Regulating Speech with Bayesian Audiences


Defamation law fines speakers who make certain false statements, because such statements mislead their audience. It is commonly thought that stricter defamation laws offer better protection against misleading statements. Here, we study the audience's equilibrium behavior and beliefs in the presence of defamation laws of varying strictness.

We find that both lax and strict defamation laws have undesirable consequences. Strict and lax regulation of information make speech largely unreliable by deterring truthful negative remarks and failing to deter frivolous statements, respectively. Under a large set of circumstances, the optimal regulation of communications should be moderate, to facilitate the effective communication of private information and to balance a trade-off between deterring defamation, chilling truthful criticisms, and litigation costs. The court's competency plays a key role in determining the optimal defamation regime, and when courts are sufficiently capable of sorting out frivolous defamation claims, the optimal defamation regime leads to a separating equilibrium where statements always accurately inform the audience.

Although our analysis focuses specifically on defamation law, it is illustrative of the dynamics present in many other contexts where the law regulates disclosure by interested private parties to a Bayesian audience. We discuss some of these contexts, including securities regulations, whistle-blowers, jury trials, and reports of criminal activity.