Peer Learning, Enforcement, and Reputation

Abstract

We consider a two-period collective experimentation model featuring self-interested agents that test a sequentially-rational principal’s enforcement propensity through their misconduct and a principal that disciplines them to build a reputation for strict enforcement. We find that a transparent enforcement regime yields more misconduct than an opaque enforcement regime because enforcement externalities, which heighten agent misconduct, dominate information externalities, which weaken agent misconduct. We also find a principal with a longer decision horizon may induce higher agent misconduct. Our results apply to various institutional settings, including relations between headquarters and divisions, common owners and portfolio firms, and regulators and firms.

Keywords: Experimentation; Peer Learning; Reputation; Enforcement Externality; Enforcement Transparency

JEL Classification: D82, D83, G18, M41, M48