United States Court of Appeals
for the Federal Circuit
______________________
BMW OF NORTH AMERICA LLC,
Plaintiff-Appellant
v.
UNITED STATES, TIMKEN COMPANY,
Defendants-Appellees
______________________
2018-1109
______________________
Appeal from the United States Court of International
Trade in No. 1:15-cv-00052-JCG, Judge Jennifer Choe-
Groves.
______________________
SEALED OPINION ISSUED: May 9, 2019
PUBLIC OPINION ISSUED: June 14, 2019 *
______________________
KAVITA MOHAN, Grunfeld, Desiderio, Lebowitz, Silver-
man & Klestadt LLP, Washington, DC, argued for plaintiff-
appellant. Also represented by NED H. MARSHAK, MAX
FRED SCHUTZMAN, New York, NY.
TARA K. HOGAN, Commercial Litigation Branch, Civil
* This opinion was originally filed under seal and has
been unsealed in part with the remaining sealed portions
redacted from the public opinion.
2 BMW OF NORTH AMERICA LLC v. UNITED STATES
Division, United States Department of Justice, Washing-
ton, DC, argued for defendant-appellee United States. Also
represented by CLAUDIA BURKE, JEANNE DAVIDSON, JOSEPH
H. HUNT.
GEERT M. DE PREST, Stewart & Stewart, Washington,
DC, argued for defendant-appellee Timken Company. Also
represented by WILLIAM ALFRED FENNELL, LANE S.
HUREWITZ, PATRICK JOHN MCDONOUGH, TERENCE PATRICK
STEWART.
______________________
Before PROST, Chief Judge, O’MALLEY and STOLL, Circuit
Judges.
Opinion for the court filed by Circuit Judge STOLL.
Opinion dissenting-in-part filed by Chief Judge PROST.
STOLL, Circuit Judge.
BMW of North America LLC (“BMW”) appeals the final
judgment of the United States Court of International
Trade, sustaining the United States Department of Com-
merce’s application of an adverse facts available (“AFA”)
rate of 126.44% against BMW. We conclude that Com-
merce did not set forth its reasoning in sufficient detail to
allow review of whether the selected AFA rate was unduly
punitive. We therefore vacate the Court of International
Trade’s decision and remand for further proceedings con-
sistent with this opinion.
BACKGROUND
I
Among its many duties, Commerce is charged with pro-
tecting domestic producers from foreign exporters’ unfair
trade practices. To this end, Commerce must investigate
and impose antidumping duties on imported merchandise
sold in the United States at less than fair value, causing
BMW OF NORTH AMERICA LLC v. UNITED STATES 3
material injury or threatening to cause material injury to
an industry in the United States. See 19 U.S.C. § 1673.
Generally, exporters of subject merchandise that have not
been individually investigated during an antidumping
duty investigation are subject to an “all-others rate” im-
posed in Commerce’s final determination.
Id.
§ 1673d(c)(1)(B). If an exporter or other interested party
properly requests administrative review and is selected for
individual examination, however, Commerce will reassess
the antidumping duties applied to the exporter’s subject
merchandise using actual dumping margins for the period
of review. See
id. § 1675.
With this brief background, we turn to the relevant
facts in this case. On May 2, 2011, Commerce posted a no-
tice in the Federal Register permitting interested parties
to request an administrative review of antidumping duty
orders on ball bearings and parts thereof from France, Ger-
many, Italy, Japan, and the United Kingdom, for the period
of May 1, 2010, through April 30, 2011. Antidumping or
Countervailing Duty Order, Finding, or Suspended Investi-
gation; Opportunity to Request Administrative Review,
76 Fed. Reg. 24,460, 24,460–61 (May 2, 2011). BMW
timely requested administrative review of the antidumping
duties on its imports of ball bearings and parts thereof from
the United Kingdom.
In the meantime, the antidumping duty orders on ball
bearings and parts thereof from Japan and the United
Kingdom, which were first imposed on May 15, 1989, were
undergoing sunset review. To satisfy international treaty
obligations, Commerce and the International Trade Com-
mission (“ITC”) must review antidumping duty orders
every five years. See Understanding Five-Year (Sunset) Re-
views, UNITED STATES INTERNATIONAL TRADE COMMISSION
(Jan. 11, 2019). To maintain an antidumping duty order,
Commerce must determine every five years that revoking
the order would be likely to lead to continuation or recur-
rence of dumping, and the ITC must determine that
4 BMW OF NORTH AMERICA LLC v. UNITED STATES
revoking the order would be likely to lead to the continua-
tion or recurrence of material injury to an industry in the
United States.
Id. Here, both agencies made the requisite
determinations, but the Court of International Trade va-
cated and remanded the ITC’s determination several times
between 2006 and 2011. The ITC eventually determined
that revocation of the orders would not be likely to lead to
the continuation or recurrence of material injury to a U.S.
industry within a reasonably foreseeable time, and the
Court of International Trade affirmed the ITC on April 20,
2011. The judgment was appealed to our court and stayed
until July 6, 2011.
On July 15, 2011, Commerce published a notice in the
Federal Register stating that it was “revoking the anti-
dumping duty orders on ball bearings and parts thereof
from Japan and the United Kingdom.” Ball Bearings and
Parts Thereof from Japan and the United Kingdom: Revo-
cation of Antidumping Duty Orders, 76 Fed. Reg. 41,761,
41,762 (July 15, 2011). The notice further indicated that,
“[a]s a result of this revocation, the Department is discon-
tinuing all unfinished administrative reviews immediately
and will not initiate any new administrative reviews of the
orders.”
Id. Commerce also noted that “the suspension of
liquidation on all entries of ball bearings . . . will continue
until there is a ‘final and conclusive’ court decision.”
Id.
at 41,763. On May 16, 2013, our court reversed the Court
of International Trade, and on November 18, 2013, the
Court of International Trade reinstated the ITC’s affirma-
tive material injury determination.
