As most separation of powers/administrative law scholars know, the Roberts Court has consistently applied the unitary executive branch theory (UET) across Democratic and Republican administrations.  That theory maintains, at a minimum, that the President must have the power to remove at will the heads of any federal agencies or commissions that wield substantial executive power.   I think the Roberts Court is wrong about the UET. But if it’s wrong, it’s been consistently wrong, regardless of who the President is.

I bring this up only because a recent New York Times essay, to which I contributed, contains a flatly wrong claim about this issue from Stanford political scientist Adam Bonica.  I will first briefly describe the history of this issue, then address Bonica’s disturbingly incorrect claim.

The Supreme Court started on the path toward endorsing the unitary executive branch theory long before the first Trump administration.   President Obama was in office when the Roberts’ Court first endorsed this conception of strong presidential power in Free Enterprise Fund v. Public Company Oversight Board (2010) (full disclosure:  I filed an amicus brief in that case defending the Sarbanes-Oxley Act on behalf of all former Chairmen of the SEC).  That decision held the structure of the Public Company Accounting Oversight Board unconstitutional because Congress had left the President too little direct or indirect control over the Board’s members. 

During the first Trump administration, the Court then invalidated the structure of the Consumer Financial Protection Bureau because Congress had sought to limit the President’s power to remove the head of the CFPB only on certain limited grounds.   The next major decision in this line, United States v. Arthrex, further empowered the President during the Biden administration.  Indeed, that decision expanded the UET into new terrain by holding that the President also had to have effective control over administrative adjudications.  

President Biden then took a number of actions based on the UET.  Shortly after taking office Biden fired the head of the Social Security Administration, despite statutory protections that insulated his removal from direct presidential control.  He also fired a number of other officials who had long been thought to be independent because they served fixed terms of office.  Indeed, a legal commentator in Slate characterized Biden as “the first unitary executive.”  None of these firings reached the Supreme Court. 

Entering the second Trump administration, what remained against the UET was the 1935 case, Humphrey’s Executor v. United States, which held that Congress could create independent, multi-member agencies, such as the Federal Trade Commission, whose heads the President could not fire at will, but only for “good cause.” But it was widely expected that the Trump administration would challenge that decision and that the Roberts Court would take the next step in its commitment to the UTE by agreeing with the President and overturning Humphrey’s Executor.   

When President Trump put the issue squarely before the Court by firing the heads of various independent agencies, the lower courts held these firings illegal. Those courts are obligated to follow Humphrey’s Executor. In two cases on the emergency docket, the Court has stayed these lower court rulings on procedural grounds.  Technically, the Court has not yet overruled Humphrey’s Executor, but it’s widely recognized that these actions make it virtually inevitable that the Court, unsurprisingly, is going to overrule Humphrey’s Executor.  Indeed, Justice Kagan’s dissent for three Justices asserts that the Court has in effect already overruled Humphrey’s Executor, even before formally interring it.  One can argue about procedural aspects of these stays, but on the substance of the UTE, these are probably the least unexpected decisions on executive power of the current Court. 

Indeed, the most important brake the Court has suggested on the UET has come during Trump’s second term, in statements from individual Justices that the President does not have the power to fire at will the Chairman of the Federal Reserve.  Critics have questioned whether there’s a principled distinction between the Federal Reserve and other agencies, but the conservative Justices have made these statements in direct response to President Trump’s threats to remove Chairman Powell.

Bonica, however, asserts that the Court has been partisanly manipulative and opportunistic in how it applies the UTE.  He levels this charge in extremely strong terms: “If the unitary executive theory were a genuine constitutional principle, its adherents would apply it equally to all presidents, regardless of party. The evidence shows it is a principle of convenience, invoked to amass power and discarded when it might constrain allies.”

To back this up, he offers two pieces of evidence.  First: “This situational approach is most visible in the actions of the theory’s own advocates. When President Biden took office and fired a handful of Trump-appointed holdovers, he was met with a flurry of lawsuits from the very conservatives who champion UET. Suddenly, these activists argued for the sanctity of the statutory tenure protections they otherwise seek to destroy.”

But the positions lawyers take in litigation obviously tells us nothing about whether the Court is applying the UET consistently.  On top of that, Bonica doesn’t tell us (or link to) who the lawyers are he has in mind behind this alleged “flurry of lawsuits” from conservatives.  In poking around a bit, the only suit I’ve come across that fits Bonica’s description is one that Russell Vought brought, in which Stephen Miller’s law firm represented him, when Biden fired Vought from the Naval Academy Board of Visitors.  But it doesn’t take a lot of strategic sense to recognize that Vought brought that suit hoping to lose it as a way of more deeply entrenching the UET.  Vought surely cares more about solidifying that doctrine than he cares about being on the Naval Board of Visitors.  In any event, whatever the motivation for this one suit, it has nothing to do with whether the Roberts Court has applied UET consistently.

