Does the Death Penalty Save Lives?: New Evidence From State Panel Data, 1977 to 2006

• NCJ Number: 229090
• Journal: Criminology & Public Policy Volume: 8 Issue: 4 Dated: November 2009 Pages: 803-843
• Author(s): Tomislav V. Kovandzic; Lynne M. Vieraitis; Denise Paquette Boots
• Date Published: November 2009
• Length: 41 pages
• Publication Format: Article
• Publication Type: Report (Study/Research); Legislation/Policy Analysis
• Language: English
• Country: United States of America

Annotation

Using well-known econometric procedures for panel data analysis, this study adds to the other studies that have used modern econometric methods to test the hypothesis that "capital punishment deters homicide."

Abstract

This study found that the death penalty (DP) is not a significant deterrent to homicide. This finding is consistent with research by economists Katz et al. (2003) and Donohue and Wolfers (2005); however, the current finding differs sharply from the strong prodeterrence findings of many recent DP studies by economists. The most likely explanation for the latter difference in findings is the failure of the prodeterrence studies to address adequately omitted variable bias by failing to include year dummies or State-specific trends in the regression model. Prodeterrence studies also failed to adjust standard errors to correct for serial correlation. In addition, they did not use reliable and valid instruments in addressing potential simultaneity bias between execution risk and homicide. The findings of the current study are also largely consistent with the considerable body of research on how offenders make decisions in the initiation of their criminal behaviors as well as research on the nature of homicide. This article reviews such research and offers guidance for future deterrence research. Similar to recent economic papers on capital punishment, the current study used annual State panel data. The study period began in 1977 (post-Gregg era) and extended the study period used in recent studies from 2000 to 2006. The primary advantage of the panel design, as opposed to the more commonly used time-series design in earlier DP deterrence research, is that it provides a comparison group by treating non-DP States as a control group for DP States. DP measures are explained, along with homicide rates, control variables, and statistical methods for panel data. 4 tables and 98 references