The most important legal development from Learning Resources v. Trump, the Supreme Court case that overturned President Donald Trump’s tariffs under the International Emergency Economic Powers Act (IEEPA), is not about tariffs themselves. The key but overlooked news is that three conservative justices—Chief Justice John Roberts and Justices Neil Gorsuch and Amy Coney Barrett—endorsed the “major questions doctrine” (MQD) not only for bureaucratic agencies but also for the president. They did so even in a case involving foreign affairs, an area of core executive responsibility.
Under the MQD, government officials who undertake novel, “unheralded” administrative initiatives of unusual economic and political significance must be able to cite statutes that authorize their initiatives “clearly.” Merely relying on the broad wording of a statute which, read literally, might encompass the challenged action is not enough.
In the case of tariffs, IEEPA’s permission to “regulate … the importation” of goods was too vague to authorize presidential tariffs. The three conservative justices in the majority agreed with their three liberal colleagues that standard guides to statutory interpretation pointed to a victory for Trump’s challengers. But Roberts, along with Gorsuch and Barrett, went further, supporting their analysis with the MQD. They argued that IEEPA’s text was insufficient because no president had previously used it to impose tariffs, and Trump’s actions would represent a “transformative expansion” of presidential authority over tariffs and the economy.
But tariffs are not the only area where Trump has exercised administrative power in “unheralded” and “transformative” ways that are of exceptional economic and political significance. The reasoning of three conservative justices in the Learning Resources majority—call them “the MQD3”—now implicitly challenges Trump’s unilateral restructuring of the civil service and reorganization of executive branch agencies. A test of their logic will likely be provoked by a recent change in civil service rules by the Office of Personnel Management (OPM) that allows the at-will firing of any government career professional whose work affects government policymaking. When the rule’s legality reaches the Supreme Court, as is likely, the votes of the MQD3 will provide the true measure of Learning Resources v. Trump.
By statute, Congress has divided the federal civil service into three segments: the competitive service, which is hired through competitive examination; the excepted service, which is hired without a competitive exam; and the senior executive service, about 8700 senior personnel, mostly career professionals, who manage agencies just below the level of presidential appointees. The excepted service, in turn, is divided into “schedules” based on the rationale for exempting members from the exam. (Schedule A, to take one example, includes positions for which an examination is impracticable.) Every civil service segment includes career employees who are likely to serve in government for years and across administrations. The excepted service Schedule C includes most political employees, those whose positions almost always change with administrations.
To grasp Trump’s goals, it’s crucial to recognize that, under federal civil service law, most members of both the competitive and excepted services have job protection. Actions, such as dismissal or even long-term suspension, entitle job-holders to advance notice and formal hearing rights. Congress has provided, however, that the president may remove such protections if two conditions are met: The employee is a member of the excepted service and their “position has been determined to be of a confidential, policy-determining, policy-making or policy-advocating character.” As I will explain, that phrase—“of a confidential, policy-determining, policy-making or policy-advocating character”—is one of two statutory phrases that will determine the scope of Trump’s authority under civil service law, similarly to how “regulate … importation” was pivotal to his authority under the IEEPA.
What Trump ordered OPM to do, and what OPM has published final rules to accomplish, is to allow the transfers of any career professional out of the competitive service whose work could affect policymaking by others. It will be said that their work, no matter how nonpartisan, is of “a confidential, policy-determining, policy-making or policy-advocating character.” On that basis, these professionals could be dismissed at will. Such employees may include climatologists who refuse to deny climate change, epidemiologists who emphasize the importance of vaccination, statisticians who challenge the notion that tariffs create manufacturing jobs, or museum curators who refuse to alter placards to deny that the antebellum South’s wealth depended on slavery.
Naturally, political judgment sometimes and often appropriately influences the extensive scope of administrative policymaking carried out by the federal bureaucracy. However, our administrative law framework expects that most regulatory functions of government agencies—such as the Environmental Protection Agency, the Federal Trade Commission, the Consumer Financial Protection Bureau, and others—are primarily guided by career professionals who apply reasoned judgment in a largely depoliticized manner, based on carefully compiled factual records. These include employees like the scientists, economists, and other specialists mentioned earlier.
