American Meat Institute v. United States Department of Agriculture

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued January 9, 2014 Decided March 28, 2014

No. 13-5281

AMERICAN MEAT INSTITUTE, ET AL.,
APPELLANTS

v.

UNITED STATES DEPARTMENT OF AGRICULTURE, ET AL.,
APPELLEES

Appeal from the United States District Court
for the District of Columbia
(No. 1:13-cv-01033)

Catherine E. Stetson argued the cause for appellants.
With her on the briefs were Jonathan L. Abram, Judith E.
Coleman, Mary Helen Wimberly, and Elizabeth B. Prelogar.

Daniel Tenny, Attorney, U.S. Department of Justice,
argued the cause for appellees. With him on the brief were
Stuart F. Delery, Assistant Attorney General, Ronald C.
Machen Jr., U.S. Attorney, and Mark B. Stern, Attorney.

Terence P. Stewart was on the brief for intervenors
United States Cattlemen’s Association, et al. in support of
appellees.
 2

Zachary B. Corrigan was on the brief for amici curiae
Food and Water Watch, Inc., et al. in support of appellees.

Jonathan R. Lovvorn and Aaron D. Green were on the
brief for amicus curiae American Grassfed Association, et al.
in support of appellees.

Before: GARLAND, Chief Judge, SRINIVASAN, Circuit
Judge, and WILLIAMS, Senior Circuit Judge.

Opinion for the Court filed by Senior Circuit Judge
WILLIAMS.

WILLIAMS, Senior Circuit Judge: In 2013 the
Agricultural Marketing Service (“AMS”), a branch of the
Department of Agriculture, adopted a rule modifying its prior
rule implementing Congress’s requirements of country-of-
origin labeling (“COOL”). Mandatory Country of Origin
Labeling, 78 Fed. Reg. 31,367 (May 24, 2013) (“2013 rule”).
The rule requires retailers of “muscle cuts” of meat, i.e.,
covered meat other than ground meat (which is governed by 7
U.S.C. § 1638a(a)(2)(E)), to list (with some qualifications) the
countries of origin and production steps—born, raised or
slaughtered—occurring in each country. Id. at 31,367/3. The
AMS’s previous rule had merely required a list of the
countries of origin (again with some qualifications) preceded
by the phrase “Product of.” Mandatory Country of Origin
Labeling, 74 Fed. Reg. 2658, 2706 (Jan. 15, 2009) (“2009
rule”). The 2013 rule also eliminated the prior rule’s
allowance for commingling—a practice by which cuts from
animals of different origins, but processed on the same day,
could all bear identical labels.

The appellants, a group of trade associations representing
livestock producers, feedlot operators, and meat packers,
whom we’ll collectively call American Meat Institute
 3

(“AMI”), challenged the 2013 rule in district court as a
violation of the COOL statute and the First Amendment. AMI
moved for a preliminary injunction halting enforcement, and
the district court denied the motion. Agreeing with the district
court that AMI is unlikely to succeed on the merits of its
claims, and believing that any error in the district court’s
balancing of the other factors governing issuance of a
preliminary injunction could not on these facts outweigh the
likely outcome on the merits, we affirm.

* * *

The COOL statute, 7 U.S.C. § 1638a, adopted in 2008,
assigns retailers an obligation to inform consumers of a cut’s
country of origin. This may be quite complicated where an
animal was born, raised, and slaughtered in more than one
country. Id. § 1638a(a)(2). The statute sets forth four
categories of muscle-cut meat and how to determine the
country of origin depending on the locale of the production
steps:

(A) United States country of origin[.] A retailer . . . may
designate the covered commodity as exclusively having a
United States country of origin only if the covered
commodity is derived from an animal that was . . .
exclusively born, raised, and slaughtered in the United
States . . . .

