Separation of Powers Does Not (Necessarily) Immunize the Veto Power -- Wherein I Respond to Eugene Volokh's Reply to My Analysis of the Rick Perry Indictment

by Michael Dorf

In my post last week on the indictment of Rick Perry, I criticized the argument by the Perry camp that the use or threatened use of the governor's veto power for nefarious purposes cannot be the basis for a criminal charge because the state constitution assigns to the governor the power of vetoing legislation. I said that this argument was weak, "at least if not further qualified." I then explained both why the maximalist argument made by Perry is weak and how a more nuanced version of the claim could be stronger. Later in the week, Professor Volokh defended the maximalist view. Some of what he wrote addresses other issues raised by the Perry indictment, but to the extent that Volokh has offered a response to my prior post (which he quotes at length), I am not persuaded.

Volokh argues that the legislature cannot, through ordinary legislation, impose limits on the executive's constitutionally conferred veto power. Before explaining where I think Volokh goes wrong, I should note that although each of us has referred to Texas law, the argument is mostly interesting only insofar as it raises general questions of how executives interact with legislatures. Neither of us is an expert in Texas law, and so it's possible that there are peculiarities of Texas law that support or undermine general arguments that he or I have made. And, as I said in my original post, I don't have a view about whether the case against Perry is weak or strong, all things considered. I understood Professor Volokh to be answering my post in the same spirit -- i.e., discussing how these issues work in general in constitutional systems with separation of powers and in which the head of the executive has veto authority. I shall continue the discussion in the same spirit, making general rather than Texas-specific points, except where otherwise noted.

Volokh offers two related grounds for his conclusion that a governor's exercise of the veto power may not be the basis for a prosecution under the particular statutes he is charged with violating: 1) Ordinary legislation cannot override a constitutional provision, so insofar as the state constitution confers on the governor unfettered discretion to exercise the veto power, that constitutional grant of power takes precedence over any legislation; and 2) A governor's veto acts as a check on the legislature, so it would also violate separation of powers for the legislature to place limits on the governor's exercise of the veto power.

These propositions strike me as sound, so far as they go, but I don't think they go as far as Volokh assumes. So, where do we disagree? The core of our disagreement is over proposition 1, which strikes me as an overstatement. A governor or president has complete discretion to exercise the veto power for good or bad policy reasons, or as part of a log-rolling or horse-trading deal, but that does not mean that he or she can never be held criminally liable--at least absent something more than the constitutional assignment of a veto power--for using that power in other, corrupt ways.

Some constitutions have particular provisions providing partial or even total immunity for acts taken in various official capacities. For example, the Speech or Debate Clause of Article I, Sec. 6 of the U.S. Constitution confers such an immunity on members of Congress. The U.S. Constitution contains no such express immunity for executive actions of the President, although in Nixon v. Fitzgerald the SCOTUS relied (partly) on principles of separation of powers to fashion a judge-made doctrine giving the President absolute immunity against civil suits for acts taken in his official capacity; there may be some similar immunity against criminal charges under the federal Constitution, while the person remains in office; however, the impeachment clause makes clear that after a president is out of office he may be subject to criminal sanctions ("but the Party convicted shall nevertheless be liable and subject to Indictment, Trial, Judgment and Punishment, according to Law").

Does the Texas Constitution provide the Governor with immunity against criminal prosecution either during his tenure in office or permanently? As I've said, I'm not an expert in Texas law, so it's possible, but I doubt it.  If there were some such express or judicially implied immunity, wouldn't Perry or his defenders have invoked it? In any event, the key point in my last post and this one is that the argument made thus far-- the one I'm critiquing--doesn't rest on any special immunity from prosecution. The argument asserts that the power to veto--in itself--necessarily entails immunity from criminal prosecution under an ordinary law because such a prosecution would be inconsistent with the constitutional assignment of the veto power to the governor. That's the claim that strikes me as overstated.

To see why, consider an analogous question involving rights. It is conventional to say that there is a First Amendment right to burn a flag, per Texas v. Johnson. But it does not follow that someone exercising that right is immune from prosecution for other crimes he commits while happening to burn a flag. If Shmonson used a burning American flag to deliberately set fire to his neighbor's house, he could be prosecuted for arson. If Gronson used a burning American flag to torch his own failing candy store to collect the insurance money, he could be prosecuted for fraud.