On December 12, 2013, Commerce e-mailed counsel for
all parties that had previously requested administrative
review, stating only that it was “sending out a quantity-
and-value-questionnaire to all respondents in the
5/1/2010–4/30/2011 administrative reviews of ball bearings
and parts thereof from Japan and the United Kingdom.”
J.A. 134. The subject line of the email read “Ball Bearings
from Japan and UK: AFBs 22 (2010–11) Quantity-and-
BMW OF NORTH AMERICA LLC v. UNITED STATES 5
Value Questionnaire.”
Id. On December 16, 2013, Com-
merce published a notice in the Federal Register, indicat-
ing that it was “hereby resuming the administrative
reviews covering the period May 1, 2010 through April 30,
2011” and that the “deadline for withdrawing requests for
review . . . will be 90 days after the date of publication of
this notice.” Ball Bearings and Parts Thereof from Japan
and the United Kingdom: Notice of Reinstatement of Anti-
dumping Duty Orders, Resumption of Administrative Re-
views, and Advance Notification of Sunset Reviews,
78 Fed. Reg. 76,104, 76,105–06 (Dec. 16, 2013). Counsel
for BMW did not fill out the quantity-and-value-question-
naire, withdraw from administrative review, or otherwise
respond to Commerce’s communications.
II
During an administrative review, the “burden of creat-
ing an adequate record lies with interested parties and not
with Commerce.” Nan Ya Plastics Corp. v. U.S.,
810 F.3d 1333, 1337 (Fed. Cir. 2016) (quoting QVD Food
Co. v. United States,
658 F.3d 1318, 1324 (Fed. Cir. 2011)).
This is because “the International Trade Administration,
the relevant agency within Commerce, has no subpoena
power.”
Id. at 1338. If a respondent fails to provide re-
quested information by the deadlines for submission,
“Commerce shall fill in the gaps with ‘facts otherwise avail-
able.’” Nippon Steel Corp. v. U.S.,
337 F.3d 1373, 1381
(Fed. Cir. 2003) (quoting 19 U.S.C. § 1677e(a) (2000)). Sep-
arately, if Commerce determines that an interested party
has “failed to cooperate by not acting to the best of its abil-
ity to comply” with a request for information, it may use an
adverse inference in selecting a rate from these facts.
Id.
(quoting 19 U.S.C. §1677e(b)). This is commonly known as
the “AFA” rate.
On September 17, 2014, Commerce released its prelim-
inary decision memorandum. Decision Memorandum for
Preliminary Results of Antidumping Duty Administrative
6 BMW OF NORTH AMERICA LLC v. UNITED STATES
Reviews; Ball Bearings and Parts Thereof from Japan and
the United Kingdom; 2010-2011, A-412-801, ARP 10-11,
at 3 (Sept. 17, 2014) (“Preliminary Decision Memo”). Com-
merce noted that BMW “did not respond to [Commerce’s]
request to provide information concerning the quantity and
value of its U.S. sales.”
Id. at 6. Finding that BMW had
“failed to cooperate by not acting to the best of its ability to
comply with the Department’s requests for information,”
Commerce applied AFA to determine BMW’s weighted-av-
erage dumping margin.
Id. at 1, 6. Commerce applied a
254.25% AFA rate, which “represent[ed] the highest rate
calculated in the petition with respect to ball bearings”
from the United Kingdom. 1
Id. at 7. Commerce explained
that, “[w]hen a respondent is not cooperative, the Depart-
ment has the discretion to presume that the highest prior
dumping margin reflects the current weighted-average
dumping margin.”
Id. Commerce reasoned that if the ac-
tual dumping margins were otherwise, the “party would
produce current information showing its rate to be less.”
Id. Commerce found that the 254.25% rate was sufficiently
probative of BMW’s actual dumping margins, as it “falls
within the range of individual dumping margins which we
calculated for [the representative exporter] in the instant
administrative review concerning ball bearings from the
United Kingdom.” 2
Id. at 8. In contrast, the other parties
1 The “petition” refers to the original request filed by
Appellee Timken Co., a domestic interested party, asking
that Commerce investigate potential dumping of ball bear-
ings and parts thereof into the U.S. market. Petitions must
include all factual information relevant to the calculation
of dumping margins on the subject merchandise. See
19 C.F.R. § 351.202(b)(7).
2 Because Commerce found that it “was not practica-
ble to individually examine each company for which a re-
view was initiated,” Commerce selected NSK Bearings
Europe Ltd. and NSK Europe Ltd. (collectively, “NSK”) to
BMW OF NORTH AMERICA LLC v. UNITED STATES 7
who had requested administrative review of the U.K. anti-
dumping duty order received rates of 1.55%. Ball Bearings
and Parts Thereof from Japan and the United Kingdom:
Preliminary Results of Antidumping Duty Administrative
Review; 2010–2011, 79 Fed. Reg. 56,771, 56,772 (Sept. 23,
2014). The “all-others” rate remained 54.27%.
On October 23, 2014, BMW submitted a brief to Com-
merce, raising three main points: (1) the administrative re-
view had been improperly reinstated; (2) the procedural
anomalies, years-long delay, and changes in counsel meant
BMW was entirely unaware it even had an administrative
review pending; and (3) the AFA rate assigned to BMW was
excessive and unreasonable. In its final results and accom-
panying decision memorandum, Commerce responded
that, while this was indeed the first time it had automati-
cally reinstated an administrative review following appel-
late court review, it nevertheless had the discretion to do
so. Further, because BMW did not respond to the quantity-
and-value questionnaire, it did not cooperate, and thus ap-
plication of an AFA rate was proper. Finally, Commerce
explained that the AFA rate was not excessive and unrea-
sonable, as it is Commerce’s “normal practice in an admin-
istrative review to select as AFA the highest rate on the
record of the proceeding that can be corroborated.” Issues
and Decision Memorandum for the Antidumping Duty Ad-
ministrative Review of Ball Bearings and Parts Thereof
from the United Kingdom; 2010–2011, A-412-801, ARP 10-
11, at 14 (Jan. 21, 2015). According to Commerce, because
“a cooperative respondent, NSK, had transaction-specific
dumping margins in excess of 254.25 percent,” the “petition
margin is a reasonable rate incorporating an adverse
represent the twelve exporters who had requested admin-
istrative review of the antidumping duty order. See Pre-
liminary Decision Memo at 5.