(B) Multiple countries of origin[.] A retailer of a covered
commodity . . . that is derived from an animal that is (I)
not exclusively born, raised, and slaughtered in the
United States; (II) born, raised, or slaughtered in the
United States, and (III) not imported into the United
States for immediate slaughter, may designate the country
of origin of such covered commodity as all of the
 4

countries in which the animal may have been born,
raised, or slaughtered.

(C) Imported for immediate slaughter[.] A retailer of a
covered commodity . . . that is derived from an animal
that is imported into the United States for immediate
slaughter shall designate the origin . . . as . . . the country
from which the animal was imported; and . . . the United
States.

(D) Foreign country of origin[.] A retailer of a covered
commodity . . . that is derived from an animal that is not
born, raised, or slaughtered in the United States shall
designate a country other than the United States as the
country of origin . . . .

Id. (emphases added). The parties call meat covered by
§ 1638a(a)(2)(A) “Category A meat,” that covered by
§ 1638a(a)(2)(B) “Category B meat,” and so on. The COOL
statute also requires the Secretary of Agriculture to
“promulgate such regulations as are necessary to implement”
the statutory regime. Id. § 1638c(b).

The 2009 rule did not demand explicit identification of
the country for each of the three production steps—born,
raised and slaughtered. It called more simply for labeling
with a phrase starting “Product of,” followed by mention of
one or more countries. 7 C.F.R. § 65.400 (2010). So
Category A meat would be labeled, “Product of the United
States”; Category B meat would be labeled, “Product of the
United States and X”; Category C meat would be labeled,
“Product of X and the United States”; and Category D meat
would be labeled “Product of X.” See id.; see also id.
§ 65.300 (2010).
 5

The 2009 rule also made allowance for a production
practice known as “commingling.” This occurs when a firm
processes meat from animals with different countries of origin
on a single production day. 7 C.F.R. § 65.300(e)(2), (e)(4)
(2010). The rule allowed retailers to label commingled meat
cuts with all the countries of origin for all the commingled
animals. As a result, Category A meat processed on the same
day as Category B or C meat could be labeled “Product of
United States and X.” Id.

In the year of the 2009 rule’s adoption, Canada and
Mexico filed a complaint with the Dispute Settlement Body of
the World Trade Organization, which found the rule to be in
violation of the WTO Agreement on Technical Barriers to
Trade. 2013 rule, 78 Fed. Reg. at 31,367/2. The gravamen of
the WTO’s ruling appears to have been an objection to the
relative imprecision of the information required by the 2009
rule. See Appellate Body Report, United States—Certain
Country of Origin Labelling (COOL) Requirements, ¶ 343,
WT/DS384/AB/R (Jun. 29, 2012). A WTO arbitrator gave
the United States until May 23, 2013, to bring its COOL
requirements into compliance with the ruling. 2013 rule, 78
Fed. Reg. at 31,367/2.

The 2013 rule increased the required level of precision.
Now, except for Category D meat, each country of origin
would generally be preceded by the production step that
occurred in that country. Id. at 31,385/3. For instance,
instead of saying, “Product of the United States,” a label for
Category A meat will now read, “Born, Raised, and
Slaughtered in the United States.” Id. Similarly, Category B
meat might now have to be labeled, “Born in X, Raised and
Slaughtered in the United States,” and Category C meat “Born
and Raised in X, Slaughtered in the United States.” Id. The
2013 rule also eliminated the special allowance for
commingled meat. Id. at 31,367/3.
 6

AMI challenged the 2013 rule in district court as (1)
exceeding the authority granted by the COOL statute, and (2)
violating the First Amendment. AMI also moved for a
preliminary injunction halting enforcement of the 2013 rule,
which the district court denied. AMI contends on appeal to us
that the district court erred in its determination that AMI is
unlikely to succeed on the merits of either claim. We review
questions of law—AMI’s substantive claims—de novo.
Sherley v. Sebelius, 644 F.3d 388, 393 (D.C. Cir. 2011).
Because we disagree with AMI on its chances of success on
the merits, we affirm the district court.

