The research used vector autoregressive methodology that focused on the econometric concepts of exogeneity, cointegration, and causality. The research sought to determine the degree to which existing estimates of the effect of imprisonment on crime rates are robust to changes in data, estimation technique, and specification. The study used aggregate time-series data from 1960 to 1997 instead of data for smaller subnational units for the same period to estimate the model. The analysis also used female labor supply as a proxy for the many economic and social changes that have affected families and communities over the last 40 years. Data sources included the FBI Uniform Crime Reports, the National Criminal Victimization Survey, prison population data, and the Economic Report of the president. The short-term response of the crime rate to increased labor force participation of women was larger than the long-term effect. Findings implied that major social changes such as the increased labor supply of women may have surprising impacts and that social institutions may take some substantial period of time to adjust to such major changes. Tables, figures, and 36 references