Constats d'infraction, accidents de la route et certitude relative de la peine : une évaluation quasi-expérimentale des effets contextuels et structurels de la dissuasion policière

Abstract

Results of studies using cross-section and time series data call into question the utility of deterrence theory to explain the relationship between arrests and crime. In this study, we propose that the absence of a relationship between arrests and the crime rate is explained, in part, by the omission of notions of contextual and structural deterrence in the analysis. Contextual deterrence would be produced when the risk of arrest is increased to an unacceptable level for the population and this risk is maintained over a lengthy period of time. Structural deterrence underscores the importance of considering police jurisdictions as a unit of analysis, since the sentence would affect the exposed population. In order to demonstrate the utility of such notions, we have evaluated the effect of the Road and Traffic Safety Team of the Montreal Police Service on collisions involving injuries. Our results show that the introduction of the Road and Traffic Safety Team was followed by an immediate and sustained increase in police activity, which, in turn, led to a gradual decrease in collisions with injury. These results show the importance of including the concepts of contextual and structural effects in the study of the general deterrent effect of sentencing.