The Use and Misuse of Coordinated Punishments*

Abstract

Communication facilitates cooperation by ensuring that deviators are collectively punished. We explore how players might misuse communication to threaten one another, and we identify ways that organizations can deter misuse and restore cooperation. In our model, a principal plays trust games with a sequence of short-run agents who communicate with one another. An agent can shirk and then extort pay by threatening to report that the principal deviated. We show that these threats can completely undermine cooperation. Investigations of agents’ efforts, or dyadic relationships between the principal and each agent, can deter extortion and restore some cooperation. Investigations of the principal’s action, on the other hand, typically do not help. Our analysis suggests that collective punishments are vulnerable to misuse unless they are designed with an eye towards discouraging it. JEL: C73, D02, D70. ∗Corresponding author: Daniel Barron, Northwestern University, Kellogg School of Management, Evanston IL 60208; email: d-barron@kellogg.northwestern.edu. The authors would like to thank Nageeb Ali, Charles Angelucci, Nemanja Antic, Alessandro Bonatti, Renee Bowen, Joyee Deb, Wouter Dessein, Matthias Fahn, Benjamin Friedrich, George Georgiadis, Marina Halac, Johannes Hörner, Peter Klibanov, Ilan Kremer, Nicolas Lambert, Stephan Lauermann, Jin Li, Elliot Lipnowski, Shuo Liu, Bentley MacLeod, David Miller, Joshua Mollner, Dilip Mookherjee, Arijit Mukherjee, Jacopo Perego, Michael Powell, Luis Rayo, Jonah Rockoff, Mark Satterthwaite, Andy Skrzypacz, Takuo Sugaya, Jeroen Swinkels, Joel Watson, and audiences at many conferences, workshops, and seminars. We thank the UCSD theory reading group for comments on a draft of this paper, and Andres Espitia for excellent research assistance.