Bonica’s second and only other piece of evidence, which he calls “the most telling evidence,” is that Justice Thomas voted against the UET in the Arthrex case, during Biden’s presidency.  But Bonica doesn’t tell readers that Thomas was in dissent, along with the three more liberal Justices.  Nor does Bonica tell readers that the Court upheld the UET claim in that case, during Biden’s presidency, in a particularly expansive opinion endorsing the UET. 

After that, Bonica leaves the UTE behind and shifts to brief discussion of other issues concerning executive power.  He criticizes the Court for not overturning universal injunctions during the Biden administration and presents data on how often the Court has used the emergency docket to uphold or reverse lower courts during the Biden and Trump years.  Jack Goldsmith recently took issue with Bonica’s use of data.  I don’t want to prolong this piece with my own views about the extent to which the Court has or has not been consistent on other executive power issues.  One big test will come with how the Court applies “the major questions doctrine” during the Trump administration.  But the Roberts Court has had a clear commitment to the unitary executive branch theory for many years and it’s been consistent about that – for better or worse.

St. Louis Public Radio reports.

Bradley Smith and I talked about Callais and the future of Section 2 in the National Constitution Center’s “We the People” podcast with Jeffrey Rosen.

In this episode, Bradley Smith of Capital University Law School and Nicholas Stephanopoulos of Harvard Law School join to discuss the history and future of racial gerrymandering, including how the Court’s upcoming decision in Louisiana v. Callais could affect the Voting Rights Act. 

Sam Levine for the Guardian:

“Gerrymandering is bad enough once a decade,” said Richard Pildes, a law professor at New York University. “But if we open the door to continual efforts throughout the decade to squeeze out every additional seat based on changing calculations to the parties, it’s very bad for voters who have enough trouble developing connections with their representatives and it’s very bad for democracy more generally because it promotes cynicism about the process.” . . .

“It presumably increases the potency for gerrymandering because you can do the gerrymandering based on very recent data and the map doesn’t have to endure for as long,” said Nicholas Stephanopoulos, an election law professor at Harvard. “The worst case isn’t just a one-off mid-decade re-redistricting it’s a continuous re-redistricting. Before every election, you check out which of your side’s incumbents had a closer call than you wanted last time and you make their district three or five or seven points more Democratic or Republican.” . . .

Stephanopoulos said the current congressional map was essentially balanced between Democrats and Republicans in the aggregate. “I’d rather have fair maps in every state aggregate into a fair US House. If we can’t get that because the Congress won’t require fair maps and neither will the supreme court, then the worst case, I think, is one side gerrymanders and the other side doesn’t, and we get a highly distorted US House.

“That then means that the majority of Americans aren’t represented by a majority of legislators, and Congress passes laws that don’t reflect what the majority of Americans want and so offsetting gerrymanders at least prevents that worst-case outcome.”

Votebeat.

Ron Brownstein in Bloomberg (paywalled).

“The current House is unusual in the modern era in being very close to perfectly neutral according to various measures of partisan bias,” Nicholas Stephanopoulos, a professor at Harvard Law School and expert in redistricting, told me. Under the current lines, Stephanopoulos and his colleagues have argued, the party that wins the most votes in House elections nationwide is also very likely to win the majority — a big improvement from 2012 when severe gerrymandering allowed Republicans to amass a 33-seat House majority even as Democrats won the national popular vote. . . .

If Texas succeeds in launching a new redistricting war, it could disrupt the best (if inadvertent) feature of the current maps: their rough partisan balance. In an all-out mobilization, Republicans probably can squeeze out about half a dozen more House seats than Democrats before 2026. That would not shift the maps back all the way to their lopsided pro-GOP tilt after 2010, says Sam Wang, president of the Electoral Innovation Lab at Princeton University. But, he says, it would erase the roughly neutral playing field, and likely force Democrats to win the total House popular vote by 2 to 3 percentage points to claim the majority. . . .

The best solution would be legislation establishing comprehensive, national rules for redistricting. In 2021, the Democratic-controlled House passed the sweeping election reform bill known as HR 1 that would have required states to use independent commissions to draw Congressional district lines and applied national standards to that process. Those included avoiding partisan bias, ensuring geographic continuity in the districts, and barring mid-decade redistricting. . . .

