4. L'analyse de réseaux au service de la lutte contre le crime

We study peer effects in crime using network analysis. We first put forward the role of network in criminal behaviour and provide some theoretical explanations. Then, we present some empirical tests showing the importance of peer effects in crime. We also present one of the first empirical tests of the key player policy, which consists in finding the delinquent who, once removed from the network, generates the highest possible reduction in aggregate delinquency level. We show, in particular, that the key player policy outperforms other standard policies of crime reduction, such as targeting the most active criminals or targeting the most central delinquents in the network.