R J Reynolds Tobacco
Co. v. Food & Drug Admin., 2024 WL 1208111, — F.4th —-, No. 23-40076 (5th
Cir. Mar. 21, 2024)
The sudden shift in the political valence of the
commercial speech doctrine
strikes again! The Fifth Circuit upholds mandatory cigarette warnings as
acceptable compelled commercial speech under Zauderer, reversing the
district court’s 2022 decision. Let’s just say that, five years ago, this would
have struck me as an unlikely result, and in 2020 the decision to file in Texas
would have been much less complicated; even in 2022, I would have expected the
district court to be upheld. (It returns to the district court for an APA
challenge, about which I express no opinion.)
The Family Smoking
Prevention and Tobacco Control Act requires cigarette packages to include
“color graphics depicting the negative health consequences of smoking to
accompany the [updated] label statements.” These warnings “shall comprise the
top 50 percent of the front and rear panels of the package” of cigarettes and
“at least 20 percent of the area of [any] advertisement ….” A facial
challenge was rejected by the Sixth Circuit in 2012, but the DC Circuit struck
down the FDA’s first attempt on an as-applied challenge. Now it’s the 5th
Circuit’s turn.
In enacting the TCA,
Congress found that “efforts to restrict advertising and marketing of tobacco
products,” including existing mandatory warnings, had “failed adequately to
curb tobacco use by adolescents, [so] comprehensive restrictions on the sale,
promotion, and distribution of such products [were] needed.” The TCA’s
legislative findings included: (1) minors still often see and are exposed to
tobacco product advertising; (2) the “overwhelming majority of Americans who
use tobacco products begin using such products while they are minors and become
addicted to the nicotine in those products before reaching the age of 18” and
(3) “[r]educing the use of tobacco by minors by 50 percent would prevent well
over 10,000,000 of today’s children from becoming regular, daily smokers,
saving over 3,000,000 of them from premature death due to tobacco-induced
disease[s]” and would “result in approximately $75,000,000,000 in savings
attributable to reduced health care costs.”
Congress identified
nine new warnings to rotate regularly, which must “comprise the top 50 percent
of the front and rear panels of” each cigarette package and “at least 20
percent of the area of [any] advertisement ….” It further instructed the
Secretary of Health and Human Services to “issue regulations that require color
graphics depicting the negative health consequences of smoking to accompany the
label statements.” And Congress gave the Secretary the authority to “adjust the
type size, text and format of the label statements” for clarity,
conspicuousness, and legibility.
When the FDA made
its first attempt, the DC Circuit held that the chosen graphics were not
targeted at deception; nor were they providing “‘purely factual and
uncontroversial’ information” because the images “could be misinterpreted by
consumers” and “are primarily intended to evoke an emotional response, or, at
most, shock the viewer into retaining the information in the text warning.” It
therefore applied Central Hudson instead of Zauderer and struck
down the initial rule. Under Central Hudson, the FDA lacked even
“a shred of evidence … showing that the graphic warnings will ‘directly
advance’ [FDA’s] interest in reducing the number of Americans who smoke.”
The FDA reasoned
that its new images promoted “the Government’s interest in promoting greater
public understanding of the negative health consequences of cigarette smoking”
and also “dissipat[es] the possibility of consumer confusion or deception,”
thereby advancing the government’s interest in preventing “consumer
misperceptions regarding the risks presented by cigarettes.”
Plaintiffs here alleged
that each of the Warnings “misrepresent[s] or exaggerate[s] the potential
effects of smoking” and that, “[c]ontrary to FDA’s characterization, the peer
reviewers raised serious, substantive concerns about FDA’s studies” used to
support the selected Warnings.
The district court
reasoned that Zauderer did not apply because the imagery was
fundamentally so “prone to ambiguous interpretation” that “it is unclear how a
court would go about determining whether it[ ] … is ‘accurate’ and ‘factual’
in nature”:
In other words, the court reasoned that no photorealistic image could
ever be purely factual and uncontroversial because different viewers will
ascribe to it different meanings. The inherent ambiguity in any graphic
warning—e.g., that viewers may interpret the heart disease warning to suggest
that open-heart surgery “is the most common treatment for heart disease” or the
best—means that the Warnings cannot be “ ‘purely factual and uncontroversial’
and objectively accurate as required to allow relaxed Zauderer review.”
Further, the court found that the graphic portions of the Warnings fell beyond Zauderer’s
reach because they are inherently “provocative.”
And the warnings
weren’t narrowly tailored under Central Hudson because the
government hadn’t first tried increased
funding for antismoking advertisements, increased government anti-smoking
communications, or “test[ed] the efficacy of ‘smaller or differently placed
warnings.’ ”
(Preclusion as to
RJR’s challenge to the constitutionality of the TCA itself would have been
appropriate, but that didn’t resolve the case (there were other plaintiffs), so
the court proceeded to the merits.)
Key holding: “The
Warnings are both factual and uncontroversial, despite the emotional impact the
graphics may have.”
The court—weighing
in on an issue that divided the DC Circuit—concluded that Zauderer is a “carve-out” from, not an
application of, Central Hudson.
