Empirical Investigation of Correctional Cost Functions - Final Report

• NCJ Number: 86002
• Author(s): A D Witte; R Hofler; W McGuire; P Reid; J Bass; W Trumbull; C Chabbott
• Date Published: 1981
• Length: 747 pages
• Grant Number(s): 78-NI-AX-0059
• Sponsoring Agency: National Institute of Justice (NIJ); US Dept of Justice NIJ Pub
• Corporate Author: University of North Carolina
• Publication Format: Document
• Language: English
• Country: United States of America

Annotation

Using modern cost and production theory along with advanced statistical techniques, this analysis estimated short- and long-run cost curves for large institutions of the California and Federal prison systems, with attention to the impact of size and rehabilitative programs on costs.

Abstract

Data were compiled from 27 Federal prisons for the period November 1975 through June 1978 and from 10 California prisons for the period July 1968 through June 1978. The study constructed three types of models to analyze these data: (1) short-range cost functions which looked at variabilities of cost of inmate confinement under the assumption of fixed capital endowment; (2) long-range cost functions which examined the system's cost minimizing behavior treating both capital and labor costs as variables; and (3) frontier cost functions which attempted to model the maximally efficient prison on the basis of empirical study of the econometric behavior of the industry as a whole. The results indicated that production in large-scale prisons was best characterized by a homothetic Cobb-Douglas production function. Bigger prisons were cheaper in both systems in the short run. In the long run, bigger prisons were cheaper in the Federal system, but prison size did not appear to affect costs in the California institutions. Because both systems were using their resources in a technically inefficient manner, California's prison costs were 14 percent higher and the Federal prison system's costs 8 percent higher than necessary. The analysis also revealed that, in the short run, costs can be minimized in both systems by providing either many or very few rehabilitative activities. In the long run, costs in Federal prisons will be lower if large numbers of inmates are involved in rehabilitation programs. A supplemental analysis showed that correctional standards requiring either more space per inmate or more sanitary facilities will increase costs, but standards requiring single cells may actually decrease costs. The study also examined relationships between costs, security levels, and inmate characteristics, the models' ability to respond to major policy changes, the nature of production in the prison system, and determinants of allocative inefficiency in the Federal prisons. Tables, computer printouts, and over 100 references are included. (Author abstract modified)