It’s no secret that some people have misgivings about the administrative state—including, most notably, the Chief Justice of the United States. In fact, Chief Justice Roberts believes that the administrative state—with its “‘vast and varied federal bureaucracy’”—presents a “danger” that “cannot be dismissed.” Although “it would be a bit much” to condemn today’s regulatory scheme as “‘the very definition of tyranny,’” the Chief Justice laments that there are “hundreds of federal agencies poking into every nook and cranny of daily life.” “And,” the Chief Justice ominously reminds us, “the federal bureaucracy continues to grow; in the last 15 years, Congress has launched more than 50 new agencies. And more are on the way.”

To those who share the Chief Justice’s concerns, Seminole Rock deference no doubt is unsettling. After all, because of it, agencies may “promulgate vague and open-ended regulations that they can later interpret as they see fit, thereby frustrating the notice and predictability purposes of rulemaking.” A skeptic therefore might conclude that this deference “‘promotes arbitrary government.’” Doesn’t this sort of deference “subject[] regulated parties to precisely the abuses that the Framers sought to prevent”? Isn’t it “‘perfectly understandable’ for an agency to ‘issue vague regulations’” since “doing so will ‘maximiz[e] agency power’”? Why not reject Seminole Rock as “a dangerous permission slip for the arrogation of power”? In short, why shouldn’t Seminole Rock be overruled?

My answer: wait a second—not so fast. The situation is more complicated than one might think. Although it is counterintuitive, if you side with the skeptics, you ought to be particularly worried about overruling Seminole Rock. In fact, the more cynical you are about agency motives, the more you should fear a post-Seminole Rock world. (Of course, whether you should side with the skeptics is a question for another day.)

In a forthcoming article, Beyond Seminole Rock, I offer a take on Seminole Rock that should give pause to skeptics of the administrative state.* In particular, without Seminole Rock, agencies might stop promulgating as many regulations, and instead begin conducting more adjudications. That would be worse, not better, for regulated parties.

To understand why, it is necessary to appreciate another aspect of administrative law. The Supreme Court’s decision in SEC v. Chenery (1947) (Chenery II) holds that an agency generally has discretion whether to engage in prospective rulemaking or instead to simply enforce the statute itself in a retroactive adjudication. Chenery II was controversial when it was decided—Justice Jackson in dissent, for instance, condemned the Court for sanctioning “conscious lawlessness as a permissible rule of administrative action” and rejecting the principle that “men should be governed by laws that they may ascertain and abide by, and which will guide the action of those in authority as well as of those who are subject to authority”—but it is now settled law. As the Supreme Court announced in NLRB v. Bell Aerospace Co., an agency “is not precluded from announcing new principles in an adjudicative proceeding and … the choice between rulemaking and adjudication lies in the first instance within the Board’s discretion.” And as the Court stated in Martin v. Occupational Safety and Health Review Commission, such “adjudication [can] operate[] as an appropriate mechanism not only for factfinding, but also for the exercise of delegated lawmaking powers, including lawmaking by interpretation.” Likewise, subject to some exceptions, agencies receive Chevron deference for interpretations announced in these adjudications. To be sure, there are retroactivity limits on an agency’s ability to make law through adjudication, but those limits are, well, limited: “[A] mere lack of clarity in the law does not make it manifestly unjust to apply a subsequent clarification of that law to past conduct.”

With that background in place, consider a point I have made before:

[I]f the Supreme Court were to overrule Seminole Rock, what would happen? The intended consequence would be clearer regulations, as agencies would have one less reason to promulgate ambiguous rules. But isn’t there also an unintended consequence lurking in the background? Might agencies not promulgate clearer regulations, but instead promulgate fewer regulations? In particular, if Seminole Rock were gone, agencies might respond at the margins by retreating from rulemaking in favor of their power under Chenery II to enforce the statutes they administer through retroactive adjudication—no doubt coupled with a lot more guidance documents and “agency threats.” If that were to happen, regulated parties could easily find themselves worse off.

To meaningfully predict what a post-Seminole Rock world would look like, “we need to know an agency’s ‘cross-elasticity of demand’ between rulemaking and adjudication, and have a good sense for how that cross-elasticity would change if Seminole Rock (which makes rulemaking relatively more attractive) were no longer part of the equation.” The more distrustful one is of agencies, the more one should worry that they would shift to the unintended consequence rather than the intended consequence. After all, the cynical view is that agencies are actively looking for ways to increase their flexibility going forward. Without Seminole Rock, one way for an agency to continue to do that is through adjudications under Chenery II.

And that sort of shift away from rulemaking would be a bad thing. Almost by definition, even an ambiguous regulation provides more notice of someone’s legal duties than the statute that the regulation implements. It could not be otherwise. If the regulation covers more policy “space” than the underlying statute, the regulation is ultra vires, and if the regulation covers the same “space,” there is no Seminole Rock deference anyway because of the anti-parroting principle. The upshot, as Jason Marisam has explained, is that “regulations are typically narrower than enabling statutes.” Thus, if agencies shift away from rulemaking in favor of adjudication, regulated parties will receive even less notice of their legal obligations than they do now. At the same time, making policy through adjudication is problematic because it generally involves less public participation.

So what is the bottom line? It is important to understand the interrelated nature of “admin law”—what happens to one doctrine has repercussions for other doctrines. Until we understand those repercussions, we should move cautiously. And if the Court is inclined to overrule Seminole Rock, it should think about Chenery II too. For reasons explained in my article, the Court could mitigate the most problematic substitution away from rulemaking without overruling Chenery II (which would have unintended consequences of its own). But the Court can’t mitigate that substitution unless it first recognizes how Seminole Rock and Chenery II fit together. This means that the Court should tread carefully when it comes to Seminole Rock. Otherwise, the Justices could end up harming the very people they hope to help.

* Of course, my article should be of interest to non-skeptics too. Even if agencies do not intentionally promulgate vague regulations, it is a fact of life that it takes resources to anticipate and preemptively address potential ambiguities. A rational agency may be more willing to tolerate unknown ambiguities if it is confident that, should such an ambiguity arise, the agency will receive deference in addressing it. If Seminole Rock ceased to exist, an agency may be less willing to engage in rulemaking because—inevitably—it would have less confidence in its ability to control enforcement of its regulation down the line. Nothing about this analysis requires bad faith or a nefarious scheme on the agency’s part. But it does mean that without Seminole Rock (so long as Chenery II is not adjusted), we may get more adjudication.

This post is part of an online symposium entitled Reflections on Seminole Rock: The Past, Present, and Future of Deference to Agency Regulatory Interpretations. You can read the entire series here.