Deterrence or Backlash? The Causal Effect of Arrest on the Dynamics of Domestic Violence

Abstract

How should police officers respond to domestic violence (DV) incidents to reduce the likelihood that battered women will be victimized again? There is a vigorous debate on whether arrests will deter future abuse or create a retaliatory backlash. We study the question of arrests and the dynamics of DV using detailed administrative data for 183,000 DV emergency calls (999 calls) for West Midlands, the second most populous county in England. We take advantage of the conditionally random assignment of officers to a case by call handlers, combined with the systematic differences across police officers in their propensity to arrest suspected batterers. We find that arrests reduce future 999 DV calls in the ensuing year by 26 percentage points, which translates to a 50% drop. Using both (i) victim versus third-party reporting of repeat 999 DV calls and (ii) measures of repeat severity, we conclude the reduction is not driven by reduced reporting, but instead a decline in repeat victimization. Arrest also increases the likelihood a suspected batterer is charged with a crime and retained in custody during an investigation. Our findings argue against recent calls for a decriminalization of domestic violence and suggest the optimal police response is to lower the threshold for arrests.