* * *

At oral argument the question arose whether AMI has
standing to raise its claims. None of the appellants is a
retailer, the type of market actor expressly covered by the bulk
of the COOL requirements. See 7 U.S.C. § 1638a(a). But
§ 1638a(e) requires that upstream producers “provide
information to the retailer indicating the country of origin of
the covered commodity.” In effect, then, the appellants are
required to make the same disclosures that retailers are, only
to a different recipient. Accordingly, we are satisfied that the
challenged regulations inflict on AMI the sort of injury-in-fact
needed for Article III standing. See Lujan v. Defenders of
Wildlife, 504 U.S. 555, 560 (1992).

We thus turn to AMI’s arguments as to why the COOL
statute does not authorize the 2013 rule: (1) the rule “bans”
commingling, and therefore alters production practices over
which the COOL statute gives the Secretary no authority; and
(2) production-step labeling is both outside of and contrary to
the plain language of the COOL statute. We are not
persuaded.
 7

AMI’s argument that the rule unlawfully “bans”
commingling fails at a key first step—the 2013 rule does not
actually ban any element of the production process. It simply
requires that meat cuts be accurately labeled with the three
phases of production named in the statute. It appears that
under current practices meat packers cannot achieve that
degree of accuracy with commingled production. The
necessary changes to production are, to be sure, costly for the
packers, but, contrary to AMI’s claim, the new rule does not
“force the segregated handling of animals with varying
geographical histories,” except in the sense that compliance
with any regulation may induce changes in unregulated
production techniques that a profit-seeking producer would
not otherwise make.

This practical burden on an existing practice would be
problematic if the statute required an exception for the
practice. But AMI points, at most, to a statutory ambiguity on
the issue of commingling. AMI contends that since
§ 1638a(a)(2)(B)(i) allows retailers to designate “all of the
countries in which the animal may have been born, raised, or
slaughtered” (emphasis added), the statute expressly
contemplates an allowance, where animals processed in a
single day have traversed different countries, for listing all
such countries, as did the 2009 rule. Although the use of
“may have been” in § 1638a(a)(2)(B)(i) is perhaps ambiguous,
it by no means renders the absence of a commingling
allowance unreasonable. In a section dealing with ground
meat Congress expressly authorized retailers to provide “a list
of all reasonably possible countries of origin.” Id.
§ 1638a(a)(2)(E). This at least hints at the kind of flexibility
AMI desires—in a different context. In contrast, Category
B’s use of the words “may have been” appears next to
references to “an” animal and “the” animal. See id.
§ 1638a(a)(2)(B)(i). Congress’s use of singular articles
certainly supports the agency’s reading of the statute as
 8

allowing it to require labels reflecting the origin of the actual
animal from which a cut derives, rather than just the origin of
any animals that may have been processed on the same day.
We return to AMI’s use of the “may have been” language in
connection with its next argument—the only one in which it
actually relies on “may have been.”

AMI also contends that the entire production-step
labeling regime—the rule’s requirement that each animal have
what AMI calls a “passport”—is inconsistent with the statute.
First, AMI argues that the statute authorizes the agency only
to require a list of the countries of origin, not a breakdown of
which production step occurred where. But the statute
ubiquitously invokes distinctions between three phases of
production—where the animal from which a cut derives was
born, raised, and slaughtered—so that the agency’s choice to
require labels linking each step to the relevant country appears
reasonable.

Second, AMI contends the regulations are in direct
conflict with what it views to be permissive language
regarding Category B meat: a retailer “may designate the
country of origin . . . as all of the countries in which the
animal may have been born, raised, or slaughtered.”
§ 1638a(a)(2)(B)(i) (emphases added). But the next
subparagraph, § 1638a(a)(2)(B)(ii), reminds the reader that
“nothing in [subparagraph (i)] alters the mandatory
requirement to inform consumers of the country of origin of
covered commodities under [§ 1638a(a)(1)].” With that in
mind, it seems a stretch to read the “may” and “may have
been” language as either rendering compliance with
subsection B permissive (which even AMI seems not to
advocate) or assuring producers that they may mingle their
cattle in a such a way that they can only guess at a particular
animal’s migrations.
 9

The agency does in fact allow leeway within category B.

