This TaxProf Op-Ed on Learning Resources is by Jon Endean (Brooklyn Law School; SSRN):

Delegations of the Taxing Power

Jon Endean

The Court’s decision in Learning Resources, Inc. v. Trump is undoubtedly one of the most consequential of the term, firmly rejecting the President’s position that IEEPA delegates the power to impose tariffs to the President. According to the Court’s opinion, “Our task today is to decide only whether the power to ‘regulate . . . importation,’ as granted to the President in IEEPA, embraces the power to impose tariffs. It does not.”¹

While Learning Resources is fundamentally a case of statutory interpretation, lurking in the background of the statutory question is the distinction between two enumerated powers in Article I—the power to lay and collect taxes, duties, and imposts, on the one hand, and the power to regulate foreign and domestic commerce, on the other hand—and how those powers can, in turn, be delegated to the President. While IEEPA reflects an explicit delegation of regulatory authority over foreign commerce, the question presented can, at one level, be understood as asking whether that delegation can be read to also include an implicit delegation of Congress’ taxing power as well. Beyond the immediate relevance to President Trump’s tariff program, the Court’s answer to this question can help inform how it views the intersection between those two enumerated powers more broadly.

In ruling against the government, the Court recognized that even if tariffs perform a regulatory function, they nonetheless retain their essential character as a tax: “[T]he power to regulate commerce is entirely distinct from the right to levy taxes. That Congress did not grant those authorities separately here is strong evidence that ‘regulate’ in IEEPA does not include taxation.”² The implication of this distinction is significant: While Congress has the power to regulate commerce by means of tariffs, the Commerce Clause, standing alone, is not sufficient to confer on Congress (and, by delegation, the President) the authority to use tariffs as a tool for regulating commerce.

At least at one level, this much is probably not particularly controversial. The fact that a revenue-raising tax has an incidental effect on the flow of commerce does not mean that the power given to Congress through the Commerce Clause is sufficient to give Congress the authority to impose such a tax; to hold otherwise would be to render the Taxing Clauses largely superfluous in the commercial context. Nonetheless, in the case of tariffs, the question remaining is whether there is a point at which the regulatory purpose of such a tariff overwhelms the revenue-raising character of the tariff such that it is more properly understood as an exercise of Congress’ commerce power than it is an exercise of Congress’ taxing power.

For the majority, the answer to this question is “no.” By contrast, the principal dissent, written by Justice Kavanaugh, rests on the premise that because tariffs have long been used as a tool of regulating foreign commerce (a fact that is hardly disputable), inherent in the power to regulate also is the power to impose a tariff as a tool of regulation.³ Similarly, given that IEEPA allows the President the ability to ban imports altogether, it may seem counterintuitive to hold that something far less extreme (a de minimis tariff, for example) would not be acceptable.

In his piece yesterday, Professor Avi-Yonah gives a good example of just such a tariff that remains in place, as it was imposed under a separate statute—the 100% tariff on EVs imported from China. This is a prohibitive tariff if there ever was one. Thus, as Professor Avi-Yonah asks in his piece, can “a tariff that is so high that it operates as a ban” really be said to be an exercise of the taxing power?

The idea that IEEPA could authorize a total ban while not authorizing a tariff came up at oral argument, where Justice Kavanaugh described this tension as a sort of “donut hole” in the statute. Ben Gutman, arguing on behalf of the state challengers, responded by arguing that this wasn’t a donut hole, but “a different kind of pastry” altogether.⁴ After all, the power to impose a tariff—which not only has a regulatory effect but also redounds to the benefit of the Treasury—is a different kind of power. While Chief Justice Roberts opted not to adopt Mr. Gutman’s colorful illustration, his opinion for the Court alluded to this distinction:

[T]ariffs . . . are different in kind, not degree, from the other authorities in IEEPA. Unlike those authorities, tariffs operate directly on domestic importers to raise revenue for the Treasury. Even though a tariff is, in some sense, “less extreme” than an outright compulsion or prohibition, it does not follow that tariffs lie on the spectrum between those poles. They are instead ‘very clear[ly] . . . a branch of the taxing power,’ and fall outside the spectrum entirely.⁵

