Much of the debate over Judge Brett Kavanaugh’s nomination to the Supreme Court focuses on his view of executive power. But the discussion here actually encompasses four separate issues: the question of when a sitting president can be investigated and tried for possible violations of criminal and civil law, judicial deference to executive branch agencies’ interpretations of law, judicial deference on national security issues, and the theory of the “unitary executive.” It is important to address each of these issues separately, because Kavanaugh’s positions have very different implications for them. In my view, concern on the first issue is overblown, and Kavanaugh is actually likely to help constrain executive overreach on the second. On the other hand, there is good reason to worry about his record on national security and the unitary executive.

I. The Overblown.

Many Democrats fear that Kavanaugh might cast votes in favor of neutering the Mueller investigation of Donald Trump’s possible collusion with Russia during the 2016 campaign, and investigations of other types of wrongdoing by Trump and his associates. It is indeed true that in a 2009 Minnesota Law Review article, Kavanaugh argued that Congress should pass a statute shielding the president from investigation and prosecution until after he has left office, because “we should not burden a sitting President with civil suits, criminal investigations, or criminal prosecutions.” But as, Harvard law professor Noah Feldman and Benjamin Wittes of the Brookings Institution explain, liberal fears are likely misplaced. And neither of them, to put it mildly, are fans of Trump. The fact that Kavanaugh argued that a congressional statute would be needed to shield the president from investigation (and possibly even prosecution) suggests that he does not believe that the Constitution forbids such investigations in and of itself. Unless and until Congress enacts the type of statute Kavanaugh advocates (which seems unlikely to happen anytime soon), he would likely vote to allow Mueller to continue his investigation, and the same for other investigations into wrongdoing by Trump.

For what it is worth, I do not agree that the sort of sweeping immunity statute Kavanaugh advocates is a good idea. There is a solid case for postponing investigations into petty illegality by the president. If, for example, evidence indicates that the president smoked a marijuana joint or committed some minor violation of tax law, there may be good reason to avoid burdening him with investigation and prosecution until he leaves office. The situation is very different in cases where he may have committed a serious violation of the law, especially one that undermines constitutional constraints on government power, or threatens national security. I am not convinced that it is safe to rely solely on impeachment as the sole remedy for such wrongdoing, until the president leaves office. But, be that as it may, this is a disagreement about policy, not about the dictates of the Constitution.

Kavanaugh’s support for unitary executive theory (discussed more fully below) could potentially lead him to conclude that Mueller’s appointment as special counsel is unconstitutional. But, for reasons Wittes explains, that is unlikely, because Mueller’s authority is constrained in ways that fit what Kavanaugh himself has advocated in the past. Moreover, as prominent conservative lawyer George Conway (ironically, also known for being the husband of Trump adviser Kellyanne Conway) has explained, the Mueller investigation is constitutional even under a very rigorous application of unitary executive theory, because Mueller is subject to control and removal by his superiors in the Justice Department. The question of the legality of Mueller’s investigation may never reach the Supreme Court. But if it does, I think there is little reason to think Kavanaugh would vote to immunize Trump from further investigation.

II. The Good.

On the issue of Chevron deference, Kavanaugh’s skepticism of judicial deference to executive agencies’ interpretation of law might very well help constrain abuses of executive power. The 1984 Chevron decision states that courts must defer to agency interpretations of federal law in most cases where the law is ambiguous, and the agency’s view of its meaning is “reasonable,” which in practice is often a very low standard. In a 2016 Harvard Law Review article, Kavanaugh argued that, “[i] many ways, Chevron is nothing more than a judicially orchestrated shift of power from Congress to the Executive Branch” and that it encourages the executive to play fast and loose with the law because its “inherent aggressiveness is amped up significantly” by the expectation of judicial deference. He fears that “executive branch agencies often think they can take a particular action unless it is clearly forbidden.” For reasons discussed here (in analyzing Justice Neil Gorsuch’s similar views), I largely agree.