There might be other ways to address the gerrymandering problem. Though the US Supreme Court ruled in the 2019 Rucho case that federal courts can’t overturn partisan gerrymanders, state legal action may still offer some opportunities to curb excesses: Stephanopoulos and his colleagues at the Harvard Election Law Clinic recently helped craft a novel lawsuit in Wisconsin that argues the extreme lack of competition in the state’s Congressional map “makes a mockery of” the state constitution’s promise of equal protection and the right to vote. In the long run, some election reformers believe that returning to the early 19th-century model of electing multiple members from a single district offers the best chance to ensure both greater fairness and competition.

From the WSJ, which is paywalled:

Five years ago, Georgia’s Republican Secretary of State Brad Raffensperger was banished to the political wilderness. 

Donald Trump blamed the mild-mannered election chief for his narrow 2020 Georgia defeat, branding him a RINO (Republican in Name Only), “incompetent and strange.” Death threats poured in. GOP senators demanded he resign for reaffirming, after recounts and audits, that Trump lost the battleground state.

The attacks stunned Raffensperger, a businessman and devout Christian who came late to public life. One consultant dubbed him “Dead Brad Walking.” 

Yet something unexpected happened on the way to his demise: He not only survived, winning re-election in 2022, but has become a serious contender in Georgia politics.

Now Raffensperger, 70 years old, a multimillionaire construction magnate, is considering a run for higher office next year, likely for governor to replace term-limited GOP Gov. Brian Kemp, or U.S. Senate to challenge Democrat John Ossoff. A May Cygnal poll found Raffensperger slightly ahead of Ossoff in a theoretical matchup, with other declared or potential GOP candidates trailing Ossoff.

That’s the title of a talk I’ve delivered while at the University of Melbourne on a Miegunyah Distinguished Visiting Fellowship. It’s my effort to explain succinctly how the United States got to the current dire situation and to point the way out of it.

LA Times:

With Democrats lining up for a bare-knuckle match on redistricting, Gov. Gavin Newsom has offered President Trump a cease-fire proposal: No redrawing congressional maps in red states and California will stand down, he wrote in a letter sent to Trump on Monday morning.

“If you will not stand down, I will be forced to lead an effort to redraw the maps in California to offset the rigging of maps in red states,” he said. “But if the other states call off their redistricting efforts, we will happily do the same. And American democracy will be better for it.”

Newsom’s latest play comes as the drama around redistricting heightened over the weekend. Democratic leaders in other blue states argued on Sunday morning political shows that they were ready to battle head-to-head over the congressional district maps, which are normally tied to the census taken once a decade….

“You are playing with fire, risking the destabilization of our democracy, while knowing that California can neutralize any gains you hope to make,” he told Trump in the letter. “This attempt to rig congressional maps to hold onto power before a single vote is cast in the 2026 election is an affront to American democracy.”

In light of Rick P’s post about David Shor’s current lament about the failure of Congress to redress gerrymandering back when Democrats were consumed by their efforts to enact the massive S1/HR1 reforms, I recall this column I wrote for The Washington Post on February 4, 2021, less than one month after the insurrection on January 6: Congress should make a deal to end partisan gerrymandering.

The main point of the column was that Democrats in Congress should trade away issues like voter ID, vote-by-mail, and other wish-list reforms relating to election administration in order to secure the much more important structural change of eliminating partisan gerrymandering.

The column noted, as Shor did the following year, that altering election administration rules wouldn’t make a significant difference in the capacity of the electoral system to translate voter preferences accurately into electoral results, whereas ending partisan gerrymandering would. (“Easy vote-by-mail isn’t a must for Democrats; they just need sufficient opportunities to cast a ballot — and reasonably drawn districts — for good candidates to have a chance.,” the column said.) To make this point, the column cited the then-recent Georgia Senate runoffs: “The lesson of the Georgia Senate runoffs is that Democrats don’t need their preferred set of voting rules in order to win. No voting rights advocate thinks Georgia’s electoral system, run by Republican Secretary of State Brad Raffensperger, is ideal. But it was good enough. It didn’t cause disenfranchisement that prevented voters from getting what they wanted.”

On the other hand, looking specifically at Texas, the column observed: “For reasons of self-interest alone, Democrats should see ending gerrymanders as Job One.”

At the time, I thought Democrats should strive to find 10 Republican Senators–just 3 more than voted to convict Trump in his January 6 impeachment trial–to agree on the necessity to end partisan gerrymandering, and since Senators don’t benefit from gerrymandering themselves, I thought the strategy should be to negotiate with them solely to achieve gerrymandering reform, without weighing it down with all the other provisions in S1/HR1 that were inevitably objectionable to all Republicans, including those willing to risk their political careers to cross Trump. The column ended this way: “Democrats should stay focused on what’s most important in electoral reform. Right now, that’s restoring sanity to redistricting.”

