Will SCOTUS allow Congress to protect non-partisan administration of the laws from plebiscitary presidentialism?

It is President's Day, so naturally presidential power and legitimacy are on my mind. In particular, I am wondering how far the Roberts Court will go in vindicating what Noah Rosenblum and I call "plebiscitary presidentialism." By "plebiscitary presidentialism," I mean the view that presidential elections give the President a uniquely democratic legitimacy in the American constitutional system, because presidents alone have a national constituency. As Roberts put it in Seila Law, "the Framers made the President the most democratic and politically accountable official in Government" because "[o]nly the President (along with the Vice President) is elected by the entire Nation. And the President’s political accountability is enhanced by the solitary nature of the Executive Branch, which provides “a single object for the jealousy and watchfulness of the people.” Sure, members of Congress are also elected, but they have parochial state- or district-specific constituencies and therefore cannot embody the will of the nation as does the President. Presidential elections, on this view, are national plebiscites that legitimize the executive branch in a way that congressional elections can never legitimize the legislature.

It follows that the President's "executive power" should be read generously to permit detailed supervision of every official in the executive branch. As Madison said (and Roberts repeatedly quotes), “the lowest officers, the middle grade, and the highest” all “depend, as they ought, on the President, and the President on the community.” In a paper that we recently posted on SSRN, Noah and I argue that United States v. Arthrex took a radical turn towards vindicating this vision of a plebiscitary presidency by protecting not merely the President's power to fire agency chiefs but also to tell every civil servant in the executive branch how to do their job. Arthrex is unique among the Roberts Court's decisions defending the unitary executive from Congress in providing the remedy of override rather than removability: The Arthrex Court re-wrote the Leahy-Smith America Invents Act to confer on the politically appointed Director of the Patent & Trademark Office the power to review and reverse the decisions of the politically insulated Patent and Trademark Boards. Only such a judicial re-write of the statute, the Court reasoned, would ensure that the latter constituted "inferior officers" who were under the direction and control of the President.

Noah and I argue that this extraordinary remedy has important and dangerous implications for the idea of administration of the laws by a non-partisan civil service. It is one thing to say that agency chiefs must serve at the pleasure of the President. It is altogether more radical to say that the President must be able to direct and control every decision by every inferior officer and employee in the executive branch. Statutes, after all, often require that technical criteria be implemented by properly qualified experts, ranging from peer review of NIH grants to recommendations of FDA scientific advisory committees to Bureau of Labor Statistics economists. Is Arthrex really serious that some politically appointed agency chief is constitutionally entitled to revise such decisions according to the chief's own ideas of, say, disease etiology, clinical controls, or statistical error?

After the jump, I will explain (1) why Arthrex might have precisely such an implication but (2) how SCOTUS might avoid plunging into the abyss of such politicized administration of the law by embracing an old concept of apolitical administration. Here's the punch line: That concept was precisely the idea defended by William Howards Taft, author of Myers v. United States and granddaddy of the "unitary executive."

1. The Potential Radicalism of Arthrex

First, consider how Arthrex radically threatens the commonplace idea that apolitical experts frequently have the last word on the implementation of the laws. According to Arthrex, "[t]he Constitution...forbids the enforcement of statutory restrictions on the Director that insulate the decisions of [Administrative Patent Judges] from his direction and supervision," because "[i]n this way, the President remains responsible for the exercise of executive power—and through him, the exercise of executive power remains accountable to the people." Congress, however, frequently requires that politically accountable agency chiefs defer to bureaucratic experts. Consider, for example, the statutory standards for review of drugs' clinical trials. 21 USC § 355(d) of the Food Drug & Cosmetic Act (FDCA) requires the Secretary of HHS to base such reviews on “substantial evidence,” defined as “evidence consisting of adequate and well-controlled investigations . . . by experts qualified by scientific training and experience to evaluate the effectiveness of the drug involved.” Further, §§ 355(n)(1) and (7) requires the Secretary to consider and make part of the administrative record "expert scientific recommendations" from dozens of scientific advisory committees.

As a practical matter, these statutory standards preclude Robert F. Kennedy Jr., our new Secretary of HHS, from reversing the approval of vaccines over the recommendations of his own staff and advisory committees: Under State Farm, any such decision would be arbitrary and capricious because it would ignore, or respond inadequately, to "important" evidence in the administrative record. (On the obvious importance of an agency staff's recommendations, see Dan Deacon's fine article here).

Do these requirements of scientific expertise in the FDCA violates the Arthrex principle that the bureaucratic experts must be under "the direction and supervision" of politically accountable superiors like RFK Jr.? One can argue that, so long as RFKJ can weigh in before courts decide the case by "reversing" his staff's recommendations and offering his own, the Arthrex principle is preserved. Noah and I would indeed be delighted by such a "minimalist" reading of Arthrex. Such a reading, however, reduces Arthrex to a fairly trivial requirement that agency chiefs be entitled to submit their own views along with the views of the civil servants that they supervise. If the statute directs courts to defer to the civil service rather than the agency chiefs, then in what sense do the chiefs "direct[] and supervis[e]" the civil service?