Kavanaugh’s concerns about Chevron echo those recently expressed by Justice Anthony Kennedy; so his elevation to the Court may not change the balance on this issue much. Chevron deference has also begun to attract skepticism from many judges across the political spectrum. Cutting back on judicial deference to agencies has obvious appeal to conservative and libertarian critics of the administrative state. But it should also appeal to liberals who fear that Republican agency heads are likely to be ideologically biased, and bend the law to their own preferences. See, for example, this praise of Kavanaugh’s critique of Chevron by prominent liberal legal scholar Jed Shugerman. Cutting back Chevron could also help strengthen the rule of law in ways that should appeal to people with a wide range of political preferences. As Neil Gorsuch put it in a well-know opinion written when he was a lower court judge, judicial deference to executive agencies allows the latter to “reverse its current view 180 degrees anytime based merely on the shift of political winds and still prevail [in court].” Surely there’s something wrong with that.

It unlikely that Kavanaugh’s confirmation would lead to the complete overruling of Chevron. But it would likely continue – and perhaps accelerate – the trend towards tighter, and less deferential judicial review of agency decisions.

III. The Seriously Problematic.

Kavanaugh’s record on judicial deference to the executive on national security issues is far less reassuring. As legal scholar Stephen Vladeck explains in the Washington Post, he has extended broad deference in several cases dealing with the rights of Guantanamo detainees. In that respect, Vladeck emphasizes, he is very different from Justice Kennedy, who was much more willing to enforce legal limits on executive national security policy. Kavanaugh’s deferential attitude may also come through in his concurring opinion to the DC Circuit’s denial of en banc review of a decision upholding the NSA’s collection of a vast amount of “bulk data” from American electronic cell phone records. While some of Kavanaugh’s reasoning was based on direct application of Supreme Court precedent, he also emphasized that “[t]he Government’s program for bulk collection of telephony metadata serves a critically important special need – preventing terrorist attacks on the United States…. In my view, that critical national security need outweighs the impact on privacy occasioned by this program.” This argument implies that other national security justifications might well also qualify as “special needs” that justify searches that would otherwise violate the Fourth Amendment.

Wide-ranging judicial deference on national security cases is not justified by the text of the Constitution. Unlike some other constitutions, ours has very little in the way of “emergency powers” provisions that diminish or cancel out constitutional rights in time of war or other national security threats. The major exception is the power to suspend the writ of habeas corpus, which has not been invoked in any of our current conflicts. The standard expertise-based arguments for special judicial deference are seriously flawed, and have led to severe abuses in the past.

The other area where Kavanaugh’s views on executive power are potentially troubling is his support of “unitary executive” theory – the idea that the Constitution requires nearly all executive power to be concentrated in the hands of the president. I used to support this view myself. But I now believe I was wrong to do so. Unitary executive theory was sound in a period where the scope of executive power was confined to its comparatively narrow original bounds. But, for reasons I summarized here, it is both dangerous and contrary to the original meaning to concentrate so much authority in one person’s hands in an era when the executive wields vastly greater power than was granted to the federal government in the original Constitution.

Judge Kavanaugh’s views on national security deference and the unitary executive are well within the mainstream of modern legal thought. He is no wild-eyed radical, or partisan apologist for Trump. His controversial opinions on national security issues actually came in cases challenging Obama administration policies.

But the fact that these ideas are mainstream is not as comforting as it may seem. Historically, most of the worst Supreme Court decisions came about precisely because mainstream legal thought went wrong on some important issue – not because the justices who voted for them were fools or extremists. A misguided “mainstream” is far more dangerous than the occasional fluky extremist ruling. When it comes to national security and the unitary executive, Kavanaugh is an especially thoughtful and articulate defender of positions on which mainstream conservative jurisprudence has gone wrong.

The fact that he may be wrong about these two questions doesn’t necessarily mean he would be an undesirable Supreme Court justice,overall. Kavanaugh’s flaws here should be weighed against his excellent record on many other issues, such as his strong support for freedom of speech. At this point in our history, neither Republicans nor Democrats are likely to give us an ideal Supreme Court justice who gets every important issue right. Very far from it, in fact. We could easily do far worse than Kavanaugh, who is a very solid choice in many ways. Still, Judge Kavanaugh’s positions on executive power are an important aspect of his overall record, and they deserve serious scrutiny.