The Court’s answer, then, is that even those tariffs that are driven by a regulatory purpose are, for the purposes of understanding the authority to impose such tariffs, taxes. The tariffs on Chinese EVs are illustrative, because while they have proved prohibitive in terms of allowing Chinese EV makers widespread access to American markets, importing them is not illegal in the same way that a ban would be. (In fact, over 1,000 Chinese EVs have been imported since President Biden levied the 100% tariff.) Moreover, unlike an outright ban, every vehicle that does get imported results in a collection of revenue to the government—in this case, over $30 million.⁶ Ultimately, the dispositive characteristic of a tariff is the legal structure that results in an exaction accruing to the benefit of the fisc, not on whether the economic effect resembles the different legal structure of a ban.

Looking forward, one key takeaway from this decision is to recognize two important doctrinal pieces. First, at least when it comes to tariffs, a regulatory purpose driving imposition of a tariff does not remove it from the ambit of Congress’ taxing power. While not every governmental charge is a tax (fines being a familiar example), the opinion suggests that regulating imports by imposing duties cannot be justified solely by reference to a general grant of authority to regulate imports; rather, the imposition of any such duties necessarily implicates the taxing power.

Second, “[w]hen Congress grants the power to impose tariffs, it does so clearly and with careful constraints,” and a statute that fails to do either cannot be read as containing the authority to impose tariffs.⁷ On one level, this framing from the opinion could be viewed as requiring a “clear statement” for Congress to delegate its taxing authority. However, while Chief Justice Roberts wrote that “the President must ‘point to clear congressional authorization’ to justify his extraordinary assertion of the power to impose tariffs,”⁸ this portion of the opinion was not joined by Justices Kagan, Sotomayor, and Jackson, and so the Court as a whole stopped short of imposing a “clear statement” rule on delegations of the taxing power.⁹

Learning Resources is certainly not going to be the last word on tariffs, but it represents an important milestone for this Court’s jurisprudence when it comes to the taxing power. In short, it reinforces the distinct constitutional status of the taxing power by treating tariffs—even regulatory ones—as exercises of the taxing power for purposes of delegation. The Constitution did not collapse commercial regulation and revenue generation into a single undifferentiated power. Rather, it granted the two powers separately, and the Court’s reasoning in Learning Resources preserves that separation when it comes to delegating those powers to the President.

1. Learning Res., Inc. v. Trump, No. 24-1287, slip op. at 16 (U.S. Feb. 20, 2026). ↩︎
2. Id. at 15. ↩︎
3. Left unaddressed in Justice Kavanaugh’s dissent is whether a tariff imposed for the primary purpose of raising revenue would still be permitted under the text of IEEPA. ↩︎
4. See Tr. of Oral Arg. at 165–66, Learning Res., Inc. v. Trump, No. 24-1287 (U.S. argued Nov. 5, 2025). ↩︎
5. Learning Res., slip op. at 17 (quoting Gibbons v. Ogden, 9 Wheat. 1, 201 (1824)). ↩︎
6. To be sure, this figure is a rounding error on a rounding error in terms of the size of the federal budget. But the amount of revenue that a tax generates—or could generate—is not dispositive as to its fundamental character. That observation, of course, was part of what buttressed the Court’s (controversial) holding in NFIB v. Sebelius, 567 U.S. 519 (2012). ↩︎
7. Learning Res., slip op. at 17. ↩︎
8. Id. at 13 (Roberts, C.J.) (quoting Biden v. Nebraska, 600 U.S. 477, 506 (2023)). ↩︎
9. Indeed, Justice Kagan expressly disclaimed the need for such a rule in her concurrence, which was joined by Justices Jackson and Sotomayor: “The use of a clear-statement rule here is unnecessary because ordinary principles of statutory interpretation lead to the same result.” Learning Res., slip op. at 3 (Kagan, J., concurring). ↩︎

Other TaxProf Blog posts in this series:

• Avi-Yonah on Learning Resources and Regulatory Taxation (Feb. 23, 2026)
• The Supreme Court Strikes Down Trump’s IEEPA Tariffs (Feb. 21, 2026)