Case Snapshot

• Court: U.S. Supreme Court
• Citation: 604 U.S. ___ (2025)
• Date: April 2, 2025
• Opinion: Justice Alito for a unanimous Court; Justice Sotomayor concurring
• Parties: Food and Drug Administration (Petitioner) v. Wages and White Lion Investments, LLC, dba Triton Distribution, et al. (Respondents)
• Disposition: Fifth Circuit vacated and remanded

Introduction

FDA v. Wages & White Lion Investments sits at the intersection of tobacco regulation and administrative law. The dispute emerged after the FDA denied premarket authorization for respondents’ flavored e-liquid products used in open-system e-cigarettes. Under the Family Smoking Prevention and Tobacco Control Act of 2009 (TCA), manufacturers of “new tobacco products” must secure FDA authorization by demonstrating that marketing the product would be “appropriate for the protection of the public health.” In the wake of surging youth use of flavored electronic nicotine delivery systems (ENDS), the FDA deemed these products “tobacco products” in 2016, thereby consigning most e-cigarette products to the TCA’s premarket regime.

Respondents challenged the FDA’s denials as arbitrary and capricious under the Administrative Procedure Act (APA), arguing that the agency changed course from its predecisional guidance by: (1) tightening evidentiary demands; (2) insisting on comparative efficacy against tobacco-flavored products; (3) effectively disregarding device-type distinctions; and (4) abandoning its own insistence that “marketing plans” were “critical,” by not reviewing them. Sitting en banc, the Fifth Circuit agreed and set aside the denials, applying a strict reading of Calcutt v. FDIC to reject the FDA’s claim that its failure to review marketing plans was harmless.

The Supreme Court’s decision vacates and remands. It rejects most of the Fifth Circuit’s arbitrary-and-capricious findings under the “change-in-position” doctrine, clarifies how that doctrine applies when agencies have issued nonbinding guidance, and narrows the Fifth Circuit’s reading of Calcutt in light of the APA’s instruction to take “due account” of the “rule of prejudicial error.” The Court declines to reach a distinct, not-fully-briefed question: whether the TCA or APA required the FDA to set PMTA standards via notice-and-comment rulemaking rather than adjudication.

Summary of the Opinion

• The Court assumes without deciding that the change-in-position doctrine applies to positions first articulated in nonbinding guidance, and holds the FDA did not impermissibly change position on three key fronts:
  • Scientific evidence: The denials were consistent with guidance and the TCA’s allowance of either “well-controlled investigations” or other “valid scientific evidence” sufficient to evaluate the product. Applicants lacking RCTs or longitudinal studies had to present robust, product-relevant evidence; respondents did not.
  • Comparative efficacy: Requiring comparisons showing benefits of dessert-, candy-, and fruit-flavored products over tobacco-flavored comparators was a natural consequence of prior guidance and the statute’s inherently comparative, population-level standard.
  • Device type: The FDA’s focus on flavor risks across device types, despite earlier enforcement emphasis on cartridge-based products, did not establish a “safe harbor” for open-system e-liquids and was reasonably supported by evidence of youth migration across product types driven by flavor.
• Marketing plans and harmless error: The FDA concedes it did not consider respondents’ marketing plans despite earlier describing them as “critical.” The Supreme Court holds the Fifth Circuit read Calcutt too broadly in rejecting harmless-error arguments categorically; it clarifies that there are recognized exceptions to the remand rule (e.g., Massachusetts Trustees) and remands for the Fifth Circuit to apply an appropriately calibrated harmless-error analysis in the first instance.
• Unreached issues: The Court expressly declines to decide whether the FDA was required to proceed by rule rather than adjudication, and it does not reach constitutional and “major questions” arguments raised by amici.

