Observing Actions in Global Games

TL;DR

This paper analyzes how private information quality affects equilibrium uniqueness in Bayesian coordination games, with implications for phenomena like bank-runs and riots.

Contribution

It introduces a model of global games with private information over actions, showing how information precision influences equilibrium multiplicity.

Findings

Low-quality private information leads to unique equilibrium.

High-quality private information allows multiple equilibria.

Results apply to real-world coordination crises like bank-runs and riots.

Abstract

We study Bayesian coordination games where agents receive noisy private information over the game's payoffs, and over each others' actions. If private information over actions is of low quality, equilibrium uniqueness obtains in a manner similar to a global games setting. On the contrary, if private information over actions (and thus over the game's payoff coefficient) is precise, agents can coordinate on multiple equilibria. We argue that our results apply to phenomena such as bank-runs, currency crises, recessions, or riots and revolutions, where agents monitor each other closely.