Reply Brief — Food and Drug Administration, Petitioner v. Wages and White Lion Investments, L.L.C., dba Triton Distribution, et al.

No. 23-1038

In the Supreme Court of the United States

FOOD AND DRUG ADMINISTRATION, PETITIONER

v.

WAGES AND WHITE LION INVESTMENTS, L.L.C.,

DBA TRITON DISTRIBUTION, ET AL.

ON WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE FIFTH CIRCUIT

REPLY BRIEF FOR THE PETITIONER

ELIZABETH B. PRELOGAR

Solicitor General

Counsel of Record

Department of Justice

Washington, D.C. 20530-0001

SupremeCtBriefs@usdoj.gov

(202) 514-2217

TABLE OF CONTENTS

Page

A. FDA properly evaluated applications for

authorization to market flavored e-liquids ..................... 2

B. FDA’s decision not to evaluate respondents’

marketing plans was harmless....................................... 15

C. The remaining arguments for setting aside FDA’s

denial orders lack merit .................................................. 19

TABLE OF AUTHORITIES

Cases:

American Academy of Pediatrics v. FDA,

399 F. Supp. 3d 479 (D. Md. 2019) .................................... 22

Atlantic Marine Construction Co. v. U.S.

District Court, 571 U.S. 49 (2013) ..................................... 22

Avail Vapor, LLC v. FDA, 55 F.4th 409

(4th Cir. 2022), cert. denied,

144 S. Ct. 277 (2023) ............................................. 6, 7, 12, 22

BMW of North America, Inc. v. Gore,

517 U.S. 559 (1996)................................................................ 3

Calcutt v. FDIC, 598 U.S. 623 (2023) ............................. 16, 17

Department of Commerce v. New York,

588 U.S. 752 (2019) ......................................13, 16, 17, 21, 22

Electric Clouds, Inc. v. FDA, 94 F.4th 950

(10th Cir. 2024) ....................................................... 6, 9, 11-13

Erie-Lackawanna R.R. Co. v. United States,

279 F. Supp. 316 (S.D.N.Y. 1967),

aff ’d with modifications, 389 U.S. 486 (1968) ................... 17

ExxonMobil Pipeline Co. v. Department of

Transportation, 867 F.3d 564 (5th Cir. 2017) .................... 3

General Electric Co v. EPA, 53 F.3d 1324

(D.C. Cir. 1995) ..................................................................... 3

Johnson v. United States, 576 U.S. 591 (2015) ..................... 3

(I)

II

Cases—Continued:

Page

Liquid Labs LLC v. FDA, 52 F.4th 533

(3d Cir. 2022) ............................................................. 6, 12, 14

Little Sisters of the Poor Saints Peter & Paul Home

v. Pennsylvania, 591 U.S. 657 (2020) ......................... 15, 16

Lotus Vaping Technologies, LLC v. FDA,

73 F.4th 657 (9th Cir. 2023), petition for cert.

pending, No. 23-871 (filed Feb. 9, 2024) ............................. 6

Magellan Technology, Inc. v. FDA, 70 F.4th 622

(2d Cir. 2023), petition for cert. pending, No. 23-799

(filed Jan. 22, 2024) ............................................................. 12

National Security Archive v. CIA, 752 F.3d 460

(D.C. Cir. 2014) ................................................................... 12

NLRB v. Bell Aerospace Co., 416 U.S. 267 (1974) ........... 4, 5

Prohibition Juice Co. v. FDA, 45 F.4th 8

(D.C. Cir. 2022) ............................................2, 6, 9, 11, 12, 15

Radio Athens, Inc., (WATH) v. FCC, 401 F.2d 398

(D.C. Cir. 1968) ..................................................................... 3

Satellite Broadcasting Co. v. FCC, 824 F.2d 1

(D.C. Cir. 1987) ................................................................. 3, 4

SEC v. Chenery Corp., 318 U.S. 80 (1943)..................... 16, 17

United States v. Chemical Foundation, Inc.,

272 U.S. 1 (1926) ................................................................. 13

United States v. Title Insurance & Trust Co.,

265 U.S. 472 (1924).............................................................. 16

Vermont Yankee Nuclear Power Corp. v. NRDC,

Inc., 435 U.S. 519 (1978) ....................................................... 4

Constitution and statutes:

U.S. Const. Amend XIV (Due Process Clause) .................... 3

Family Smoking Prevention and Tobacco Control

Act, Pub. L. No. 111-31, Div. A, 123 Stat. 1776 ................. 1

5 U.S.C. 706 ............................................................................ 15

21 U.S.C. 387j(a)(1) ................................................................. 4

III

Statutes—Continued:

Page

21 U.S.C. 387j(b)(1)(A) ............................................................ 7

21 U.S.C. 387j(c)(2) .................................................................. 7

21 U.S.C. 387j(c)(4) .............................................................. 6, 8

Miscellaneous:

FDA, Technical Project Lead (TPL) Review of

PMTAs (June 21, 2024), https://www.fda.gov/

media/179501/download ..................................................... 19

In the Supreme Court of the United States

No. 23-1038

FOOD AND DRUG ADMINISTRATION, PETITIONER

v.