If an animal is raised in the United States as well as
another country (or multiple countries), the raising
occurring in the other country (or countries) may be
omitted from the origin designation except if the
animal was imported for immediate slaughter as
defined in § 65.180 or where by doing so the muscle
cut covered commodity would be designated as having
a United States country of origin (e.g., “Born in
Country X, Raised and Slaughtered in the United
States” in lieu of “Born and Raised in Country X,
Raised in Country Y, Raised and Slaughtered in the
United States”).

2013 rule, 78 Fed. Reg. at 31,385/3 (quoting 7 C.F.R.
§ 65.300(e) as amended by the 2013 rule). It thus assures
flexibility, bounded mainly by precluding attribution entirely
to the United States in cases where another country has also
played a role in the three-step process. Despite AMI’s two
objections, AMS’s interpretation of the statute is a reasonable
one, and thus entitled to be upheld. See Entergy Corp. v.
Riverkeeper, Inc., 556 U.S. 208, 218 (2009); Am. Elec. Power
Serv. Corp. v. FCC, 708 F.3d 183, 186 (D.C. Cir. 2013).

* * *

AMI argues that compulsion to make the disclosures
required by the 2013 rule violates its First Amendment rights.
Its first step in this contention is that we should apply the
general test for commercial speech formulated in Central
Hudson Gas & Electric Corp. v. Public Service Commission,
447 U.S. 557, 566 (1980), rather than that of Zauderer v.
Office of Disciplinary Counsel, 471 U.S. 626, 651 (1985), a
 10

standard that applies only to requirements that a commercial
actor disclose factual and non-controversial information.

To begin, all parties agree that the rule involves
commercial speech. In addition, it restricts speech only in the
sense of requiring a disclosure, a prerequisite to invoking
Zauderer. See id. at 650-51. Finally, the disclosure is purely
factual and non-controversial. Unlike the challengers in
United States v. United Foods, Inc., 533 U.S. 405, 411 (2001),
or R.J. Reynolds Tobacco Co. v. FDA, 696 F.3d 1205, 1212,
1216-17 (D.C. Cir. 2012), AMI has not articulated an
objection to the content of the message conveyed by the
mandated speech. While it has objected to the term
“slaughtered,” it has not expressed any problem with the
euphemism that the 2013 rule allows retailers to substitute—
“harvested.” 78 Fed. Reg. at 31,368/2.

AMI invokes International Dairy Foods Association v.
Amestoy, 92 F.3d 67 (2d Cir. 1996), in which the court
invalidated a Vermont statute requiring dairy manufacturers to
disclose treatment of their cows with recombinant Bovine
Somatotropin (“rBST”), a treatment that the U.S. Food and
Drug Administration had found to have no significant effect
on the milk. The government (although disagreeing with the
case) suggests that the disclosure required there might have
been seen by consumers “as a concession that the treatment
might affect the quality of the milk,” Resp. Br. at 31, and thus
a more significant intrusion on First Amendment rights than
the disclosure here. Although the government later seeks to
justify the COOL requirements as possibly reassuring
consumers who are anxious about potentially lax foreign
practices, it seems a good deal less likely that consumers
would draw negative hints from COOL information than from
the required declarations about use of rBST. Reference to an
apparently novel additive on milk cartons might well lead to
an inference that the additive might have a dangerous effect,
 11

whereas the appearance of countries of origin on packages of
meat seems susceptible to quite benign inferences, including
simply that the retailers take pride in identifying the source of
their products. Accordingly, without resolving whether
Amestoy was correctly decided, we find it distinguishable and
(if correct) no obstacle to characterizing the disclosure here as
purely factual and non-controversial.

