Learning Resources, Inc. v. Trump: An Empirical Breakdown of the Court’s IEEPA Tariff Decision

A statistical and comparative romp through the opinions, briefs, and oral argument to assess the impact of this landmark decision

The Decision

We decide whether the International Emergency Economic Powers Act (IEEPA) authorizes the President to impose tariffs.

That is the first sentence of Chief Justice Roberts’s opinion for the Court in Learning Resources, Inc. v. Trump, No. 24-1287, decided February 20, 2026. The case arose from a challenge to broad tariffs the executive branch imposed pursuant to IEEPA’s grant of authority to “regulate . . . importation.” The Court’s answer was unambiguous:

The President asserts the extraordinary power to unilaterally impose tariffs of unlimited amount, duration, and scope. In light of the breadth, history, and constitutional context of that asserted authority, he must identify clear congressional authorization to exercise it. IEEPA’s grant of authority to “regulate . . . importation” falls short. IEEPA contains no reference to tariffs or duties. The Government points to no statute in which Congress used the word “regulate” to authorize taxation. And until now no President has read IEEPA to confer such power. We claim no special competence in matters of economics or foreign affairs. We claim only, as we must, the limited role assigned to us by Article III of the Constitution. Fulfilling that role, we hold that IEEPA does not authorize the President to impose tariffs.

The decision resolved companion case Trump v. V.O.S. Selections by affirming the lower court judgment there, and vacated and remanded the district-court judgment in Learning Resources itself for dismissal on jurisdictional grounds. The merits holding, however, is unqualified: IEEPA is not a tariff statute.

II. The Court’s Coalition Geometry

The Vote and Join Structure

Chief Justice Roberts announced the judgment and wrote the principal opinion, but that opinion carries two distinct coalition tiers — a feature you can measure directly from the join statement.

Parts I, II-A-1, and II-B constitute the Opinion of the Court, joined by Justices Sotomayor, Kagan, Gorsuch, Barrett, and Jackson — a six-Justice coalition. These sections do the statute’s core work: reading the text, examining the constitutional backdrop, and concluding IEEPA authorizes no tariffs.

Parts II-A-2 and III, which contain the major questions doctrine analysis and the conclusion requiring “clear congressional authorization,” were joined only by Justices Gorsuch and Barrett — making those portions a three-Justice plurality, not a majority holding.

Beyond Roberts, six separate writings were filed:

• Gorsuch concurred (joined the Roberts opinion in full, wrote separately on nondelegation)

• Barrett concurred (joined the Roberts opinion in full, wrote briefly on major questions scope)

• Kagan concurred in part and in the judgment, joined by Sotomayor and Jackson (agreed on result; rejected the major questions framing)

• Jackson concurred in part and in the judgment (agreed on result; relied on legislative history)

• Thomas dissented separately

• Kavanaugh dissented, joined by Thomas and Alito

The Citation and Length Data

The volume and character of separate writings is itself a data point. The chart below plots each writing by page count against case-citation occurrences — counting “X v. Y” style citations with Id. citations removed.

Figure 1. Each point represents one writing. Kavanaugh’s dissent (63 pages, 87 citation occurrences) and Gorsuch’s concurrence (46 pages, 46 occurrences) dominate the citation landscape. Roberts’s Opinion of the Court (21 pages, approximately 30 occurrences) is comparatively lean. Barrett, Kagan, Jackson, and Thomas cluster near the origin.

The raw numbers tell a story about where the Court’s reasoning load is concentrated. Kavanaugh’s dissent is the longest writing and carries the heaviest citation burden — 53 unique cases and 87 total occurrences. Gorsuch’s concurrence is second-longest at 46 pages and 44 unique cases. Roberts’s Opinion of the Court cites 28 unique cases in 21 pages. Barrett, Kagan, and Jackson are brief and lightly cited.

The semantic/textual similarity (Jaccard measure) between the Roberts opinion’s case set and Kavanaugh’s dissent’s case set — a measure of citation overlap between 0 and 1 — is approximately 9.3 percent. That reflects a narrow shared core of framework precedents (Biden v. Nebraska, West Virginia v. EPA, Algonquin SNG, Youngstown, Utility Air, and Yoshida Int’l) set against very different wider citation universes. The majority and the lead dissent are largely talking past each other at the level of precedent.

III. How Well Did My Prior Prediction Hold Up?

In my prior Legalytics piece, “The $133 Billion Question: Inside the IEEPA Tariff Case,” I made a series of empirical predictions before argument. With the decision now in hand, here is the scorecard:

The core predictions held. The vote count and Kavanaugh’s position were both secondary possibilities and the length of the opinion correlation with why it took longer than expected seems correct as well (170 pdf pages).

