Edited by Bruce L. Benson and Paul R. Zimmerman
Chapter 7: Prison Population and Crime
* Thomas B. Marvell INTRODUCTION The literature survey and data analysis here have a narrow focus: the impact of prison populations on crime. Researchers reach extremely disparate estimates concerning this impact – from elasticities of zero to over minus 2. That is, at one extreme prison population growth might not affect crime trends, and at the other extreme crime rates decline by 2 percent for each 1 percent increase in prison populations, an extraordinary impact. Research results are spread across this range. Consequently, the research to date offers little help in answering the question of how much, if any, of the recent crime declines are due to the prison buildup in the past 35 years. I attempt to discern why there is such a large spread of results, and from that I attempt to give a rough estimate of the impact of prison on crime – an elasticity of about minus 1, near the midpoint of prior findings. Although most studies use the same basic methodologies, details of regression procedure make a difference. These details, in turn, are largely based on underlying theory about how the crime enterprise operates and, consequently, how imprisonment might affect crime rates. A second feature of this chapter is a new empirical analysis, in the Appendix, using a time-series–cross-section (TSCS) design with a panel of state data over 1973–2007. This is the dominant design in crime/prison studies, and the Appendix presents results with the latest data available. The main purpose of the Appendix, though, is only...
You are not authenticated to view the full text of this chapter or article.