8 BMW OF NORTH AMERICA LLC v. UNITED STATES
inference for a company that did not respond to our request
for information.”
Id. at 15.
BMW filed a complaint in the Court of International
Trade. The Court of International Trade held that Com-
merce had the authority to resume the discontinued review
without having to initiate a new review. BMW of N. Am.
LLC v. United States,
208 F. Supp. 3d 1388, 1393–94
(Ct. Int’l Trade 2017). The Court of International Trade
also concluded that Commerce’s use of AFA was supported
by substantial evidence. The court identified the fact that
“every other respondent in this review responded to the
[quantity-and-value] questionnaire,” coupled with BMW’s
“failure to monitor the status of the litigation that led to
the revocation of the very order that [BMW] requested to
be reviewed” as substantial evidence supporting “Com-
merce’s conclusion that Plaintiff failed to act to the best of
its ability.”
Id. at 1395.
The Court of International Trade found that the AFA
rate of 254.25%, however, was not corroborated or sup-
ported by substantial evidence.
Id. at 1398. The court
noted that “the purpose of AFA is to ‘provide respondents
with an incentive to cooperate, not to impose punitive, ab-
errational, or uncorroborated margins.’”
Id. at 1396. The
court further noted that while the petition Commerce re-
lied on “alleged dumping at 254.25 percent, the two indi-
vidually investigated parties received rates of 61.14
percent and 44.02 percent, respectively, while the all oth-
ers rate was calculated at 54.27 percent.”
Id. at 1397. The
court rejected Commerce’s attempt to corroborate this rate
based on the representative exporter’s “transaction-specific
dumping margins in excess of 254.25 percent,” reasoning
that these margins represented [redacted] of the repre-
sentative exporter’s total transactions, and were thus ab-
errational.
Id.
On remand, Commerce applied an AFA rate of
126.44%, which it explained was:
BMW OF NORTH AMERICA LLC v. UNITED STATES 9
the highest transaction-specific dumping margin
that forms part of a closely-connected range of
transaction-specific margins (i.e., it falls on a [re-
dacted] of transaction-specific dumping margins in
the sense that [redacted]).
Results of Remand Determination, A-412-801, ARP 10-11,
at 6 (May 11, 2017) (“Remand Determination”). Commerce
found “no circumstances indicating that the selected rate
is aberrational or otherwise not appropriate as AFA” as the
quantity, price, and circumstances surrounding the trans-
actions were not unusual.
Id. at 6–7. As to BMW’s argu-
ment that the rate was “punitive” and inconsistent with
BMW’s “level of culpability,” Commerce responded that “it
appears that BMW is actually seeking to relitigate the De-
partment’s determination to apply facts available with an
adverse inference in the first instance.”
Id. at 16. Com-
merce further noted that a “respondent who has not coop-
erated in a review should not benefit from its non-
cooperation.”
Id. at 17. Continuing, Commerce explained
that “[a]ssigning to BMW a rate that is higher than its ex-
isting cash deposit rate [54.27 percent], or than NSK’s cal-
culated 1.43 percent rate, is adverse, not ‘punitive.’”
Id.
The Court of International Trade affirmed, concluding
simply that:
[Commerce’s] analysis indicates that the
126.44 percent rate, while high, was neither aber-
rational nor punitive. Commerce exercised its stat-
utory discretion in applying AFA and selected a
rate from the highest, non-aberrational, transac-
tion-specific margin. The selected rate strikes a
reasonable balance, as it serves the goal of inducing
cooperation with Commerce’s administrative re-
view procedures and is not based on an aberra-
tional rate.
BMW of North America LLC v. United States,
255 F. Supp. 3d 1342, 1347 (Ct. Int’l Trade 2017).
10 BMW OF NORTH AMERICA LLC v. UNITED STATES
DISCUSSION
BMW makes two primary arguments on appeal:
(1) that Commerce unlawfully reinstated the administra-
tive reviews after they had been terminated; and (2) that
Commerce’s selection of the 126.44% AFA rate was unduly
punitive and unsupported by substantial evidence. We ad-
dress each issue in turn.
I
We first address whether, after it discontinued the ad-
ministrative reviews, Commerce had the authority to re-
sume them following our court’s reversal of the Court of
International Trade’s revocation of the antidumping duty
orders. Commerce’s authority to conduct an administra-
tive review stems from 19 U.S.C. § 1675, and entails four
steps:
(1) Commerce publishes a notice of Opportunity to
Request an Administrative Review for the [period
of review] at issue; (2) upon receipt of such request,
Commerce publishes a notice of Initiation of an Ad-
ministrative Review in the Federal Register;
(3) Commerce, in order to obtain pertinent infor-
mation, distributes or makes available question-
naires to those entities Commerce designated in
the notice of Initiation; and (4) on the basis of the
information gathered, Commerce determines the
antidumping duty rates applicable to each entry or
type of entries and publishes these determinations
in the Federal Register.
Transcom, Inc. v. United States,
123 F. Supp. 2d 1372,
1374 (2000) (first citing 19 U.S.C. § 1675(a) (1994); then
citing 19 C.F.R. §§353.22, 353.31 (1995)). Under 19 U.S.C.
§ 1675(a)(3)(A), Commerce must “make a preliminary de-
termination . . . within 245 days after the last day of the
month in which occurs the anniversary of the date of the
publication of the order [for which review is requested].”
BMW OF NORTH AMERICA LLC v. UNITED STATES 11
BMW argues that Commerce’s notice that it was “dis-
continuing all unfinished administrative reviews” effec-
tively terminated the reviews, and, as such, Commerce had
to restart the four steps outlined in Transcom to conduct a
new administrative review. BMW further argues that
Commerce was required to publish its preliminary deter-
mination 245 days after the last date of the month in which
the order was published, which was January 31, 2012, and
thus could not resume the discontinued administrative re-
views on December 16, 2013, over a year after the statutory
deadline.