IV. What Each Opinion Said and How They Compare

The Roberts Opinion of the Court — Six Justices

The six-Justice opinion does the statute’s work. It reads “regulate . . . importation” against its textual neighbors and statutory context, observes that IEEPA contains no reference to tariffs or duties, and notes the Government could not identify a single statute in which Congress used “regulate” to authorize taxation. The opinion draws on Gibbons v. Ogden for the proposition that tariffs are “a branch of the taxing power,” sitting categorically outside the compel-to-prohibit verb spectrum the Government proposed. It also uses an export-tax symmetry: if “regulate . . . exportation” included taxation, IEEPA would authorize what Article I expressly forbids.

The Roberts Plurality — Major Questions — Three Justices

The narrower Roberts-Gorsuch-Barrett plurality (Parts II-A-2 and III) applies the major questions doctrine. It requires “clear congressional authorization” for the claimed power, holds there is no emergency exception to that requirement, holds that foreign-affairs implications do not displace the doctrine, and treats as “telling” that in IEEPA’s fifty-year existence no President had invoked it to impose tariffs.

The Concurrences

Gorsuch (46 pages) joins Roberts in full and writes the document’s second-longest opinion to defend the major questions framework and address nondelegation concerns.

Barrett (4 pages) joins Roberts in full and writes briefly to address one conceptual point: that the major questions doctrine is not a “magic words” test, engaging with the Biden v. Nebraska framework.

Kagan (7 pages, joined by Sotomayor and Jackson) joins the six-Justice portions but not the major questions plurality. Her core position is stated explicitly: ordinary tools of statutory interpretation are sufficient; she describes proceeding “as I did in West Virginia and Nebraska“ — meaning she does not need the major questions thumb on the scale to reach the same result.

Jackson (5 pages) joins the same portions as Kagan and writes separately to rely on the legislative record — House and Senate Reports — to anchor congressional intent when IEEPA was enacted. This is the opinion’s primary legislative-history-positive writing, and it stands in direct methodological tension with Gorsuch’s explicit caution against those same materials.

The Dissents

Thomas dissented separately.

Kavanaugh (63 pages, joined by Thomas and Alito) is the longest writing in the case and the most comprehensive defense of the Government’s position. His dissent:

• Structures the case using Justice Jackson’s Youngstown framework, placing the President in category one — acting pursuant to congressional authorization

• Argues “regulate . . . importation” and “adjust . . . imports” (the Section 232 language upheld in Algonquin) are not meaningfully distinguishable

• Contends the Nixon-era TWEA tariff episode means the claimed authority was not “unheralded”

• Argues the major questions doctrine has never been applied to a foreign affairs statute and should not be applied here, citing Department of Navy v. Egan

• Warns that applying the majority’s new approach would likely have altered outcomes in prior decisions including Dames & Moore and Hamdi

The intra-Court methodological conflict across the writings is stark and measurable. The phrase “major questions” and “clear congressional authorization” function as decisive analytical tools in Roberts, Gorsuch, and Barrett. Kagan explicitly states those tools are unnecessary. Jackson relies on legislative history that Gorsuch expressly cautions against. Kavanaugh argues the doctrine should not apply in this domain at all.

V. Opinions vs. Briefs: What Was Adopted, What Was Rejected

The merits briefs — the Government’s opening brief, the State Respondents’ brief, and the Private Respondents’ brief (Learning Resources) — allow a document-grounded uptake analysis. Here is what the Court adopted and what it rejected from each side.

What the Court Adopted (from challengers’ briefs)

• Tariffs as taxing power. The challengers argued that tariffs are “a branch of the taxing power” under Gibbons v. Ogden and therefore categorically different from regulatory tools like quotas and embargoes. The Roberts Opinion of the Court adopted this framing directly.

• Major questions applies; no emergency exception. The challengers argued the claimed authority is “breathtaking” and “unprecedented” and requires clear statutory authorization. The plurality adopted this framing and expressly held there is no emergency-statute carveout.

• No foreign-affairs exception to MQD. The challengers argued the foreign-affairs context does not flip the interpretive presumption. The plurality agreed and rejected the Government’s proposed carveout.

• Algonquin is contextually distinct. The challengers argued Section 232 sits within a framework that explicitly references “duties” and uses broader discretion-conferring language. Roberts adopted this distinction and declined to extend Algonquin’s “expressly limited” holding.

• Lack of historical precedent is telling. No President had used IEEPA for tariffs in fifty years. The plurality treated this as an affirmative indicator against the Government’s reading.

What the Court Rejected (from the Government’s brief)

• The “poles/spectrum” theory. The Government argued “regulate” sits between “compel” and “prohibit,” capturing less extreme tools including tariffs. The Court rejected this directly, finding tariffs “operate directly on domestic importers to raise revenue” and are not simply a milder embargo.

• Both anti-MQD carveouts. The Government argued MQD should not apply to emergency statutes and that foreign affairs context flips the interpretive presumption. Both were expressly rejected.

