CP Kelco Oy et al v. United States

OCR transcription for search; preserved printed pages are authoritative

Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 1 of 11

Slip Op. 15-24

UNITED STATES COURT OF INTERNATIONAL TRADE

CP KELCO OY and CP KELCO US, INC.,

Plaintiffs, Before: Richard W. Goldberg, Senior Judge
Court No. 13-00079

UNITED STATES, PUBLIC VERSION

Defendant,
and

ASHLAND SPECIALTY INGREDIENTS,
GP.,

Defendant-Intervenor.

OPINION

Matthew L. Kanna, Nancy A. Noonan, and Tina Termei, Arent Fox LLP, of Washington,
DC, for plaintiff.

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
Joyce R. Branda, Acting Assistant Attorney General, Jeanne E. Davidson, Director, and
Reginald T. Blades, Jr., Assistant Director. Of counsel on the brief was Aman Kakar, Attorney,
Office of the Chief Counsel for Trade Enforcement & Compliance, U.S. Department of
Commerce, of Washington, DC.

Edward M. Lebow, Haynes and Boone, LLP, of Washington, DC, for defendant-
intervenor Ashland Specialty Ingredients, G-P.

Goldberg, Senior Judge: This case returns after a succinct remand. In its previous
opinion, the court invalidated the decision of the Department of Commerce (“Commerce” or “the
agency”) to calculate antidumping duties for plaintiffs CP Kelco Oy and CP Kelco US, Inc.

(collectively “Kelco”’) using the average-to-transactional methodology. The court found

AUTHENTICATED
U.S. GOVERNMENT
INFORMATION

GPO

Court No. 13-00079 	Page 2
Commerce acted arbitrarily by failing to explain why Kelco’s “targeted” sales were sufficient to
merit the average-to-transactional treatment. See CP Kelco Oy v. United States, 38 CIT __, __,
978 F. Supp. 2d 1315, 1327í29 (2014). Now the agency has furnished the explanation required,
and the court sustains the decision to use the average-to-transactional method to craft Kelco’s
antidumping rate.
BACKGROUND
The court presumes familiarity with its prior opinion, including the exposition of
Commerce’s margin calculation methods and the targeted dumping test. The abridged facts that
follow will suffice for the sake of this decision.
In 2011, Commerce began an administrative review of an antidumping duty order on
carboxymethylcellulose from Finland. Initiation of Antidumping and Countervailing Duty
Administrative Reviews, 76 Fed. Reg. 53,404, 53,405 (Dep’t Commerce Aug. 26, 2011). During
the review, Commerce considered whether to apply its default average-to-average methodology
(“A-A”), or its exceptional average-to-transactional methodology (“A-T”), to render Kelco’s
dumping margins. To guide its decision, Commerce followed the statutory framework in
§ 777A(d)(1)(B) of the Tariff Act of 1930, 19 U.S.C. § 1677f-1(d)(1)(B) (2012). See Purified
Carboxymethylcellulose from Finland, 78 Fed. Reg. 11,817, 11,817 (Dep’t Commerce Feb. 20,
2013) (final admin. review) (“Final Results”); Issues & Decision Mem. (“I&D Mem.”) at 6, 8–
10, PD 80 (Feb. 6, 2013).
That provision, known colloquially as the “targeted dumping” statute, reads as follows:
(B) Exception
The administering authority may determine whether the subject merchandise is
being sold in the United States at less than fair value by comparing the weighted
average of the normal values to the export prices (or constructed export prices) of
individual transactions for comparable merchandise, if--
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 2 of 11