Commerce responds that the administrative reviews
were expressly “discontinued,” not “terminated.” Com-
merce further asserts that the revocation notice clearly in-
dicated that there was a pending appeal, that it did in fact
complete all four steps required in Transcom, and that it
had the discretion to suspend the statutory deadlines pend-
ing appeal.
We evaluate Commerce’s interpretation of a statute in
accordance with Chevron U.S.A., Inc. v. Natural Resources
Defense Council, Inc.,
467 U.S. 837 (1984). Under the
Chevron framework, we first examine “whether Congress
has directly spoken to the precise question at issue.”
Id.
at 842. If so, then both we and Commerce must defer to
Congressional intent.
Id. at 843. However, “if the statute
is silent or ambiguous with respect to the specific issue, the
question for the court is whether the agency’s answer is
based on a permissible construction of the statute.”
Id. We
will deem an “unreasonable resolution of [statutory] lan-
guage” impermissible. See Thai Plastic Bags Indus. Co. v.
United States,
746 F.3d 1358, 1364 (Fed. Cir. 2014) (quot-
ing United States v. Eurodif S.A.,
555 U.S. 305, 316 (2009)).
In Timken Co. v. United States, we held that Commerce
must publish notice of an adverse Court of International
Trade decision within ten days of issuance of the judgment,
and suspend liquidation of imports “until there is a
12 BMW OF NORTH AMERICA LLC v. UNITED STATES
conclusive decision in the action.”
893 F.2d 337, 341
(Fed. Cir. 1990). Timken, however, addressed the interpre-
tation of a statute that spoke directly to such matters. See
id. at 339 (explaining that the dispute centered around the
parties’ interpretation of the term “final decision” in
§ 1516(a)(e)). Here, there is no such statutory scheme that
addresses the effect of an adverse decision on an adminis-
trative review. While the statute does cover deadlines for
providing the results of an administrative review, it does
not state how revocation of the underlying order and sub-
sequent reinstatement would affect these deadlines.
Given that the relevant statutory framework is silent
on these issues, we proceed to Chevron step two and con-
sider whether Commerce’s interpretation—allowing it to
resume the discontinued reviews—is a permissible reading
of the statute. We conclude that it is. First, the statute
expressly grants Commerce the authority to conduct these
reviews upon request. See 19 U.S.C. § 1675. And the stat-
ute does not restrict Commerce’s authority on this point,
instead broadly directing Commerce to “review, and deter-
mine . . . the amount of any antidumping duty.” 19 U.S.C.
§ 1675(a)(1)(B). Next, there is no dispute that an underly-
ing antidumping order survives revocation and subsequent
reinstatement. It is reasonable, therefore, to interpret the
related administrative review thereof as similarly surviv-
ing revocation. Commerce, moreover, properly complied
with its administrative requirements in reinstating the re-
view by, for example, sending out a quantity-and-value-
questionnaire and publishing notice of the resumption of
administrative review in the Federal Register. Finally, al-
lowing Commerce to resume a prior-revoked review is con-
sistent with the language of § 1675(f), governing
suspension agreements. Under § 1675(f), when the reason
for a suspended investigation no longer exists, “the suspen-
sion agreement shall be treated as not accepted, beginning
on the date of publication of the Commission’s determina-
tion, and the administering authority and the Commission
BMW OF NORTH AMERICA LLC v. UNITED STATES 13
shall proceed . . . as if the suspension agreement had been
violated on that date.” (emphasis added). Similar to a neg-
ative determination of a suspension agreement, reinstate-
ment of a prior-revoked review should cause Commerce to
proceed as if the order had not been revoked. See King v.
Burwell,
135 S. Ct. 2480, 2489 (2015) (“[W]e must read the
words in their context and with a view to their place in the
overall statutory scheme.” (internal quotation marks omit-
ted)). We thus conclude that Commerce’s interpretation al-
lowing it to resume the discontinued reviews was not an
“unreasonable resolution.” Thai
Plastic, 746 F.3d at 1364
(quoting
Eurodif, 535 U.S. at 316). As such, Commerce’s
decision to resume was permissible.
To be sure, BMW’s stated belief that the revocation no-
tice terminated the administrative review is not unreason-
able. Commerce did not have guidelines on the process for
conducting an administrative review in the event of rein-
statement of a revoked antidumping duty order. Despite
both parties’ protestations, the word “discontinue” does not
have a clear meaning in this context, and the term is not
itself defined in any relevant statute or regulation. It is
unclear that Commerce even contemplated the possibility
of resuming the reviews at the time the revocation notice
was written.
Regardless, BMW’s subjective belief that its adminis-
trative review had been permanently terminated does not
impact the authority of Commerce to resume that review.
The sole question before us is whether the relevant statu-
tory framework permits such an action. With due defer-
ence to Commerce’s interpretation of the statute as
required by Chevron, we hold that it does.
II
BMW next argues that the 126.44% AFA rate selected
by Commerce is excessive, aberrational, unduly punitive,
and unsupported by substantial evidence. In particular,
BMW argues that because it was completely unaware that
14 BMW OF NORTH AMERICA LLC v. UNITED STATES
it had an administrative review pending, in large part be-
cause of the procedural anomalies that occurred during the
review, the 126.44% rate does not reflect its level of culpa-
bility nor its commercial reality.
When reviewing Commerce’s anti-dumping determina-
tions, this court “applies anew the standard of review ap-
plied by the Court of International Trade in its review of
the administrative record.” F.lli De Cecco Di Filippo Fara
S. Martino S.p.A. v. United States,
216 F.3d 1027, 1031
(Fed. Cir. 2000). “In doing so, we uphold Commerce’s de-
termination unless it is ‘unsupported by substantial evi-
dence on the record, or otherwise not in accordance with
law.’”
Id. (quoting 19 U.S.C. § 1516a(b)(1(B)(i)(1994)).