Analysis

1) Precedents Cited and Their Influence

• FCC v. Fox Television Stations (2009) and Encino Motorcars v. Navarro (2016): These cases frame the change-in-position doctrine: agencies may change policy if they “display awareness” of the change, offer “good reasons,” and consider “serious reliance interests.” The Court uses this framework to ask first whether the FDA actually changed position, and second, if so, whether it adequately justified that change. Here, the Court finds no actionable change for scientific evidence, comparative efficacy, or device type.
• Motor Vehicle Manufacturers Ass’n v. State Farm (1983): Provides the general arbitrary-and-capricious standard requiring a rational connection between facts found and choices made. It also supplies language about reversals of course. The Court invokes State Farm’s core principle of reasoned explanation.
• SEC v. Chenery Corp. (Chenery I and II) (1943; 1947) and their progeny: Establish the remand rule—courts do not affirm agency action based on post hoc rationalization or alternative grounds not embraced by the agency. The Court reaffirms the rule but explains it must be harmonized with the APA’s prejudicial-error instruction.
• Calcutt v. FDIC (2023) (per curiam): Recently emphasized the remand rule. The Fifth Circuit read Calcutt to permit harmless error only when the agency would be legally required to reach the same outcome. The Court rejects that reading as too narrow.
• Shinseki v. Sanders (2009): The APA imports the “rule of prejudicial error.” The Court recognizes tension between Sanders and Calcutt and clarifies that harmless-error review survives in administrative law and must be adapted to the agency-review context.
• Massachusetts Trustees v. United States (1964): Illustrates a longstanding exception to remand: when an agency’s decision rests on several independent factual findings and one is unsound, remand may be pointless if the error had “no bearing” on the decision. The Court uses this to show Calcutt does not foreclose harmless-error analysis beyond a single narrow exception.
• Citizens to Preserve Overton Park v. Volpe (1971); Dept. of Commerce v. New York (2019); Arlington Heights v. MHDC (1977): Emphasize the presumption of regularity and the high bar for probing agency motives or considering extra-record evidence; the Court relies on these principles to discount any inference that the FDA secretly applied a “fatal flaw” requirement found in an internal memo.
• National Cable & Telecommunications Ass’n v. Brand X (2005) and Long Island Care at Home v. Coke (2007): Stand for the idea that what agencies ultimately do may be a logical or predictable outgrowth of prior communications—relevant to the Court’s holding that the FDA’s comparative-efficacy requirement was a natural consequence of its guidance.
• Camp v. Pitts (1973) and Biestek v. Berryhill (2019): Confirm record-based review and the substantial-evidence threshold; the Court rejects respondents’ reliance on post-application data and upholds the FDA’s reasonable inferences about flavor-driven youth demand.
• Smiley v. Citibank (1996): Limits reliance interests; beliefs about the likelihood of enforcement are not “serious reliance interests” warranting heightened protection here.
• Morgan Stanley Capital Group v. PUD No. 1 of Snohomish County (2008) and INS v. Ventura (2002): Provide context for remand principles and “required action” scenarios, which the Court distinguishes in clarifying harmless-error possibilities.

2) Legal Reasoning

The Court structures its analysis around the change-in-position doctrine and the harmless-error/remand rule tension. The key pillars are:

• Scientific evidence:
  • The TCA allows the FDA to rely on either “well-controlled investigations” (such as RCTs or cohort studies) when appropriate, or other “valid scientific evidence” sufficient to evaluate the product. This grants broad discretion to calibrate evidentiary demands.
  • Across 2018–2019 guidance, the FDA never promised that a particular study type would be required or sufficient. It stated that nonclinical studies alone are generally not sufficient; literature reviews are “less robust” support; and any “bridged” or indirect evidence must be tied to the applicant’s product with a sound scientific rationale.
  • Respondents’ submissions lacked RCTs or cohort studies. Their “other evidence” (a literature review and one cross-sectional survey) was not robustly product-specific or comparative in the way the FDA had repeatedly signaled would be necessary for flavored products. That is consistent with the guidance and statutory scheme.
  • Internal FDA memoranda from July and August 2021 do not prove a secret hardening of standards. The later memo undercut the earlier “fatal flaw” language and the FDA represented it did not rely on either memo in adjudications. Presumption of regularity controls; absent strong evidence of bad faith, courts do not pierce the agency’s explanations.
• Comparative efficacy:
  • The TCA itself requires applicants to provide comparative information and directs the FDA to assess population-level risks and benefits, inherently requiring comparisons across products and flavors.
  • The FDA’s 2019 guidance recommended comparators within and across categories and singled out flavors as a key driver of youth appeal. The 2020 enforcement guidance indicated that fruit and candy flavors especially drive youth use, while tobacco and menthol skew more adult.
  • Against that backdrop, the FDA’s insistence that flavored products demonstrate a benefit over tobacco-flavored comparators was a logical extension—not an unexplained policy reversal.
  • Respondents’ own applications attempted the very comparisons they later claimed were unforeseen, underscoring fair notice.
• Device type:
  • The 2020 guidance prioritized enforcement against flavored, cartridge-based products but also expressly prioritized action against any products whose marketing is likely to promote youth use or that lack adequate youth-access measures. It did not create a “safe harbor” for open-system e-liquids.
  • The FDA reasonably concluded youth demand had migrated from cartridge-based to disposable flavored products, supporting an inference that flavor—more than device type—drives youth uptake. That inference justified focusing on flavored products across device types.
  • Even if there was some evolution, the FDA had “good reasons” for any shift, and regulated parties could not claim deep reliance interests in the very first wave of tobacco PMTA denials.
• Marketing plans and harmless error:
  • The FDA admits it did not review respondents’ marketing plans “for the sake of efficiency,” despite previously stressing that such plans were “critical.” The Fifth Circuit deemed that arbitrary and capricious; the FDA did not seek review of that threshold error.
  • However, the Supreme Court holds the Fifth Circuit read Calcutt too broadly. Harmless-error analysis remains available beyond the narrow scenario where the agency was “required” to take the same action; long-recognized exceptions (e.g., Massachusetts Trustees) demonstrate that not every error mandates remand.
  • At the same time, harmless error must not swallow Chenery’s remand rule. The Court quotes Judge Friendly: where the agency rests on an unsustainable reason, courts should generally remand even if there’s a strong possibility the agency could reach the same result on a different rationale.
  • Because the FDA ultimately asked only for clarification of the standard, not for the Supreme Court to decide harmlessness on the existing record, the Court vacates and remands for the Fifth Circuit to apply a correct, balanced harmless-error framework in the first instance.
• Unreached issues:
  • The Court declines to decide whether the TCA or APA required rulemaking (rather than adjudication) to set PMTA evidentiary standards. It also declines to reach constitutional or major-questions arguments raised by amici but not pressed by the parties.
  • Justice Sotomayor, concurring, underscores that the FDA consistently hewed to its statutory duty to permit marketing only where products are “appropriate for the protection of the public health,” and that demanding rigorous evidence for flavored products should not have surprised applicants.

3) Impact

• Administrative law—Change-in-position doctrine and guidance: The decision meaningfully cabins “agency flip-flop” arguments when prior communications were nonbinding and couched in flexible terms, and when later adjudications flow naturally from statutory text and earlier guidance. Agencies need not reduce every expectation to a rigid checklist in guidance to avoid accusations of surprise.
• Harmless error vs. remand: The Court recalibrates the post-Calcutt landscape. It rejects a categorical rule that only legally compelled outcomes can be harmless. Circuits must harmonize Chenery’s remand rule with the APA’s prejudicial-error command, preserving space for harmless-error review—while guarding against post hoc rationalization and ensuring meaningful agency decisionmaking on the right grounds.
• FDA tobacco regulation and PMTAs:
  • Applicants should expect the FDA to demand robust, product-specific evidence of adult-switching benefits that outweigh youth risks, especially for flavors other than tobacco and menthol. Literature reviews and weakly “bridged” studies will rarely suffice.
  • Comparisons to tobacco-flavored products are effectively the baseline for flavored PMTAs, aligning with the statute’s population-level test and the agency’s youth-risk concerns.
  • Device type will not insulate flavored products; flavor drives the risk calculus. Applicants must frame evidence accordingly.
  • Marketing plans remain relevant, but unless they materially and credibly mitigate youth appeal and access, they are unlikely to salvage an otherwise inadequate application. On remand, the Fifth Circuit’s harmless-error analysis will test how much weight such plans can carry in judicial review.
• Agency process choices: Although the Court does not decide whether rulemaking was required, the opinion implicitly affirms agencies’ latitude to proceed by adjudication, particularly under statutory schemes that vest broad, standard-based discretion and where guidance has already signaled key considerations.
• Litigation strategy and records: Parties face a high bar to pierce the presumption of regularity or to rely on extra-record materials. Challenges should be grounded in the administrative record and framed to meet the standards for showing bad faith or improper behavior if extra-record inquiry is sought.