In the case of a rule mandating such a disclosure,
Zauderer found Central Hudson review—particularly its
“least restrictive alternative” element—to be unnecessary.
Zauderer, 471 U.S. at 651 & 651-52 n.14. Reasoning that
commercial speech warrants protection mainly due to its
information-producing function, the Supreme Court found that
a commercial actor has only a “minimal” First Amendment
interest in not providing purely factual information with
which the actor does not disagree. Id. at 651. Such mandates
do not violate an advertiser’s First Amendment rights, it said,
“as long as disclosure requirements are reasonably related to
the State’s interest in preventing deception of consumers.” Id.

AMI would read that formula as excluding all other
justifying interests. Neither party has called our attention to
any Supreme Court case extending Zauderer beyond
mandates correcting deception, and we have found none.
Other circuits, however, have extended it to, for example,
government interests in telling buyers that mercury-containing
light bulbs do contain mercury and may not be disposed of
until steps have been taken to “ensure that [the mercury] does
not become part of solid waste or wastewater,” Nat’l Elec.
Mfrs. Ass’n v. Sorrell, 272 F.3d 104, 107 n.1 (2d Cir. 2001),
and in alerting health benefit providers of the background
decisions made by pharmacy benefit managers in their sales to
the providers, Pharm. Care Mgmt. Ass’n v. Rowe, 429 F.3d
294, 298-99, 308-10 (1st Cir. 2005) (Torruella, J.); id. at 316
(Boudin, C.J. & Dyk, J.) (giving Zauderer a very broad
 12

reading); id. at 297-98 (per curiam) (explaining that the
opinion of Chief Judge Boudin and Judge Dyk is controlling
on the First Amendment issue). Although AMI’s preferred
analysis has an appealing symmetry (deception as the evil to
be corrected, disclosure of purely factual and non-
controversial information as the permissible cure), Zauderer’s
characterization of the speaker’s interest in opposing forced
disclosure of such information as “minimal” seems inherently
applicable beyond the problem of deception. See, e.g., N.Y.
Rest. Ass’n v. N.Y. City Bd. of Health, 556 F.3d 114, 133 (2d
Cir. 2009) (applying Zauderer to requirement that restaurant
menus include calorie content information).

AMI argues, however, that our prior decisions in
Reynolds and National Association of Manufacturers v.
NLRB, 717 F.3d 947, 959 n.18 (D.C. Cir. 2013) (“NAM”),
held that Zauderer applied only to disclosure mandates aimed
at correcting deception. Indeed those opinions contain
language quoting or echoing Zauderer’s reference to that
specific interest. Reynolds, 696 F.3d at 1213; NAM, 717 F.3d
at 959 n.18. We do not believe that these passages are
correctly construed as holdings.

In the first place, both decisions pointed to features of
those cases that render wholly inapplicable Zauderer’s
characterization of the speaker’s interest as “minimal”: they
rejected any idea that the mandated disclosures were of
“purely factual and uncontroversial” information. Reynolds,
696 F.3d at 1212 (quoting Zauderer, 471 U.S. at 651). In
Reynolds we found that the “inflammatory images and the
provocatively-named hotline [could not] rationally be viewed
as pure attempts to convey information to consumers.” 696
F.3d at 1216-17. And in NAM we approvingly cited
plaintiffs’ description of the notice they were required to post
“as one-sided, as favoring unionization,” because the required
notice made no mention of other worker rights that were
 13

highly relevant to those required to be highlighted in the
mandated notices. 717 F.3d at 958. In cases where there was
clearly no basis for classifying the speaker’s interest as
minimal, it is hard to read the court’s use of Zauderer’s
language (which formulated the rule in terms of the facts
before it, i.e., a mandate aimed at curing deception) as a
holding that would preclude Zauderer’s application to
mandates justified by other interests. Indeed, in Reynolds, in
the very paragraph quoting Zauderer’s language about
deception, we went on to incorporate language from Pacific
Gas & Electric Co. v. Public Utilities Commission, 475 U.S.
1, 15-16 n.12 (1986), precluding Zauderer’s application to
messages “biased against or . . . expressly contrary to the . . .
views” of the entity subject to the mandate. 696 F.3d at 1213-
14. Reynolds’s amalgamation of distinctions—the problem to
be cured and the character of the mandate—militates against
viewing it as a holding that the first alone was fatal to
Zauderer review.