Court No. 13-00079 	Page 3
(i) 	there is a pattern of export prices (or constructed export prices) for
comparable merchandise that differ significantly among purchasers,
regions, or periods of time, and
(ii) 	the administering authority explains why such differences cannot be taken
into account using a method described in paragraph (1)(A)(i) or (ii) [i.e.,
the A-A or transactional-to-transactional methods].
19 U.S.C. § 1677f-1(d)(1)(B).
To perform this inquiry, Commerce first applied the so-called Nails test to Kelco’s U.S.
sales. The Nails test identifies targeted transactions, or patterns of export prices that differ
significantly among purchasers, regions, or time periods. See id. § 1677f-1(d)(1)(B)(i); Mid
Continent Nail Corp. v. United States, 34 CIT __, __, 712 F. Supp. 2d 1370, 1373í74 (2010)
(explaining the Nails test). Then, after finding that Kelco targeted [[ 	]] percent of its sales
by quantity and [[ 	]] percent by value, Commerce concluded that the targeting was more than
de minimis, comprising a fraction of total U.S. sales sufficient to merit further analysis. See
Analysis of Data Submitted by CP Kelco Oy & CP Kelco U.S. Inc. at 2, CD 195 (Feb. 11, 2013)
(naming ratios of Kelco’s targeted to total U.S. sales); I&D Mem. at 9–10 (finding sufficient
volume of sales passed Nails test). The agency did not name the quantum of sales needed to
clear the de minimis bar.
Despite this ambiguity, Commerce moved to the second step of the statutory inquiry,
which asks whether A-A cannot account for targeted sales. See 19 U.S.C. § 1677f-1(d)(1)(B)(ii).
The agency found it could not. A-A yielded a “meaningfully” lower antidumping rate than A-T,
so Commerce, in its discretion, applied A-T to Kelco’s sales to form a 12.06 percent final rate.
Final Results at 11,817 (final rate); I&D Mem. at 9 (meaningful difference test).
Kelco filed suit at the Court of International Trade on February 26, 2013. Summons,
ECF No. 1. In its brief Kelco raised a host of claims, only one of which survived judicial review.
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 3 of 11

Court No. 13-00079 	Page 4
After dismissing Kelco’s arguments regarding Commerce’s authority to conduct targeting
inquiries in reviews and the legality of the Nails test, the court invalidated Commerce’s de
minimis finding as arbitrary. Kelco, 38 CIT at __, 978 F. Supp. 2d at 1327–29. The court’s
criticism was twofold. First, the court noted that “Commerce never explained what purpose the
de minimis test serves in the statutory scheme.” Id. at __, 978 F. Supp. 2d at 1328. It was
unclear from the record whether the de minimis analysis helped to identify a pattern of targeting
under § 1677f-1(d)(1)(B)(i), or whether the test guided the agency’s discretion under § 1677f-
1(d)(1)(B)(ii) to use A-T once targeting was found. Second, the court faulted Commerce for
failing to identify “the quantum of an exporter’s sales that must be targeted to fall above or
below the de minimis threshold.” Id. Because Commerce never articulated “the basis on which
the [agency] exercised its expert discretion,” the court remanded for Commerce to explain its
construction and application of the de minimis test. Id. at __, 978 F. Supp. 2d at 1327 (quoting
Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 167 (1962)) (internal quotation
marks omitted).
On remand, Commerce provided the explanation required. See Final Results of
Redetermination Pursuant to Ct. Remand, ECF No. 72 (“Remand Results”). First, Commerce
said the de minimis test served both to identify a pattern of targeting, as required under § 1677f-
1(d)(1)(B)(i), and to guide Commerce’s discretion to apply A-T, as allowed under § 1677f-
1(d)(1)(B)(ii). Id. at 9–15. Second, in response to the court’s request to identify the quantum of
sales that cleared the de minimis bar, Commerce declared that targeted sales were more than de
minimis if they comprised more than five percent of total U.S. sales during the review period. Id.
at 15–19. Commerce drew this threshold from statutory provisions and regulations that use the
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 4 of 11

Court No. 13-00079 	Page 5
same figure in other contexts, and noted that the threshold corroborated with Commerce’s
experience in administering the targeted dumping test. See id. at 18–19.
Having rendered this explanation, Commerce again held that Kelco targeted its sales in a
volume sufficient to merit consideration of the A-T remedy. After comparing Kelco’s A-A rate
with its A-T rate, and finding a meaningful difference between the two, Commerce applied A-T
to all of Kelco’s sales for a 12.06 percent final rate. See id. at 9í10, 16, 20–21.
JURISDICTION AND STANDARD OF REVIEW
The court has jurisdiction to hear this appeal under 28 U.S.C. § 1581(c). The court will
sustain the agency's decisions unless they are “unsupported by substantial evidence on the
record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(1)(B)(i).
DISCUSSION
The court now sustains the Remand Results. In its previous opinion, the court invalidated
Commerce’s targeted dumping decision not because of patent flaws in the agency’s reasoning,
but because the agency offered no reasoning at all regarding the de minimis test’s quantitative
contours and role in the statute. Kelco, 38 CIT at __, 978 F. Supp. 2d at 1327–29. Now
Commerce has explained itself to the court’s satisfaction, and though reasonable minds may
differ over the wisdom of Commerce’s choice, the decision merits deference.
I. 	Commerce Reasonably Explained the De Minimis Test’s Role
To begin, the court holds that Commerce reasonably “explained what purpose the de
minimis test serves in the statutory scheme.” Id. at 1328. In the agency’s view, the role is
twofold. First, the inquiry helps to identify “a pattern of prices that differ significantly by
purchaser, region or period of time” among respondent’s U.S. sales. Remand Results 9. Of
course, the statute does not say whether Commerce must search out the pattern among all
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 5 of 11

