Ninety-one years ago, when the Supreme Court held in Humphrey’s Executor v. United States that Congress could make agencies like the Federal Trade Commission “independent” from total presidential control, many were shocked – including President Franklin Roosevelt himself.
FDR was “outraged by the Humphrey ruling,” one historian recounted, because not a decade earlier, in 1926’s Myers v. United States, the court had broadly endorsed presidential power to fire agency heads. FDR “had strong ground for believing he had acted properly” in firing FTC Commission William Humphrey. Worse still, even Justices Louis Brandeis and Benjamin Cardozo had joined the court’s unanimous ruling, which FDR found “bewildering.”
By contrast, no one should have been bewildered by the Roberts court’s approach to agency independence last week. The outcomes in Trump v. Slaughter, on the Federal Trade Commission, and Trump v. Cook, on the Federal Reserve, had been telegraphed for months, even years.
Nearly all of Roberts’ tenure on the court has been marked by clear skepticism of Humphrey’s Executor. The court has been skeptical of its bottom-line ruling that Congress can give “quasi-judicial and quasi-legislative” regulatory commissions some independence from the president by providing the commissioners a fixed term of years and preventing presidents from firing them early, except “for cause,” sometimes only for “inefficiency, neglect of duty, or malfeasance in office.” And, even setting aside Humphrey’s description of those agencies in 1935, the court has also been skeptical of arguments that modern regulatory agencies’ work, regardless of nominal labels, is in substance anything other than executive action.
This has played out in a series of prominent decisions, written mostly by the chief justice himself. In 2009, Roberts wrote the court’s majority opinion in Free Enterprise Fund v. PCAOB, declaring that Humphrey’s Executor’s precedent would not be extended to new, novel forms of agency independence. Then, in 2020, he wrote the court’s opinion in Seila Law v. CFPB, voiding the Consumer Financial Protection Bureau’s for-cause removal protection because the CFPB, unlike the FTC in Humphrey’s, is not a multi-member commission – and, more fundamentally, because the CFPB wields “significant executive power.” The court reiterated this principle a year later as it pertains to the Federal Housing Finance Agency, in Collins v. Yellen.
When the end finally came for Humphrey’s Executor on June 29, and the court held that the modern FTC’s powers are overwhelmingly executive, Roberts’ majority opinion reiterated what it had been saying for years – that, whatever nominal labels Congress might attach to the FTC as a whole, or to its specific powers (such as “rulemaking” or agency “adjudication”), in substance the FTC’s powers are quintessentially executive:
First, the FTC has the power to promulgate substantive rules that carry the force of law. … The power to flesh out such statutory regimes—and to do so through discretionary actions, largely outside the remit of courts—is executive through and through.
Second, the FTC not only investigates businesses to ensure they comply with its statutes and rules … but enforces those statutes and rules through in-house adjudications. … This power, too, is executive.
And third, the FTC files civil suits on behalf of the United States in federal court. … As we have said many times, the “discretionary power to seek judicial relief” lies at the very core of executive authority.
That analysis, filled with cites to Seila Law, should surprise no one. Nor did the court’s holding in Trump v. Cook, that the Constitution allows Congress to protect the Federal Reserve System’s independence by prohibiting presidents from firing Federal Reserve governors without “cause,” come out of the blue. The Roberts court has been signaling this, too, for years – in Seila Law’s observation that the Federal Reserve System is not simply analogous to regulatory agencies like the CFPB; and last year, in an interim docket order involving the National Labor Relations Board, where the court sketched out its sense that the Federal Reserve System rests on much different constitutional footing than other administrative agencies, tracing back to the founding-era Bank of the United States.
To be sure, not all the justices see things this way. In last week’s Cook decision, Justices Samuel Alito (joined by Neil Gorsuch) and Amy Coney Barrett each wrote dissents, explicitly declining to reach this issue of Fed constitutionality, even though they had not dissented from the court’s discussion of the Fed in last summer’s per curiam order on the NLRB. More significantly, Justice Clarence Thomas, who also did not dissent last summer, dissented loudly this time, laying out a strong argument that the Federal Reserve Board does, in fact, wield significant executive power, because Congress has empowered it to regulate the private sector.
The chief justice’s argument to the contrary – with which, as I’ve indicated in previous columns, I’m strongly inclined to agree – begins from fundamentally different premises. Where he sees the FTC (and CFPB, and surely most other agencies) as doing virtually nothing other than executing laws through regulatory actions against the private sector, he sees the Federal Reserve System – its Federal Reserve Board of Governors, its Federal Open Market Committee, and its Federal Reserve Banks – as being predominantly engaged in monetary activities. Or, as President George Washington and Alexander Hamilton would have said, banking activities – which, as Roberts’ opinion for the court emphasizes, has never been constitutionally committed to sheer executive power.
But, as Thomas emphasizes, and Roberts recognizes, the Federal Reserve System does have regulatory powers. And that points toward one of the next developments that we should await: a constitutional challenge to the Federal Reserve System’s power to regulate banks and other private actors.