Moreover (and not
unrelatedly), Zauderer applies to all compelled commercial
speech, not just deception-preventing speech. The Fifth Circuit held in NetChoice
that the state’s interest in “enabling
users to make an informed choice regarding whether to use [social media]
Platforms” was sufficient to survive review under Zauderer. Similarly, Chamber
of Commerce of the USA v. SEC, 85 F.4th 760 (5th Cir. 2023), recently held that
“the disclosure of a company’s rationale for a stock buyback was purely factual
and uncontroversial commercial speech” (although it still struck down the SEC’s
action because it was the SEC, I mean because of the APA).
First, the warnings
were “purely factual.” What is factual? Well, it’s not an opinion. Moreover,
the government may not demand a private party “undertake contextual analyses,
weighing and balancing many factors … that depend on community standards,” to
determine the speech it must “parrot.” Book People, Inc. v. Wong, 91 F.4th 318,
340 (5th Cir. 2024). “Factual” needs to involve “information” that is “[c]oncerned
with what is actually the case rather than interpretations of or reactions to
it” and “actually occurring.” (Lots of dictionaries invoked here.) Thus,
“factual” must mean “falsifiable material and inferences fairly drawn from it,
rather than one’s non-falsifiable ‘interpretations[,] … reactions,’ or
opinions.”
Crucially, “factual”
does not mean “true,” because that would make “purely factual information”—the
language of Zauderer—surplusage. (This seems to ignore the idea that
opinions aren’t statutes, but here we are.) Thus, the required warnings would
be factual if they were comprised of “only (a) information supported by facts
and (b) conclusions driven by those facts, and (2) not akin to unfalsifiable
statements of opinion.”
Plaintiffs argued
that the new warnings “misleadingly exaggerate smoking risks” and improperly
“focus on conditions that less frequently arise from smoking,” even though the
existing warnings were concededly purely factual. “Consequences supported by
scientific findings, even if exaggerated or non-modal, are still, by
definition, factual.” The factual content of the text was
undisputed.
What about the images? Images can be factual. “The addition of images to the textual
warnings makes no difference to the constitutional analysis of factuality.” In
the FDA’s own words: “FDA used a certified medical illustrator to design images
that depicted common visual presentations of the health conditions and/or
showed disease states and symptoms as they are typically experienced, and that
present the health conditions in a realistic and objective format devoid of
non-essential elements.” Each of the images was “a straightforward,
science-based, objectively truthful depiction of the accompanying text,” “no
different from those a medical student might see in a textbook.”
Merely because the
images might convey “an ideological or provocative message” does not make them
nonfactual:
A fact does not become “value-laden” merely because the fact drives a
reaction. But even if it did, ideological baggage has no relevance to the first
Zauderer prong. Any number of factual messages are, of course,
ideological. Similarly, emotional response to a statement is irrelevant to its
truth. That someone may have to declare bankruptcy [in order to get debt
relief] is likely to engender strong emotions. But the Court never even
discussed that aspect of the mandatory disclosures [in its case upholding required
disclosures about bankruptcy by certain debt relief providers].
[Footnote] … We offer the following example: “The Nazis committed
genocide.” That is a factual statement. It is also a statement that denounces
the Nazi’s actions and beliefs as morally repugnant. That is an ideological
message. Though the government may not be able to compel Volkswagen to include
that message in its advertising without justification, a court would likely
still review any such attempt under Zauderer.
[Somebody is
thinking about abortion disclosures.]
Plus, these images
were “meant to be interpreted literally.” They weren’t “primarily intended to
evoke an emotional response” but instead “to draw attention to the warning and
depict a possible medical consequence of smoking. Thus, at most, the emotional
response of viewers is incidental to their retention of information about the
health risks.”
What about the
argument that the images might be subject to several different interpretations,
and the FDA didn’t test for consumer takeaway? “[W]hen each image is paired
with a fact-based, textual warning, any reasonable viewer interprets the image
in light of the words.” It was error to ignore the words.
Also, the government
need not choose only the most common side-effect or consequence of the disease
or injury discussed in a warning. “People may interpret ‘debt relief agency’ in
many ways, but disclosing that a business is one is still purely factual.” Nor
were cigarette companies required to make difficult contextual judgments
weighing multiple factors to determine the warning, since the FDA did it for
them.
For similar reasons,
the warnings were uncontroversial. NIFLA says that abortion is a
controversial topic, making disclosures about abortion controversial; but NetChoice
said that “disclosures of social media
censorship decisions” were not controversial. Thus, a factual disclosure is
“controversial” under Zauderer “where the truth of the statement is not
settled or is overwhelmingly disproven or where the inherent nature of the
subject raises a live, contentious political dispute.” Content moderation isn’t
inherently contentious, even though it was connected to “a live, contentious, political issue.” [Wow,
this might be even dumber than the statements in NetChoice
itself. Because it is about content that some people want and some people
don’t, content moderation policy is the definition of inherently contentious—as
abortion is not, even if people living in Texas today think it must be. This is
a fake argument; the real reason—inconsistent with NIFLA’s dicta, which
should be ignored the way all abortion-related First Amendment pronouncements
should be ignored—comes next.]