WAGES AND WHITE LION INVESTMENTS, L.L.C.,

DBA TRITON DISTRIBUTION, ET AL.

ON WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE FIFTH CIRCUIT

REPLY BRIEF FOR THE PETITIONER

In the Family Smoking Prevention and Tobacco

Control Act (Tobacco Control Act or Act), Pub. L. No.

111-31, Div. A, 123 Stat. 1776, Congress directed the

Food and Drug Administration (FDA) to authorize a

new tobacco product only if an applicant could show that

the marketing of its product would be appropriate for

the protection of the public health. Applying that standard, FDA denied respondents’ applications for authorization to market e-liquids with flavors such as cotton

candy, pink lemonade, and milk and cookies because

such flavored products pose a serious risk of attracting

youth to the use of tobacco without sufficient offsetting

benefits for adults. Seven other courts of appeals have

upheld similar denial orders, each time unanimously.

In the decision below, however, the en banc Fifth

Circuit held that FDA’s denial of respondents’ applica(1)

2

tions was arbitrary and capricious. The Fifth Circuit

offered five rationales for its decision, but respondents

abandoned one of them before this Court granted the

petition for a writ of certiorari. Respondents now retreat from the remaining four rationales, significantly

rewriting some of them and defending another only in a

footnote. Many of respondents’ amici jettison the Fifth

Circuit’s rationales altogether, instead offering their

own alternative grounds for setting aside FDA’s orders.

All of that just confirms that the Fifth Circuit’s outlier

decision is wrong. This Court should reverse.

A. FDA Properly Evaluated Applications For Authorization To Market Flavored E-Liquids

In its most significant ruling, the Fifth Circuit held,

contrary to the decisions of seven other courts of appeals, that FDA unfairly surprised manufacturers of

flavored e-cigarettes during the application process.

See Gov’t Br. 17. Respondents offer no meritorious defense of that aberrational holding.

1. At the outset, respondents apply the wrong legal

test in analyzing their unfair-surprise claims. An

agency acts arbitrarily if it announces one standard, and

then applies a different standard, without explaining

the change and considering serious reliance interests

engendered by its previous policy. See Gov’t Br. 18. To

make that sort of claim, respondents need to show that

the agency “affirmatively misdirected” them. Prohibition Juice Co. v. FDA, 45 F.4th 8, 21 (D.C. Cir. 2022).

It is not enough for them to show that the agency did

not forecast how it would evaluate their applications or

that they had a reasonable, if mistaken, interpretation

of the agency’s guidance about the process.

In framing their argument, respondents instead contend that FDA was required to “communicate with ‘as-

3

certainable certainty’ ” the standard that applicants

would need to meet in order to obtain authorization. Br.

28 (citation omitted). But an agency generally has no

affirmative obligation to issue guidance in the first

place. See Gov’t Br. 25-28. Nor was the Fifth Circuit

correct in concluding that an agency is bound by a private party’s purportedly reasonable misreading of its

guidance. See id. at 28-31.

To support their demand for greater pre-application

certainty, respondents invoke principles of due process

and fair notice. Of course, “[e]lementary notions of fairness” require the government to provide a person with

“fair notice” of “the conduct that will subject him to

punishment.” BMW of North America, Inc. v. Gore,

517 U.S. 559, 574 (1996) (emphasis added). This Court

has thus held that the Due Process Clause prohibits the

enforcement of “a criminal law so vague that it fails to

give ordinary people fair notice of the conduct it punishes.” Johnson v. United States, 576 U.S. 591, 595

(2015). And in the cases that respondents cite (Br. 2829, 34), courts of appeals have extended “this ‘no punishment without notice’ rule” to “the civil administrative

context,” requiring agencies to provide fair notice before imposing civil penalties. General Electric Co. v.

EPA, 53 F.3d 1324, 1327, 1329 (D.C. Cir. 1995); see ExxonMobil Pipeline Co. v. Department of Transportation,

867 F.3d 564, 568 (5th Cir. 2017) (“civil administrative

penalties”); Satellite Broadcasting Co. v. FCC, 824 F.2d

1, 3 (D.C. Cir. 1987) (“sanction”); Radio Athens, Inc.,

(WATH) v. FCC, 401 F.2d 398, 404 (D.C. Cir. 1968)

(“sanction”).

But in this case, the government did not seek to punish respondents for violating the law. It simply found

that they were not entitled to a benefit for which they

4

had applied: permission to sell an otherwise unlawful

product.

Respondents cite (Br. 35) a case in which the D.C.

Circuit required an agency to provide fair notice before

dismissing an application as a “sanction” for violating a

procedural rule. Satellite Broadcasting, 824 F.2d at 3.

But FDA did not dismiss respondents’ applications to

penalize them for procedural violations. Rather, FDA

found that respondents were not entitled to the benefit

they sought because they had failed to show that their

products would be appropriate for the protection of the

public health. Respondents cite no case (apart from the

decision below) requiring an agency to provide such detailed notice before evaluating an application for a benefit.

Nor do the courts have the power to impose such a

requirement. The reviewing court’s role is to ensure

that agencies comply with the requirements prescribed

by Congress—not to enforce its “own notions of proper

procedures.” Vermont Yankee Nuclear Power Corp. v.

