William & Mary Law Review


Jamal Greene


One underappreciated cost of constitutional rights enforcement is moral hazard. In economics, moral hazard refers to the increased propensity of insured individuals to engage in costly behavior. This Essay concerns what I call “constitutional moral hazard,” defined as the use of constitutional rights (or their conspicuous absence) to shield potentially destructive behavior from moral or pragmatic assessment. What I have in mind here is not simply the risk that people will make poor decisions when they have a right to do so, but that people may, at times, make poor decisions because they have a right. Moral hazard is not about how individuals behave in general but on the margins. It concerns the incentive effect of holding security against worst-case scenarios.


As Part I discusses, the danger of moral hazard varies with the proportion of risk borne by third parties. Speech can be produced at a low cost relative to other activities. And even bracketing the potential harm to those who feel burdened or threatened by racist speech, the potential for violence in response to such speech imposes, at times, significant economic costs on universities and law enforcement agencies. Those costs are not internalized by the speakers, who are often seeking exactly the attention that visible security measures produce.

Part II establishes what is already well-known to constitutional scholars, namely that the First Amendment is not absolute. Neither the constitutional text, nor constitutional history, nor even constitutional doctrine establish that the First Amendment is incapable of being sensitive to the institutional context in which speech takes place. Moreover, and contrary to some conventional wisdom, unbridled freedom of speech is ill-suited to a university setting. As Part III elaborates, from admissions decisions, to pedagogical choices, to the hiring of faculty and administrators, universities are quintessential curators of speech. In an information environment in which citizens can expect to be flooded with data that is either indiscriminate or calculated to mislead, it is more important than ever for educational institutions to serve their essential function.

Finally, and most significantly, Part IV argues that universities nonetheless should permit provocative speakers on campus. Many of the reasons asserted for permitting such speaker--the linedrawing problems, the capacity for incidents to act as teachable moments, the importance of inculcating tolerance, and so forth--are persuasive. The point, though, is that these considerations are best understood not as interpretive arguments about constitutional free speech guarantees but rather as merits arguments in favor of a permissive campus speech policy. Colleges and universities, whether public or private, should have the discretion to reject these arguments in favor of more interventionist approaches. They should also have the discretion to experiment with different approaches to security costs for provocative speech. That is, rejecting or canceling a speaker because of a school’s inability to afford security born of a predictive concern about the behavior of protesters should be permitted under the First Amendment, notwithstanding longstanding suspicion of allowing a “heckler’s veto.” This posture creates an incentive for individuals and groups to engage in or threaten disruptive acts, and so carries its own risk of moral hazard, but whether to abide that risk should fall within a university’s discretion.

This abstract has been adapted from the author's introduction.


This essay is based on the 2017 Cutler Lecture delivered at William & Mary Law School.

Publication Information

61 William & Mary Law Review 223-251 (2019)