Advancing Methods for Analyzing Coordination: New Developments in Global Game Theory.

Massachusetts Institute Of Technology, Cambridge MA

Investigators

Abstract

This award funds research in economic theory, with a focus on coordination games. This theory can be used to analyze a wide variety of economic problems, including financial crises and business cycles. While the standard coordination game approach offers insight into these settings, it allows multiple equilibria and thus does not offer a complete theory in the absence of a theory of equilibrium selection. In response, a global games approach has been widely used in applied economics. By assuming noisy information about payoffs, one can remove the multiplicity of equilibria and develop a wide variety of predictions that can then be tested against data. However, there are a number of important limitations to the global games approach. The four components of this research proposal address four different limitations. The components are relevant for all global game applications, but are developed in the context of applications in macroeconomics, political economy and finance. The project therefore contributes to developing new theories and methods that can help individuals, businesses, and policymakers at all levels of government make better decisions. Global game analysis has made extreme informational assumptions: there is common knowledge of payoffs and multiple equilibria, or small noise and globally unique equilibria. The project has four components, each designed to address one aspect of these assumptions. The first component examines richer classes of information structures to develop novel predictions about when global game predictions arise. Global game analysis has been primarily static. The second component examines dynamic environments with learning to develop predictions about when global game predictions occur. Global game analysis has focused on binary action games. The third component analyzes a political economy application where feasible set of available actions is itself a choice variable, and characterizes when the action set is or is not binary. Global game analysis has mostly relied on simple analytic solutions to deliver qualitative insights. The fourth component develops a finance application with a rich set of heterogeneous players in order to develop quantitative framework that that can be used in empirical work. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.