This study examines the applicability of several theoretically derived accounts used to explain the variations in police killings across 39 of Canada’s largest cities over a 15-year period. Pooled time-series negative binomial regression results are consistent with the ethnic threat hypothesis by indicating that lethal police action is associated with the size of the ethnic minority population in each city. Political accounts are supported as non-linear specifications suggest that once ethnic minorities reach a numerical majority in our sample of cities there is a decline in police killings. Findings also support expectations that greater female representation within policing will reduce the use of lethal force by changing the overall culture of the department. Theoretical implications of our findings are discussed.