NRDC, Inc., 435 U.S. 519, 525 (1978). And a court may

not abridge an agency’s discretion to develop regulatory principles through “case-by-case evolution” rather

than through “a general rule” issued in advance. NLRB

v. Bell Aerospace Co., 416 U.S. 267, 293 (1974) (citation

omitted).

Respondents’ contrary approach would prove especially unworkable in the context of the Tobacco Control

Act. The Act requires manufacturers to obtain authorization before marketing new tobacco products. See 21

U.S.C. 387j(a)(1). In evaluating new products, FDA will

inevitably encounter new issues. It would make little

sense to require an agency to announce ahead of time

how it will resolve issues that it has not yet confronted.

5

See Bell Aerospace, 416 U.S. at 293 (“[P]roblems may

arise in a case which the administrative agency could

not reasonably foresee, problems which must be solved

despite the absence of a relevant general rule.”) (citation omitted).

Separately, respondents emphasize (Br. 42-45) the

principle that an agency may change its policy only if it

considers serious reliance interests engendered by its

previous policy. But that principle applies only when an

agency actually changes its policy. Respondents, like

the Fifth Circuit, assert (Br. 43-44) that an agency must

also consider a private party’s reliance on a purportedly

“reasonable” interpretation of the agency’s guidance,

regardless of whether that interpretation turns out to

be “ ‘correct’ or ‘incorrect.’ ” But respondents cite no

statutory provision or decision of this Court imposing

such a requirement.

Respondents’ approach, again, would be particularly

unworkable in the context of the Tobacco Control Act.

Manufacturers have applied to FDA for authorization

to market millions of new tobacco products. See Pet.

App. 129a. Different applicants may interpret the Act,

the regulations, and the guidance in different ways. Respondents’ theory would seemingly require the agency

to follow a bespoke review process for each applicant,

tailored to that particular applicant’s purportedly “reasonable” reading of the relevant requirements and guidance. Resp. Br. 43.

2. Regardless of how the legal test is framed, respondents’ claim of unfair surprise fails for a more basic

reason: They did not lack notice of the relevant considerations. In this Court, respondents primarily contend

(Br. 27) that FDA unfairly surprised manufacturers of

flavored e-liquids by requiring them to submit evidence

6

“comparing [their] flavored products to tobaccoflavored products.” In fact, respondents stake almost

their entire case on that premise. See, e.g., id. at 29

(“[FDA] made no public suggestion of any need to compare flavored and tobacco-flavored [e-liquids].”); id. at

30 (“FDA made no mention of any requirement * * * to

compare one [e-cigarette] product against another.”);

id. at 36 (“[FDA] never told applicants that they needed

to compare their flavored products to tobacco-flavored

products.”).

Although respondents emphasized the same theory

below (e.g., C.A. Resp. Br. 39), the Fifth Circuit did not

rely on it. And the five other courts of appeals to consider that contention have correctly rejected it. See

Liquid Labs LLC v. FDA, 52 F.4th 533, 542-543 (3d Cir.

2022); Avail Vapor, LLC v. FDA, 55 F.4th 409, 419 (4th

Cir. 2022), cert. denied, 144 S. Ct. 277 (2023); Lotus

Vaping Technologies v. FDA, 73 F.4th 657, 670-672 (9th

Cir. 2023), petition for cert. pending, No. 23-871 (filed

Feb. 9, 2024); Electric Clouds, Inc. v. FDA, 94 F.4th

950, 958-960 (10th Cir. 2024); Prohibition Juice, 45

F.4th at 23-24 (D.C. Cir.).

The most serious problem with respondents’ theory

is that the statute itself unambiguously requires FDA

to make comparative judgments. See, e.g., Electric

Clouds, 94 F.4th at 958-959. The Act directs FDA, in

evaluating whether a new product would be appropriate

for the protection of the public health, to consider “the

increased or decreased likelihood that existing users of

tobacco products will stop using such products” and

“the increased or decreased likelihood that those who

do not use tobacco products will start using such products.” 21 U.S.C. 387j(c)(4) (emphases added). Those

factors are “inherently comparative.” Lotus Vaping, 73

7

F.4th at 669. The only way to gauge whether a new

product “increases” or “decreases” the likelihood of particular benefits or risks is to compare that product to

other products. “[N]othing can ‘increase’ or ‘decrease’ in

a vacuum.” Avail Vapor, 55 F.4th at 428 (citation omitted).

If that were not enough, the Act specifically requires

the manufacturer to submit, as part of its application,

information about “whether [its] tobacco product presents less risk than other tobacco products.” 21 U.S.C.

387j(b)(1)(A) (emphasis added). The Act then directs

FDA to consider “the information submitted” as part of

the application—including such information comparing

the applicant’s product to other products—when reaching its decision. 21 U.S.C. 387j(c)(2).