Near the end of his first term, Trump sought to purge such career professionals by directing OPM to create a Schedule F that would remove them from the competitive service and deny them their previous job protections. Under President Joe Biden, OPM revoked the Schedule F program and implemented a 2024 rule to protect career professionals from arbitrary treatment. Trump’s OPM has revoked the Biden rule and resuscitated the Schedule F idea, now dubbed “Schedule Professional/Career.” Schedule P/C, however, will be administered in one profoundly different way from Schedule F: Trump will implement it personally.
Easily lost within the 78 pages of triple-column prose that OPM’s new rule and its preamble occupy in the Federal Register is this telling paragraph:
In a structural difference with [Trump’s Schedule F order], the President—not OPM—will now move positions into Schedule Policy/Career. Pursuant to [Trump’s new order] E.O. 14171, agencies will assess their workforces and petition OPM to recommend that the President move specific positions into Schedule Policy/Career. OPM will review these petitions and make the recommendations it deems appropriate. However, the President will make the final decision about which positions go into Schedule Policy/Career. That decision will be effectuated by a new executive order issued under Presidential—not OPM—authority.
In essence, Trump will make it clear that he alone determines the career outcomes of any federal employee he targets. Moreover, shifting the ultimate decision-making power from OPM to the president will create a procedural complication for legal challengers, since the president, unlike OPM, is not classified as an “agency” under the Administrative Procedure Act.
What does the Learning Resources tariff case have to do with all this? At one level, Learning Resources was quite straightforward. Although the justices’ opinions will occupy 170 pages of the United States Reports when published, only 14 pages, including the statement of facts, were needed for six justices to explain why Trump overstepped. What proved fatal to Trump’s tariffs were the so-called traditional rules of statutory construction. The “ordinary meaning” of “regulate” does not include duties. The verbs surrounding “regulate”—such as “investigate,” “block,” and “nullify,” among others—have nothing to do with raising revenue. In the IEEPA, if “regulate” permits tariffs on imports, it would also permit tariffs on exports, which would be unconstitutional. To be sure, there were nontrivial arguments to the contrary—aired at length in Justice Brett Kavanaugh’s principal dissent—but this portion of the majority opinion drew exclusively on tools familiar to first-year law students.
What occupied most of the prose in the justices’ analyses wasn’t the meaning of “regulate” but how to decide the meaning of “regulate.” This is where the MQD comes in. Its animating idea is that, when Congress wants the executive branch to undertake a pathbreaking administrative task, it will spell it out. Though the proposition has roots in earlier cases, the Court’s 2022 decision in West Virginia v. EPA crystallized what sounded like a new formula for assessing the legality of especially ambitious administrative uses of the powers Congress grants to the executive branch via statute.
West Virginia reviewed President Barack Obama’s clean air regulations, which could have caused a major restructuring of the national electric power market. The Court needed to determine whether it qualified as a “major question” warranting extra scrutiny. This required examining the “history and the breadth of the authority claimed by the agency” and the regulation’s economic and political importance. If the initiative would lead to an “unheralded” and “transformative expansion” of the EPA’s regulatory powers, then the agency had to find “clear congressional authorization” for its actions in the statute.
The West Virginia majority concluded that the challenged regulations were sufficiently novel and far-reaching to warrant MQD review and that the Clean Air Act language lacked the clarity necessary to sanction the EPA’s action. In the same manner, according to Roberts, Gorsuch, and Barrett, the MQD framework was also fatal to Trump’s tariffing.
The consensus on the Court disintegrates when discussing the necessity, relevance, and meaning of the MQD. The three liberal Justices, exemplified by Justice Elena Kagan’s main concurrence in the tariffs case, see the MQD as unnecessary. Kagan prefers to rely on “standard rules of statutory construction,” which she summarizes as “reading text in context.” On the other hand, Justice Barrett, a conservative, considers the MQD a useful interpretive tool to help judges determine what a statute likely intended to communicate. She describes it as “an ordinary application of textualism,” used “to ascertain a text’s ‘most natural meaning.’”