“This court reviews de novo the trial court’s answers to all
questions of law, including statutory interpretation ques-
tions; evidentiary decisions . . . are reviewed for abuse of
discretion.”
Id. “Although we review the decisions of the
CIT de novo, we give great weight to the informed opinion
of the CIT and it is nearly always the starting point of our
analysis.” Nan
Ya, 810 F.3d at 1341.
Under 19 U.S.C. § 1677e(b)(1), if Commerce “finds that
an interested party has failed to cooperate by not acting to
the best of its ability to comply with a request for infor-
mation” it “may use an inference that is adverse to the in-
terests of that party in selecting from the facts otherwise
available.”
Id. § 1677e(b)(1). Under § 1677e(b)(2), such an
adverse inference “may include reliance on information de-
rived from,” inter alia, “the petition” or “any other infor-
mation placed on the record.” We have recognized that the
“purpose of section 1677e(b) is to provide respondents with
incentive to cooperate, not to impose punitive, aberra-
tional, or uncorroborated margins.” De
Cecco, 216 F.3d
at 1032. The AFA rate is intended “to be a reasonably ac-
curate estimate of the respondent’s actual rate, albeit with
some built-in increase intended as a deterrent to non-com-
pliance.”
Id. While “a higher adverse margin creates a
BMW OF NORTH AMERICA LLC v. UNITED STATES 15
stronger deterrent,” Commerce must not “overreach reality
in seeking to maximize deterrence.”
Id.
We have also previously held that Commerce has wide
discretion to assign the “highest calculated rate” to unco-
operative parties. KYD, Inc. v. United States,
607 F.3d 760,
766 (Fed. Cir. 2010); see also Ta Chen Stainless Steel Pipe,
Inc. v. United States,
298 F.3d 1330, 1339 (Fed. Cir. 2002);
De
Cecco, 216 F.3d at 1029. We have sustained the use of
the highest calculated rate as “it reflects a common sense
inference that the highest prior margin is the most proba-
tive evidence of current margins because, if it were not so,
the importer, knowing of the rule, would have produced
current information showing the margin to be less.” Rhone
Poulenc, Inc. v. United States,
899 F.2d 1185, 1190
(Fed. Cir. 1990) (emphasis in original). The use of the high-
est rate is not automatic, however, and “will depend upon
the facts of a particular case.” Nan
Ya, 810 F.3d at 1347.
The Court of International Trade has held that Com-
merce must be “mindful of the purpose of § 1677(e)(b)”
when determining an AFA rate, and thus, an unusually
high rate is permissible when it is “necessary to serve the
purpose of deterrence[.]” Papierfabrik Aug. Koehler AG v.
United States,
180 F. Supp. 3d 1211, 1231–32 (Ct. Int’l
Trade 2016), aff’d sub nom. Papierfabrik Aug. Koehler SE
v. United States, 710 F. App’x 889 (Fed. Cir. 2018). In Pa-
pierfabrik, the Court of International Trade affirmed Com-
merce’s selection of an unusually high 75.36% AFA rate
when it was imposed to deter serious misconduct by the pe-
titioner, including deliberate falsity and intentional con-
cealment.
Id. In reaching this decision, the Court of
International Trade specifically looked to the facts sur-
rounding the uncooperative party’s failure to cooperate,
noting that the Department of Commerce’s finding that the
party had intentionally and fraudulently undermined the
entire proceeding was “an integral part of the Depart-
ment’s reasoning for imposing the highest rate in any pre-
vious segment of the proceeding.”
Id. The Court of
16 BMW OF NORTH AMERICA LLC v. UNITED STATES
International Trade further explained that “the serious-
ness of the type of misconduct Commerce was seeking . . .
to deter cause[d] the court to conclude that Commerce did
not overreach in assigning the 75.36% rate.”
Id. The court
thus based its decision on the complete factual record and
the seriousness of the conduct committed. We agree, and
common sense dictates, that Commerce should consider the
overall facts and circumstances of each case, including the
level of culpability of the non-cooperating party in an AFA
analysis. 3 And, contrary to the Dissent’s view that our case
law does not contemplate an inquiry into the seriousness of
the uncooperative party’s conduct when selecting a non-pu-
nitive AFA rate, see Dissent Op. at 3, we affirmed the Court
of International Trade’s decision in Papierfabrik, wherein,
3 Our decision relies on case law that interprets stat-
utory language that is identical in both versions of the stat-
ute pre-and-post-dating the amendments made by
Congress in the Trade Preferences Extension Act (“TPEA”),
and thus we need not reach BMW’s argument that Com-
merce improperly applied the TPEA to its May 2017 re-
mand determination. See Trade Preferences Extension Act
(TPEA) of 2015, Pub. L. No. 114-27, 129 Stat. 362, 383–87
(2015). In any event, the amended version of the statute
further supports our view that Commerce should consider
the level of culpability of the uncooperative party. Now,
under § 1677e(d)(2), “the administering authority may ap-
ply any of the countervailable subsidy rates or dumping
margins specified under [paragraph (1)], including the
highest such rate or margin, based on the evaluation by the
administering authority of the situation that resulted in the
administering authority using an adverse inference in se-
lecting among the facts otherwise available.” (emphasis
added). This language undermines the Dissent’s view that
it is unnecessary for Commerce to consider surrounding
facts and circumstances when selecting an appropriate
rate. See Dissent Op. 3.
BMW OF NORTH AMERICA LLC v. UNITED STATES 17
as discussed, the AFA rate selected was specifically based
on “the seriousness of the type of misconduct” committed
by the uncooperative party. Papierfabrik, 710 F. App’x 889
(Fed. Cir. 2018).
It is clear from these cases that Commerce has wide
discretion when it comes to selecting an AFA rate for non-
cooperating parties. See De
Cecco, 216 F.3d at 1033 (“Com-
merce, not the courts, must make determinations as to
proper anti-dumping margins.”). It is also clear, however,
that this discretion is not without limits. The appropriate
rate “will depend upon the facts of a particular case,”
Nan
Ya, 810 F.3d at 1347, cannot be “punitive, aberra-
tional, or uncorroborated,” De
Cecco, 216 F.3d at 1032, in-
cludes “some built-in increase” to deter non-compliance,
id.
