicited product grades where applicable, complete
specifications for each grade reported, and purchase contracts, supplier’s invoices, packing lists,
and certificates of assay. Id. The questions were calibrated to establish the chemical makeup of
the billet beyond the basic description in the initial response.
Chengde’s answers were less than complete, however. For a technical description of the
billet, Chengde offered only that its inputs complied with standards from the American Society
of Mechanical Engineers (“ASME”). See First Suppl. Resp. (“First Suppl. Resp.”) at Ex. S1-15
(S1-4)&',9í -DQ . The ASME standards gave a range of chemical content
for compliant billet, but they did not confirm whether the billet was carbon or alloy steel. See
Resp. to Section A Questionnaire at Ex. A-&',9í 2FW Chengde also
furnished a few purchase contracts from a billet supplier. The contracts said the billet had to
conform to the seller’s “Product Quality Certificate” or the “technical agreement between [the]
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 4 of 21

Court No. 13-00029 	Page 5
two parties,” but neither of these side items were given to the agency. First Suppl. Resp. at Ex.
S1-16 (S1-5). The first supplemental response added little, if anything, to the initial reply.
Commerce sent a second round of supplemental questions shortly thereafter, but the line
of inquiry differed from the first round. In an earlier response, Chengde had given the agency a
few of its OCTG sales contracts. The contracts mentioned that Chengde would send ATP
“product quality certificates” and “mill test reports” when it shipped the goods. Id. at Ex. S1-5.
In the second supplemental questionnaire, Commerce asked Chengde to submit such certificates
for each unique OCTG product sold. 1 The agency wrote, in its own words,
Please submit sample product quality certificates and mill test reports/certificates
for all control numbers (“CONNUMS”) sold during the [review period]. Submit
one product quality certificate and one mill test report for each CONNUM for
each month during the [review period] in which that CONNUM was produced.
Second Suppl. Questionnaire (“Second Suppl. Q.”) at 4, CD IV 47 (Feb. 29, 2012). So unlike
the inquiry in the first supplemental questionnaire, the request in the second supplemental
questionnaire did not focus on the chemical makeup of Chengde’s billet. Instead the questions
centered on the OCTG sales contracts and Chengde’s promise to deliver mill certificates to ATP.
In response, Chengde furnished the first page of ten mill test reports, but for only six of
the nine CONNUMs sold during the review period. It was also unclear whether Chengde had
submitted a report for each CONNUM for each month in which the CONNUMs were produced.
See Second Suppl. Resp. (“Second Suppl. Resp.”) at Exs. S2-13, S2-&',9í 0DU
2012). Even so, contrary to the respondent’s initial declaration, the mill certificates showed that
the tube tested therein was made of high or low carbon steel, not alloy steel. High carbon steel
1 The agency identifies distinct products by assigning each a control number, or “CONNUM.” A
CONNUM is a set of numeric digits representing “a hierarchy of specified physical characteristics” that may vary
from one proceeding to the next. Union Steel v. United States, 36 CIT __, __, 823 F. Supp. 2d 1346, 1349 (2012).
“All products whose product hierarchy characteristics are identical are deemed to be part of the same CONNUM and
are regarded as ‘identical’ merchandise” for the purpose of comparing export prices to NV. Id.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 5 of 21

Court No. 13-00029 	Page 6
has a carbon content of 0.25% or more by weight, and low carbon steel has a carbon content of
less than 0.25% by weight. See HTS íPlaintiffs urged the agency to value all of
Chengde’s billet as carbon steel in light of the mill reports. ATP Revised Case Br. DWí&'
IV 73 (Aug. 3, 2012); Chengde Revised Case Br. DWí3',, $XJ. 2, 2012).
The agency wove these threads of data into surrogate values for billet in the Final
Results. As a first step, Commerce had to decide whether the inputs were alloy or carbon steel.
It relied chiefly on Chengde’s declaration in the initial response that the billet was alloy. See
Issues & Decision Mem. (“I&D Mem.” DWí, PD II 164 (Dec. 6, 2012). The agency also
credited a website which suggested that some of Chengde’s pipe was alloy metal. This evidence
led Commerce to conclude that [[ 	]] of the billet was alloy steel. See &RQI6OLS2Sí
Nevertheless, because billet bears the same chemical signature as finished OCTG—and because
the OCTG sampled in the mill reports was carbon steel—Commerce deemed that [[ 	]] of
the billet was carbon steel. Id. By doing so, the agency rejected Plaintiffs’ argument that all of
the billet was carbon metal, and that Chengde reported using alloy billet in error. Yet Commerce
never said why the mill reports failed to establish the chemical makeup of billet that had not been
tested. The omission seemed odd, because Chengde’s mill reports sampled OCTG from six of
the nine CONNUMs sold during the review period. Presumably, the chemical signature of
OCTG that had been tested would be the same as that of untested yet commercially identical
merchandise.
B. The Agency Selects Surrogate Values for Carbon Steel
Next, after deciding the chemical makeup of Chengde’s billet, Commerce had to forge a
surrogate value for each type of billet consumed. For carbon steel billet, the agency crafted a
composite surrogate using prices for low and high carbon steel. Commerce drew price data from
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 6 of 21