Thus, the Act alone provided respondents with ample notice. But FDA’s 2019 Guidance also reminded applicants to submit comparative evidence. In a section

titled “Comparison Products,” the guidance encouraged

applicants to “compare” their products to other products and to justify their chosen “comparators,” explaining that “comparative” data is “an important part” of

FDA’s analysis. J.A. 30. Other parts of the guidance

likewise discussed the importance of submitting comparative evidence. See, e.g., J.A. 51 (“relative health

risks of the new tobacco product * * * compared to

other tobacco products”); J.A. 53 (“assessment of whether

the product will have a positive impact on the health of

the population * * * as compared to other tobacco products”); J.A. 58 (“comparison to other e-liquids”); J.A.

101 (“comparative assessments”).

Respondents insist (Br. 27, 31 n.25, 36) that the Act

and FDA’s guidance left applicants “free to select a

comparator product of their choosing”; that applicants

8

need not show that “flavored products” work better

than “tobacco-flavored products”; and that applicants

may instead obtain authorization to sell fruit-, candy-,

and dessert-flavored e-liquids simply by showing that eliquids are less harmful than “combustible cigarette[s].”

But that is an implausible interpretation of the Act and

the guidance. The application process is not a game that

tests an applicant’s skill at drawing comparisons to just

any other product. The point of the Act’s comparativeevidence requirement is instead to enable FDA to assess a product’s marginal “risks and benefits” relative

to the range of potential alternatives—and thus to determine whether the product is “appropriate for the

protection of the public health.” 21 U.S.C. 387j(c)(4).

Here, FDA found that fruit-, candy-, and dessertflavored e-liquids pose a higher risk of attracting youth

than do tobacco-flavored e-liquids. See Pet. App. 183a197a, 237a-251a, 291a-305a. It explained that the “majority of youth who use [e-cigarette products] report using a flavored [e-cigarette product]”; that youth are

“more likely to use flavored [products]” than adults;

that youth users “consistently select flavors as a top

reason” when asked why they use e-cigarettes; and that

“flavoring” makes tobacco products “more palatable for

novice youth and young adults.” Id. at 187a-190a, 241a244a, 295a-298a. Because flavors thus pose an added

risk to public health, FDA sensibly looked for evidence

that they provide an “added benefit” that could “outweigh” that risk. Id. at 168a, 228a, 280a. But FDA “did

not find such evidence” in respondents’ applications. Id.

at 168a, 228a, 263a.

Respondents insist (Br. 36) that FDA never specifically told applicants that “they needed to compare their

flavored products to tobacco-flavored products.” Even

9

if that were true, it would not establish unfair surprise.

FDA’s 2019 Guidance was directed to e-cigarette and

e-liquid applicants in general, not to makers of flavored

e-liquids in particular. See J.A. 19. It should hardly

come as a surprise that guidance addressed to a general

audience would speak in general terms.

In any event, respondents’ characterization is incorrect. In a section of the 2019 Guidance captioned “Flavors,” FDA flagged its concerns about the “impact of

flavors” on “appeal to youth and young adults”; stated

that it was “important for [applications] for flavored

products to examine the impact of the flavoring”; and

encouraged applicants to “examin[e] adult appeal of

such flavors.” J.A. 87-88. The 2019 Guidance, in short,

notified applicants of the need to compare “e-liquids

with and without flavoring.” Electric Clouds, 94 F.4th

at 960.

Indeed, respondents’ own applications reflected a

recognition of their need to compare the risks and benefits of products with and without flavors. Respondents

asserted that “flavors are crucial to getting adult smokers to make the switch and stay away from combustible

cigarettes”; that certain survey participants “started

out using tobacco or menthol flavors but now always or

almost always use other flavors”; and that “[t]his

change in flavor preference * * * has powerful implications” for “the role of flavors.” J.A. 319, 321. Respondents thus attempted to show the very thing that they

claim was an undisclosed requirement: that flavors provide added benefits relative to other types of products.

Accordingly, any assertion of unfair surprise is “far off

base.” Prohibition Juice, 45 F.4th at 23. And FDA’s

conclusion that respondents had failed to provide

enough evidence to back up their comparative claims,

10

see Pet. App. 168a, 228a, 263a, is also unsurprising—

especially when respondents’ own review of the scientific literature found that “no conclusions can be made

about the association of e-cigarette flavors and smoking

cessation,” J.A. 476.

3. As noted above, the Fifth Circuit did not rely on

the theory that FDA surprised applicants by applying

the Act’s comparative-evidence requirement. See p. 6,

supra. The court instead reasoned that FDA surprised

applicants by requiring particular types of studies. See

Pet. App. 36a. Respondents briefly raise that argument

as well, but their contentions lack merit.

Respondents repeat (Br. 37) the Fifth Circuit’s claim

that FDA unexpectedly “required applicants to submit

a randomized controlled trial or longitudinal cohort

study.” That claim conflicts with the plain text of FDA’s

denial orders, which stated: “This evidence could have

been provided using a randomized controlled trial

and/or longitudinal cohort study[.] * * * Alternatively,

FDA would consider other evidence[.] * * * We did not

find such evidence in your [application].” Pet. App.

167a-168a (emphases added); see id. at 227a-228a; id. at

263a-264a. Respondents object (Br. 37 n.28) that “any

‘other evidence’ must still be of comparative efficacy,”

but that just repeats their meritless comparativeevidence theory.

