Emotional Agents Make a (Bank) Run

Abstract

Agent-based Computational Economics (ACE) is an area that has gained significant attention, since it offers the possibility to model economic phenomena in a more fine-grained manner than other approaches. One such phenomenon is “bank panic” in which the term “panic” implies the existence of emotional bias towards to the sudden withdrawal of deposits from financial institutions (simultaneous bank runs). However, research towards complex emotional agents in ACE has not been extensively conducted. The paper employs a formal state-based model enhanced with explicit emotional states, mood and personality characteristics in order to describe the agents behavior. A NetLogo simulation of a multi-agent system in a limited economic environment is attempted in order to study the effects of emotions, emotion contagion and the role of various players in the genesis of a bank panic crisis. The aim is to investigate further whether such agent models that are already used in other areas, such as evacuation simulation, could also provide a better insight on the evolution of such economic phenomena.