Court No. 13-00079 	Page 6
reviewed sales, as it did below, or only among sales to alleged targets. See 19 U.S.C. § 1677f-
1(d)(1)(B)(i). Commerce has taken conflicting positions on this issue in the past.1 But whatever
its prior views, the agency acted sensibly here to require that the volume of targeted sales
comprise more than a token part of U.S. sales. To form a “pattern,” a “mode of behavior or
series of acts,” such as selective low-cost sales, must be “recognizably consistent.” Black’s Law
Dictionary 1308 (10th ed. 2014). By checking to see whether Kelco’s targeting encompassed
more than a de minimis share of total sales, Commerce carried its duty to find a recognizable,
and hence remediable, pattern of targeting among all reviewed transactions.
Second, the agency explained that the de minimis inquiry informs its choice to impose A-
T after finding targeting. Remand Results 13í15. Under the statute, Commerce “may
determine” dumping margins using A-T if a respondent’s sales meet the criteria in § 1677f-
1(d)(1)(B)—in short, the law commits the decision to apply A-T to the agency’s judgment. See
Timken Co. v. United States, 38 CIT __, __, 968 F. Supp. 2d 1279, 1287í88 (2014). But even
given this discretion, it was wise for Commerce to consider the ratio of targeted to untargeted
sales before invoking A-T. Having withdrawn a regulation called the limiting rule in 2008,
Commerce applied A-T to all sales, both targeted and untargeted, during the review in question.
See I&D Mem. at 9í10; see also Apex Frozen Foods Private Ltd. v. United States, 38 CIT __,
__, 37 F. Supp. 3d 1286, 1301í03 (2014) (explaining that Commerce applied A-T to all sales
after 2008). To avoid imposing A-T on untargeted sales without cause, the agency first ensured
1 In some proceedings, Commerce found a pattern of significantly differing prices only where targeted sales
comprised a sufficient part of total U.S. sales. See, e.g., Certain Stilbenic Optical Brightening Agents from Taiwan,
77 Fed. Reg. 17,027, 17,027í28 (Dep’t Commerce Mar. 23, 2012) (final determination); Ball Bearings and Parts
Thereof from France, Germany, and Italy, 77 Fed. Reg. 73,415 (Dep’t Commerce Dec. 10, 2012) (final admin.
review) and accompanying I&D Mem at cmt. 1 (“Ball Bearings”). In other proceedings, Commerce found a pattern
of prices without regard to the ratio of targeted sales to total sales. See, e.g., High Pressure Steel Cylinders from the
People’s Republic of China, 77 Fed. Reg. 26,739 (Dep’t Commerce May 7, 2012) (final determination) and
accompanying I&D Mem. at cmt. IV; Certain Steel Nails from the United Arab Emirates, 77 Fed. Reg. 17,029
(Dep’t Commerce Mar. 23, 2012) (final determination) and accompanying I&D Mem. at cmt. 3. The latter formula
is not under scrutiny here, however, and the former approach is reasonable in its own right.
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 6 of 11

Court No. 13-00079 	Page 7
that Kelco’s targeting was more than de minimis, or sufficient to merit a broad remedy. This was
a fair approach, which Kelco does not protest as a general matter. See Pls.’ Objections to
Remand Redetermination 2í5, ECF No. 80 (“Pls.’ Objections”) (objecting to the five-percent
threshold, not the de minimis test itself). The court sustains the de minimis test as a reasonable
interpretation of the law and a prudent guide to Commerce’s discretion.
II. 	Commerce Reasonably Defined the De Minimis Threshold at Five Percent
The court also sustains the five-percent threshold Commerce chose for its de minimis test.
See Remand Results 15í19; Kelco, 38 CIT at __, 978 F. Supp. 2d at 1330 (ordering agency to
define quantum of sales that are more than de minimis). One should recognize, of course, that
this threshold is a line in the sand: Commerce might have picked a different number to
effectuate the statute’s purpose, with reasonable results. See Chevron, U.S.A., Inc. v. Natural
Res. Def. Council, Inc., 467 U.S. 837, 844í45 (1984) (warning courts not to substitute their own
interpretation of statute for agency’s reasonable construction). Yet because the agency’s choice
does not run afoul of the statute and is not arbitrary, the court will defer to Commerce despite the
possibility of alternatives. See Mid Continent Nail, 34 CIT at __, 712 F. Supp. 2d at 1378í79.
On remand, Commerce reported that it measures significance with five-percent tests in
many contexts. Remand Results 15í16. For example, it uses a five-percent threshold to decide
whether home market sales can make a viable normal value. See 19 U.S.C. § 1677b(a)(1)(C). If
a respondent’s home market sales amount to less than five percent of its U.S. sales by volume or
value, then Commerce normally deems that market too thin to use in finding dumping margins.
If home market sales amount to five percent or more of U.S. sales by volume or value, however,
then the agency presumes the home market prices are comparable to U.S. export prices. See id.;
Uruguay Round Agreements Act, Statement of Administrative Action, H.R. Doc. No. 103-316,
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 7 of 11

