The Impact of a Road Safety Unit on Accidents with Injuries: A Quasi-experimental Study.

Several authors claim that the deterrence theory explains poorly the arrest-crime relationship, even when rates of arrest are quite high and odds of arrest are about 30%. In this paper we argue that that the so-called absence of relationship between the probability of being arrested and crime rates fails to take into account the contextual and structural effect of sanctions. Contextual deterrent effects occur when crime levels react to significant and permanents shits in arrests rates. Structural deterrent effects are likely in settings where enforcement practices can be viewed as homogeneous but unlikely in jurisdictions (cities or states) where enforcement strategies vary across police organizations. Using interrupted time series analysis, we estimate the effect of a police strategy designed to reduce traffic violations. Segmented regressions analyses show that the increase in the issuance of traffic tickets by police officers caused a gradual decrease in accidents only in Montreal. No effect was observed on the control series. So, our results support the preservation of this unit and its activities to maintain this downward trend. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]