Court No. 13-00029 	Page 7
2011 Indonesian import statistics in the Global Trade Atlas (“GTA”). Id. at 10. The GTA listed
all imports of high and low carbon steel products to Indonesia by country, quantity, and price.
With this information in hand, the agency proceeded to exclude any data from nonmarket
economies, from countries with generally available export subsidies, and without quantity or
price terms. See Final Analysis Mem., CD IV 80 (Dec. 5, 2012). Commerce also excluded
import prices that it deemed aberrational. For low carbon steel, this meant the agency deleted
data from six countries, which each had low import quantities and average unit values (“AUVs”)
more than five times the highest remaining import price. For high carbon steel, by contrast,
Commerce excluded only one aberrant value. The omitted value comprised imports from
Malaysia, which exceeded the next highest import value by 2.4 times. With these data scrubbed
from the set, Commerce rendered a weighted AUV of $566.64 per metric ton (“MT”) for low
carbon steel. The weighted AUV for high carbon steel was $1,149.40/MT. Together, the
weighted average surrogate value for carbon steel was $813.86/MT.
Finally, to check whether the AUVs were reasonable, Commerce compared its surrogates
to “benchmarks,” or average import prices from countries economically similar to the PRC. For
low carbon steel, Commerce found the Indonesian surrogate fell between benchmarks from the
Philippines ($1740.68/MT), South Africa ($1270.96/MT), and Ukraine ($600.57/MT). I&D
Mem. at 11. Furthermore, though the Indonesian value for high carbon steel exceeded those
from Thailand ($567.55/MT) and Ukraine ($653.84/MT), the agency said the surrogate was not
so high as to be “distortive or misrepresentative.” Id. Hence Commerce used the Indonesian
data to value both low and high carbon steel over Plaintiffs’ objections.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 7 of 21

Court No. 13-00029 	Page 8
II. 	The Court Case
On appeal, Plaintiffs contested the decision to value most of Chengde’s billet as alloy
steel. Pls.’ Mot. for J. on Agency R. í, ECF No. 39-1 (“Pls.’ Br.”). They also argued that
the Indonesian surrogate for high carbon steel was aberrantly high. Id. at í.
The United States, for its part, requested a voluntary remand to decide if its carbon steel
surrogates were substantiated in evidence. Def.’s Resp. to Pls.’ Rule 56.2 Mot. for J. on Agency
R. 19, ECF No. 67 (“Gov’t Br.”). In the I&D Memo, the agency used petitioner’s data as
benchmarks, but it forgot to delete prices from nonmarket economies, from countries with
generally available export subsidies, and with missing quantities or values. See Remand Results
í. Commerce wished to redo the analysis to make sure its carbon steel surrogates looked
reasonable compared to properly manicured benchmarks.
The court granted a remand on both issues. To begin, the court scrutinized how the
agency determined the chemical makeup of Chengde’s billet. It divided the analysis by each
OCTG contract sold during the review period. 2 For instance, the court sustained the conclusion
that OCTG sold under contracts [[ 	]] was made of alloy steel billet. None of
the mill reports tested the tube sold in these contracts, but a website confirmed that the goods
were alloy. See Am. Tubular Prods., 2014 WL 4977626, at *6. The court also sustained the
decision that the OCTG sampled in the mill certificates was carbon steel. Id. But the court
reversed the determination that billet used to make OCTG in contract [[ 	]] was alloy steel.
A U.S. Customs and Border Protection (“CBP”) entry summary said this OCTG was carbon
2 Please see the court’s prior confidential opinion at Table 1 for a concise description of the thirteen sales
contracts exchanged during the review period. Table 1 lists each contract number, the model of OCTG sold under
the contract, the CONNUM sold under the contract, the amount of OCTG from the contract sampled in the mill
reports, and the total amount of OCTG sold under the contract. See Conf. Slip Op. 7. Unfortunately, Table 1 as
rendered in the public version is less useful owing to redactions.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 8 of 21