Court No. 13-00079 	Page 8
vol. 1, at 821í22 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4161í62 (“SAA”). The statute
outlines a similar test to use when Commerce assembles normal value from third-country data.
See 19 U.S.C. § 1677b(a)(1)(B)(ii)(II); SAA at 822. In like manner, it was reasonable for
Commerce to find a representative “pattern” of targeting when targeted sales exceed five percent
of total U.S. sales by volume. See 19 U.S.C. § 1677f-1(d)(1)(B).
Other five-percent tests support this view. The regulations, for instance, permit
Commerce to set normal value using home-market data from an exporter’s affiliate, but only if
the affiliate’s sales make up five percent or more of all home sales. 19 C.F.R. § 351.403(d)
(2015); see also China Steel Corp. v. United States, 28 CIT 38, 46, 306 F. Supp. 2d 1291, 1300
(2004) (allowing use of affiliate data to form normal value where affiliate sales exceeded five
percent, “a significant percentage”). Commerce also uses a five-percent threshold to ferret out
targeting under the Nails test. If Commerce locates a “pattern” of prices below one standard
deviation of the mean by control number, it calls those price differences “significant” if the gap
between the weighted-average price to the alleged target and the weighted-average price to the
next higher untargeted group exceeds the average gap between untargeted groups for five percent
of sales to the alleged target. See I&D Mem. at 9. Commerce sensibly used a similar test to
decide if Kelco’s sales formed a pattern of differing prices meriting the A-T remedy. 2
2 Commerce adds that it forged its five-percent threshold in past cases. If it did, the court has no way of
knowing. In prior proceedings, Commerce scrupulously avoided naming the quantum of targeted sales needed to
clear the de minimis bar. The agency also failed to reveal the ratio of targeted to total U.S. sales from individual
companies, probably to preserve confidentiality. See, e.g., Corrosion-Resistant Carbon Steel Flat Products from the
Republic of Korea, 78 Fed. Reg. 16,247 (Dep’t Commerce Mar. 14, 2013) (final admin. review) and accompanying
I&D Mem. at cmt. 1(C)(6); Polyethylene Terephthalate Film, Sheet, and Strip from India, 78 Fed. Reg. 9670 (Dep’t
Commerce Feb. 11, 2013) (final admin. review) and accompanying I&D Mem. at cmt. 1; Tapered Roller Bearings
and Parts Thereof, Finished and Unfinished, from the People’s Republic of China, 78 Fed. Reg. 3396 (Dep’t
Commerce Jan. 16, 2013) (final admin. review) and accompanying I&D Mem. at cmt. 1; Ball Bearings, 77 Fed.
Reg. at 73,415 and accompanying I&D Mem. at cmt. 1; Circular Welded Carbon Steel Pipes and Tubes from
Turkey, 77 Fed. Reg. 72,818 (Dep’t Commerce Dec. 6, 2012) (final admin. review) and accompanying I&D Mem. at
cmt. 1. Because none of these proceedings revealed the contours of Commerce’s de minimis test, they do little to aid
judicial review. See Burlington, 371 U.S. at 167.
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 8 of 11

