Toward a “Unitary Executive” Vision of Article II?

The U.S. Supreme Court made a significant move toward a “unitary executive” vision of Article II in Seila Law LLC v. CFPB. In this 5-4 decision, the Court struck down the statutory provision granting for-cause removal protection to the director of the Consumer Financial Protection Bureau (CFPB). To reach that conclusion, the majority had to address the Court’s unanimous 1935 decision in Humphrey’s Executor v. United States, which upheld an identical removal provision governing the Federal Trade Commission (FTC). Chief Justice John Roberts, writing for the majority, distinguished Seila Law from Humphrey’s Executor because the FTC is headed by five commissioners, and the CFPB is headed by only one director. Although it is clear that five agency heads are different than one, what is not clear is why that distinction matters. One could very well distinguish the two agencies on the ground that the CFPB’s name is composed of four words whereas the FTC’s has only three. The Court, understandably, does not rely on that distinction. One could also distinguish the agencies on the ground that the CFPB was established in September, whereas the FTC was established in July. The Court, again understandably, does not rely on this distinction either. Why are some obvious distinctions irrelevant to the analysis, while another distinction—the number of individuals leading an agency—is legally dispositive? The government, which made the unusual move of siding with the CFPB’s challengers, had embraced this distinction and offered the Court three possible ways in which agencies with a single head differ functionally from agencies with multiple heads. First, the government argued that the President could control multi-member agencies whose commissioners are protected from removal through his supposedly “unfettered ability” both to appoint and to remove the chairs of those multi-member bodies. As I pointed out in a brief filed on behalf of several administrative law scholars supporting the constitutionality of the CFPB’s removal provision, that claim is both not true across the board, because the President has such powers for some multi-member agencies but not for others, and not particularly relevant, because chairs do not have the power necessary to control an agency. Second, the government argued that a President can exert more influence on a multi-member agency because he can appoint “at least some” of its commissioners. But as my brief noted, a President cannot appoint a controlling majority to every agency in one four-year term, particularly given the partisan balance requirements that govern appointments for some agency heads. Finally, the government argued that removal protections for heads of multi-member agencies “facilitate a frank and open exchange of views,” making the agency’s decisions less likely to diverge from the President’s preferences. My brief pointed out, however, that robust social science research suggests that the opposite could well be the case. To its credit, the Court did not embrace the bulk of the government’s functional arguments, perhaps sensing that these arguments were flawed. In particular, the majority opinion said nothing about a President’s ability to appoint agency chairs or about whether multi-member agencies are likely to be more closely aligned with a President’s policy preferences. Yet vestiges of the government’s incoherent arguments can be found in the majority opinion. For example, the Court stated that, “Because the CFPB is headed by a single director with a five-year term, some Presidents may not have any opportunity to shape its leadership and thereby influence its activities.” But the majority never explained why this aspect of the CFPB’s structure is legally relevant. If, for example, a President gets to appoint only two of the seven members of the Federal Reserve Board during a four-year term, as could well be the case, and the other five are hostile to the President’s views, which could also be the case, to what extent has the President meaningfully influenced the agency’s activities? Along these lines, would the President’s control really be different if the CFPB head had only a three-year term, so that every President could appoint a director? Even with a shortened term, it would still be the case that the President would have no more influence on the CFPB’s activities during the tenure of the holdover director from the prior administration. If lack of presidential control is the problem, how would the ability to appoint an agency head later in a presidential term cure a constitutionally insufficient level of control up until that point? Years ago, Kirti Datla and I pointed out that administrative agencies differ along many dimensions in ways that affect the degree of presidential control available over those agencies. One such difference is the number of people in charge of an agency. The argument that this difference is the most relevant, or perhaps the only, distinction cognizable for constitutional purposes just does not hold up. As Justice Elena Kagan indicated powerfully in her dissent, presidential control “can operate through many means—removal to be sure, but also appointments, oversight devices (e.g., centralized review of rulemaking or litigating positions), budgetary processes, personal outreach, and more.” She added that the “effectiveness of each of those control mechanisms, when present, can then depend on a multitude of agency-specific practices, norms, rules, and organizational features. In that complex stew, the difference between a singular and plural agency head will often make not a whit of difference.” In the end, the majority’s revisionist history about three foundational cases of the administrative state—Myers v. United States, Humphrey’s Executor, and Morrison v. Olson—does most of the work in its opinion. The Court adopted a novel reading of Myers, a case in which the Court struck down a significant restriction on the President’s power to remove a Postmaster. In Seila Law, the Court explained that constraints on the President’s removal power were impermissible in Myers but permissible in Humphrey’s Executor because the latter addressed a multi-member agency. But this reading of history contradicts what has been, for decades, the standard view among scholars of the administrative state, a view that the Court itself reaffirmed in 1986 in Bowsher v. Synar and in 1988 in Morrison v. Olson. This standard view was that the Myers removal provision was problematic because it explicitly empowered the U.S. Senate to play a role in the agency head’s removal—not because a single individual headed the agency. The Court’s explanation for distinguishing Morrison v. Olson was similarly implausible. There, the Court upheld a for-cause removal provision for a single independent counsel who could be appointed to investigate and prosecute misconduct by high-level government officials, including the President. The Morrison Court asked whether the removal restrictions were “of such a nature that they impede the President’s ability to perform his constitutional duty.” In analyzing the very narrow duties, jurisdiction, and tenure of the independent counsel, the Court concluded that the “good cause” removal provision did not create such an imposition. But the Court distinguished the independent counsel from the CFPB director on the ground that the latter’s “duties are far from limited.” In striking down the CFPB director’s for-cause removal provision, the Court in Seila Law apparently believed that the CFPB director is more of a threat to a President’s power than an independent counsel with broad powers to investigate and even prosecute the President. Such a conclusion would certainly come as a surprise to most Presidents. Furthermore, the Court’s distinction leads to absurd results. Imagine an independent counsel with just a little more authority—such as prosecuting ordinary citizens for misdemeanor offenses in national parks. Would this extension of authority turn a constitutionally permissible prosecution of the President, by an individual protected by a for-case removal provision, into an impermissible encroachment on the President’s Article II powers? It would be truly odd to think that the constitutional line was crossed by the power to prosecute the misdemeanors but not by the power to prosecute the President. Seila Law embraces a “unitary executive” vision of Article II, but it leaves open how much further the Court might move the law in that direction. Will the Court eventually overrule Humphrey’s Executor, as the government urged if the Court was not persuaded by the one head versus multiple heads distinction? Will it eliminate rules against presidential interference in formal adjudication? Will it strike down partisan balance requirements in multi-member agencies? Will it decide that independent budget authority is independently problematic, as suggested in Seila Law? Because the Supreme Court in Seila Law failed to grapple seriously with the complexities of the administrative state, the Court provided us with little guidance on how to think about these questions