(emphasis added), and, as noted above, reflects the serious-
ness of the non-cooperating party’s misconduct.
It is with this framework in mind that we consider
whether Commerce’s application of the 126.44% AFA rate
against BMW was proper. As Commerce correctly noted,
because the AFA rate was selected from primary, rather
than secondary, information, there is no corroboration re-
quirement. See Remand Determination at 8; see also
19 U.S.C. § 1677e(c). In addition, Commerce explained in
detail that the selected rate is the highest of a [redacted] of
the representative exporter’s transaction-specific margins
and that the quantity, price, and other circumstances sur-
rounding the transactions were not unusual. Remand De-
termination at 6–7. This is sufficient to demonstrate that
the selected rate is, in a literal sense, “non-aberrational.”
BMW also argued, however, that the selected AFA
rate was unduly punitive and inconsistent with its level of
culpability. Commerce largely ignored this argument, in-
stead reiterating that a “respondent who has not cooper-
ated in a review should not benefit from its non-
cooperation,” and that “[a]ssigning to BMW an AFA rate
that is higher than its existing cash deposit rate [54.27
18 BMW OF NORTH AMERICA LLC v. UNITED STATES
percent], or than [the representative exporter]’s calculated
1.43 percent rate, is adverse, not ‘punitive.’”
Id. at 17.
Though Commerce is at liberty to exercise its judgment and
select a rate it finds appropriate to deter non-compliance,
there is an extremely large range of rates between 1.43%
and 126.44%. Commerce did not consider or address
BMW’s argument regarding its mitigating circumstances
or explain why it determined that the 126.44% rate was
appropriate given the unique factual circumstances sur-
rounding BMW’s failure to return the quantity-and-value
questionnaire.
While we are mindful of the great deference given to
Commerce in making these determinations, without any
articulation of its rationale for finding that the selected
rate was not unduly punitive, we are limited in our review.
Our case law establishes that Commerce must consider the
totality of the circumstances in selecting an AFA rate, in-
cluding, if relevant, the seriousness of the conduct of the
uncooperative party. Here, Commerce did not address how
the procedural irregularities surrounding the administra-
tive review process affected its view of BMW’s level of cul-
pability. On this record, we cannot ascertain whether
Commerce properly selected an AFA rate that was not un-
duly punitive. See SEC v. Chenery Corp.,
318 U.S. 80, 94
(1943) (explaining that reviewing courts “cannot exercise
their duty of review unless they are advised of the consid-
erations underlying the action under review”). As we are
unable to affirm Commerce’s determination, we vacate the
judgment of the Court of International Trade and remand
for further proceedings.
CONCLUSION
We conclude that Commerce lawfully resumed BMW’s
administrative review after reinstatement of the related
antidumping duty order, but must explain its consideration
of the particular factual circumstances surrounding
BMW’s failure to cooperate with Commerce’s request for
BMW OF NORTH AMERICA LLC v. UNITED STATES 19
information when considering whether its selected AFA
rate of 126.44% was unduly punitive. Accordingly, we va-
cate the judgment of the Court of International Trade and
remand for further proceedings consistent with this opin-
ion.
VACATED AND REMANDED
COSTS
No costs.
United States Court of Appeals
for the Federal Circuit
______________________
BMW OF NORTH AMERICA LLC,
Plaintiff-Appellant
v.
UNITED STATES, TIMKEN COMPANY,
Defendants-Appellees
______________________
2018-1109
______________________
Appeal from the United States Court of International
Trade in No. 1:15-cv-00052-JCG, Judge Jennifer Choe-
Groves.
______________________
PROST, Chief Judge, dissenting-in-part.
After considering the facts and circumstances leading
to BMW’s failure to cooperate with the Department of Com-
merce’s (“Commerce”) request for information, Commerce
applied adverse facts available (“AFA”) to BMW, then se-
lected an AFA rate from actual sales information. In the
Majority’s view, Commerce did not do enough to justify its
selection of the AFA rate applied. According to the Major-
ity, Commerce must also explain, based on those factual
circumstances, why the AFA rate it selected was not un-
duly punitive. See Majority Op. 19. This request may seem
reasonable at first blush, for who could disagree with a sim-
ple request for further explanation? But the problem is
that Commerce simply has no such obligation under the
2 BMW OF NORTH AMERICA LLC v. UNITED STATES
statute or our case law. To be sure, the factual circum-
stances of a case may be relevant to deciding whether to
apply an AFA rate at all—a decision not contested here on
appeal. 1 But Commerce need not re-consider those facts
when setting the rate.
In my view, the Majority has erred by imposing new,
extra-statutory limits on the discretion that Congress
granted to Commerce. Worse yet, those limits are ill-de-
fined and amorphous, making Commerce’s job harder and
our review function difficult, if not impossible. See Nan Ya
Plastics Corp. v. United States,
810 F.3d 1333, 1345 (Fed.
Cir. 2016) (“[A] court errs in adopting an interpretation of
a statute that would create a tremendous burden on Com-
merce that is not required or suggested by the statute.” (in-
ternal citation and quotation marks omitted)). For
example, what are the subjective criteria to be applied in
assessing whether one AFA rate versus another is “unduly
punitive”? “Congress decided what requirements Com-
merce must fulfill in reaching its determinations,” and this
court should not “impose conditions not present in or sug-
gested by the statute’s text.”
Id. at 1347. Because the Ma-
jority does not heed this instruction, I respectfully dissent
from section II of the Majority’s opinion.
I
I begin with the language of the statute. Section
1677e(b) requires Commerce to perform fact-finding as a
condition precedent to using an adverse inference. See
19 U.S.C. § 1677e(b) (2012) (“If [Commerce] finds that an
interested party has failed to cooperate by not acting to the
best of its ability to comply with the request for infor-
mation . . . [Commerce] . . . may use an inference that is
adverse to the interests of that party in selecting from
1 Oral Argument at 1:54–59 (Q: “There’s no dispute
that an AFA rate is justified, right”? A: “That’s correct.”).