NAM in fact did not apply the First Amendment at all, but
rested instead on 29 U.S.C. § 158(c), which, it carefully
explained, goes significantly beyond merely incorporating the
First Amendment. 717 F.3d at 955. In a footnote, to be sure,
we addressed an NLRB footnote invoking Zauderer, and there
we relied on the anti-deception purpose present in Zauderer.
Id. at 959 n.18. But in a case turning on a statute, a footnote
response to a party’s footnote on a constitutional issue
altogether lacks the earmarks of a constitutional holding.

Finding that Zauderer is best read as applying not only to
mandates aimed at curing deception but also to ones for other
purposes, and that neither Reynolds nor NAM represents a
holding to the contrary, we adopt that reading, with the
 14

incidental advantage of avoiding the creation of a split with
the First and Second Circuits.1

What then are the government interests here? AMI
argues that the rule merely satisfies consumers’ curiosity. But
we can see non-frivolous values advanced by the information.
Obviously it enables a consumer to apply patriotic or
protectionist criteria in the choice of meat. And it enables one
who believes that United States practices and regulation are
better at assuring food safety than those of other countries, or
indeed the reverse, to act on that premise. See, e.g., 148
CONG. REC. H1538 (daily ed. Apr. 24, 2002) (statement of
Rep. Hooley, co-sponsor of COOL amendment to 2002 Farm
Bill) (asserting possible consumer interests in food safety and
in favoring American producers); 149 CONG. REC. S14,117
(daily ed. Nov. 6, 2003) (statement of Sen. Johnson) (same).
We cannot declare these goals so trivial or misguided as to fall
below the threshold needed to justify the “minimal” intrusion
on AMI’s First Amendment interests. Thus AMI has failed to
show a likelihood of success on the merits.

* * *

Besides the plaintiff’s likelihood of success on the merits,
grant of a preliminary injunction also turns on the existence of
irreparable harm, the balance of equities, and the public
interest. Sherley, 644 F.3d at 392. This circuit has repeatedly

1
We recognize that reasonable judges may read Reynolds as
holding that Zauderer can apply only where the government’s
interest is in correcting deception. Accordingly, we suggest that the
full court hear this case en banc to resolve for the circuit whether,
under Zauderer, government interests in addition to correcting
deception can sustain a commercial speech mandate that compels
firms to disclose purely factual and non-controversial information.
 15

declined to take sides in a circuit split on the question of
whether likelihood of success on the merits is a freestanding
threshold requirement to issuance of a preliminary injunction.
Id. at 393. We need not take sides today. Even if the sliding
scale approach to assessing eligibility for preliminary
injunctions survived Winter v. Natural Resources Defense
Council, Inc., 555 U.S. 7 (2008), a plaintiff with a weak
showing on the first factor would have to show that all three
of the other factors “so much favor the plaintiffs that they
need only have raised a ‘serious legal question’ on the
merits.” Sherley, 644 F.3d at 398. Given that plaintiffs’ lack
of success on the merits turns on the regulation’s surviving
Zauderer’s balancing test, it would be remarkable if we could
find an abuse of discretion in the district court’s finding
against plaintiffs. There is, moreover, a public interest factor
that we did not consider in our constitutional analysis, that of
allowing the United States’s effort to comply with the WTO
ruling to take effect. We are clearly in a poor position to
assess the effects of any noncompliance. Accordingly, the
judgment of the district court is

Affirmed.