Professor Volokh attempts to show that a governor is immune from prosecution for exercising the veto power by giving a number of examples in which the legislature specifically outlaws particular uses of the veto power. Yet his examples are plainly irrelevant to the question of whether the governor can be prosecuted under a general statute when the governor happens to violate that general statute by exercising the veto power. They are like the Texas law that specifically targeted flag desecreation in Texas v. Johnson, not like the arson law or the fraud law.

Volokh does give one hypothetical example in which a governor is prosecuted under a general statute, but it's a peculiar one.  Here's his best example:
Say that a governor vetoes a bill appropriating money for some public health measure. Someone dies, allegedly because of the veto (i.e., the person wouldn’t have died had the measure been funded). A prosecutor then charges the governor with “depraved heart” murder, meaning (roughly) that the governor acted despite his knowledge that there was a very grave and unjustified risk that his veto would cause death. Or perhaps the prosecutor charges the governor with negligent manslaughter, meaning (roughly) that the governor was grossly negligent in vetoing the law.
I agree with Volokh's intuition that charging the governor with murder or manslaughter in this example would violate the best conception of separation of powers, but not with his conclusion. For one thing, given the libertarian cast of American criminal law, in this context the veto looks like a non-culpable failure to rescue rather than an affirmative act. Moreover, the particular consequences here are inextricably tied up in the governor's policy judgment about whether the public health measure is wise. Presumably the proponents of the health measure argue that absent the measure, people would die, whereas opponents argue either that they wouldn't or that if they do, that's unfortunate but that funding the measure would have other harmful consequences, such as raising taxes, increasing the deficit, etc. Volokh's example really only shows that under the best conception of separation of powers, a governor cannot be prosecuted for the foreseeable consequences of a questionable or even bad policy decision.

But that is not enough. My claim, and presumably the claim of the prosecutor in the Perry case, is not that every time a governor vetoes a law he can be prosecuted for doing so. I am simply denying the Perry camp's equally sweeping contrary claim, namely that a governor can never be prosecuted for exercising the veto power.  Even one example of a circumstance in which separation of powers is not offended by prosecuting a governor for his exercise of the veto power would suffice to rebut the maximalist view of Perry's defenders. And we already have one: bribery. As I noted in the earlier post, a governor who vetoes a bill as part of a quid pro quo for a a bribe can be prosecuted for bribery, notwithstanding the fact that the governor has unfettered discretion to make a good or bad policy choice in deciding whether to sign or veto a law.

Volokh does not deny that a governor could be prosecuted for bribery in connection with the veto power, but instead argues that bribery is somehow unique--an exception that proves the supposed rule that a governor cannot be prosecuted for the exercise of the veto power. He attempts to distinguish bribery from all other crimes in three ways. Let's consider them in turn.

1) Volokh cites the 1972 SCOTUS decision in United States v. Brewster, which said that the federal statutory crime of bribery consists in taking the bribe, regardless of "performance of the illegal promise." That's true. It's even possible to read Brewster as relying on a principle of constitutional avoidance for that reading of the statute--although it would be very odd for the Court to apply constitutional avoidance to make it easier to convict a criminal defendant. But let's suppose that we read Brewster for all it is worth. It would mean that the federal Constitution requires that, as applied to members of Congress, the crime of bribery must refer only to the acceptance of a bribe, not the performance of the act for which the bribe is paid. But only for members of Congress, because the constitutional question in Brewster was whether the crime of bribery violated the Speech & Debate Clause, which, as I noted above, is a special immunity for legislators. It is not a general entailment of separation of powers, so it does not apply to the federal executive, much less to executives in all governments with separation of powers.

2) Volokh next says that "there is a special provision of the Texas Constitution that expressly forbids accepting a bribe by an official (including the governor), and authorizes criminal punishment for such behavior." I think Volokh means by this to suggest that, in the absence of this special provision, separation of powers would forbid bribery prosecutions of the Texas governor for the exercise of the veto power. Yet (other than the addition of a line-item veto), the veto provision of the Texas Constitution is very similar to its federal counterpart, which does not need a "special provision" to authorize bribery prosecutions. How do I know the federal veto clause does not create the need for such a special provision? Because the impeachment clause recognizes that the President and other officials may be impeached (and later prosecuted) for "Treason, Bribery, or other High Crimes or Misdemeanors" without specially enumerating what those sundry other crimes are.