Court No. 13-00029 	Page 9
steel, so the court made the agency reassess the makeup of these billets in light of the evidence.
Id. at *7. The court also asked Commerce to consider whether the mill reports established the
chemical properties of OCTG that had not been tested. In their brief, Plaintiffs noted that OCTG
sampled in the mill reports represented six of the nine CONNUMs and ten of the thirteen
contracts sold during the review period. See Pls.’ Br. 12. One might assume, then, that sampled
OCTG bore the same chemical traits as unsampled OCTG in the same CONNUM and contract.
The court remanded so the agency could explain whether this assumption was correct. 3
Commerce would then recalculate the percentage of Chengde’s billet that was alloy or carbon
steel based on its assessment. See Am. Tubular Prods., 2014 WL 4977626, at *7.
The court also remanded the high carbon steel surrogate issue at the agency’s request.
Commerce was ordered to reconsider whether the Indonesian surrogates were “the best available
information on the record compared to other carbon steel billet surrogate data.” Id. at *9.
III. 	The Remand Results
On remand, the agency tackled both issues that the court returned for reconsideration.
The effort yielded major changes in the way Commerce viewed the balance between carbon and
alloy steel billet. As an initial matter, the agency continued to find that OCTG sold in contracts
[[ 	]] was alloy steel. See 5HPDQG5HVXOWVí,WDOVRconceded that the
OCTG sampled in the mill reports was carbon steel. See id. at 5. But in a departure from its
previous position, Commerce decided that the OCTG sold in contract [[ 	]] was carbon
steel. The agency cited the CBP entry summary in support of the change. Id. at 8.
Commerce also took a fresh look at the OCTG not tested in the mill certificates. Though
the Final Results held that untested OCTG was alloy steel, the agency now said it could not
3 Commerce had to address the mill test issue for OCTG sold in contracts [[
]]. Am. Tubular Prods., 2014 WL 4977626, at *7.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 9 of 21

Court No. 13-00029 	Page 10
determine the chemical traits of unsampled OCTG with the proof at hand. While some data
suggested that the billet was alloy—e.g., Chengde’s initial questionnaire response and the
website—the evidence did not stretch to cover all of the unsampled OCTG. The mill tests
debunked the notion that all the OCTG was alloy, and the website only established the status of
pipe sold in contracts [[ 	]]. See id. at í&RQYHUVHO\, though the mill
tests showed that some of the billet was carbon steel, they did not prove that all the unsampled
billet was carbon. Commerce asked for a direct account of the billet’s chemical composition in
the first supplemental questionnaire, but Chengde never supplied one. And the mill reports,
which did not cover every CONNUM in the month produced, failed to explain how the tests
were run or whether they represented the chemical properties of untested merchandise. See id. at
í*LYHQWKHVH ambiguities, Commerce chose to value the untested billet using a simple
average of the alloy and carbon steel surrogates. Id. at 9.
Finally, Commerce decided if its surrogates for carbon steel billet were reasonable. It
first purged the datasets of information from nonmarket economies, from countries with
generally available export subsidies, and with missing quantities or values. Id. at 10. Then it
compared the two carbon steel surrogates to benchmarks. For low carbon steel, the agency
found that the surrogate fell just below the least expensive comparable benchmark. For high
carbon steel, the surrogate landed in the middle of its class—above the Thailand at $567.55/MT
and below the Philippines at $2211.14/MT. Commerce inferred from these comparisons that
neither of its carbon steel surrogates were aberrational. Id. at 11.
After making its revisions on remand, Commerce unveiled a new, 137.62% weighted-
average dumping margin for Chengde’s OCTG. Id. at 24. This was a 20.24% reduction from
the 172.54% rate assigned in the Final Results. See Final Results at 74,645.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 10 of 21

