In 2021, after a multi-year regulatory process, the Food and Drug Administration denied more than a million applications from tobacco manufacturers seeking to sell various flavored e-cigarette products. After that process concluded, an en banc Fifth Circuit vacated the FDA’s denial of those products’ applications, concluding that the agency performed a “surprise switcheroo” by relying on criteria for evaluating applications that the Fifth Circuit found were inconsistent with the agency’s pre-application industry guidance. But as has happened more than once in recent years, in FDA v. Wages and White Lion Investments, LLC (No. 23-1038), the Supreme Court concluded that the Fifth Circuit was getting a bit too far over its skis. In a lengthy and thorough decision by Justice Alito, the Court unanimously vacated the Fifth Circuit, concluding that it had misapplied administrative law’s “change-in-position” doctrine by being too eager to read ambiguous and caveat-filled guidance from the FDA as stating a definitive agency position, meaning the FDA’s later rejection of manufacturers’ applications was not inconsistent with any actual position announced by the agency in the run-up to its decisions.

For nearly a hundred years, the FDA has had the statutory authority to regulate drugs, which it does primarily through an extensive premarket authorization process. Until 2009, however, the FDA’s regulatory authority did not extend to tobacco products. That changed when Congress enacted the Family Smoking Prevention and Tobacco Control Act of 2009, which gave the FDA the authority to regulate the marketing, sale, and distribution of tobacco products. Among other regulatory tools, that law subjected any “new tobacco products”—defined as products not marketed in the United States before February 15, 2007—to a premarket authorization process not unlike the one the FDA uses for drugs. Initially, this new regulatory authority didn’t mean a whole lot, because the most commonly used tobacco products were hardly “new.” But in the 2010s, the use of e-cigarettes, particularly flavored e-cigarettes, exploded among younger Americans. In 2016, the FDA responded by issuing a rule finding that e-cigarettes and e-liquids (the liquid one buys to refill some varieties of e-cigarettes) constituted “new tobacco products” under the TCA, subjecting them to the Act’s pre-authorization requirements even though these products had by then been on the market for several years.

To give manufacturers of e-cigarettes time to obtain “premarket” authorization for the products they had already been selling, the FDA announced that it would be delaying enforcement of any ban on e-cigarette products for a few years. Any manufacturers who wanted to continue to sell their products after that, though, would have to apply for and receive preauthorization. To help them with the application process, the FDA issued various pieces of external guidance and developed internal memoranda addressing how applications would be reviewed and evaluated. While the FDA’s guidance as to its thought process is too long to summarize here, it addressed four main topics: (1) evidence manufacturers would need to provide regarding their products’ health effects, (2) comparisons of the risk posed by their products compared with other tobacco products, (3) the FDA’s enforcement priorities, and (4) information about manufacturers’ marketing plans for their products.

The FDA ultimately received applications from more than 500 companies covering 6.5 million e-cigarette products, including several from respondents Wages and White Lion Investments. The FDA denied a substantial portion of these applications, including essentially all applications for flavored e-cigarette products, finding that these products were particularly attractive, and hence harmful, to young adults. Wages and White Lion Investments’ products were among those rejected by the FDA. It sought review of the FDA’s denials in the Fifth Circuit, where a divided merits panel affirmed the FDA’s decision. But the Fifth Circuit then took up the case en banc and reversed, with that court’s majority concluding that the FDA had acted arbitrarily and capriciously. In the en banc majority’s view, the FDA had performed a “surprise switcheroo” (their term) by changing its position on things like the kind of scientific evidence it expected manufacturers to provide in support of their application and by not even considering manufacturers’ marketing plans. Because that decision conflicted with those of other circuits, the Court granted certiorari.

In an opinion written by Justice Alito, the court unanimously vacated the en banc Fifth Circuit’s decision. Alito began by trying to pin down exactly what this case was about. While the en banc Fifth Circuit’s decision was complex, it ultimately boiled down to the concern that the FDA had effectively changed the requirements for premarket authorization between the time when it issued guidance to manufacturers and the time when it denied their applications. Administrative law addresses that concern with the “change-in-position” doctrine, which essentially asks whether an agency has changed its course and if so whether it offered satisfactory reasons for the change. For the most part, the Fifth Circuit’s criticisms of the FDA’s application decisions failed on the first prong: The FDA had never taken a “position” in the first place, so there was no “change” it needed to explain. Take, for example, the FDA’s stance on the type of scientific evidence it expected manufacturers to provide regarding their products’ health risks. While the respondents pointed to various snippets in FDA pronouncements suggesting that one or another type of study would or would not be looked on favorably, none of these statements amounted to a “hard-and-fast commitment” from the agency; they were full of the hems, haws, and caveats one typically sees in regulatory guidance. The same was true for the Fifth Circuit’s conclusions regarding comparative studies and agency priorities: For each, the Supreme Court saw no clear pre-application statement from the FDA about what it was expecting that was at odds with the ultimate grounds for its application decisions. On all these points, the basic flaw with the Fifth Circuit’s analysis was too much tea-leaf reading: The change-in-position doctrine requires a true change in a definitive policy, not an inconsistency in the hints one might draw from vague agency pronouncements.

That said, there was one aspect of the FDA’s decisionmaking where the Court was inclined to see a change of position: marketing plans. As the FDA itself conceded, it had changed its position on that point, as it indicated in its guidance documents that it would consider manufacturers’ plans, but when it came time to act on their applications, it ignored the plans entirely. But that was not a basis to vacate its decisions, the FDA argued, because any change in position was harmless: nothing in those ignored marketing plans could have saved applications, like Respondents’, that were deficient in other respects. But is harmless error applicable when courts review agency action? The en banc Fifth Circuit concluded it was not, interpreting the Court’s per curiam decision in Calcutt v. FDIC (2023) to mean that when an agency has erred, a court must remand the case to the agency for re-consideration unless the agency was legally required to take the action it ultimately took. That reading of Calcutt was too strong, Alito concluded, as prior decisions had recognized that remand to an agency is not required when the agency’s error “had no bearing on the procedure used or the substance of the decision reached.” So when exactly is an agency’s error harmless? Alito declined to say, deciding that the better course was to simply vacate the Fifth Circuit’s decision, which relied on its misreading of Calcutt and let that court decide in the first instance whether the Respondents had shown enough prejudice from the FDA’s change in position.

Justice Sotomayor concurred with one-paragraph intended to “clarify” a point. While Alito’s opinion sometimes characterized the FDA as an agency “feeling its way toward a final stance” and “unable or unwilling to say in clear and specific terms precisely what applicants would have to provide,” she thought its guidance was better seen as reasonably giving manufacturers flexibility to decide for themselves what sort of evidence was necessary.