A Dynamic Agent-Based Model of Corruption

Abstract

The author builds an agent-based model wherein the societal corruption level is derived from individual corruption levels optimally chosen by heterogeneous agents with different risk aversion and human capital. The societal corruption level, in turn, affects the risk-return profile of corruption for the individual agents. Simulating a multi-generational economy with heterogeneous agents, the author shows that there are locally stable equilibrium corruption levels with certain socio-economic determinants. However, there are situations when corruption can rise until it stifles all economic activity. “You live in a society where everybody steals. Do you choose to steal? The probability that you will be caught is low ... and, even if you are caught, the chances of your being punished severely for a crime so common are low. Therefore you too steal. By contrast, if you live in a society where theft is rare, the chances of your being caught and punished are high, so you choose not to steal.” (Mauro, 1998, p. 12)