*This is the fourth post in a series on Cass Sunstein and Adrian Vermeule’s new book Law and Leviathan: Redeeming the Administrative State. For other posts in this series, click here.

Law and Leviathan is a comforting read. The modern regulatory state, Cass Sunstein and Adrian Vermeule tell us, is not in tension with the rule of law, however much its critics may fulminate about incipient tyranny. Look closely and you’ll find that various doctrinal safeguards address the rule-of-law concerns that most worry those critics. Those safeguards aren’t as formidable as the Justice Gorsuches of the world might prefer. They’re only “second best.” But stronger measures are “inconsistent with pillars of modern administrative law” and “would radically destabilize too many extant arrangements.” (p. 118).

At most, we’ll see change at the margins. The nondelegation doctrine will stay dead, though some statutes may be read to foreclose excessive delegations. Chevron will remain intact, even as the major questions doctrine limits its reach. Heckler v. Chaney will afford agencies latitude to choose where to devote their enforcement resources, except when nonenforcement becomes systematic. Baltimore Gas-style deference will remain the norm, unless an agency’s rationale is completely divorced from the action the agency took.

It’s a comforting story—a bid for sanity in a world gone mad—and I appreciate it for that. The administrative state is neither as bad as you fear nor as good as you think. Like all human creations, it needs tending, and the courts aren’t doing so badly.

Still, I wonder. Sunstein and Vermeule may be right that even a Supreme Court packed with skeptics won’t radically change administrative law. But they may scant the cumulative detrimental effects of multiple small changes, especially in state courts and the lower federal courts. A lot of harm can be done out of the limelight.

* * *

I’ll use the nondelegation doctrine as an example, but the point generalizes. Although the Supreme Court hasn’t yet endorsed it, Gorsuch’s Gundyfied version of the doctrine is already causing mischief. Last October, on the cusp of the second wave of COVID-19, the Michigan Supreme Court declared unconstitutional a 75-year-old statute granting emergency powers to the governor. This February, a Texas federal judge provisionally did the same with respect to an Affordable Care Act provision requiring insurers to cover certain preventive services without extra charges. And this past Friday, five judges on the Fifth Circuit dissented from the denial of rehearing en banc because they believed a complicated agency rule governing payment of an ACA tax also violated the nondelegation doctrine.

Arresting as these cases are, they may be outliers: constitutional avoidance offers a subtler and equally effective means of trimming agency authority. A couple of weeks ago, the Sixth Circuit refused to stay a district court injunction of CDC’s eviction moratorium, agreeing with the district court that a broad construction of the agency’s statutory authority raised “concerns about the delegation of power to the legislative branch.” The Wisconsin Supreme Court has twice construed public health laws to prohibit the state from exercising emergency powers during the pandemic. Both times, justices concurred separately to flag nondelegation concerns.

For the lower federal courts, avoidance is especially appealing because it reduces the risk of Supreme Court review. Although the Justice Department has a defeasible institutional commitment to defending the constitutionality of federal laws, it lacks a commensurate responsibility to defend a grudging interpretation of a statute. Sometimes it may be better to lump the loss rather than take the case to a very conservative Supreme Court.

Nondelegation avoidance was available before Gundy, of course. But the increased salience of nondelegation concerns will make it more tempting to invoke. That will in turn affect the behavior of risk-averse and cash-strapped agencies. Why devote resources to necessary but controversial measures that might prompt judges to clip their statutory wings, especially when a dysfunctional Congress can’t be counted on to clarify the breadth of their authority?

The incentive structure for ambitious conservative judges reinforces the point. They noticed when Gorsuch was elevated to the Supreme Court because of the ardency of his nondelegation and anti-Chevron beliefs. The growth of partisan litigation from state attorneys general, strategically filed in venues before like-minded jurists, guarantees that there will be no shortage of opportunities to make similarly vocal demonstrations of fealty to the cause.

* * *

Nothing I’ve said is really inconsistent with Sunstein’s and Vermeule’s thesis. They’d agree that administrative law is contested at the margins; they just think the margins are relatively small. Taking the long view, they’ve got a point. Pendulums swing one way and then they swing back. Maybe we should all stop freaking out.

But I fear the margins here—and the stakes—are larger than they allow. The Supreme Court doesn’t need to make grand gestures to cripple agencies. Surrogate safeguards in the hands of highly motivated judges can do the trick. You sidestep a statutory bar on judicial review on Monday, withhold Chevron deference on Tuesday, remand an order for want of explanation on Wednesday, and invalidate a rule for notice-and-comment violations on Thursday. On Friday, all that’s left is to pity the poor agency.

The Supreme Court’s rhetorical bombast, paired with modest increases in the intensity of surrogate safeguards, may even be the ideal strategy to enfeeble the administrative state. Judges who come to see agencies as the bastard stepchildren of a damaged constitutional system will have less compunction about reaching for their second-best tools. Over time, as practice develops in the lower courts, it could ease the way for the more radically destabilizing steps that Vermeule and Sunstein believe the Supreme Court will never take.

So it’s hard for me to be complacent. The risks posed by an increasingly complex and interdependent world—climate change chief among them, but also financial crises, industrial pollution, and new pandemics—are not ones that the government can fight with one hand tied behind its back.

Maybe I’m wrong to worry. Maybe I need to take a deep breath, pour myself a stiff drink, and read Law and Leviathan again. Because Sunstein and Vermeule are right: come hell or high water, the administrative state is here to stay. It’s just that the water is rising awfully fast.

Nicholas Bagley is a Professor of Law at the University of Michigan Law School.