Prison Downsizing and Public Safety

• DOI: http://doi.org/10.1111/1745-9133.12203
• Author: Steven Raphael,Magnus Lofstrom
• Published date: 01 May 2016
• Date: 01 May 2016

POLICY ESSAY

DOWNSIZING PRISONS

Prison Downsizing and Public Safety

Evidence From California

Magnus Lofstrom

Public Policy Institute of California

Steven Raphael

University of California, Berkeley

Since the mid-1970s, the United States has experienced explosive growth in the

incarceration rate and now incarcerates adults at a higher rate than any other country

in the world (Raphael and Stoll, 2013). State and local budgets primarily carry the

economic burden as most inmates are held in state prisons and local jails. The social

costs of incarceration are largely borne by poor and minority households whose members

disproportionately experience incarceration directly or indirectly through the incarceration

of a family member. Not surprisingly, many states, as well as the federal government, are

actively seeking alternative strategies to manage public safety. Recent reforms have put

California at the forefront of broad efforts across the country to address the reliance on

costly incarceration.

California’s recent history presents unique opportunities to study large, exogenous

changes in incarceration rates. The 2011 public safety realignment (or AB 109) examined

by Jody Sundt, Emily Salisbury, and Mark Harmon (2016, this issue) caused one of the

largest declines in a state’s incarceration rate in U.S. history with the absolute magnitude

of the decline comparable with those caused by the Italian periodic collective amnesties.

The article by Sundt et al. (2016) addresses one of the key issues and challenges facing

efforts to reduce reliance on costly incarceration; can it be done without jeopardizing public

safety? The Sundt et al. article complements earlier studies, and the findings are, broadly

speaking, consistent with the existing research that has employed California’s realignment

reform to study the relationship between incarceration and crime (Lofstrom and Raphael,

2013b, 2015, 2016). Several different empirical strategies focusing on realignment now

paint a picture of noticeable declines in incarceration rates with no measurable effect on

Direct correspondence to Magnus Lofstrom, Public Policy Institute of California, 500 Washington Street, Suite

600, San Francisco, CA 94111 (e-mail: Lofstrom@ppic.org).

DOI:10.1111/1745-9133.12203 C2016 American Society of Criminology 349

Criminology & Public Policy rVolume 15 rIssue 2

Policy Essay Downsizing Prisons

violent crime and only a modest effect on property crime, which is driven by an increase

in motor vehicle thefts. These findings are also consistent with a growing body of research

showing diminishing marginal crime preventive effects of incarceration (Johnson and

Raphael, 2012; Liedka, Piehl, and Useem, 2006).

We generally agreewith the empirical findings in Sundt et al. (2016) and, in particular,

with the magnitudes of the effect estimates. However, we do have a few areas of subtle dis-

agreement with Sundt et al. that warrant additional discussions. In particular, Sundt et al.

assert that the reform they study should not be characterized as a decarceration effort but

as a decentralization of incarceration responsibilities. This characterization by Sundt et al.

conflicts with the sharp overall decrease in incarceration in the state and fails to appreciate

the particular population dynamics unleashed by the 2011 reforms. Second, Sundt et al.

argue that auto theft rates are highly volatile, that one should not interpret their findings as

evidence of an impact of the reform on auto theft rates, and that the effect they do find dis-

appears by 2014. We disagree. The evidence of an impact in the first few post-realignment

years is strong, although the effect is arguably small. The increase in auto theft rates relative

to a carefully chosen set of comparison states is robust, is evident in cross-county analysis as

well as in interstate comparisons, and persists through 2014. That being said, we do observe

that the auto theft effect narrows by 2014, which suggests that local law enforcement or local

probation departments were able in part to address this impact of realignment 3 yearsinto the

reform. In any case, the effects are small. Finally, the findings and conclusions from our ear-

lier work, in particular,how they situate within the larger prison–crime literature, are slightly

mischaracterized. When situated within a broad set of research studies, the results presented

by Sundt et al., as well as the findings from our research, are strongly suggestive of diminish-

ing marginal effects of incarceration on crime levels. This in turn means that the marginal

benefits associated with the relatively high pre-realignment incarceration levels are also small

and likely do not justify the high fiscal and social costs of typical U.S. incarceration levels.

California’sDecarceration Path

After decades of dramatic growth in its state prison population—primarily driven by “tough-

on-crime policies”—California faced severe overcrowding and lawsuits alleging inadequate

mental health care (Coleman v. Brown,originally filed in 1991) and later inadequate medical

care (Plata v. Brown, originally filed in 2001). These lawsuits resulted in the appointment

by a federal court of a Special Master and a Receiver with budgetary authority to oversee

mental health care and medical care, respectively. As a result of a lack of progress under

this remedy, a federal three-judge panel was convened in 2007 to oversee prison reforms in

the state with the aim of bringing the state system into compliance with the federal court

orders. In 2009, the three-judge panel ordered the state to reduce the prison population

to a level that it deemed would be consistent with adequate mental health and medical

care. The order effectively required a reduction in the population from close to 190.0% of

350 Criminology & Public Policy