Yesterday (November 16, 2022), Yale Law School (YLS) announced that it would no longer be cooperating with U.S. News & World Report (USNWR) on law school rankings. Within a few hours, Harvard Law School (HLS) made the same announcement. Earlier today, Berkeley made a similar statement. My view on the merits is that the USNWR rankings scheme is bad for legal education, for many of the reasons articulated by Deans Gerken, Manning, and Chemerinksy. It’s not that rankings are necessarily bad—giving students, employers, and others information on law schools is important. The problem is that USNWR places weight on arbitrary and manipulable factors, which in turn pressure schools to allocate resources in ways that are detrimental to legal education, equity, and ultimately society at large. So sign me up for the project of breaking USNWR’s spell.

Ever since yesterday’s announcements, folks have been asking me whether there is a potential antitrust problem with any of this. Before answering, let me preempt future efforts to use anything I say here as factual evidence with respect to collusion. I have had nothing whatsoever to do with my own institution’s consideration of these issues—indeed, I don’t know whether my own institution is considering doing anything. So I am only offering an objective scholarly analysis of the problem.

First observation, which could be dispositive of everything: Any school that wants to make a unilateral decision not to cooperate with USNWR is free to do so. The antitrust laws impose no duty to cooperate with a third-party (or third-rate) media outlet or ratings organization. So any school that wants to walk away on its own, more power to ya.

The problem, of course, is that very few schools have the, uhm, fortitude to do this on their own. Even hegemonic YLS and HLS seemed to gain courage from one another’s resolve. Do we seriously think that there were no prior conversations and that John Manning happened to see Heather Gerken’s announcement come across the transom and instantly decided “oh, let’s do that too?” Whether or not HLS and YLS “agreed” in antitrust sense I don’t purport to say, but it’s hard to believe that there was no interdependence to their decisions.

And that raises the threshold antitrust question: is there agreement within the meaning of Section 1 of the Sherman Act? That, in turn, has a substantive (what is agreement?) and procedural (how can you prove it?) component.

Substantively, agreement doesn’t have to be formal or explicit. Understandings or tacit agreements can suffice. A conversation in which the upshot is “I will, if you will” is agreement. On the other hand, mere “conscious parallelism” isn’t agreement. Suppose the following happens: YLS moves unilaterally, HLS learns of it and follows suit, and then the rest of the T14 do the same, with the certain knowledge that they will be ok because everyone else is certain to do the same. That is not agreement for antitrust purposes. Home free … but can you prove it (or, really, prevent the plaintiff from getting to a jury on the converse theory)?

The procedural/evidentiary angle is familiar to any proceduralists who teach Twombly: What evidence (or pleading at the motion to dismiss stage) is sufficient to disprove the possibility that parallel conduct is the product of independent decision-making, and is instead the product of agreement? If we’re thinking realistically about the antitrust risks facing defecting schools, the most immediate and salient question is whether a plaintiff could get past a motion to dismiss and get Deans Gerken, Manning, and Chemerinksy in the deposition chair (armed, of course, with their emails extracted from discovery). I won’t speculate about that here.

One strategic observation for law school deans still on the fence: You’re safest if you wait a while before joining the parade. If you announce two weeks from now, it’s much less likely that a plaintiff can plausibly interpret your move as the evident consequence of agreement. On the other hand, precisely because interdependence is key to this working, the longer individual schools wait, the less likely it is that the snowball effect will occur. There’s a moment right now where a bunch of deans may be waiting to see what happens. If everyone seems to be doing it, they would be happy to join, but if it sputters, they don’t want to be left out in the cold. So precisely the impetus that can make this work—a cascade of schools doing it all at once—is also the pattern most likely to suggest the plausibility of horizontal agreement.

But suppose that a plaintiff could prove agreement—indeed, suppose the schools got the courage of their convictions and explicitly and openly agreed to stop cooperating with USNMR. What then? Forget the “boycott” label. Some antitrust amateur might throw that out, but it’s the wrong doctrinal category. “Boycotts” in an antitrust sense have to be directed at competitors, and this one is not.

What it could be, believe it or not, is a form of horizontal price fixing. Of course no one’s agreeing directly or even indirectly on price, but antitrust law defines the “price fixing” category to include any tampering with market-mediated forces that affect the vectors of competition. (Antitrust mavens may want to think of Catalano, where the agreement to stop extending short-term credit was price fixing, even though the defendants continued to compete along many other dimensions). What’s more, if there is an agreement here, it’s clearly horizontal—the laws schools are obviously competitors. When a group of dentists agreed to stop supplying x-rays to insurance companies, SCOTUS found them liable. If you want to think about how this could be modeled, suppose that a group of financial institutions agreed to stop supplying data to credit rating agencies. Yup, serious liability.

But not so fast. Antitrust distinguishes between “naked” horizontal agreements and those that are “ancillary” to some other bona fide efficiency-enhancing function. The former are per se illegal, the latter judged under a much more permissive rule of reason. Further, in making these assessments, the courts are typically reluctant to apply the per se rule to professional or ethical self-regulation by “the learned professions.” So doctors, dentists, lawyers, engineers, etc., who engage in self-regulatory processes typically do not face per se liability. If this were all done through the ABA or the AALS, it wouldn’t be immune from liability, but much more likely that the rule of reason would apply. But, for now, it isn’t being done that way.

And then it gets more complicated (with apologies to those whose eyes have already glazed over). When courts excuse defendants from per se liability because of the “learned profession” exception, they don’t necessarily go straight to full rule of reason analysis. Rather, they sometimes apply something called the “quick look,” one of the most misunderstood (especially by judges) antitrust categories. I won’t bore you with all of the details, but the upshot is that, in a quick look case, the defendant bears the prima facie burden of offering a procompetitive justification for the challenged restraint. If it’s accepted, then we’re off to full rule of reason analysis. If rejected, the restraint is treated as effectively per se illegal.

So suppose that (1) two or more law schools were found to have agreed, (2) a court believed that they were doing something like professional or ethical self-regulation, and (3) they were then put to the obligation of offering a procompetitive justification. Now things get even trickier. What counts as a procompetitive justification? In particular, do efforts to increase educational access and equity count? The leading case is the 3rd Circuit’s 1993 decision in U.S. v. Brown University, where the court held that the Ivy Leagues’ “overlap” program on student financial aid could be justified on the grounds that it was designed to maximize “educational access and opportunity.” The law schools would argue something similar here. But if they were agreeing casually instead of as part of some more formal and deliberative process, it’s not clear that they would even get to quick look analysis. I shiver to say it, but the possibility of per se condemnation is real.

There are many more bells and whistles to antitrust liability, of course. Private plaintiffs would have trouble showing standing, because it’s far from clear that anyone has suffered cognizable injury. Governmental enforcers wouldn’t have that problem. And all of this is assuming that there is sufficient proof of horizontal agreement, which there may well not be.

To sum up, if all of the law schools act unilaterally, no problem. If they coordinate but through a formal “self-regulatory” process, there’s antitrust risk, but they have good defenses. If they simply agree informally not to supply data to USNWR…. Have I mentioned that I know some good antitrust lawyers?

Daniel A. Crane is the Frederick Paul Furth, Sr. Professor of Law at the University of Michigan Law School.