Recidivism and survival time: racial disparity among jail ex-inmates.

Author:Jung, Hyunzee

Incarcerated men, most of whom are recidivists, are disproportionately black. Much literature about prison ex-inmates reports on this disparity, yet little is known about racial disparity in recidivism rates among jail ex-inmates. This study examined recidivism rates and survival time (period from release date to rearrest) among male ex-inmates released from the Allegheny County Jail in Pennsylvania during 2003 who were tracked for three years (N = 12,545). Recidivism rates were compared across race using chi-square tests. Survival time was analyzed using survival analysis including Kaplan-Meier estimation with log-rank test and Cox's proportional hazard model. Findings showed that the overall three-year recidivism rate was 55.9%, and black men recidivated at a significantly higher rate than white men. Survival analysis also attested to racial disparity in recidivism, and black men recidivated in a shorter time frame than their white peers with the covariates age at release and length of stay in jail controlled. Interaction effects of race and the covariates were found to be significant. Research, policy, and practice implications are presented.

KEY WORDS: incarceration; jail; racial disparity; recidivism; survival time


The increasing incarceration and recidivism rates in the United States marked by pronounced racial disparity are recognized economic and social burdens that need to be reversed (Pew Center on the States, 2009).Twenty-five percent of the world's inmates are in U.S. prisons and jails--despite the United States comprising only 5% of the total world population (International Center for Prison Studies, 2008). Most inmates are recidivists, who are disproportionately black men (Sabol, Minton, & Harrison, 2007). The U.S. aggregated corrections costs rose from $12 billion in 1987 to $49 billion in 2007--$20 billion of which was spent on jails and local criminal justice (Hughes, 2006).

Annually, an estimated 12 million people cycle in and out of nearly 3,500 U.S. jails (Beck, 2006). In 2008, 785,556 individuals were in jail at midyear, up from an average daily population of 403,019 in 1990 (Jail Inmates in 1994, 1994; Minton & Sabol, 2009). Of the 2008 jail population, 42.5% were white men and 39.2% were black men, even though of the total U.S. population only about 6% were black men and 28% were white men (McKinnon & Bennett, 2005). More than 97% of prison and jail inmates are eventually released and need to reintegrate into their communities (Sabol & Couture, 2008; Sabol &Minton, 2008). Despite the substantial cost and numbers of people affected by the jail system and its unique role in criminal justice, the preponderance of literature about recidivism and racial disparity focuses on former prison inmates (Yamatani, 2008). Scant information is known about these factors among jail ex-inmates, which this study begins to address.

Because of the dearth of literature about jail recidivism, we had to draw predominantly from studies of former prison inmates to investigate recidivism and racial disparity. It is important to explain that jails and their inmates differ substantially from prisons, so prison literature findings cannot be generalized to jails (Center for Therapeutic Justice, 2000).Jails are short-term incarceration facilities that primarily hold offenders of minor crimes, who typically stay for less than one year. Jails are generally larger, more structured, and closer to home, with a rapid turnover of inmates, many of whom are never convicted. Prisons are more violent and chaotic, and draw from large areas. Many jail inmates cycle in and out, escalating their criminal activity until they are committed to prison. Effective jail and post-jail interventions can deter people from ending up in prison (Roman & Chalfin, 2006).

Prison-focused studies have reported that two-thirds of prison ex-inmates were rearrested (a measure of recidivism) within three years post-release, with the highest rates occurring during the first year (Langan & Levin, 2002; Mears, Wang, Hay, & Bales, 2008). A 15-state prison recidivism study showed that within the first year, 44.1% of ex-inmates were rearrested for a new offense, with the number rising to 59% by the second year and 67.5% cumulatively by year 3. One jail study showed a one-year recidivism rate of 48% (Lyman & LoBuglio, 2006).

Gendreau, Little, and Goggin (1996) concluded in their meta-analysis that gender, age, criminal history, and race were key predictive individual-level risk domains for recidivism. Langan and Levin (2002) found that, within three years of release, younger age and having a longer criminal history were both associated with a greater likelihood of rearrest. Race also is a major factor. Black ex-inmates have significantly higher recidivism rates than their white peers (Bonczar, 2003; Kansal, 2005). Langan and Levin reported that 72.9% of black ex-inmates were rearrested compared with 62.7% of their white counterparts within three years postrelease. Some prison studies found that the effects of race remained after controlling for individual-level risk factors for recidivism, including prior criminal record, length of prison time, and type of offense (for example, Mears et al., 2008).

This study examined recidivism in reference to racial disparity among jail ex-inmates and was informed by prison literature showing associations among factors, including race and recidivism. Our goal was to build on this empirical literature by investigating these associations with jail ex-inmates, because characteristics of jail differ from those of prison systems and populations. The proposed two hypotheses are as follows: (1) A higher percentage of black men will recidivate within one, two, and three years after release compared with white men. (2) Black men will recidivate in a shorter time frame than white men after holding constant age at release and length of jail stay.

In post hoc analyses, we explored the interaction effects between black and white male ex-inmates and two covariates--age at release and length of jail stay. Prison literature indicates that these individual-level factors tend to affect recidivism rates differently by race (Mears et al., 2008).The question of whether race interacts with these factors is minimally addressed in prison studies and unavailable in jail studies--further adding to the value of this study to the literature.

The magnitude of the jail population and its racial disparity speaks to the need for new criminal justice system and practice interventions (Pew Center on the States, 2009). Such efforts require greater knowledge of the correlates of recidivism. As social work researchers, policy advocates, and practitioners, we posit that it is our ethical obligation to explore and address this apparent injustice. This study contributes to a further understanding of these issues by...

To continue reading