Court No. 13-00029 	Page 11
DISCUSSION
Now in their remand comments, Plaintiffs and U.S. Steel contest the decision to use an
alloy-carbon average as a surrogate for some of Chengde’s billet. Plaintiffs also challenge the
agency’s reliance on Indonesian import data to represent high carbon steel. But none of these
arguments withstand scrutiny. On remand, the agency took sensible positions that were firmly
rooted in substantial evidence. See 19 U.S.C. § 1516a(b)(1)(B)(i). The court thus sustains them.
I. 	Commerce Reasonably Determined the Composition of Steel Billet
To begin, the court examines whether the agency properly decided the chemical makeup
of Chengde’s billet. But before diving into this analysis, we first name aspects of the remand
decision that the parties do not contest. Neither Plaintiffs nor U.S. Steel challenges the finding
that tube in contracts [[ 	]] was made of alloy billet. And neither dispute that
the OCTG in contract [[ 	]] was carbon metal. See Pls.’ Comments on Commerce’s Draft
Results of Redetermination Pursuant to Remand 2 n.1, ECF No. 106 (“Pls.’ Cmts.”); Comments
of U.S. Steel CRUSRQ)LQDO5HVXOWVRI5HGHWHUPLQDWLRQ3XUVXDQWWR&W5HPDQGí(&)1R
104 (“U.S. Steel Cmts.”). U.S. Steel hints that OCTG tested in the mill reports was not really
carbon steel, but it did not raise this issue as a legal claim. See U.S. Steel Cmts. 9. That leaves
one key item for the court to review: the choice to value billet from untested OCTG using a
simple average of alloy and carbon steel prices.
On that score, Plaintiffs take an all-or-nothing stance. They argue now, as before, that all
OCTG not proven to be alloy was made of carbon steel. By requesting only sample mill reports
in the second supplemental questionnaire, Commerce implied that sample data could prove the
chemical content of the billet, tested or not. Because each report analyzed OCTG from just one
contract—and because each contract sold only one CONNUM—Plaintiffs argue that the mill
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 11 of 21

Court No. 13-00029 	Page 12
reports established the chemical traits of all OCTG sold under the contracts with mill reports.
See 3OV¶&PWVí,QWKHLUYLHZ Commerce failed to rebut this argument on remand.
U.S. Steel also takes an extreme position. Though it agrees with Commerce that the mill
reports do not represent goods not specifically tested, U.S. Steel goes a step further: It insists
that any billet not specifically tested was alloy steel. As proof, it points to Chengde’s initial
statement that its billet was alloy—a statement that went unchallenged until the case brief. It
also highlights Chengde’s failure to submit mill reports for contracts that later proved to sell
alloy steel tube866WHHO&PWVí)URPWKHVHWZRIDFWVU.S. Steel infers “that Chengde
selectively submitted some mill certificates showing it used carbon steel billets but withheld mill
certificates that would have shown it used alloy steel billets.” Id. at 12. And the only reason
Chengde would have submitted certificates selectively, U.S. Steel opines, is if “the merchandise
not covered by the mill certificates . . . was produced using alloy steel billets.” Id.
The court can endorse neither this argument nor the Plaintiffs’. In its previous opinion,
the court asked Commerce to “explain whether Chengde’s mill certificates prove the chemical
properties of OCTG not specifically tested in those certificates.” Am. Tubular Prods., 2014 WL
4977626, at *7. Commerce answered with an unequivocal “no” on remand, and the court finds
the agency adequately explained its reasoning. The court also finds U.S. Steel’s argument too
speculative to support valuing the billet used to make the unsampled tube as alloy steel.
The reasonability of the agency’s redetermination comes into focus by breaking down the
task on remand. At its core, the court’s previous opinion asked Commerce to confirm or reject
two assumptions underpinning Plaintiffs’ argument. The first is that OCTG in same CONNUM,
or sold in same contract, bears a uniform chemical makeup. The second is that the chemical
makeup was accurately depicted in the mill reports. If all merchandise sold under a contract or
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 12 of 21

