'As time goes by': the effectiveness of protection orders in reducing intimate partner violence recidivism

Abstract

This study investigates the effectiveness of protection orders in reducing intimate partner violence (IPV) recidivism. Protection orders are legal measures imposing restrictions on aggressors to prevent them from approaching or contacting their victims. These legal measures have seen rapid international expansion in recent years. However, understanding the extent to which they work remains an ongoing challenge. To contribute to this debate, we employ Cox standard proportional hazards models and Prentice-Williams-Peterson gap-time models for recurrent events using a new dataset, which includes comprehensive registry data from an entire cohort of more than 6000 aggressors sentenced for at least one IPV crime over a five-year period, followed for at least four years. Our findings unveil an initial, average adverse effect of protection orders on recidivism. This effect becomes non-significant after excluding breaches of the orders and after controlling for judges' tendency to impose these measures on high-risk offenders. Moreover, we observe that the effects of protection orders vary over time: While there is an adverse immediate impact of holding an order, a positive temporal cumulative effect counteracts it after a few months.