Criminometrics, latent variables, panel data, and different types of crime

A behavioural model of crime is developed and applied to panel data on the number of crimes and clear-ups for the 53 police districts in Norway for the period 1970-78. Data on both total crime and on 12 different types of crime is employed. The model consists of behavioural relations of the offenders and the police, and of measurement relations allowing for random and systematic errors in the registered crimes and clear-ups. A theoretical analysis reveals that the model is identified under certain conditions, and our empirical analysis supports the hypothesis that these conditions are satisfied. Detailed empirical results on deterrence elasticities and other structural parameters are presented. Keywords: Economics of crime, deterrence, equilibrium models, panel data, latent variables, measurement errors