Court No. 13-00029 	Page 13
CONNUM had the same chemical signature, and if the mill reports showed that chemical
signature was carbon steel, then all merchandise in that CONNUM or contract must be carbon
steel. To fulfill the court’s order, Commerce had to explain whether these assumptions were
supported by record evidence.
The agency did precisely that on remand. First, it found no evidence to confirm that
merchandise sold in same contract or CONNUM bore a uniform chemical signature. It observed
that the billet used to make OCTG for each contract came from a handful of suppliers.
Commerce also noted that tube sold in the same contract was made in multiple production runs
or “heats.” This suggested that pipe in a single contract could have different chemical properties.
See Remand Results 15. Furthermore, as stated in response to the remand comments, none of the
CONNUMs used to identify unique tubular products included digits to represent chemical
composition. Def.’s Resp. to Comments on Remand Redetermination 6 n.4, ECF No. 118. So
even if merchandise in a single CONNUM was identical in some respects, the goods were not
necessarily identical in chemical makeup. Of course, this evidence does not prove that each
contract and CONNUM contained both alloy and carbon goods. But it shows there is no factual
basis to assume that goods in the same CONNUM or contract were chemically equivalent.
Commerce also deconstructed Plaintiffs’ second assumption: that the mill reports
accurately represented the chemical traits of untested OCTG. In general, for a sample to
represent the attributes of an object or objects, it must possess “the desired properties” of the
object, and be of sufficient quantity to be analyzed. G. Kateman & L. Buydens, Quality Control
in Analytical Chemistry 18 (2d ed. 1993). But as Commerce observed, the mill reports gave no
context or description of the procedures used to run the tests. Remand Results íIt was
impossible to discern whether the mill tests took samples from all of, or just a fraction of, the
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 13 of 21

Court No. 13-00029 	Page 14
OCTG in a given contract. Furthermore, judging from the first page of the reports, it seems
Chengde took only one sample per heat of OCTG produced. Yet according to rules promulgated
by the International Organization for Standardization, the manufacturer must test two tubes per
heat to get a valid chemical reading. See id. at 16; First Suppl. Resp. at S1-9. With no outline of
the test procedures, and with no inkling of the number of tubes sampled, Commerce could not
decide whether the mill reports established the chemical properties of unsampled OCTG. Thus
both assumptions supporting Plaintiffs’ claim lacked a foundation in fact.
Beset with ambiguity, Commerce sensibly valued [[ 	]] of Chengde’s billet using a
simple average of the carbon and alloy steel surrogates. The agency knew Chengde consumed
some carbon steel billet. The mill certificates and CBP entry form proved as much. Commerce
also knew that Chengde used some alloy steel billet. Chengde’s declaration in the initial
response and the company website supported that fact. Because the agency lacked adequate data
to value the untested billet as all alloy or all carbon—and because it knew that Chengde used
both types of billet—the agency reasonably chose to value half of the unsampled billet as alloy
steel and half as carbon. See Remand Results 13. This Solomonic solution was based in
substantial evidence—far more than if Commerce had invoked Plaintiffs’ baseless assumptions
to hold that the unsampled tube was carbon steel.
In rejoinder, Plaintiffs argue that Commerce asked only for sample mill certificates, not
certificates for every bundle of OCTG sold during the review period. Because Commerce asked
for sample reports, Plaintiffs say it was wrong for the agency to refuse using sample data to value
the billet. See 3OV¶&PWVíBut as it explained on remand, Commerce requested the sample
mill certificates to corroborate Chengde’s U.S. sales data. It did not ask for the certificates to
establish the chemical signature of the billet. See Remand Results 18. That was the purpose of
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 14 of 21