Roberts signals this freely, in both Trump v. Cook itself and, it seems to me, in his discussion of the FTC in Trump v. Slaughter. In footnote 6 of the Cook decision, when Roberts responds to Thomas’ account of the Fed’s regulatory powers, he writes: “In upholding the constitutionality of the Federal Reserve as currently structured and with its existing enforcement authorities, we do not suggest that Congress could assign the Federal Reserve additional regulatory powers that are attenuated from monetary policy.” This footnote invites the question – perhaps in future litigation brought by Fed-regulated plaintiffs – of whether specific aspects of the Fed’s regulatory powers are in fact unconstitutional, because they are being wielded by a Federal Reserve that is independent from total presidential control.
For Roberts and the court, the clear solution for such a problem would not be to end the Fed’s independence, but rather to sever those discrete regulatory powers from the Fed.
As it happens, Roberts suggests such an approach (albeit not in direct reference to the Fed) in another footnote – this time in his opinion for the court in Trump v. Slaughter. In footnote 3’s description of how to remedy a constitutional flaw in one small part of a much larger statutory framework, Roberts writes: “In our prior cases, ‘when confronting a constitutional flaw in a statute’ like the one before us, we have sought to limit ‘the solution to the problem,’ severing the invalid removal provision ‘while leaving the remainder intact.’”
I think this is precisely how Roberts – along with Justice Brett Kavanaugh (who joined both Cook and Slaughter in full) and perhaps others – would ultimately approach a challenge to the Fed’s regulatory powers: not by questioning the Fed’s independence, but by severing that particular aspect of the Fed’s powers.
In that respect, when Roberts and Kavanaugh see Fed independence differently from Thomas, it is because they are looking through opposite ends of the institutional telescope: Roberts and Kavanaugh see rightly independent central banking system with perhaps unconstitutional regulatory powers; Thomas sees a regulatory agency with unconstitutional independence. (Looking ahead to such arguments, we would all be well-served by reading Professors Aaron Nielson’s and Aditya Bamzai’s excellent Cornell Law Review article, “Article II and the Federal Reserve.”)
In the meantime, here are a few more issues that could arise in these cases’ wake, among the FTC and other previously independent regulatory agencies:
Appointments: The Slaughter case was about removing regulatory commissioners, but its effects could also be felt on appointments.
Multi-member agencies like the FTC operate under a statutory requirement for bipartisanship. At the five-member FTC, for example, Congress requires that “Not more than three of the Commissioners shall be members of the same political party.” This leaves presidents responsible for appointing members representing the opposing political party – a process that has worked, even in our bitterly partisan era, under an informal norm that the president will cooperate with the Senate’s opposition leader to pick minority-party nominees.
That approach makes sense when the president cannot unilaterally fire the other party’s members for any reason or no reason at all. But will it work in the post-Slaughter era? We should not assume that the opposing party’s Senate leader will invest much time or effort in recruiting or vetting new appointees.
And it is not hard to imagine President Donald Trump or his successors reading the bipartisanship requirements narrowly, even cynically – say, by appointing three Republicans and two others who are not Democrats but not registered Republicans. Or perhaps presidents will simply appoint a bare majority of three members to each commission and leave two seats perpetually vacant.
Which raises the next question …
Quorums: Given the large number of vacancies on multimember commissions recently, many of us have spent time looking up a lot of agency-specific quorum requirements: say, three members for the Federal Communications Commission, maybe three or two for the FTC, and perhaps just one member for the Commodity Futures Trading Commission?
If we are entering an era of bare-majority quorums, then we are also entering an era when a single commissioner’s death, debilitation, or lucrative private-sector job opportunity would render his agency inert, unable to promulgate regulations or take other substantive regulatory actions requiring a quorum of the agency’s leaders, until the president can nominate a new commissioner and get the Senate’s confirmation.
And speaking of vacancies …
Waking up to “midnight firings”: As a smart friend pointed out to me, the Slaughter decision radically changes the incentives for presidents in the last days of their administration, at least when the next president comes from the other party.
Simply put: A president who see his successor with something less than equanimity will have a great opportunity to kneecap the new administration by firing all the remaining commissioners from the other party, so that the incoming administration will not have readily available commissioners to serve as acting chairmen.
That is, when a new president is inaugurated, he appoints one of the carry-over members of each regulatory commission – that is, one of the last president’s opposite-party appointees – to serve as acting chairman, to keep the agency on track until other commissioners can be appointed and Senate-confirmed.
It is not hard to see this play out. Would Trump, on his last day in office, fire all remaining Democratic members of the FCC and other agencies – assuming any Democratic members would still be in place – before a new Democratic president takes the inaugural oath? (Would President Joe Biden have done so in January 2025, had the Supreme Court ended Humphrey’s Executor years earlier?) If so, this new kind of “midnight firings” could render many agencies effectively moot for the new administration’s first months or even longer.
These are just a few things I’ll be curiously watching, in the months and years ahead. Still, I hesitate to make any great predictions of what Slaughter and Cook might mean for the future of regulation and administration. I’m old enough to remember just a couple years ago, when very smart people were confidently declaring that the end of Chevron deference would incapacitate government agencies. Surely, two years from now, Slaughter and Cook will seem as distant a memory as Loper Bright v. Raimondo seems today.