Seeking to overcome the text of the denial orders,

respondents cite (Br. 23) the scientific-review form that

FDA used when assessing applications. But that form

stated: “This review determines whether the subject

[applications] contain evidence from a randomized controlled trial, longitudinal cohort study, and/or other evidence regarding the impact of the new [product].” J.A.

617 (emphasis added). Respondents emphasize (Br. 23)

11

that the form contained boxes, for “Criterion A” and

“Criterion B,” titled “Randomized Controlled Trial”

and “Longitudinal Cohort Study,” but those two boxes

were immediately followed by another box, for “Criterion C,” titled “Other evidence in the [application(s)] related to potential benefit to adults.” J.A. 617, 619-620

(emphases omitted). Thus, the form that FDA used to

identify applicants’ evidence showed—as the denial orders themselves did—FDA’s willingness to “look[] beyond randomized trials and longitudinal studies.” Prohibition Juice, 45 F.4th at 22; see Electric Clouds, 94

F.4th at 961. Moreover, since the title for Criterion C

expressly addressed the “potential benefit to adults,”

J.A. 20, respondents err in asserting (Br. 39 n.30) that

the “scientific review forms” made “no reference” to respondents’ need to substantiate their applications’

claims that “flavors are crucial to getting adult smokers

to * * * stay away from combustible cigarettes.”

Respondents next cite (Br. 18-19) an internal FDA

memorandum, issued on July 9, 2021, which proposed

treating the lack of a randomized trial or longitudinal

study as a “fatal flaw” in an application. J.A. 243; see

J.A. 241-244. They also cite (Br. 19-20) another internal

FDA memorandum, issued on August 17, 2021, which

identified standards for FDA to apply in evaluating applications. See. J.A. 245-280. But the July memorandum was “superseded” by the August memorandum,

and “there’s no evidence that the FDA ever applied” it

when evaluating applications. Electric Clouds, 94 F.4th

at 960. The August memorandum itself was then “rescind[ed]” only a few days after it was issued. J.A. 282.

FDA had “reconsidered the process” for evaluating applications, decided that “it will not consider or rely on”

the August memorandum in that process, and deter-

12

mined that the memorandum was therefore “no longer

needed.” Ibid. The Fifth Circuit thus did not rely on

either of the two internal documents, and other courts

have uniformly rejected arguments based on them. See

Electric Clouds, 94 F.4th at 960-961; Magellan Technology, Inc. v. FDA, 70 F.4th 622, 630 (2d Cir. 2023), petition for cert. pending, No. 23-799 (filed Jan. 22, 2024);

Avail Vapor, 55 F.4th at 424; Liquid Labs, 52 F.4th at

540 n.7; Prohibition Juice, 45 F.4th at 22.

That consensus is correct. Agencies “should be judged

by what they decided, not for matters they considered

before making up their minds.” National Security Archive v. CIA, 752 F.3d 460, 462 (D.C. Cir. 2014) (Kavanaugh, J.) (citation omitted). A contrary approach

would inhibit candor during “internal discussions”;

would promote “gridlock” by identifying “a point where

agency deliberations become frozen in time”; and would

deprive agencies of “the value of ongoing dialogue,”

hampering their ability to change course even if they

conclude that previously considered approaches would

raise legal concerns that might be avoided under alternative approaches. Avail Vapor, 55 F.4th at 424. Here,

FDA ultimately agreed to consider other types of evidence, not just randomized trials or longitudinal studies. FDA’s denial orders nowhere stated that the

agency treated the absence of such trials or studies as a

“fatal flaw.” Whether FDA had once contemplated a

different approach, before it decided respondents’ applications, is immaterial.

Respondents invite (Br. 18-20, 23, 29) this Court to

infer that, although FDA stated that it was not relying

on the July and August memoranda, it surreptitiously

did so anyway. That argument conflicts with the “presumption of regularity” that courts owe to executive

13

agencies, which form part of a coordinate branch of the

federal government. United States v. Chemical Foundation, Inc., 272 U.S. 1, 14 (1926). While an agency

must “ ‘disclose the basis’ ” of its action, a court must

generally accept “an agency’s stated reasons for acting.” Department of Commerce v. New York, 588 U.S.

752, 780-781 (2019) (citation omitted). “[F]urther judicial inquiry into ‘executive motivation’ represents ‘a

substantial intrusion’ into the workings of another branch

of Government and should normally be avoided.” Id. at

781 (citation omitted). A court may look behind an

agency’s stated reasons only in extraordinary cases involving a “strong showing of bad faith.” Ibid. (citation

omitted). Respondents have not seriously tried to make

such a showing here.

Regardless, FDA’s internal documents do not “reveal a secret plan” to “nix any application” that lacks a

randomized trial or longitudinal study. Electric Clouds,

94 F.4th at 961. Although the July memorandum referred to the absence of such evidence as a “fatal flaw,”

the same sentence said that “any application lacking

this evidence will likely receive a marketing denial

order”—not that it definitely will. J.A. 243 (emphasis

added). The August memorandum, meanwhile, stated

that randomized trials and longitudinal studies are “the

strongest types of evidence,” but it also said that FDA

“would also consider evidence from another study design” if it “reliably and robustly” demonstrates the

product’s benefits. J.A. 247 & n.ix.