Court No. 13-00029 	Page 15
the first supplemental questionnaire, where Commerce asked for the “chemical specifications” of
the billet and for certificates of assay in support. See First Suppl. Q. at 6. Had Chengde wished
to prove that its billet was carbon steel, it could have answered these requests with exactness.
Instead, it regurgitated that its billet complied with ASME standards (which do not distinguish
between alloy and carbon steel), and withheld the certificates of assay that Commerce wanted.
See First Suppl. Resp. at Exs. S1-15 (S1-4), S1-16 (S1-5). Hence the agency used the OCTG
mill certificates as a secondary means of discerning the billet’s chemical makeup. Commerce
never intended the certificates to establish the chemical composition of all the billet.
U.S. Steel’s rebuttal fails too. As explained above, U.S. Steel insists that any billet not
specifically tested in the mill certificates was alloy steel. It bases its argument on Chengde’s first
response that the billet was alloy metal. It also cites Chengde’s failure to submit mill reports for
tube that later proved to be alloy. See 866WHHO&PWVí+DGWKHXQWHVWHGtube been made
of carbon steel, U.S. Steel says Chengde would have provided proof to that end instead of
retaining mill reports as it did for contracts [[ 	]].
But the court’s standard of review precludes this argument. For an agency decision to
gain the court’s stamp of approval, that determination must be rooted in substantial evidence.
See 19 U.S.C. § 1516a(b)(1)(B)(i). Substantial evidence is “such relevant evidence as a
reasonable mind might accept as adequate to support a conclusion.” Consol. Edison Co. v.
NLRB, 305 U.S. 197, 229 (1938). Speculative claims that are plausible in theory but
unsupported in fact do not make the cut. See Thai Plastic Bags Indus. Co. v. United States, 37
CIT __, __, 904 F. Supp. 2d 1326, 1332 (2013) (citing Lucent Techs., Inc. v. Gateway, Inc., 580
F.3d 1301, 1327 (Fed. Cir. 2009) (“It is well established that speculation does not constitute
substantial evidence.”)).
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 15 of 21

Court No. 13-00029 	Page 16
U.S. Steel’s argument is the very soul of speculation. The defendant-intervenor posits
that all the billet was alloy steel because Chengde said it was so in its initial response. But the
mill certificates and the CBP entry form decisively disprove this. Furthermore, one cannot infer
from the failure to submit comprehensive mill reports that Chengde was hiding something. It is
certainly possible that Chengde withheld select reports because they showed that the billet tested
was alloy. But it is also possible that Chengde provided incomplete reports because it tested just
a few of its products. Or maybe Chengde misplaced the missing certificates—another benign
possibility. In short, one cannot infer from the patchwork record that Chengde withheld mill
reports to cut its dumping rate. 4 A man is not a thief simply because he is wearing a trench coat.
After considering the parties’ comments, the court sustains the decision to value the
untested billet using a simple average of the alloy and carbon steel surrogate prices. Commerce
was working with an imperfect record, and the decision it rendered on remand dealt reasonably
with the evidence available.
II. 	Commerce Reasonably Determined the Surrogate Price for High Carbon Steel
The court now turns to the second contested issue: the agency’s use of Indonesian
surrogate data to value high carbon steel. In its previous opinion, the court granted Commerce’s
request to reconsider its surrogates for high and low carbon steel. Am. Tubular Prods., 2014 WL
4977626, at *íThen on remand, the agency found that the surrogates were reasonable
compared to benchmarks from several developing countries. See Remand Results í
Plaintiffs now argue that the high carbon steel surrogate was aberrational, even though it fell
among a range of third-country benchmarks. To ensure the high carbon steel surrogate reflects
4 U.S. Steel says the Chengde company website also shows that the untested tube was alloy steel. U.S.
Steel Cmts. 13. The court disagrees. The website established that pipe sold in contracts [[ 	]] was
alloy steel, but it accomplished nothing beyond that. See Am. Tubular Prods., :/DW í
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 16 of 21