Court No. 13-00079 	Page 9
Kelco counters that it would have been more appropriate for Commerce to set the de
minimis threshold at twenty percent. Pls.’ Objections 5. Under 19 U.S.C. § 1677b(b)(1),
Commerce may exclude from its normal value calculation any below-cost home market sales
made “within an extended period of time in substantial quantities.” The statute defines
“substantial quantities” of below-cost sales as transactions encompassing twenty percent “or
more of the volume of sales under consideration for the determination of normal value.” 19
U.S.C. § 1677b(b)(2)(C)(i). Yet as the United States suggests in its response on remand, the
twenty-percent threshold does not measure the representativeness of normal value data, like 19
U.S.C. § 1677b(a)(1) (market viability) or 19 C.F.R. 351.403(d) (affiliate sales). Def.’s Resp. to
Pls.’ Cmts. on Final Results of Remand Redetermination 8–9, ECF No. 84. Instead, the twenty-
percent threshold indicates whether below-cost sales render home market prices unrepresentative
of normal value. See NSK Ltd. v. United States, 28 CIT 1535, 1551í53, 346 F. Supp. 2d 1312,
1327–29 (2004) (suggesting that purpose of cost-of-production provision is to “evaluate the
rationality of exporters [sic] pricing practices”). Furthermore, even if a twenty-percent threshold
correctly identified significant targeting, that would not mean Commerce’s choice was
unreasonable. When Commerce is met with equally plausible alternatives, it may choose the
approach that it thinks best effectuates the statute’s purpose. See Chevron, 467 U.S. at 844í45.
Here, the agency used a five-percent test instead of a twenty-percent test to identify a pattern of
targeting, and the court will defer to the agency’s reasonable preference.
III. 	Kelco Waived Any Argument Regarding the Meaningful Difference Test
Finally, Kelco claims that the agency failed to explain the “meaningful difference” test
conducted on remand. Pls.’ Objections 5–6. After finding that a sufficient number of Kelco’s
U.S. sales were targeted, Commerce had to decide whether A-A could account for that targeting
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 9 of 11

Court No. 13-00079 	Page 10
under 19 U.S.C. § 1677f-1(d)(1)(B)(ii). See Remand Results 16. To do so, Commerce
compared Kelco’s weighted-average dumping margin as rendered under A-A and A-T. Then,
after discovering a meaningful difference between the two rates, the agency applied A-T to yield
a final rate. See id. Kelco claims that Commerce acted arbitrarily by neglecting to define what
“meaningful” means in this context.
The court will not consider this claim, however, because Kelco did not raise it before the
agency, in the complaint, or in the original briefs. See I&D Mem. at 7–8; Complaint 6–7, ECF
No. 4; Mem. of L. in Support of Pls.’ R. 56.2 Mot. For J. upon Agency R. 1–2, ECF No. 28.
Commerce made similar “meaningful difference” findings in the post-preliminary results and
Final Results, see Post-Prelim. Targeted Dumping Analysis Mem. at 3, PD 68 (Dec. 26, 2012);
I&D Mem. 9–10, and if Kelco wished to challenge the test, it should have done so at the earliest
opportunity, see 28 U.S.C. § 2637(d) (requiring exhaustion of administrative remedies); Casa de
Cambio Comdiv S.A., de C.V. v. United States, 291 F.3d 1356, 1366 (Fed. Cir. 2002) (rejecting
claim not raised in complaint).
The court also disagrees that the meaningful difference test became “subject to judicial
review” when Commerce used it to set a rate on remand. Pls.’ Objections 6. Though Kelco
claims otherwise, remand is not the time to litigate new claims that could have been raised in
prior proceedings. See Bridgestone Americas, Inc. v. United States, 34 CIT __, __, 710 F. Supp.
2d 1359, 1365–67 (2010) (rejecting argument raised after remand). What is more, the court did
not ask Commerce to revisit the meaningful difference test in the remand order. See Kelco, 38
CIT at __, 978 F. Supp. 2d at 1330. The court thus declines to entertain Kelco’s meaningful
difference argument—whatever its merits—because it was made too late.3
3 Kelco’s meaningful difference challenge would fail in any event. The court has found that the disparity
between A-A and A-T rates is “meaningful” if it is more than de minimis, see Apex, 38 CIT at (footnote continued)
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 10 of 11

Court No. 13-00079 	Page 11
CONCLUSION
In its prior opinion, the court faulted Commerce for failing to explain the purpose and
contours of the de minimis test. Commerce has now provided an explanation that is in harmony
with the law and evidence. The court sustains the agency’s remand redetermination, and
judgment will enter accordingly.
/s/ Richard W. Goldberg
Richard W. Goldberg
Senior Judge
Dated:
New York, New York
__, 37 F. Supp. 3d at 1299í1300 (sustaining “meaningful difference” of 0.5 percentage points), and the difference
between Kelco’s A-A rate and A-T rate (applied to all sales) certainly clears this bar, see Final Results at 11,817
(giving Kelco’s A-T rate); Pls.’ Objections 6 (giving Kelco’s A-A rate); Remand Results 9í10 (noting Commerce
applies A-T to all sales when finding A-T rate).
March 25, 2015
Case 1:13-cv-00079-RWG Document 89 Filed 03/31/15 Page 11 of 11