Oregon’s Transitional Leave Program and Recidivism

• Published date: 01 January 2024
• DOI: http://doi.org/10.1177/00938548231208192
• Author: Mark G. Leymon,Christopher M. Campbell,Kris Henning
• Date: 01 January 2024

CRIMINAL JUSTICE AND BEHAVIOR, 2024, Vol. 51, No. 1, January 2024, 43 –65.

DOI: https://doi.org/10.1177/00938548231208192

Article reuse guidelines: sagepub.com/journals-permissions

© 2023 International Association for Correctional and Forensic Psychology

43

OREGON’S TRANSITIONAL LEAVE PROGRAM

AND RECIDIVISM

MARK G. LEYMON

CHRISTOPHER M. CAMPBELL

KRIS HENNING

Portland State University

The State of Oregon has operated an early prison release program called Short-Term Transitional Leave (STTL) since 1989.

The program was designed to improve prison releasee reentry planning and reintegration in the community. In the last 10

years, Oregon expanded STTL several times, with individuals now being released up to 120 days early. We assessed whether

differences in recidivism existed between STTL and those completing their planned sentence and if the length of their leave

is associated with differential recidivism. We used propensity-score matching (PSM) to create quasi-experimental models.

The results indicate no appreciable observed associations between STTL and rearrest, reconviction, or reincarceration.

However, there were higher rates of technical violations among STTL participants, especially for those with the longest

release time.

Keywords: recidivism; sentencing; propensity-score matching; prison reentry; transitional leave

Shifting public sentiment and the fiscal challenges of maintaining the world’s highest

incarceration rate are leading many U.S. states to explore new strategies for decarcerat-

ing. To be successful, these efforts must reduce the number of people being sent to prison

and/or limit how long they stay there (Clear & Austin, 2009). Strategies addressing the

former include prison diversion programs, revisions to mandatory sentencing laws, decrim-

inalization or defelonization of drug offenses, and alternative responses to technical revoca-

tions among probationers. Clear (2021) argues that these “front-end” initiatives can be

effective in the short term. Still, they yield diminishing returns over time as most people

diverted from prison have relatively short sentences. Those with longer sentences, includ-

ing people convicted of serious violent crimes, are rarely eligible for these initiatives.

Hence, the potential value of concurrent efforts targeting the second factor: sentence length.

Average and median sentences in the United States are longer than those found in most

European nations (Kazemian, 2022), and there is limited evidence that we benefit from

AUTHORS’ NOTE: The Oregon Criminal Justice Commission funded this project under Grant No. PSU-20-

3. Correspondence concerning this article should be addressed to Mark G. Leymon, Department of Criminology

and Criminal Justice, Portland State University, Portland, OR 97207-0751; e-mail: mleymon@pdx.edu.

1208192CJBXXX10.1177/00938548231208192Criminal Justice and BehaviorLeymon et al. / Short-Term Transitional Leave and Recidivism

research-article2023

44 CRIMINAL JUSTICE AND BEHAVIOR

lower crime rates as a result (Pfaff, 2017). Nor have studies demonstrated a clear advantage

of longer sentences when adults in custody (AICs) are eventually released to the community

(Leymon et al., 2022; Loughran et al., 2009; Rhodes et al., 2018; Snodgrass et al., 2011;

Weisburd et al., 2017). This has led some jurisdictions to investigate “back-end” policies for

granting individuals early release from prison (Clear, 2021). The current study evaluates

one such policy in the state of Oregon.

As part of the state’s Justice Reinvestment Initiative in 2013 (JRI, HB 3194), legislators

significantly expanded Short-term Transitional Leave (STTL). This program allows eligible

AICs to leave prison up to 90 days early to establish housing, employment, and social con-

nections in the community. It is estimated that the STTL program saved Oregon 544,167

prison bed days from January 2014 through August 2019 (Criminal Justice Commission,

2020), amounting to a potential savings of more than 50 million dollars. The central ques-

tion for the present research is whether these benefits come at a cost in the form of increased

recidivism measured through rearrest, reconviction, reincarceration, and technical viola-

tions within 3 years of prison release. It should be noted that the Oregon Legislature

increased the maximum STTL to 120 days in August 2017 via HB 3078. The AICs in the

current study were ineligible for this extension, so the current focus is on the policy estab-

lished in 2013.

Allowing individuals to leave prison early could reduce the deterrent value of prison,

leading to higher rates of rearrest, reconviction, and reincarceration. Alternatively, extended

periods of incarceration may have a harmful effect via severed family relations, disrupted

housing, reduced employment opportunities, exposure to traumatic events, and increased

association with deviant peers. Releasing individuals early might mitigate these harms and

lead to lower recidivism rates. Finally, relatively small reductions in prison stays may have

no bearing on recidivism. This study investigated these potential outcomes by comparing a

large sample of persons granted STTL with a matched sample that did not participate in the

program.

EARLY RELEASE PROGRAMS

Compared with research on many other topics in corrections, we know relatively little

about the impact of leaving prison early. The so-called “good-time” credits have been used

for decades to incentivize positive behavior in prison, but rigorous research on these poli-

cies is lacking. A small number of studies have looked at the effect of leave in the context

of establishing employment (i.e., work release) or reconnecting with family (i.e., home

release). Cheliotis (2008) systematically reviewed the early literature in this field and con-

cluded that these programs lead to small reductions in recidivism. Several methodological

limitations qualify this assertion. First, most of these studies failed to account for other fac-

tors connected to program participation that might impact outcomes. In other words, indi-

viduals granted work or home release might have been at lower risk from the outset. Second,

the available studies vary greatly with regard to the length of leave granted and the follow-

up period for assessing recidivism. Third, some studies combined transitional leave with

additional programming requirements (e.g., vocational training and drug treatment), raising

questions about the independent effect of early release. Fourth, a few studies included tem-

porary leave programs, where releasees must return to custody each night or on weekends.

Finally, many of the studies reviewed (see Joo et al., 1995; Sims & O’Connell, 1985; K. A.