There’s no good-faith debate that the warnings aren’t
truthful. Thus, they are uncontroversial.
Next, the warnings
must be “reasonably related to the State’s interest” and not “unjustified or
unduly burdensome.” “Zauderer does not require the state to assert an
anti-deception interest.” No court of appeals majority has ever held otherwise,
and the Fifth Circuit has previously referred to valid interests in “promoting
the free flow of commercial information”; we ruled that was “more than enough
to satisfy this prong of Zauderer” and “promoting the ethical integrity
of the legal profession.” “Increasing public understanding of the risks of
smoking, particularly given the ‘long history of deception concerning consumer
health risks in the cigarette industry,’ is a legitimate state interest.” [Now
do the long history of state and private discrimination against nonwhites.]
The warnings were
not unjustified. Plaintiffs argued that the interest at issue was too amorphous
and that the warnings hadn’t been shown to be effective.
The images served an
informational interest. Zauderer itself explained that “[t]he use of illustrations or pictures in
advertisements serves important communicative functions: it attracts the
attention of the audience to the advertiser’s message, and it may also serve to
impart information directly.” The FDA even tested their effectiveness in
raising consumer awareness and then refined them based on those results.
NIFLA says that a compelled disclosure is
justified only if it will “remedy a harm that is ‘potentially real[,] not
purely hypothetical,’ and … ‘extend[s] no broader than reasonably necessary.’
” Plaintiffs argued that current Surgeon General’s warnings are sufficient, but
that ignored “significant evidence that consumers do not notice, much less
internalize, the text-only warnings in the status quo. The updated warnings
serve to remedy the harm that buyers might (1) not know about tobacco’s harms
or (2) ignore the existing Surgeon General’s warnings.”
And here the Fifth
Circuit engages in what is all too common in rejecting plaintiffs’ arguments:
it makes up a contradiction that doesn’t exist. This isn’t to say the Fifth
Circuit is wrong about the weighing here, but it’s a bad look to claim logical
flaws that are themselves illogical: “Plaintiffs inconsistently claim that the
disclosure requirements are overly emotional and ideological such that they
become non-factual speech, while also asserting that FDA’s informational
interest does not justify the Warnings because they will not be effective. In
other words, plaintiffs suggest consumers will simultaneously notice and not
notice the warnings.” But of course, consumers could both notice
and not be informed or change their behavior because of the warnings. I
notice a ton of stuff to which I am indifferent every day. The underlying
question is whether the government ought to have to show some real likelihood
of changed decisionmaking in order to justify mandatory disclosures, and I have
to admit that I am leaning more towards “yes,” at least for a noticeable
percentage of consumers. In the absence of any need for effectiveness,
disclosure becomes a compromise where regulators/lawmakers tell themselves
they’re protecting consumers while blaming the ones who continue to make “bad”
decisions for the consequences of continued marketing.
Effectiveness:
Plaintiffs argued that the FDA’s studies were flawed, but all that was required
constitutionally was a reasonable relation to a legitimate state interest. “Whether
FDA’s use of the studies survives APA review is a question we consider separately
from our Zauderer review.” This move too sets up the ability to approve
state mandates (not subject to the APA) like in NetChoice while
still invalidating anything the Fifth Circuit doesn’t like as a policy matter.
The warnings were not unduly burdensome, despite their size
and offputting content. Even if they wouldn’t survive Central Hudson
review, that wasn’t enough to invalidate them. The Sixth Circuit already upheld
the required size and the court here wasn’t going to revisit that. “Undue
burden” means that “the regulation
cannot impose a burden excessive or disproportionate to the benefits gained.”
In NIFLA, the burden was undue because the disclosures were “wholly
disconnected from California’s informational interest”; allowed for no
consideration of “what the facilities say on site or in their advertisements”;
and “cover[ed] a curiously narrow subset of speakers.” But in NetChoice,
decision disclosure/appeal and biannual transparency disclosure couldn’t
possibly burden protected speech, so that was ok, and the SEC’s buyback
disclosures weren’t unduly burdensome because they “neither burden[ ] issuers’
protected speech nor drown[ ] out their message” given that they occurred only
“within the narrow confines of SEC filings.” [Gotta admit, these don’t sound
promising for disclosures that have to be a big part of every package.]
Here, the benefits
were to alleviate “information asymmetry regarding the harms tobacco causes and
consumers’ suboptimal awareness of and response to those harms,” and reducing
those harms would be a significant benefit. [If it occurs.]
The claimed harms
were to plaintiffs’ free speech rights and to their finances. But “plaintiffs
can still speak on 80% of their advertisements, and they still control more
than 50% of the total surface area of their cigarette packages.” That allowed “ample
room for manufacturers to distinguish their products from other products” and
not be “drown[ed] out” or deterred from advertising at all. And plaintiffs have
at most a “minimal” interest in withholding useful and factual information;
harm suffered from an infringement on that interest was limited. Thus, the
burdens were not undue in comparison to the benefits.
from Blogger http://tushnet.blogspot.com/2024/03/5th-circuit-allows-image-based-tobacco.html
Rebecca Tushnet 