Social Networks and Crime Decisions: The Role of Social Structure in Facilitating Delinquent Behaviour

We develop a model in which delinquents compete with each other in criminal activities but may benefit from being friends with other criminals (by learning and acquiring proper know-how in the crime business). We first study the Nash equilibrium of this game by taking the social network connecting agents as given. We show that this game always has a pure strategy Nash equilibrium for generic values of the parameters. Ex ante identical individuals connected through a network can end up with very different equilibrium outcomes: either employed, or isolated criminal or criminals in networks. We also show that multiple equilibria with different number of active criminals and levels of involvement in crime activities may coexist and are only driven by the geometry of the pattern of links connecting criminals. We then consider a two-stage network formation and crime decisions game. We show that the multiplicity of equilibrium outcomes holds even when we allow for endogenous network formation.