Working Paper: CEPR ID: DP6130
Authors: Eleonora Patacchini; Yves Zenou
Abstract: We develop a model where the decision to commit a crime in a neighboring area is a positive function of the percentage of same-race individuals residing in that area since they can provide crucial information on crime possibilities. The model then predicts a positive spatial correlation in crime between different contiguous areas; this correlation is higher the closer the distance between the areas. We empirically investigate these relationships using data from the crime statistics that are recorded by the police in Britain. We find results that are consistent with the model. In particular, the agglomeration of a given ethnic minority group is positively related to its crime activity and this effect declines quite sharply with distance between areas.
Keywords: crime; ethnic minorities; social interactions
JEL Codes: C23; K42; R12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| percentage of same-race individuals residing in an area (R23) | crime rates in that area (K14) |
| concentration of individuals of the same race (J15) | probability of committing a crime in a neighboring area (K14) |
| concentration of a particular ethnic minority group in an area (R23) | crime activity in that area (K42) |
| distance (R12) | effect of concentration on crime (K42) |
| crime rates in adjacent areas (K14) | correlation in crime rates (K00) |