But for the five conservatives on the Court besides Barrett—namely, Roberts, Gorsuch, and, in other cases, Kavanaugh, along with Justices Samuel Alito and Clarence Thomas—the MQD means something more. As Gorsuch lays out in his 46-page law review-like concurring opinion, the MQD protects a particular understanding of the separation of powers. Because the Constitution explicitly vests legislative power only in Congress, Gorsuch and most of the conservative wing habitually doubt any grant of administrative power that shifts significant policymaking from Congress to the executive branch. It is Congress, not the executive, which ought to make the policy judgments that they say amount to “legislating.” These include judgments as to whether, under what circumstances, and for how long to impose tariffs. Giving the executive branch too much leeway would, in their eyes, mean abdicating power to the executive that should be exercised only by Congress.
Learning Resources provides a template for assessing Trump’s intended changes to the civil service. At some point, agencies will send lists of vulnerable positions to OPM, OPM will recommend that Trump include them under a new Schedule P/C, and employees or their union will sue. The Justice Department will argue that the civil service law sustains Trump’s initiative. One provision authorizes the president to “prescribe rules governing the competitive service,” including “necessary exceptions of positions from the competitive service,” but only if the “conditions of good administration warrant.” That last phrase will be the first text the eventual lawsuit targets. Trump’s lawyers will argue, as OPM has, that the “good administration” standard allows moving positions to the excepted service to address “problems of poor performance, misconduct, and corruption in the civil service.” Challengers will argue that, in context, the “good administration” standard precludes undermining Congress’s goal of preventing patronage and partisanship in the hiring and firing of career employees.
Regarding stripping Schedule P/C members of job protections, Trump plans to rely on the above provision that allows the President to remove protections from excepted service employees if their role is considered “confidential, policy-determining, policy-making, or policy-advocating.” This will be the second point of contention. The Biden rule asserted that, in context, Congress gave the president authority to deny job protections only to political appointees—the category of employees expected to leave at the start of a new presidency. Trump is reading the phrase more broadly to cover any career employee whose work could influence policy. Trump is entitled to rid the civil service, OPM says, of any policy-influencing employee who would “obstruct the democratic process” by intentionally resisting the president’s agenda.
If the Learning Resources majority follows the logic of the MQD3, however, the civil service challengers in my hypothetical lawsuit will likely prevail. Trump’s order is unprecedented and of exceptional economic and political significance. The authority to create federal offices and to establish working conditions for the federal workforce is as foundational to legislative power as the power to set tariffs. The justices should block Trump’s overreach here as they did with trade policy.
But here’s the rub: In curbing Trump’s tariff authority, the Learning Resources majority produced a result entirely in line with conservative deregulatory orthodoxy. As pointed out by Josh Marshall of Talking Points Memo and a Washington Monthly contributing editor: “The architect of the current Court—the Federalist Society’s Leonard Leo—was behind the litigation that undid the tariffs. That tells you all you need to know.” In other words, Trump may fume that Learning Resources broke faith with him, but the decision is squarely in line with the conservative legal movement’s values.
Limiting presidential control of the civil service, however, pushes against the Roberts Court’s love affair with unitary executive theory. Roberts once wrote: “Thousands of officers wield executive power on behalf of the President in the name of the United States. That power acquires its legitimacy and accountability to the public through ‘a clear and effective chain of command’ down from the President, on whom all the people vote.” Such dicta implicitly cast the civil service more as a tool of the president than of Congress, which depends on the civil service to implement its legislation. Champions of Schedule P/C will no doubt quote Roberts in their briefs.
In principle, the MQD analysis embraced by the conservative justices in the Learning Resources majority should curb Trump’s effort to make civil servants his apparatchiks. Whether the MQD3 conservatives stick with their reasoning when it hampers the realization of conservative goals remains to be seen. But if the Roberts wing of the Roberts Court concludes that the case against Trump’s civil service overreach is as strong as the case against tariffs, critics of the Roberts Court will have to rethink at least some of their cynicism.
