Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Canada exhibits no correlation between income and victimization, rich neighborhoods are less exposed to property crime, rich households are more victimized than their neighbors, and rich households and neighborhoods invest more in protection. We provide a theory consistent with these facts. Criminals within city choose a neighborhood and pay a search cost to compare potential victims, whereas households invest in self‐protection. As criminals' return to search increases with neighborhood income, households in rich neighborhoods are likelier to enter a race to greater protection driving criminals toward poorer areas. A calibration reproduces the Canadian victimization and protection pattern by household/neighborhood income.