Association for Public Policy Analysis & Management

Deterrence-oriented crime and policing policies are predicated on the assumption that individuals’ beliefs regarding the certainty and severity of punishment can influence their decision to engage in criminal behavior. Yet, there are likely disparities between objective levels of policing and sanctioning in a certain region and individuals’ subjective beliefs about these potential risks to offending in their region. Moreover, policy shifts such as changes in policing tactics are potentially inefficient if individuals’ perceptions are not updated to reflect additional information from macro-level policies changes, and could potentially contribute to dipartites in community beliefs about procedural justice and legitimacy.

This paper considers the impact of a large scale policy change—a new crime fighting strategy initiated by the Philadelphia, PA Police Department—on both offending behavior and subjective perceptions of individual offenders across different police precincts over a seven year period from 2003-2010. Specifically, we analyze panel data on 700 serious adolescent offenders, in which we observe both self-reported offending as well as self-reported beliefs regarding perceived risk of detection and procedural justice. We explore how these outcomes change in response to shifting policies at the police district level, measured as the Philadelphia Police Department shift staffing and overtime resources to nine particularly violent police districts, during the observational period. We also link these individuals to precinct-level data on expenditures, policing and other control-related factors which also vary both within and across precinct over the time span. We analyze these data in relation to geographic boundaries accounting for individual and precinct-level fixed effects, to study whether individual perceptions of risks and procedural justice vary across different police precincts, and if these changes related to changes in policing, spending and other policy shifts.

Preliminary results suggest that there is important variability in individual perceptions that can be explained by police district, and there are important within-individual changes across the time period. We consider multiple implications for criminal justice policy.