BMW OF NORTH AMERICA LLC v. UNITED STATES 3
among facts otherwise available.”). Thus, Commerce must
fully consider the circumstances leading to a respondent’s
noncooperation before it decides to draw an adverse infer-
ence in the first instance. Nippon Steel Corp. v. United
States,
337 F.3d 1373, 1382 (Fed. Cir. 2003) (“Before mak-
ing an adverse inference, Commerce must examine re-
spondent’s actions and assess the extent of respondent’s
abilities, efforts, and cooperation in responding to Com-
merce’s requests for information.”).
The statute does not require Commerce, contrary to the
Majority’s view, to re-consider those facts and circum-
stances when selecting an appropriate, non-punitive AFA
rate. See 19 U.S.C. § 1677e(b) (“Such adverse inference
may include reliance on information derived from (1) the
petition, (2) a final determination in the investigation un-
der this title, (3) any previous review under section 751 [19
U.S.C. § 1675] or determination under section 753 [19
U.S.C. § 1675b], or (4) any other information placed on the
record.”); see also Nippon
Steel, 337 F.3d at 1383 (Fed. Cir.
2003) (“[S]ection 1677e(b) does not by its terms set a ‘will-
fulness’ or ‘reasonable respondent’ standard, nor does it re-
quire findings of motivation or intent. Simply put, there is
no mens rea component to the section 1677e(b) inquiry.”).
Nor does our case law contemplate an inquiry into the
“seriousness of the conduct of the uncooperative party,”
Majority Op. 18, when selecting a non-punitive AFA rate,
as the Majority would require, see, e.g., Papierfabrik Au-
gust Koehler SE v. United States,
843 F.3d 1373, 1378–82
(Fed. Cir. 2016) (considering reasons for failure to cooper-
ate when determining the propriety of assigning an AFA
rate in the first place, but not when determining whether
substantial evidence supports Commerce’s selection of a
specific AFA rate); PAM, S.p.A v. United States,
582 F.3d
1336, 1339–40 (Fed. Cir. 2009) (same). To avoid selecting
a punitive AFA rate, Commerce must simply ensure that it
does not “overreach reality in seeking to maximize deter-
rence.” F.Lii de Cecco di Filippo Fara S. Martino S.p.A. v.
4 BMW OF NORTH AMERICA LLC v. UNITED STATES
United States,
216 F.3d 1027, 1032 (Fed. Cir. 2000). That
is all that is required. 2
With this legal framework in mind, I turn to the facts
of this case. In my view, Commerce easily satisfied its ob-
ligations under the statute and our case law.
II
First, Commerce considered the totality of circum-
stances leading to BMW’s unresponsiveness and found,
among other things, that there was “no reason to believe or
suspect that the email [Commerce sent to BMW’s counsel]
was not delivered,” that “every other respondent in this re-
view responded to the Q&V questionnaire,” and that
BMW’s “failure to cooperate need not be intentional” for an
AFA rate to apply. J.A. 738–39. 3 Accordingly, Commerce
2 The Majority seems to suggest that our case law
contemplates an inquiry into the seriousness of the unco-
operative party’s conduct when selecting an AFA rate be-
cause a panel of this court summarily affirmed under
Federal Circuit Rule 36 a Court of International Trade de-
cision that included such reasoning. Majority Op. 17. But
a Rule 36 judgment simply affirms the prior tribunal’s
judgment. TecSec, Inc. v. IBM,
731 F.3d 1336, 1350 (Fed.
Cir. 2013). “It does not endorse or sustain any specific part
of the prior tribunal’s reasoning.”
Id. Moreover, the spe-
cific issues raised in that appeal were unrelated to whether
the selected AFA was punitive. See Papierfabrik Aug.
Koehler SE v. United States, No. 16-2425, ECF No. 32 at 3
(Appellant’s Opening Brief).
3 “The statutory trigger for Commerce’s considera-
tion of an adverse inference is simply a failure to cooperate
to the best of respondent’s ability, regardless of motivation
or intent.” Nippon
Steel, 337 F.3d at 1383.
BMW OF NORTH AMERICA LLC v. UNITED STATES 5
decided, based on the specific circumstances of this case, to
assign an AFA rate to BMW. 4
Second, substantial evidence supports that BMW’s
AFA rate was a “reasonably accurate estimate of [BMW’s]
actual rate, albeit with some built-in increase intended as
a deterrent to non-compliance.” See De
Cecco, 216 F.3d at
1032. By definition, such a rate is adverse, not punitive.
See
id. At the outset, I note that BMW’s failure to cooper-
ate deprived Commerce of the most direct evidence of
BMW’s actual dumping margin. See KYD, Inc. v. United
States,
607 F.3d 760, 767 (Fed. Cir. 2010). And because
Commerce had never before reviewed BMW, Commerce
could use only the information it had at its disposal to de-
termine an appropriate AFA rate for BMW. Logically, it
used the sales information it received from a mandatory
respondent, NSK.
Further, there was more than sufficient basis for Com-
merce’s conclusion that 126.44% was reliable and well-
grounded because that margin was supported by “Com-
merce’s calculation of ‘high-volume transaction-specific
margins for cooperative companies which are both higher
than . . . and . . . close to that rate.’” See
KYD, 607 F.3d at
766 (quoting Universal Polybag Co. v. United States, 577 F.
Supp. 2d 1284, 1300–01 (Ct. Int’l Trade 2008)). Commerce
selected NSK’s “highest transaction-specific dumping mar-
gin that forms part of a closely-connected range of transac-
tion-specific margins,” 126.44%, as BMW’s AFA rate.
J.A. 1580. Specifically, Commerce selected NSK’s [re-
dacted] highest transaction-specific margin, and several
4 Because BMW did not contest on appeal Com-
merce’s decision to apply an AFA rate, the question before
this court was limited to whether the selected AFA rate
was supported by substantial evidence and was in accord-
ance with law.
