Are Private Prisons More Cost-Effective Than Public Prisons? A Meta-Analysis of Evaluation Research Studies
Author: Travis C. Pratt
Author: Jeff Maahs
Abstract: The need to reduce the costs of incarceration to state and federal correctional agencies has allowed the movement to privatize correctional institutions to gain considerable momentum. The empirical evidence regarding whether private prisons are more costeffective than public institutions, however, is inconclusive. To address this question, a meta-analysis was conducted of 33 cost-effectiveness evaluations of private and public prisons from 24 independent studies. The results revealed that private prisons were no more cost-effective than public prisons, and that other institutional characteristics—such as the facility's economy of scale, age, and security level—were the strongest predictors of a prison's daily per diem cost.
Date: July 1, 1999
URL: https://doi.org/10.1177/0011128799045003004
Accessed: 2/3/2022, 2:39:13 PM
Volume: 45
Pages: 358-371
Publication: Crime & Delinquency
Issue: 3
Reading Notes:
Objective: To look at a number of studies to determine if private prisons are more cost-effective than public prisons of a similar type
Importance: The individual reviews have not settled the debate between advocates of private prisons and the detractors
Data: 33 evaluations of public vs private prisons from 24 different studies, including only adult male prisons
Key variables: Daily per diem cost, ownership of prison, # of inmates (considered measure of economy of scale), age of the facility (newer better designed), security level, year of data collection
Background: Public choice theory - privately run firms are typically more cost-effective than publicly-run firms
Methodology: Meta-analysis - running a regression on the results from a compilation of other studies
Results: Private prisons may have a small cost savings, but most of the difference in cost between prisons come from economies of scale, age of facility, and security level
Comments: I am unfamiliar with meta-analysis, so don't know how to judge the validity of the choices made with respect to the method
Key Table/Figure:
