What is known about the impact of police arrest policies relative to domestic violence is that most cases brought to police attention involve lower income and minority group households. Findings from research in six cities (Minneapolis, Omaha, Charlotte, Milwaukee, Colorado Springs, and Miami) show both deterrent and backfiring effects of arrest. Arrest cured some abusers but made others worse; arrest eased the pain for victims of employed abusers but increased it for those with unemployed partners; and arrest assisted white and Hispanic victims but fell short of deterring further violence among black victims. The authors conclude that arrest has a differential effect on suspects from different household types, that arrest reduces domestic violence in some cities but increases it in others, and that arrest reduces domestic violence among employed people but increases it among unemployed people. In addition, the authors indicate that police can predict which couples are most likely to suffer future violence, even though society values privacy too highly to encourage preventive action. Policy recommendations are offered that focus on repealing mandatory arrest laws, substituting structured police discretion, allowing warrantless arrests, encouraging the issuance of arrest warrants for absent offenders, and focusing on chronically violent couples. 1 reference and 1 table