As for the Texas Constitution, I'm not sure what "special provision" Volokh is referring to, but the portion of the Texas Constitution setting out the executive powers contains no reference to bribery. There is a general provision of the Texas Constitution that forbids someone who obtained his office via bribery from continuing in office, but that's not at issue with respect to Governor Perry. His case would appear to be governed by another provision in the same Article of the state Constitution obligating the legislature to enact laws that "exclude from office persons who have been convicted of bribery, perjury, forgery, or other high crimes." The reference to unspecified other crimes strongly implies that the background assumption is that gubernatorial misconduct--whether in the use of the veto or some other way--is subject to prosecution, regardless of whether it takes the form of taking bribes or something else. Far from showing the uniqueness of bribery, this "special provision" shows that bribery is merely one of a large class of crimes for which governors can be prosecuted as a consequence of their abuse of power.

3) Volokh then adds that "the risk of prosecution for bribery is far less of a deterrent to a veto (assuming the governor hasn’t been bribed) than the risk of prosecution for violating [other laws], precisely because a prosecution for bribery has to show an agreement to take a bribe — not just a supposedly improper intent behind the veto." This seems to me both wrong and irrelevant.

It's wrong because sophisticated corrupt government officials have ways of taking bribes that are proved in much the same way as other kinds of corruption charges. Unless they are idiots, corrupt government officials do not receive checks with "bribe" written in the memo section; they do not even usually receive bags of small unmarked bills; they typically take otherwise legal campaign contributions in exchange for otherwise legal conduct; or they funnel money through intermediaries. Thus, prosecuting such an official for accepting bribery involves the same sort of evidence as is needed when prosecuting an official for other forms of corruption in which intent is critical. The current corruption prosecution of former Virginia Governor Bob McDonnell makes this point pretty clearly: it is not disputed that McDonnell and his wife took gifts from Johnnie R. Williams; the sticking point is why, just as it would be if the question were improper use of the veto power.

Moreover, even if Volokh were right that proof that a governor illegally took a bribe in exchange for a veto would be more concrete than proof that a governor committed some other crime by exercising his veto power, so what? That would simply mean that, other things being equal, it would be easier for governors to get away with committing one kind of crime than another. It would not mean that the exercise of the veto power is immune to prosecution for all crimes other than bribery.

Accordingly, I think Volokh's efforts to distinguish bribery are wholly unavailing. Still, it is worth considering why he tries so hard to distinguish bribery. He apparently shares my strong intuition that a governor who took a bribe to veto (or to sign) legislation should be subject to criminal punishment for that offense, and that to preserve his separation-of-powers theory he therefore needs some way to show that bribery is unique--that no other offenses may be committed in the exercise of the veto power. That the effort is unavailing is important, but the fact that he makes the effort at all is revealing.

I'll offer one more example of a crime at the federal level that undermines the strong separation-of-powers claim. Consider treason. (Spoiler alert!): Imagine a scenario based on The Manchurian Candidate or Homeland if Brody had become president and remained loyal to his former captors: U.S. intelligence learns of a foreign terrorist plot that would exploit a key vulnerability in domestic defenses; Congress acts swiftly to pass a bill appropriating funds for an emergency fix; but the President, secretly acting on behalf of the terrorists, allows the bill to sit on his desk for nine days, and then vetoes it, during which time the devastating terrorist attack occurs. Isn't this a pretty clear case of Presidential treason via the veto power?

Note that the two examples I've given--treason and bribery--are the two expressly enumerated grounds for impeachment of the President, and both are crimes that can be committed through the exercise of the veto power. There is nothing in the text or structure of the Constitution that would lead to a different result for the other, unnamed high crimes and misdemeanors for which a president can be impeached and subsequently prosecuted. Nor is there anything in the general principle of separation of powers that would lead to the conclusion that treason and bribery are unique in this regard.

Is it possible that a court might nonetheless construe the veto power of the Texas governor in the maximalist fashion I have criticized? Sure. And it's also possible that one or more of the Texas charges will be held inapplicable to Perry's alleged conduct on purely statutory grounds, or that he might be acquitted on the facts. But nothing I've seen thus far from Professor Volokh or anyone else supports the broad claim that a purely discretionary power to veto on policy grounds implies immunity from criminal prosecution for the exercise of the veto power on corrupt grounds.