Respondents also contend (Br. 47-49) that FDA

should have promulgated the August memorandum

through notice-and-comment rulemaking. That theory

is not properly before this Court, irrelevant to the outcome of this case, and wrong. The question presented

14

asks (Pet. I) whether FDA’s “denial orders” were “arbitrary and capricious,” not whether the August memorandum was procedurally faulty. Nor did the Fifth Circuit pass upon respondents’ theory. Further, because

FDA rescinded the August memorandum and did not

rely on it when denying respondents’ applications, any

procedural challenge to the memorandum is moot and

immaterial to the resolution of this case. In all events,

the notice-and-comment theory rests on the erroneous

premise that the memorandum imposed a categorical

“requirement” to submit a “longitudinal” study. Resp.

Br. 47. As explained above, the August memorandum

did no such thing; it was almost immediately rescinded;

and FDA’s denial orders reflected no such requirement.

4. Respondents’ remaining arguments about unfair

surprise lack merit. Respondents contend (Br. 29, 41)

that FDA assured applicants that they would not need

to submit “long-term clinical studies,” yet turned around

and faulted respondents for failing to provide evidence

of their products’ effects “over time.” Our opening brief

explained (at 22-23) why that argument is wrong: It

overlooks the difference between “long-term” studies

(which the 2019 Guidance defined to mean studies conducted over six months or longer) and evidence about a

product’s effects “over time” (which could include evidence that spans less than six months). See Liquid

Labs, 52 F.4th at 541 n.10. FDA did not require longterm studies, but it did recommend that applicants submit evidence about “the trends by which users consume

the product over time.” J.A. 53 (emphasis added). Respondents ignore those points.

Respondents also object (Br. 40) that FDA did not

invite them to amend their applications and submit new

evidence. If FDA had changed the rules after the ap-

15

plications had been filed, it would indeed have been appropriate for the agency to notify applicants and to give

them an opportunity for amendment. But, as explained

above, FDA did not change its regulatory approach.

Rather, the “final determinations were consistent with

the 2019 Guidance”—and with the Tobacco Control Act

itself. Prohibition Juice, 45 F.4th at 21. Even so, as

FDA’s denial orders make clear, respondents remain

free “to submit new applications for these products.”

Pet. App. 167a.

In sum, FDA did not unfairly surprise respondents.

FDA’s denial of their applications instead reflected a

“straightforward application” of the criteria set out in

the Act and in FDA’s guidance. Prohibition Juice, 45

F.4th at 24.

B. FDA’s Decision Not To Evaluate Respondents’ Marketing Plans Was Harmless

The Fifth Circuit also erred in its refusal to apply the

harmless-error rule to the agency’s decision not to evaluate respondents’ marketing plans. See Pet. App. 60a.

Respondents offer no good defense of that holding,

which contradicts the decisions of six other courts of appeals. See Gov’t Br. 31-32.

1. Congress has expressly instructed courts to take

“due account” of “the rule of prejudicial error.” 5 U.S.C.

706. Yet the Fifth Circuit held that “errors are only

harmless where the agency would be required to take

the same action no matter what” and that the “harmless-error rule simply does not apply to * * * discretionary administrative decisions.” Pet. App. 59a-60a.

As our opening brief explained (at 39-40), that holding conflicts with this Court’s cases, which have found

errors harmless even when no statute required the

agency to take the action at issue. See Little Sisters of

16

the Poor Saints Peter & Paul Home v. Pennsylvania,

591 U.S. 657, 684-685 (2020); Department of Commerce,

588 U.S. at 780. Respondents deny (Br. 55) that those

cases involved discretionary action, but they plainly did.

See Little Sisters, 591 U.S. at 680 (deciding the case on

the premise that the agency was not “compelled” to take

the action at issue); Department of Commerce, 588 U.S.

at 771 (explaining that the statutory provisions “le[ft]

much to the Secretary’s discretion”). Respondents also

argue (Br. 55) that, because those decisions held in the

alternative that agencies had not erred in the first place,

the Court’s discussions of harmlessness were dicta. But

when a court resolves a case on two alternative grounds,

“the ruling on neither is obiter, but each is the judgment

of the court, and of equal validity with the other.”

United States v. Title Insurance & Trust Co., 265 U.S.

472, 486 (1924) (citation omitted).

Respondents attempt (Br. 52) to ground the Fifth

Circuit’s approach in the ordinary remand rule, which

generally requires a court that detects an error in

agency action to “remand to the agency for additional

investigation or explanation.” Calcutt v. FDIC, 598 U.S.

623, 629 (2023) (per curiam) (citation omitted). But that

rule simply recognizes that, where an issue “has been

delegated to an agency,” the reviewing court “is not

generally empowered to conduct a de novo inquiry into

the matter.” Ibid. (citation omitted). If an order’s validity depends on a “policy or judgment which the agency

alone is authorized to make and which it has not made,

a judicial judgment cannot be made to do service for an

administrative judgment.” SEC v. Chenery Corp., 318

U.S. 80, 88 (1943). “[A]n appellate court cannot intrude

upon the domain which Congress has exclusively entrusted to an administrative agency,” and that is so

17

“[f ]or purposes of affirming no less than reversing [the

agency’s] orders.” Ibid.