Court No. 13-00029 	Page 17
the best available information on the record, Plaintiffs say Commerce should use an Indonesian
value, but washed of import prices that exceed the simple average of five select benchmarks. See
Pls.’ Cmts. 16. Otherwise, they ask the agency to use Ukrainian data, which they say contain no
aberrant values.
The court does not share the Plaintiffs’ concerns. While it is true that Commerce must
use only the “best available information” to make surrogates, the agency has discretion to decide
what the best information in the record is. See Nation Ford Chem. Co. v. United States, 166 F.3d
1373, 1377 (Fed. Cir. 1999) (discussing 19 U.S.C. § 1677b(c)(1)). In practice, Commerce often
chooses import data as the raw material for its surrogate values. The information usually comes
from a single surrogate country with a market economy comparable to the subject country’s. See
19 C.F.R. § 351.408(c)(2). After molding the data into a surrogate, Commerce can check to see
if the surrogate jibes with economic reality by comparing it to benchmarks from other markets.
If the surrogate falls far outside the range of benchmark prices, it may be unfit for use in the NV
formula, especially if it’s based on small, unrepresentative import quantities. See Blue Field
(Sichuan) Food Indus. Co. v. United States&,7BBBB)6XSSGí
(2013) (remanding rice straw surrogate that exceeded benchmarks by fifteen times); Mittal Steel
Galati S.A. v. United States, 31 CIT 1121, í)6XSSGí  
(remanding surrogate from Philippines because surrogate was based on small import volume and
was ten times higher than other benchmarks); Shanghai Foreign Trade Enters. Co. v. United
States, 28 CIT 480, í)6XSSGí   UHPDQGLQJVXUURJDWHIURP
India because surrogate was based on small import volume and was higher than other
benchmarks). But if a surrogate places comfortably among the benchmarks, Commerce may
find that it is the best available information to value the input.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 17 of 21

Court No. 13-00029 	Page 18
The agency reached the latter conclusion on remand, and reasonably so. To ensure that
its high carbon steel surrogate reflected the best available information on record, Commerce first
deleted from the Indonesian surrogate value any import prices from nonmarket economies, from
countries with generally available export subsidies, with zero quantity or price values and with
high prices but low volumes. This yielded a surrogate of $1149.40/MT. Then Commerce
compared the Indonesian surrogate value to third-country benchmarks, also scrubbed of import
data from nonmarket economies, et cetera. The comparison showed that the Indonesian
surrogate was about two-hundred percent of the lowest benchmark (Thailand at $567.55/MT)
and fifty percent of the highest benchmark (the Philippines at $2211.14/MT). Because the
surrogate fell between these bookends, Commerce concluded that the value was not aberrational.
See Remand Results 11. And based on its review of the record, the court agrees. Even if the
agency could have used other data to value high carbon billet, Commerce was within its
discretion to use the Indonesian values as a surrogate. See Nation Ford, 166 F.3d at 1377.
Plaintiffs still insist that the Indonesian surrogate was aberrant. In the first of four
arguments, they say the benchmark from the Philippines ($2211.14/MT) was itself aberrational,
and hence ineligible to prove that the Indonesian price was reasonable. They add that the five
least expensive benchmarks (Colombia, Peru, South Africa, Thailand, and Ukraine) better reflect
UHDOLW\DQGVKRZWKDWWKH,QGRQHVLDQVXUURJDWHLVWRRKLJK3OV¶&PWVí, Ex. 3. But
Plaintiffs’ technique begs the question. Instead of offering independent reasons why the
Indonesian and Philippines values are aberrant, they surmise that any price higher than the
Indonesian price is distortive. In essence, they assume the surrogate is aberrational in an attempt
to prove that the surrogate is aberrational. This logic is circular; it cannot prove that the values
from Indonesia and the Philippines were deviant.
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 18 of 21

Court No. 13-00029 	Page 19
Furthermore, the Indonesian and Filipino values represent only 203% to 390% of the
lowest benchmark (Thailand at $567.55/MT), and 124% to 239% of the benchmark just below
Indonesia (South Africa at $926.16/MT). See id. at Ex. 3. The court has found disparities of this
size to signal aberrations in the past, but those aberrational prices were usually based on tiny
import volumes. See Xinjiamei Furniture (Zhangzhou) Co. v. United States, 37 CIT __, __, 2013
:/DW í(Mar. 11, 2013) (remanding surrogate that exceeded benchmarks by 200%
to 400%, was based on import volume that equaled 0.047% of producer’s consumption);
Shanghai Foreign Trade, 28 CIT at í)6XSSGDWí UHPDQGLQJsurrogate
that exceeded benchmarks by between 40% and 80% and was based on small import quantity).
Here, Plaintiffs have not argued that the values from Indonesia and the Philippines were based on
small import volumes. And the price differences, though sizeable, are not outlandish. See Hebei
Metals & Minerals Imp. & Exp. Corp. v. United States, 28 CIT 1185, 1200 (2004) (holding
import price that was 8.5 times higher than average varied to a “uniquely extreme degree”). So
even if the Indonesian and Filipino surrogates exceeded other benchmarks, this does not prove
that the values were aberrant.
Second, Plaintiffs contend that the high carbon steel surrogate ($1149.40/MT) appears
aberrational next to the low carbon steel surrogate ($556.64/MT). They also note that the low
carbon Indonesian import prices are tightly clustered but the high carbon import prices are not.
In their view, this means the high carbon surrogate must contain some aberrant values. See Pls.’
&PWVíBut neither of these points discredits the high carbon surrogate. Though the high
carbon surrogate exceeds its low carbon counterpart, Plaintiffs offer no extrinsic evidence that
items in the HTS categories for high and low carbon steel sell for the same price. Even if the
two types of steel are chemically similar, they may be manufactured or used in different ways
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 19 of 21

