Over at Jotwell — the Journal of Things We Like (Lots) — Richard Murphy reviews one of Adrian Vermeule’s latest essays, “Rationally Arbitrary Decisions (in Administrative Law),” which is available on SSRN here. I should probably confess at the outset that Professor Vermeule is one of my favorite administrative law scholars, and I devour everything he writes. To be sure, I don’t always agree with him, but his arguments are provocative and force me to rethink how I view the administrative state and governance more generally.

This essay is no exception. Here’s a summary of the essay, from the SSRN abstract:

How should administrative law cope with genuine uncertainty, in which probabilities cannot be attached to outcomes? I argue that there is an important category of agency decisions under uncertainty [in] which it is rational to be arbitrary. Rational arbitrariness arises when no first-order reason can be given for the agency’s choice one way or another within a certain domain, yet the agency has valid second-order reasons to make some choice or other. When these conditions obtain, even coin-flipping may be a perfectly rational strategy of decisionmaking for agencies.

Courts should defer to rationally arbitrary decisions by agencies. There is a proper role for courts in ensuring that agencies have adequately invested resources in information-gathering, which may dispel uncertainty. Yet in some cases the value of further investments in information-gathering will itself be genuinely uncertain. If so, courts should defer to agencies’ second-order choices about informational investments on the same grounds that justify deference to agencies’ first-order choices under uncertainty.

Also with a great title — “When We Found Out That Counting Lizard Poop Is Not a Good Way to Count Lizards: Now What?” — Professor Murphy’s Jotwell review is a fun read. Here’s a sample, though the review (like the essay being reviewed) is worth the full read:

Professor Vermeule builds his argument around several judicial opinions typifying various types of uncertainty (“brute,” “strategic,” and “model”) that in his view can justify rational arbitrariness by agencies. Let’s take a quick look at his example of “brute uncertainty,” Tucson Herpetological Society v. Salazar, 566 F.3d 870 (9th Cir. 2009). This case, which by 2009 had been proceeding for 16 years, revolved around whether the “flat-tailed horned lizard,” which is a “small cryptically colored iguanid . . . that is restricted to flats and valleys of the western Sonoran desert” should be listed as a threatened species under the Endangered Species Act (ESA). A threatened species is one that “is in danger of extinction throughout all or a significant portion of its range.” 16 U.S.C. § 1532(20). The Secretary first proposed listing the lizard as threatened back in 1993. Since then, the Secretary has repeatedly issued decisions withdrawing the proposed listing, and non-profit environmental groups, such as the Tucson Herpetological Society, have persuaded courts to vacate these withdrawals. In support of a 2006 withdrawal, the Secretary “quantified the lizard’s lost range [and] explained why that range is not ‘significant’ within the meaning of the ESA.” Id. at 875. The Secretary based this conclusion in large part on a finding that “lizard populations persist across most of the species’ current range despite habitat loss and fragmentation.” Id. at 877.

But there was a very basic problem with this finding of persistence. Studies of lizard populations used to rely on “scat counts,” but all sides now agree that counting lizard poop is not a good way to count lizards. And the only available “capture-mark-recapture” study included a warning from the study’s author that it was based on “sparse data” that “should be viewed with caution.” In short, whether lizards were persisting was genuinely uncertain. Nobody but the lizards knew, and they weren’t telling. This left the agency in a bind — the administrative record did not support a finding that the lizard population was viable or non-viable. Either might be regarded as arbitrary.

Given this problem, Professor Vermeule contends that the court of appeals was wrong to throw out the agency’s finding—“[t]he Secretary had to decide in which direction to take a leap of faith, and it is a kind of pathological hyper-rationalism to demand that the Secretary give reasons for taking it one direction rather than the other.” (P. 7.) Also, targeting precautionary attitudes toward decision-making, he maintains that courts should avoid the temptation to impose conservative default rules, e.g., erring on the supposed side of “caution” to assume a low number of lizards. (PP. 7, 10-13.)

Like many administrative law professors, I’m a big fan of reason-giving in administrative law, so I’m a bit ruffled by Professor Vermeule’s attack on judicial review of agency reason-giving in the context of uncertainty. That said, Professor Vermeule makes a strong case (at 5) that “courts should defer to agencies in situations of brute uncertainty, in which well-defined facts about the world relevant to the decision cannot be ascertained (at acceptable cost); strategic uncertainty, in which interdependent choices create multiple equilibria; and model uncertainty, in which the very analytic framework to be used to assess uncertain choices is itself unclear.” The case is strengthened, I think, if the agency is candid in its decision-making by explaining why there is uncertainty and outlining the second-order reasons for why it has decided to adopt particular somewhat arbitrary approach. (It seems Professor Murphy may disagree with me on this point, suggesting perhaps that “truth [is] something that should be optimized rather than maximized in this context.”)

On the other hand, my nondelegation instincts question why we should allow the delegation of arbitrary decisions to federal agencies in the first place. Instead, when there is real uncertainty requiring an arbitrary decision, wouldn’t it be better for the agency to gather the information, provide a reasoned position for why the first-order reasons are unobtainable due to uncertainty and which way the second-order reasons cut, and then allow Congress to flip the coin through the legislative process (via bicameralism and presentment)? If the legislative process were not so plagued with gridlock, this seems like the obviously preferred approach. But, in light of gridlock, such an approach would run into status quo default problems Professor Vermeule explores in his essay.

I’m not as troubled by the status quo default if it is more liberty-enhancing and would be tempted to carve out an exception for arbitrary agency decisionmaking only when such decisionmaking enhanced liberty, not restricted it. But such a case-by-case rule wouldn’t be particularly administrable, not to mention that it’s not always easy to determine whether a rule is liberty-enhancing when multiple parties are affected, etc. So I’m left convinced that Professor Vermeule has a strong case that judicial review should be different when an agency is acting under uncertainty, dissatisfied with Professor Vermeule ultimate conclusion about how judicial review should be different, yet unable to come up with a better solution (at least not yet).

In all events, Professor Vermeule’s essay provides so much food for thought that requires much more digestion on my part. It’s a must-read for all adlaw nerds.