Socially Optimal Crime and Punishment

Speaker: Rodrigo Soares (INSPER)
When: Monday, May 12 – 1:30 PM (Santiago)
Where: Sala Consejo (401) – Beauchef 851, 4th floor | Department of Industrial Engineering

Abstract:
This paper develops a dynamic equilibrium model of crime with heterogeneous agents and human capital accumulation. Agents decide whether to commit crimes by comparing potential gains with the expected punishment. Public security policies are defined as combinations of police force size and average sentence length. We propose a micro-founded public security technology that links police expenditures to the probability of apprehending a criminal. Using an estimated version of the model with US data, we: (i) characterize the socially optimal public security policy; (ii) explore potential complementarities between security and education policies; and (iii) show that comparative statics exercises can yield different recommendations from those based on global optimality analysis under reasonable parameter values. These findings underscore the importance of equilibrium effects in the economic analysis of crime and provide a methodological foundation for integrating diverse pieces of evidence when evaluating public policy trade-offs.