Here, harmless-error analysis would not require the

reviewing court to “conduct a de novo inquiry” into the

efficacy of respondents’ proposed sales and marketing

restrictions. Calcutt, 598 U.S. at 629 (citation omitted).

Instead, the court need only recognize that FDA itself

has already found that such restrictions are “not sufficient to address this issue.” J.A. 220. The remand rule

is inapposite in such a case. As Judge Friendly explained, the rule exists to ensure that a “reviewing court

[does] not affirm an agency on a principle the agency

might not embrace”—not to prolong proceedings “while

court and agency engage in a nigh endless game of battledore and shuttlecock.” Erie-Lackawanna R.R. Co. v.

United States, 279 F. Supp. 316, 354-355 (S.D.N.Y.

1967) (three-judge court), aff ’d with modifications, 389

U.S. 486 (1968).

Respondents also invoke (Br. 50 n.34, 57) the principle that a court may not uphold agency action based on

a post hoc rationalization, but that principle does not

support the Fifth Circuit’s decision either. Respondents are correct that, in deciding whether an agency

erred in the first place, a court is “limited to evaluating

the agency’s contemporaneous explanation in light of

the existing administrative record.” Department of

Commerce, 588 U.S. at 780. But a court is not so limited

in deciding whether an error was harmless. The whole

point of the harmless-error rule is that, even when an

agency’s contemporaneous justification of its action was

defective, something else could show that the error

made no difference to the outcome that the agency

would reach if the matter were remanded for reconsideration.

18

2. This Court should reverse the Fifth Circuit’s

holding that the harmless-error rule “simply does not

apply,” Pet. App. 60a, and remand the case so that the

Fifth Circuit can apply that rule. The Court need not

address respondents’ fact-bound arguments (Br. 53-58)

that FDA’s decision not to evaluate their marketing

plans prejudiced them. At any rate, those arguments

are incorrect.

As an initial matter, respondents mischaracterize

(Br. 54) our position as requiring a court to accept an

agency’s “bald assertion that non-record, undisclosed

evidence shows an agency error was harmless.” We

have advanced no such theory. Our harmlessness argument relies instead on FDA’s 2020 Guidance, in which

the agency explained that “focusing on how the product

was sold would not be sufficient to address youth use of

these products.” Gov’t Br. 34 (quoting J.A. 215). It also

relies on publicly available orders confirming that, in

other cases, FDA has found various types of marketing

restrictions to be insufficient. See id. at 34-35. In suggesting otherwise, respondents attack a strawman.

Respondents contend (Br. 56) that their marketing

restrictions differ from the restrictions addressed in the

2020 Guidance. But that contention rests (ibid.) on the

flawed premise that the guidance addressed only some

types of e-cigarette devices. In fact, the guidance made

clear that, as a general matter, marketing restrictions

are “not sufficient to address this issue, given the most

recent data that youth use of [e-cigarette] products continues to increase.” J.A. 220-221.

Finally, respondents contend (Br. 57) that, when

FDA granted another manufacturer, NJOY, authorization to market menthol-flavored e-cigarette products in

June 2024, it acknowledged the efficacy of NJOY’s pro-

19

posed marketing restrictions. That is incorrect. FDA

has recognized that marketing restrictions may be necessary to mitigate risks to youth, but it has repeatedly

found that such restrictions are “not sufficient.” J.A.

220 (emphasis added). Consistent with that approach,

when FDA granted NJOY’s application, it repeated that

marketing restrictions, on their own, do not “sufficiently” “mitigate the substantial risk to youth from

flavored [e-cigarette products].” FDA, Technical Project Lead (TPL) Review of PMTAs 6 (June 21, 2024),

https://www.fda.gov/media/179501/download. But the

agency first concluded that NJOY had provided sufficient evidence of its flavored products’ benefits relative

to tobacco-flavored e-cigarettes (i.e., evidence of their

superior ability to help adult smokers switch from combustible cigarettes), which outweighed their risk to

youth. Id. at 9, 45. Only then did FDA observe that

marketing restrictions “may help further limit youth

exposure to the new products * * * and the potential

for youth initiation.” Id. at 9.

C. The Remaining Arguments For Setting Aside FDA’s

Denial Orders Lack Merit

1. The Fifth Circuit concluded—again contrary to

the decisions of six other courts of appeals—that FDA’s

2020 Guidance distinguished between different types of

e-cigarette devices and that FDA then changed its position without explanation when denying respondents’

applications. See Pet. App. 46a; Gov’t Br. 42-44. Respondents repeat (Br. 45-47) the same argument, but

their analysis suffers from the same flaws as the Fifth

Circuit’s.

To begin, respondents overstate (Br. 45) what FDA

said in the 2020 Guidance. FDA did not suggest in that

guidance that the marketing of flavored e-liquids would

20

be appropriate for the protection of the public health so

long as the e-liquids were used with the right type of

device. To the contrary, the guidance explained that

FDA was “concerned about the extraordinary popularity of flavored [e-cigarette] products with youth,” that

“flavors increase youth appeal,” and that “[e]vidence

continues to accumulate * * * that youth are particularly attracted to flavored [e-cigarette] products.” J.A.