Complex Concepts Simplified

• PMTA (Premarket Tobacco Product Application): A manufacturer’s request for FDA authorization to market a “new tobacco product.” The FDA must deny unless the product’s marketing would be “appropriate for the protection of the public health,” considering population-level effects.
• Appropriate for the protection of the public health: A comparative, population-focused standard. The FDA weighs whether marketing the product will likely reduce or increase tobacco use among current users and non-users, including youth initiation.
• Well-controlled investigations vs. other valid scientific evidence: The TCA allows the FDA to consider rigorous clinical or epidemiologic studies (e.g., randomized controlled trials, longitudinal cohorts) “when appropriate,” but also to rely on other scientific evidence if it is sufficiently robust and tailored to evaluate the specific product.
• Change-in-position doctrine: Courts police agency “policy flips.” If an agency changes position, it must acknowledge the change, offer good reasons, and account for serious reliance interests. If no change occurred—or if later action is a logical outgrowth—no special justification is needed beyond ordinary reasoned decisionmaking.
• Harmless error and the remand rule: Chenery’s remand rule generally forbids affirming agency action on new grounds; errors usually trigger remand. The APA, however, instructs courts to take “due account” of prejudicial error. Some errors may be harmless—e.g., where the agency’s mistake clearly had no bearing on the outcome. Courts must balance these principles to avoid both rubber-stamping and needless remands.
• Presumption of regularity: Courts generally presume agencies act properly and consider the right factors. Overcoming that presumption and probing internal motives requires a strong showing of bad faith or improper behavior.
• Open-system vs. cartridge/disposable e-cigarettes: Open systems use refillable tanks; cartridge (pod) and disposable systems are closed. The Court accepts the FDA’s data-driven inference that flavor, rather than device type, primarily drives youth appeal.

Conclusion

FDA v. Wages & White Lion is a consequential recalibration of administrative law doctrines applied to a modern public-health regime. On the merits, the Court rejects claims that the FDA “changed position” as to evidentiary expectations, comparative efficacy, and device specificity. The agency’s denials flowed from statutory text, nonbinding guidance that consistently highlighted flavor-related youth risks, and an insistence on robust, product-relevant evidence—not from a covert pivot to rigid study requirements or a surprise comparator standard.

Doctrinally, the decision narrows an expansive reading of Calcutt and instructs lower courts to harmonize Chenery’s remand rule with the APA’s prejudicial-error command. Harmless error survives in administrative review, though it cannot be stretched to ratify agency action on new rationales. The Court identifies at least one long-recognized scenario in which remand is unnecessary (Massachusetts Trustees), and it leaves the broader calibration to the Fifth Circuit on remand and to future cases.

Practically, flavored e-cigarette PMTA applicants now have a clearer map: compelling, product-specific, comparative evidence is essential; tobacco-flavored comparators are the relevant benchmark; flavor is the risk driver across device types; and marketing plans, while relevant, will not substitute for missing science. For agencies, the opinion confirms that flexible, nonbinding guidance—if consistent with statutory standards and logically developed—can guide adjudications without creating a trap under the change-in-position doctrine. For courts, it underscores a measured approach to remedies, preserving both meaningful review and administrative primacy.

The Court leaves unresolved the rulemaking-vs.-adjudication question and several constitutional issues flagged by amici. But its core guidance is clear: agencies may adapt within the bounds of reasoned decisionmaking and their statutes; regulated entities must meet the substantive standard with rigorous evidence; and reviewing courts must balance fidelity to Chenery with the APA’s directive to disregard non-prejudicial error.