Factors contributing to prison overcrowding are diminished discretion in granting parole, demographic changes, and public opinion. Methods for anticipating and controlling prison populations involve population projections with and without policy changes and the development of policy that prevents the consequences of prison congestion. Some prison population estimation methods are to use the current population as a baseline, project future prison populations as a linear extrapolation of recent trends, and apply multivariate regression. Other methods use demographic-specific incarceration rates, disaggregated flow models, and microsimulation models. The estimation of a policy change on prison population involves characterizing the subset of court cases to which the policy applies, translating the policies into corresponding values of the policy variables in the projection models, formulating the behavioral model that characterizes the court's response to the sentencing policy, and calculating the projected change in prison population resulting from response to the changed policy. The estimation of policy impact is illustrated for a Pennsylvania mandatory minimum sentencing bill. The Minnesota sentencing guidelines system is presented as a means of incorporating prison population projections and policy impact estimates into the policy process. 39 footnotes, 3 tables, and 1 figure.