151-152. In distinguishing among such products, FDA

merely stated that, because certain flavored products—

i.e., “cartridge-based” products—were especially “popular with young people” at that time, the agency would

prioritize enforcement against those types of products.

J.A. 156; see J.A. 145.

FDA’s views regarding device types had evolved by

the time it resolved respondents’ applications. But

FDA acknowledged and explained that change. When

it denied respondents’ applications, it acknowledged

that its 2020 Guidance had prioritized enforcement

against the types of devices that “were most appealing

to youth at the time.” Pet. App. 192a, 246a, 300a. But

FDA “subsequently observed a substantial rise” in the

use of other types of devices. Ibid. “This trend,” the

agency reasoned, “illustrates that the removal of one

flavored product option prompted youth to migrate to

another [device] type that offered the desired flavor options, underscoring the fundamental role of flavor in

driving appeal.” Id. at 192a, 246a-247a, 300a. FDA thus

did not act arbitrarily; it responded to, and explained its

response to, new evidence.

Respondents contest (Br. 46) FDA’s weighing of the

new evidence, arguing that “device features” rather

than “flavors” “drove youth usage.” But the Tobacco

Control Act vests the authority to evaluate applications

21

in FDA, not the courts. The arbitrary-and-capricious

standard requires courts to ensure that the agency

acted within the zone of reasonableness, not to reweigh

the evidence for themselves. See Department of Commerce, 588 U.S. at 777. In this case, FDA concluded

that “the role of flavor is consistent” across “different

device types,” Pet. App. 191a, 245a, 299a, and it detailed

the evidence supporting that conclusion, see id. at 191a192a, 245a-246a, 299a-300a. Because FDA’s decision

was reasonable and reasonably explained, a court has

no basis for setting it aside.

2. The Fifth Circuit also stated that FDA had improperly imposed a “categorical ban” on all flavored ecigarettes. Pet. App. 47a n.5. Respondents allude to

that conclusion in a footnote. See Br. 49 n.33. But we

have already explained why it is wrong. FDA’s denial

orders do not mention any such ban, and the agency has

since authorized the marketing of some flavored products. See Gov’t Br. 44-47. Respondents do not address

those points.

3. The Fifth Circuit separately concluded that FDA

had arbitrarily distinguished between menthol-flavored

products and other flavored products. See Pet. App.

24a-25a. Respondents disavowed that argument below,

see C.A. Doc. 362, at 6 (Feb. 20, 2024), and they do not

discuss it here.

4. Apparently dissatisfied with the Fifth Circuit’s

five grounds for setting aside FDA’s orders, respondents’ amici offer a surfeit of alternative theories. They

argue, among other things, that FDA has misinterpreted the Tobacco Control Act, see Vapor Technology

Ass’n Amicus Br. 11-20; that it has violated the majorquestions doctrine, see Thirteen Members of Congress

Amicus Br. 5-13; and that the Act is void for vagueness,

22

see Taxpayers Protection Alliance Amicus Br. 6-14.

Those arguments were not passed upon below, are outside the scope of the question presented, and have not

been briefed by the parties. They are not properly before this Court. See, e.g., Atlantic Marine Construction Co. v. U.S. District Court, 571 U.S. 49, 61 (2013)

(declining to consider an issue raised only by an amicus).

Some amici criticize (RJRV Amicus Br. 2-20) FDA’s

handling of e-cigarette applications generally. But this

is a judicial-review proceeding concerning FDA’s denial

of particular applications, not a freewheeling inquiry

into the application process in general. Amici’s contentions lack merit in any event. For example, those amici

say (id. at 5) that one district court “took a dim view of

FDA’s conduct” in extending the application deadlines.

In fact, that court commended “FDA’s laudable efforts,” while criticizing e-cigarette manufacturers for

taking “dilatory measure[s].” American Academy of

Pediatrics v. FDA, 399 F. Supp. 3d 479, 481, 485 (D. Md.

2019). Amici also contend (RJRV Amicus Br. 10-11)

that FDA faced “political pressure” from the “White

House” to address youth vaping in 2019-2020. But there

is nothing wrong with that; “[a]gency policymaking is

not a ‘rarefied technocratic process, unaffected by political considerations or the presence of Presidential

power.’ ” Department of Commerce, 588 U.S. at 781 (citation omitted).

* * * * *

In the Tobacco Control Act, Congress gave FDA

“the daunting task of ensuring that another generation

of Americans does not become addicted to nicotine and

tobacco products.” Avail Vapor, 55 F.4th at 428. It left

FDA with flexibility in performing that task, allowing

23

FDA to determine whether the marketing of a new tobacco product is appropriate for the protection of the

public health. FDA found that respondents’ e-liquids—

which, again, come in flavors such as cotton candy, pink

lemonade, and milk and cookies—pose a serious risk of

attracting youth to e-cigarettes, yet do not offer sufficient offsetting benefits for adults. That common-sense

determination was not arbitrary and capricious.

This Court should reverse the judgment of the court

of appeals.

Respectfully submitted.

ELIZABETH B. PRELOGAR

Solicitor General

NOVEMBER 2024