Evaluating Realignment

• Author: Shawn D. Bushway
• Date: 01 May 2016
• Published date: 01 May 2016
• DOI: http://doi.org/10.1111/1745-9133.12215

EDITORIAL INTRODUCTION

DOWNSIZING PRISONS

Evaluating Realignment

What Do We Learn About the Impact of Incarceration

on Crime?

Shawn D. Bushway

University at Albany, SUNY

In the main article, Jody Sundt, Emily Salisbury, and Mark Harmon (2016, this issue;

hereafter SSH) evaluate the impact of the California Public Safety Realignment Act of

2011. They find that Realignment, which created a dramatic 17% decline in the prison

population in California, is associated with no increase in violent crime and only a modest

increase in property crime. They make it clear in their conclusions that their result should be

interpreted as the impact of the policy writ large, not about the underlying mechanisms that

might explain the result, including the effectiveness of alternative community interventions

and local law enforcement, wiser use of jail space, or diminished returns from incarceration

itself.

The three policy essays that accompany SSH’s (2016) article are in agreement about

the importance of the SSH article for the ongoing policy debate. If, as many people believe,

we are seeing a fundamental reorientation away from the use of prisons as a crime-fighting

tool in the United States, then SSH’s evaluation of Realignment is of signature importance.

SSH show clearly that it is possible to implement a policy that moves away from the use of

state prison without experiencing a large, across-the-board increase in crime.

Magnus Lofstrom and Steven Raphael (2016b, this issue) engage in the conversation

as fellow evaluators of this policy, having recently published a similar evaluation (Lofstrom

and Raphael, 2016a). Like SSH, they use data from the other 49 states, albeit with a

different estimation strategy, to create a relevant counterfactual. SSH use regression point

displacement design, and Lofstrom and Raphael use a synthetic cohort design that puts

more weight on data from states that share important characteristics with California to

create an effective control group. Both approaches are relatively new to criminology, and

Direct correspondence to Shawn D. Bushway, Department of Public Administration and Policy, Rockefeller

College of Public Affairs and Policy, 220 Milne Hall, 135 Western Avenue, University at Albany, Albany, NY

12203 (e-mail: sbushway@albany.edu).

DOI:10.1111/1745-9133.12215 C2016 American Society of Criminology 309

Criminology & Public Policy rVolume 15 rIssue 2