Crime and Social Interactions: A Developing Country Case Study

Abstract

The paper investigates the relationship between social interactions and crime for the case of a developing country that has shown increasing crime rates in recent years. Social interactions are defined as any type of relationship of an individual with other individuals that can affect his/her behavior. We use a unique data set to investigate whether interaction among individuals of the same family and/or a neighborhood or community had any influence on the criminal behavior of individuals who are already convicted. The data set was constructed through interviews at the Papuda State Jail in Brasilia, Brazil. The main finding is that social interaction seems to be an important factor to explain criminal behavior. Second, different sets of social interaction variables are required to explain different types of crime. Finally, some variables affect different types of crime in opposite directions. For example, a more educated person has a lower probability of committing homicide, but has a higher probability of being a drug dealer. In the same way, when an individual uses drugs his probability of committing homicide increases, but his probability of being a drug dealer decreases. The results contribute to a growing literature on the study of criminal behavior and demonstrate that strenghtening family and community relationships can be a desirable way of fighting crime in the context of developing countries.