Overall, the results indicate that public psychiatric hospital capacity has a significant negative effect on crime and arrest rates, mainly through its impact on homelessness. Mentally ill offenders were found to be overrepresented among those incarcerated in jails and prisons, especially for violent crimes. However, city-apportioned total mental health expenditures were found to be unrelated to overall crime or arrest rates. The findings suggest that community-based services may not have a significant impact on community crime and arrest rates. The findings have implications for public policy considerations that would further reduce public psychiatric capacity. The research involved a macro-level approach to examining the relationships between psychiatric hospital capacity, homelessness, and crime and arrest rates. The sample included 81 U.S. cities with populations of greater than 50,000 that had city-level estimates of homelessness from a variety of sources. Data examined were from 1989 through 1990. The city-level measure of psychiatric hospital capacity was drawn from the annual Guide to the Healthcare Field (1990) publication from the American Hospital Association. Crime and arrest data were drawn from the Federal Bureau of Investigation’s (FBI’s) Uniform Crime Reports. Data analysis techniques included bivariate correlations and regression models. Future research might consider testing the relationships among these same variables over time. Footnotes, tables, references