6 BMW OF NORTH AMERICA LLC v. UNITED STATES
transaction-specific dumping margins fell above and close
to 126.44%. 5
Moreover, the Court of International Trade has upheld
AFA rates “based upon a transaction-specific margin
‘within the mainstream’ of the cooperating respondent’s
sales, particularly when the agency has not previously re-
viewed the uncooperative party and could not rely on that
party’s own deficient data to determine a rate,” as was the
case here. 6 iScholar, Inc. v. United States, 2011 Ct. Int’l
Trade LEXIS 3, *7, slip op. 2011-4 (Ct. Int’l Trade Jan. 13,
2011); see, e.g., Shanghai Taoen Int’l Trading Co. v. United
States,
360 F. Supp. 2d 1339, 1348 (Ct. Int’l Trade 2005)
(finding a party’s 223.01% AFA rate non-punitive because
it reflected recent commercial activity of a respondent who
cooperated in the review and the party’s failure to
5 Although the Majority opinion states that using
the highest transaction-specific margin as the AFA rate
will depend upon the facts of a particular case, Majority
Op. 17, the AFA rate that Commerce selected, 126.44%, is
far from the highest transaction-specific margin. It was
NSK’s [redacted] highest transaction-specific margin—[re-
dacted]% less than NSK’s highest transaction-specific mar-
gin ([redacted]%). J.A. 1560. To the extent that the
Majority relies on changes to the statute that became effec-
tive after Commerce issued its decision, Majority Op. 16–
17 n.3, the relevant portion of the statute is titled “Discre-
tion to apply highest rate.” 19 U.S.C. § 1677e(d)(2) (2015)
(emphasis added). Section 1677e(d)(2) is therefore limited
to application of the highest rate, which Commerce did not
select, and irrelevant here.
6 “A respondent’s ‘mainstream’ sales constitute those
‘transactions that reflect sales of products that are repre-
sentative of the broader range of models used to determine
normal value.’” iScholar, 2011 Ct. Int’l Trade LEXIS at *7
n.3 (citation omitted).
BMW OF NORTH AMERICA LLC v. UNITED STATES 7
accurately respond to Commerce’s questions resulted in an
egregious lack of record evidence to suggest an alternative
rate), aff’d, No. 10-1219, slip op. at 2 (Fed. Cir. Dec. 13,
2010). Yet the Majority opinion states that Commerce
failed to provide “any articulation of its rationale for find-
ing that the selected rate was not unduly punitive.” Major-
ity Op. 18. That, however, is not so.
Beyond selecting a rate within the mainstream of
NSK’s sales, Commerce reasoned that 126.44% does not
overreach reality because BMW argued that NSK’s overall
weighted-average dumping margin, 1.43%, reflected
BMW’s commercial reality. J.A. 1588. From BMW’s argu-
ment, Commerce concluded that NSK’s transaction-spe-
cific dumping margins—especially those falling on a
[redacted] that comprise that weighted-average dumping
margin—must have also reflected BMW’s commercial real-
ity.
Id.
Commerce also determined that, despite BMW’s claim
that the all-others rate of 54.27% would have a reasonable
connection to its commercial reality, J.A. 1586, assigning
BMW the all-others rate would not ensure that BMW “does
not obtain a more favorable result by failing to cooperate
than if it had cooperated fully.” J.A. 1591. Indeed, the all-
other’s rate is the same rate received by parties who did
not request administrative review and is lower than the
61.14% rate assigned to an individually-investigated re-
spondent who fully cooperated with Commerce. See BMW
of N. Am. LLC v. United States,
208 F. Supp. 3d 1388, 1397
(Ct. Int’l. Trade 2017). Even BMW concedes that it should
not receive a rate lower than the all-others rate. Oral Ar-
gument at 9:55–10:06 (Q: “It can’t be that when adverse
facts are inferred that the party that didn’t respond gets to
8 BMW OF NORTH AMERICA LLC v. UNITED STATES
have a rate that’s lower than it would have had had it not
asked for review, right”? A: “Well no, not lower than.”). 7
At bottom, “Commerce is in the best position, based on
its expert knowledge of the market and the individual re-
spondent, to select adverse facts that will create the proper
deterrent to non-cooperation with its investigations and as-
sure a reasonable margin.” De
Cecco, 216 F.3d at 1032.
While choosing an AFA rate “with an eye toward deter-
rence, Commerce acts within its discretion so long as the
rate chosen has a relationship to the actual sales infor-
mation available.” Ta Chen Stainless Steel Pipe, Inc. v.
United States,
298 F.3d 1330, 1340 (Fed. Cir. 2002). The
Majority’s additional requirement, factoring in a party’s
“level of culpability” when choosing a particular AFA rate,
is therefore misguided.
While that information may be relevant to Commerce’s
decision to apply an AFA rate in the first place, it need not
inform Commerce’s actual rate selection decision.
Nippon,
337 F.3d at 1383 (explaining that § 1677e(b) does not re-
quire Commerce to evaluate overall patterns of behavior,
7 While BMW does take issue with the fact that its
AFA rate is “more than twice as high as the 54.27 percent
all others rate,” Appellant’s Br. 48, that fact alone is insuf-
ficient to deem its AFA rate punitive. For example, in De
Cecco—the case on which the Majority relies for the propo-
sition that an AFA rate must not be punitive—this court
determined that 24.31% was adverse, not punitive, even
though the all-others rate was only
12.09%. 216 F.3d at
1034. And, in Viet I-Mei Frozen Foods Co. v. United States,
this court upheld an AFA rate that was five times the all-
others rate even though the disadvantaged party lodged
several justifications for its failure to cooperate.
839 F.3d
1099, 1109–10 (Fed. Cir. 2016).
BMW OF NORTH AMERICA LLC v. UNITED STATES 9
examine reasons for initial mistakes, or to permit explana-
tions of the extenuating circumstances).
Because substantial evidence supports Commerce’s
AFA rate selection, and it was in accordance with the law,
I would affirm the Court of International Trade’s judg-
ment.
