Social Interactions and Crime Revisited : An Investigation Using Individual Offender Data in Dutch Neighborhoods

Using data on the age, sex, ethnicity and criminal involvement of 14.3 million residents aged 10-89 residing in 4,007 neighborhoods in the Netherlands, this article tests if an individual’s criminal involvement is affected by the proportion of criminals living in their neighborhood of residence. We develop a binomial discrete choice model for criminal involvement and estimate it on individual data. We control for the endogeneity that may be related to the unobserved neighborhood characteristics and take into account possible biases that may result from sorting behavior. We find strongly significant social interaction effects and explore their implications for the presence of multiple equilibria and social multipliers