• Algonquin as near-controlling. The Government argued “regulate . . . importation” and “adjust . . . imports” are functionally equivalent and Algonquin is essentially controlling. Rejected by the majority; fully embraced by Kavanaugh’s dissent.

• Dames & Moore as broad emergency authority. The Government cited it to support expansive foreign-affairs power. The Court treated it as an illustration of statutory narrowness and used it against the Government’s reading.

The Kavanaugh dissent is the primary vehicle for the Government’s brief architecture — adopting the textual argument, the Algonquin a fortiori framing, the Yoshida historical-practice argument, and the foreign-affairs MQD carveout in both sequence and emphasis.

VI. Oral Argument to Written Opinions

The Court heard argument on November 5, 2025. The transcript provides a measurable bridge between what the Justices pressed at argument and where they landed in the written opinions.

Figure 2. Measured by count of “?” appearing in each Justice’s speaking turns in the official transcript. Barrett (49), Gorsuch (38), and Alito (38) were the highest-volume questioners; Thomas (6) and Kagan (12) the lowest.

Question volume does not directly track final vote, but it identifies where doctrinal and factual pressure concentrated at argument. Barrett led at 49 question marks, followed by Gorsuch and Alito at 38 each, Roberts at 35, Jackson at 28, Sotomayor at 26, Kavanaugh at 17, Kagan at 12, and Thomas at 6.

Four Question Forms — and Where They Landed

Questions in the transcript fall into four observable forms based on their linguistic structure. Each maps onto specific opinion passages:

• Authority-demand — “JUSTICE BARRETT: Can you identify any statute that used that phrase to confer tariffs?”

  Justice Barrett’s challenge to the Solicitor General from oral argument in this exact form became a load-bearing beam in the Roberts opinion, which expressly notes the Government “points to no statute” where “regulate” authorizes taxation.

• Skeptical framing — “Why should we think ‘regulate importation’ includes taxing importations?” States a proposition and asks why it does not cut against the Government. Justice Sotomayor’s export-tax symmetry question — pressing the point that if exporting cannot be taxed under the Constitution, why treat importation differently — maps directly onto the constitutional-signpost argument the Court adopted.

• Precedent confrontation — Extended setup, direct quotation of a case, then a pointed demand. Justice Kavanaugh’s Algonquin sequence at argument — quoting the decision’s language about monetary versus quantitative methods and pressing the Government on why “adjust imports” should be read differently from “regulate importation” — tracks precisely into the Algonquin analysis anchoring his dissent. Notably, Justice Barrett also pressed Algonquin hard at argument (asking specifically why the license/fee distinction matters given IEEPA’s own licensing authority) but ultimately joined the majority’s Algonquin-distinguishing reasoning rather than Kavanaugh’s a fortiori position.

• Doctrinal framing — Justice states a test, then interrogates within it. Justice Thomas’s direct question asking why the major questions doctrine does not apply to the President maps onto the plurality’s holding that it does. Justice Gorsuch’s use of “unheralded” as a framing device at argument, and his questions about what Congress understood when it enacted IEEPA relative to Nixon and Yoshida, appear in substantially the same form in his concurrence.

A fifth form — consequence and remedy — pressing the practical fallout of a ruling, including refunds and administrative disruption, appears at argument and lands almost exclusively in Kavanaugh’s dissent in the written opinions.

Argument Volume and Opinion Length

The argument-to-opinion correlation is strongest for the Justices who wrote extensively. Gorsuch and Kavanaugh both asked high volumes of doctrinal-framing and precedent-confrontation questions at argument, and both produced long written opinions anchored in the same frameworks they were probing. Barrett asked the highest volume of questions but produced the shortest concurrence; her argument questions focused on clarifying the scope of Algonquin and the “magic words” problem, and her brief writing addresses exactly those scope questions. Kagan and Thomas asked the fewest questions; Kagan’s concurrence is brief and methodologically focused, consistent with having entered argument already confident in her analytical path.

VII. What Comes Next: Refunds and Remaining Uncertainty

The Court’s holding resolves the statutory authority question. It does not resolve the refund question. The majority’s operative holding is that IEEPA does not authorize the President to impose tariffs; the opinion does not set out a refund mechanism, does not order restitution, and does not address the administrative processes by which duties already paid might be recovered.

The only explicit discussion of refunds in the slip opinion appears in Justice Kavanaugh’s dissent. He warns that the United States “may be required to refund billions of dollars to importers who paid the IEEPA tariffs,” and describes the refund process as likely to be a “mess” — citing the oral argument transcript for that characterization. He also flags the pass-through problem: whether importers who passed tariff costs to consumers can recover at all is a further downstream complication.

That the refund discussion is concentrated entirely in the dissent, rather than anywhere in the majority’s rule statement, is itself a measurable feature of the opinion: the Court resolved the legal question of authority and left the remedial mechanics entirely to future proceedings.