Court No. 13-00029 	Page 20
and command different prices. Furthermore, the variation among import prices in either
subheading was largely the agency’s own doing. Before calculating the surrogates, Commerce
cleaned the datasets of aberrational import prices, or any price more than about four times the
price at the bottom of the dataset. For low carbon steel, Commerce eliminated six import prices
as aberrational; for high carbon steel, it deleted one. See id. DW([VíThis left a tightly
clustered set of prices for low carbon steel, but a more variegated range for high carbon steel.
The differences between the variances, then, do not prove that the high carbon surrogate
contained aberrant import prices. Instead it reflects the agency’s judgment that prices below four
times the lowest import value were not aberrational—a judgment that the court finds to be
reasonable on this record.
Third, Plaintiffs argue that the high carbon steel price appears strangely when stacked
next to the alloy steel surrogate price ($1120.13/MT). They claim that alloy steel is a higher
value product than carbon steel, and that the high carbon steel price should not exceed the alloy
steel price. 3OV¶&PWVíYet Plaintiffs fail to back their position with firm evidence. For
example, they offer no proof—other than their word alone—that alloy steel is more refined than
high carbon steel. If it were, then perhaps the court would hold that the high carbon surrogate
was aberrational. See Blue Field&,7DWBB)6XSSGDWí ILQGLQJULFHVWUDZ
surrogate aberrational because it exceeded price of rice grain); Baroque Timber Indus.
(Zhongshan) Co. v. United States, 37 CIT __, __, 925 F. Supp. 2d 1332, í  
(finding core veneer surrogate aberrational because it exceeded price of face veneer). Plaintiffs
also fail to show that alloy steel prices always exceed high carbon steel prices. They note that
the weighted average high and low carbon steel price ($813.86/MT) is less than the alloy price
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 20 of 21

Court No. 13-00029 	Page 21
here. But even if the composite carbon price is lower than the alloy price, this does not mean
that high carbon steel, standing alone, must also be less expensive than alloy steel.
Finally, Plaintiffs argue that the high carbon surrogate was aberrant because it exceeded
weekly average prices on the London Metals Exchange (“LME”) for high carbon steel. They
admit that the LME price was unfit as a surrogate, yet they claim that the agency erred to reject
the LME prices as benchmarks. See 3OV¶&PWVí FLWLQJFDVHVDSSURYLQJEHQFKPDUNVWKDW
were not appropriate as surrogates). But Plaintiffs forget that the LME data comprised prices
from nonmarket economies and countries with generally available export subsidies. The agency
usually purges these values from the import data to ensure its surrogates and benchmarks reflect
free market prices. See Remand Results 22. So if Commerce had compared the high carbon
surrogate to the LME prices, there was a risk that the surrogate would appear aberrant, even if it
was not. The agency was right to exclude the LME prices as a benchmark for high carbon steel.
In sum, Plaintiffs have not shown that the Indonesian high carbon steel surrogate was
aberrational. So even if there were other viable high carbon surrogate on the record, Commerce
properly exercised its discretion to choose the value it thought best. See Nation Ford, 166 F.3d
at 1377. The court sustains both the high carbon and low carbon steel surrogate values as
substantiated in evidence and otherwise in accordance with law.
CONCLUSION
The decisions in the Remand Results are sustained. Judgment will enter accordingly.
/s/ Richard W. Goldberg
Richard W. Goldberg
Senior Judge
Dated:
New York, New York
August 28, 2015
Case 1:13-cv-00029-RWG Document 124 Filed 09/08/15 Page 21 of 21