The US Department of Treasury has released a working paper (paper) that examines ways in which agent-based models can explain how the behaviour of firms can affect financial stability. Existing models of financial instability tend to be based on top-down, partial-equilibrium views of markets and their interactions; they are unable to incorporate the complexity of behavior among heterogeneous firms or the tendency for all types of firms to change their behavior during a crisis. The paper argues that agent-based models (ABMs)—which seek to explain how the behavior of individual firms or “agents” can affect outcomes in complex systems—can make an important contribution to our understanding of potential vulnerabilities and paths through which risks can propagate across the financial system. 

The paper has five sections. Section I introduces ABMs and illustrates their application in social interactions and vehicular and pedestrian traffic flow, areas in which they have led to valuable insights. Section II highlights the differences between traditional economic models and the ABM approach, focusing on the assumptions that can be weakened as one moves from traditional models to the ABM approach. Section III gives some examples of applications of ABMs in finance and economics. ABMs are outside the academic mainstream in economics and finance; as a result, the literature is sparse, with much of what has been done coming out of industry and central banks rather than academia. Section IV discusses some of the practical issues in implementing ABMs.  Section V concludes with a discussion of how ABMs might fit into the agenda of the OFR and be used to address general questions of policy analysis.