This was a theme I continued to stress in additional Washington Post columns in 2021. For example, on March 29, 2021, I repeated that Democrats should focus on the “anti-gerrymandering” provisions in S1/HR1 and jettison all provisions that were dealbreakers from a GOP perspective. The last sentence of that column: “Democrats are in danger of missing the moment, by going too big and too far.”

On May 27, 2021, I again stressed the need to focus on gerrymandering and wrote: “Senate Democrats are at risk of blowing their chance at meaningful electoral reform. Rather than ending up with nothing, because they spent too long trying to shoot the moon with S. 1, they should compromise with 10 reasonable Republicans on a set of simple measures to ensure that congressional elections genuinely implement voter preferences.”

I won’t belabor the point by quoting the additional columns along the same lines as the Democrats continued their efforts to enact their omnibus bill over unified Republican opposition.

As the effort to protect American democracy from the forces of authoritarianism continues, and indeed grows more urgent, I continue to believe that the potential window of opportunity of 2021-2022 was wasted by a misguided approach on the part of congressional Democrats.

“The majority of today should always fear that it may find itself in the minority tomorrow and that its rules can be used against it. . . . What happens when this breaks down? What happens if the majority of today sees this as the last chance to take it all?”

–Sam Issacharoff, quoted in Adam Liptak’s must-read piece, “In Election Cases, Supreme Court Keeps Removing Guardrails.”

I address this Supreme Court history, and why I told Adam I think we may be heading back to the early 1960s in terms of judicial protection of voting and elections, in Richard L. Hasen, The Stagnation, Retrogression, and Potential Pro-Voter Transformation of U.S. Election Law, 134 Yale Law Journal 1673 (2025).

for blogging this last week, which should have been quiet at the beginning of August and was anything but.

As we all know, the Callais reargument potentially poses the momentous question of Section 2’s constitutionality. But it’s worth noting how many steps the Court would have to take to reach that question. Each of these steps represents an offramp for the Court that would allow it to resolve this case without having to address Section 2’s validity.

1. Race or Politics: First off, if the Court concludes that politics better explains District 6’s formation than race, then strict scrutiny wouldn’t apply and the district would almost certainly be constitutional. There’s a good argument that politics — in particular, protecting Republican incumbents Mike Johnson, Steve Scalise, and Julia Letlow, and sacrificing the seat of Garret Graves — is why District 6’s shape is relatively unusual. If one of the other Republican incumbents’ seats had been sacrificed, District 6 could have been substantially more “reasonably configured.”
2. Narrow Tailoring: If the Court determines that District 6 was designed for a racially predominant reason, the Court could still avoid grappling with Section 2’s constitutionality by holding that the district is an insufficiently tailored remedy for any Section 2 violation. Again, District 6 arguably isn’t “reasonably configured,” and the Court has previously said that only a reasonably-compact minority-opportunity district can remedy a Section 2 violation. Per Bush v. Vera, “if a reasonably compact district can be created, nothing in § 2 requires the race-based creation of a district that is far from compact.”
3. Compelling Interest: Lastly, even if the Court rules that race did predominate in the creation of District 6, and that District 6 is narrowly tailored to remedy a Section 2 violation, the Court could still refrain from reaching Section 2’s constitutionality by holding that compliance with Section 2 isn’t a compelling state interest. The Court has long assumed that Section 2 compliance is a compelling interest. But the Court has never actually held as much. And if Section 2 compliance isn’t a compelling interest, then it can’t save a district drawn for a racially predominant reason — and the Court would have no occasion to comment on Section 2’s validity. Note that this resolution of Callais wouldn’t bar future Section 2 vote dilution claims since it wouldn’t change any aspect of the Gingles framework. Nor would it mean that any vote dilution remedy is necessarily an unlawful racial gerrymander. As Chief Justice Roberts made clear in Milligan, the critical line as far as racial gerrymandering is concerned is “between [race] consciousness and [racial] predominance.” Both plaintiffs’ demonstrative maps at Gingles‘s first step and jurisdictions’ ultimate remedies can — and should — exhibit race-consciousness but not racial predominance, in which they don’t trigger strict scrutiny.

Of course, if the Court is intent on addressing Section 2’s constitutionality, it can find a way to do so. The point here is just that the Court has several options at its disposal that wouldn’t raise the stakes in Callais to the maximum possible size. (Disclosure: Other attorneys in HLS’s Election Law Clinic represent the Robinson intervenors in Callais, but I haven’t been involved in that representation.)