SCOTUS will be pretty busy, however, if it decides to invalidate statutory requirements that the politicians defer to the civil service. Such requirements are ubiquitous in the U.S. Code. The Bureau of Labor Statistics is statutorily charged with collecting "full and complete statistics of the conditions of labor and the products distribution of the products of the same." Such a standard obviously connotes that statisticians, not the Secretary of Labor will call the shots on statistical method with their Handbook of Methods. The National Institutes of Health are required to award grants based on peer review by "scientific peer review groups." Again, there is no way to read that standard without deferring to those peer groups of scientists appointed according to statutory procedures. The power of experts to call the shots extends beyond federal civil servants: By statute, Congress handed off power to define standards to "voluntary consensus standards bodies" like the American National Standards Institute.

Will the Court go full populist and say that Congress is constitutionally barred from requiring the executive branch to honor apolitical expertise in the name of some unwritten principle of plebiscitary presidentialism? Noah and I hope not: For practical reasons, such a step would seriously erode American governing capacity. (Consider this evidence from the USA's own Gilded Age from Abhay Aneja and Guo Xu). But misalignment between Republican and Trumpist politicians and a mostly Democratic civil service could lead to a purge that the Roberts Court might facilitate by striking down laws that insulate the career service from such attacks. Chief among these laws are simply those statutory standards that require decisions to follow the dictates of expertise rather than an agency chief's whim or partisan agenda.

2. Our Arthrex Compromise: Revive Taft's Distinction Between Political Control over "Policy," Bureaucratic Insulation for "Administration"

Noah and I would like to give the Roberts Court a third option between trivializing Arthrex (which we doubt they would do) and metastasizing it (which we wish they wouldn't do). That option is to revive William Howard Taft's distinction between "policy," which must be controlled by the President, and "administration," which can and indeed should be controlled by an apolitical and insulated civil service. "Administration" refers to the implementation of statutory standards that are defined by scientific expertise, technical competence, or practical know-how. "Policy" refers to implementation of statutory standards that leave a gap for presidents to fill with their own enforcement policies. As an example to illustrate "administration," consider the example of the Bureau of Labor Statistics' calculation of the consumer price index. The BLS must use "full and complete statistics," a standard that completely fills up any gap in the statutory standard with implementation rooted in statistical expertise. Politically accountable chiefs can, of course, oversee their economists' development of survey standards for assessing prices, but any such interventions must themselves be rooted in the science of statistics that precludes political control. By contrast, statutes routinely leave giant gaps on enforcement discretion -- i.e, the timing, target, and intensity of enforcement against particular persons. As Heckler v. Chaney noted, this enforcement discretion lies at the heart of the President's Article II power to "take Care that the Laws be faithfully executed," precisely because there is no statutory standard to serve as "a focus for judicial review."

Congress cannot delegate to bureaucrats the task of implementing a non-existent statutory standard, because inventing a standard to fill gaps in statutes is precisely what it means to execute a law, and Article II bars Congress from taking that power away from the President. The President is, in this sense, the gap-filler in chief. Where statutory standards fill that gap by reference to expert criteria, however, then the President is charged with executing only that standard -- and deferring to the civil servants with the relevant expertise is how presidents and their appointees carry out the statute.

Why should the Roberts Court limit Arthrex with such a distinction? The Court is, after all, committed to plebiscitary presidentialism: Should not Article II be read to give the president carte blanche to override their subordinates in the name of We the People?

One reason for the Roberts Court to hesitate in such a populist enterprise is that the administration/policy distinction we defend here was endorsed by William Howard Taft, the inventor of the unitary executive and author of Myers that SCOTUS also purports to defend. In Myers, Taft distinguished between "political" duties and "duties so peculiarly and specifically committed to the discretion of a particular officer as to raise a question whether the President may overrule or revise the officer's interpretation of his statutory duty in a particular instance.” Regarding the latter, the President did not have plenary power to revise subordinates' judgments but instead only “general administrative control.” That control included the ability to “supervise and guide [officers’] construction of the statutes under which they act in order to secure [the] unitary and uniform execution of the laws” and to evaluate and remove lower-level officers if they are “negligent and inefficient.” It did not, however, include any power to remove subordinates for any reason whatsoever -- for instance, to conduct a personal vendetta against the prior administration. Such control also did not allow the President to revise the decisions of the civil service in ways that contradicted the statutory criteria that governed such subordinates.

It should hardly be a surprise that Taft endorsed the administrative autonomy of a professional civil service: Like other Republicans in the early twentieth century, Taft was an enemy of patronage-driven political machines associated with New York City's Democratic Tammany organization. Moreover, as Jesse Tarbert has shown in an important recent book, Taft was a leading member of a group of 1920s Republicans who wanted to build state capacity by professionalizing government. Acolytes of this position included not only Frank Goodnow and Woodrow Wilson, well-known architects of the modern administrative state, but also leading Republicans like Henry Stimson, Elihu Root, and Taft.

The Roberts Court's effort to revive Taft's vision in Myers, therefore, ought to be faithful as well to Taft's ideal of an independent civil service administering statutes according to statutory criteria that the President is legally bound to respect. Whether the Court will jettison that ideal to advance its own vision of a plebiscitary presidency remains to be seen. If it does, however, it will have no one to blame but itself: That plebiscitary vision has nothing to do with the tradition of presidential control in Myers that the Roberts Court claims to revive.