The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
If there is one thing that most conservative and libertarian originalists agree on, it is that executive power under the Constitution must be "unitary." That doesn't necessarily mean that the president's power is unlimited, or even particularly broad. It does mean that whatever authority the executive has must be controlled by the president. He is free to hire, fire, and issue orders to all other executive branch officials, as he sees fit—except in a few cases specifically noted in the Constitution, such as the requirement of Senate confirmation for cabinet officers. This issue came up recently when the Senate Judiciary Committee voted to approve a bill that would protect special Counsel Robert Mueller from being fired by President Donald Trump—the man whose activities and assciates Mueller is charged with investigating. The bill passed the Committee by a 14-7 vote. But several GOP senators—including some, like Ben Sasse, who are no fans of Trump—voted "no," citing constitutional considerations based on unitary executive theory.
For a long time, I too accepted the theory of the unitary executive. But I am no longer convinced. In this post, I try to explain where I went wrong, and why I hope others will reconsider the issue as well.
Unitary executive theory is sound up to a point. But it does not hold true in an era when much of the power wielded by the executive branch is authority that the original meaning of the Constitution never gave the federal government in the first place. Allowing such a massive concentration of authority in the hands of one man is dangerous. And it does not nothing to enforce the original meaning of the Constitution.
In some ways, the originalist case for a unitary executive is as compelling as ever. Article II of the Constitution states that "The executive Power shall be vested in a President of the United States of America." It does not say that executive power can be divided between the branches of government or given to bureaucratic agencies independent of presidential control. This strongly implies that he is supposed to have all the power given to the executive branch, except such as is specifically allocated elsewhere in other parts of the Constitution. Much evidence from the Founding era suggests that this is indeed how things were understood at the time, though some scholars argue otherwise.
If the executive branch still wielded only the relatively narrow range of powers it had at the time of the Founding, the case for the unitary executive would be very strong (at least on originalist grounds). Unfortunately, however, the current scope of executive authority goes far beyond that. To take just one particularly noteworthy example, the president now presides over a vast federal law-enforcement apparatus, much of it devoted to waging the War on Drugs (which accounts for the lion's share of federal prosecutions and prisoners). Under the original meaning of the Constitution—and the dominant understanding of the first 150 years of American history—the federal government did not have the power to ban in-state possession and distribution of goods. That's why it took a constitutional amendment to establish federal alcohol Prohibition in 1919. Giving the president control over the waging of the federal War on Drugs is giving him a power the federal government was never supposed to have in the first place.
The same holds true for a great many other powers currently wielded by the executive branch. The original Constitution does not authorize the federal government to regulate nearly every aspect of our lives to the point where we have so many federal laws that a large majority of adult Americans have violated federal criminal law at some point in their lives (to say nothing of civil law).
There is nothing originalist about giving the president such unconstitutional powers. If "executive" power is the power to "execute" federal laws authorized by the original meaning of the Constitution, it does not apply to powers that have no such authorization. The only way to truly enforce the original meaning in such cases is to remove such authority from federal hands altogether. But if we cannot or will not do that, there is no reason to think that giving the power to president is any better—from an originalist point of view—than lodging it somewhere else. Either way, someone in the federal government will be wielding power that they are not supposed to have under the original meaning of the Constitution.
In many cases, it might be more in the spirit of the Founding Fathers to divide this overgrown authority than to give it all to the President. After all, the Founders repeatedly warned against excessive concentration of power in the hands of any one person. They would be especially appalled to see it in the hands of of an office whose occupant is now selected by a far more populist selection process than the Founders intended, and therefore more likely to be a dangerous demagogue.
In sum, at least when it comes to the distribution of power that was never supposed to be in the hands of the federal government in the first place, there is no good originalist reason to give it to the president, as opposed to anyone else (assuming that the power will be in federal hands at all). That does not answer the question of exactly which nonoriginalist powers should be insulated from presidential control and which should not. But it does suggest the question cannot and should not be answered by reference to originalist unitary executive theory. It also suggest that originalists should prioritize reducing the scope of executive power over restoring unitariness. Indeed, the former is the only safe (and originalist) way to enable the former.
Obviously, not everyone is an originalist. Many nonoriginalists comfortable with the current scope of federal power, and oppose efforts to bring it closer to its original scope. But if you are a nonoroginalist about the scope of federal power, you also have good reason to be a nonoriginalist about its distribution. Given the enormous authority of the modern executive branch, it is dangerous to concentrate all of that power in the hands of any one man—especially in an era when James Madison's warning that "Enlightened statesmen will not always be at the helm" seems more prescient than ever.
Conservatives who think that Trump can be safely trusted with such vast unitary authority should ask themselves whether they feel the same way about Barack Obama or Hillary Clinton, both of whom were condemned on the right (sometimes justly) as lawless abusers of power. I suspect most such conservatives do not expect Democratic presidential nominees in future elections to be significantly better.
The above does not, by itself, prove that the dissenting senators were wrong to object to the Mueller bill on constitutional grounds. After all, investigation and prosecution of corruption and lawbreaking within the executive branch (the main object of Mueller's investigation) is not a new federal power, but one that already existed at the time of the Founding.
But the growth of executive power far beyond its original bounds is still relevant to the Mueller question. Letting the president have near-total control over investigation of his own and his close associates' possible wrongdoing may well have been tolerable at a time when the scope of executive power was relatively modest, thereby limiting the potential opportunity for abuse and corruption. The danger posed by a corrupt or lawless president is far greater in an era when the office has vastly more power.
That makes it more defensible to insulate investigation of high-level executive wrongdoing from presidential control, even if such insulation carries risks of its own, as famously elaborated in Justice Antonin Scalia's dissent in Morrison v. Olson (cited by some of the congressional opponents of the Mueller bill). When it comes to investigation of the the executive branch's own wrongdoing, the originalist theory of the distribution of executive power may be heavily dependent on confining that power to its original scope. Enforcing the latter without enforcing the former may be both dangerous and a perversion of the Founders' work.
There might still be valid grounds for opposing the Mueller bill. For example, it could be that political pressure is enough to prevent Trump from firing Mueller or otherwise stifling the investigation (it seems to have worked so far). But the issue cannot turn on a simplistic application of unitary executive theory.
Whatever we ultimately think of the Mueller bill, the broader lesson here is that it is a mistake to enforce the originalist view of the unitary executive in an era when the power wielded by the executive branch has grown far beyond its original bounds.
UPDATE: I perhaps should have mentioned the oft-made argument that maintaining a unitary executive—even when it comes to powers beyond the scope of the original meaning of the Constitution—is desirable because it enhances political accountability. Even if true, this claim is about what is pragmatically desirable, not about the text and original meaning of the Constitution. But the claim is dubious even on its own terms. The greater the scope of executive power, the harder it is for rationally ignorant voters to keep track of more than a small fraction of it. Moreover, it becomes difficult to figure out how to weigh the president's performance in one area against what he does in others (assuming there is variation in quality, as will often be the case). It is therefore unlikely that concentrating a vast range of power in the hands of one person does much to enhance accountability. I discuss the tradeoff between accountability and scope of government power in somewhat more detail in my book Democracy and Political Ignorance: Why Smaller Government is Smarter.
UPDATE #2: While not directly addressing the issue of whether executive power should be "unitary," this recent article by John Dickerson is an good discussion of how the growth of presidential power, combined with rising public expectations, has created a range of pathologies.