Report

Recidivism Patterns Among Those Released from Prison in Illinois

    By
  • Name
    Dave Olson ·
On

REPORT-SENTENCES-PRISON-RECIDIVISM

Key takeaways

  • The majority (75%) of those exiting prison in Illinois during 2018 and 2019 were not arrested for a violent offense within 3 years of their release from prison.
  • Roughly one in eight (12%) of those exiting prison were arrested for a domestic violence offense within 3 years of their release.
  • Less than 1% of those exiting prison were arrested for murder within 3 years of release.
  • Being younger, male and having prior arrests for violent crime were the strongest predictors of a post-prison arrest for a violent crime, but most in these higher risk groups were still not arrested for a violent crime.
  • Those released from prison in Illinois during the onset of the COVID-19 mitigation strategies were less likely to be arrested within 1 year for a non-violent crime than those released prior to COVID-19 (16% vs. 24%, respectively); the likelihood of arrest for a violent crime was only slightly lower among the COVID-19 cohort when compared to the pre-COVID group (9% vs. 11%, respectively).
  • The shortened lengths of Mandatory Supervised Release (MSR) resulting from Illinois’ Safety, Accountability, Fairness and Equity-Today (SAFE-T) Act will reduce the proportion of individuals at risk of being returned to prison for a technical violation related to a new arrest for a violent crime.

Introduction

Successful reentry following release from prison depends on many things and can be measured in many ways. Finding safe and stable housing and meaningful employment, getting needed medical and mental health services, and rebuilding positive connections to family and community are all crucial to post-release success, and can be viewed as important reentry outcomes on their own. But the most important reentry outcome from society’s point of view is the successful avoidance of renewed involvement in criminal activity. It is this last outcome—recidivism, and especially violent recidivism—that is of most concern to policymakers, and is most often the focus of reentry research.

There is no one way to define and measure recidivism. For example, the Illinois Department of Corrections uses a measure of recidivism that focuses solely on return to prison: whether someone comes back to prison within three years of release, either for a violation of the conditions of their mandatory supervised release (MSR) or because of a conviction for a new felony offense. Other recidivism measures may focus on all new arrests, on new arrests for specific types of crimes, or on new convictions.

This research brief summarizes patterns of arrest among those released from prison between 2018 and 2022, exploring the nature of the crimes for which people released from prison are subsequently arrested, the timing of the arrests, whether specific characteristics of those released are associated with the likelihood of being arrested, and the degree to which rates of post-release arrest may have changed during the COVID-19 period. Arrest was chosen as the measure of recidivism for two basic reasons. First, given how long it takes for serious criminal cases to matriculate through the court system, using a conviction-based measure of recidivism would have missed many serious crimes unless the follow-up period were exceptionally long. And second, while not all arrests are consequential or result in convictions, arrests following release from prison may result in return to prison anyway, since individuals on MSR can be returned to prison if they are arrested, and must be returned as technical violators if arrested for specific types of offenses (e.g., domestic violence offenses).

Rates and Patterns of Recidivism

Among those released from prison in Illinois during 2018 and 2019, 45% were not arrested for anything within three years of their exit from prison, and an additional 30% were arrested but not for an offense that involved violence (Figure 1). Thus, 75% of those released from prison in Illinois were not arrested for an offense that involved violence within three years of their release. Of the 25% who did have a new arrest for a violent crime, 13% were arrested for a violent crime that did not include a charge of domestic violence and 12% were arrested for a violent crime that did include a domestic violence charge (i.e., domestic battery, violating an order of protection). Less than 1% of those released from prison were charged with first degree murder (not shown in Figure 1 due to scale).



There are several individual characteristics correlated with violent crime arrests following release from prison, although the correlations are not strong. Those who were relatively young, male, with numerous prior arrests for violent crimes were more likely to be arrested for a violent crime after release from prison (Figure 2), but most people falling into these categories were not arrested for a violent crime following release. For example, while those that were between 18 and 25 years old at release had the highest arrest rate for a violent offense (36%), most (64%) were not arrested for a violent crime. Similarly, 31% of those with 3 or more prior arrests for a violent crime were arrested after release for a violent crime, but 69% were not. Obviously, preventing post-prison involvement in violent crime is an important goal, but predicting this outcome is difficult given the relative rarity of post-release arrests for violent crimes. For example, those released from prison that were male, under 26 and had 3 or more prior arrests for a violent crime accounted for only 6% of all those released from prison and 47% of that group was arrested for a violent crime within 3 years of release. Previous research by Loyola’s Center for Criminal Justice documented the dramatic increase in the number of people in Illinois sentenced to, and released from, prison for offenses involving the illegal possession of a firearm (see https://loyolaccj.org/blog/sentences-to-prison-in-illinois-for-illegal-possession-of-a-firearm). Whether someone had been sentenced to prison for a firearm possession offense, or had firearm possession arrests in their criminal background, did not increase the likelihood they would be arrested post-prison for a violent crime. In fact, after accounting for the age, race, sex, and the extent and nature of prior criminal history, those released from prison after serving a sentence for a firearm possession offense were slightly less likely to be arrested within 3 years for a violent crime than those sentenced to prison for other offenses.



COVID-19’s Impact on Recidivism

The COVID-19 pandemic resulted in substantial changes to policing as well as individual behavior. What impact did these changes have on arrests following release from prison in Illinois? To answer this question, we compared 1-year post-prison arrest rates for two groups of people: those who left prison and lived in the community before COVID mitigation strategies began (released between April 1, 2018 and March 31, 2019) and those released after COVID-19 mitigation strategies began (released from April 1, 2020 through March 31, 2021). As seen in Figure 3, those in the latter group had a much lower rate of arrest for all offenses combined than those released during the pre-COVID period. This likely reflects pandemic-related changes in policing practices (e.g., fewer traffic stops), along with a pandemic-related drop in specific types of criminal activity (e.g., fewer burglary offenses due to potential victims staying at home). While one-year post-prison arrest rates for non-violent crimes dropped from 24% to 16% between these two periods, arrest rates for violent crimes declined less dramatically. Specifically, 11% of those released from prison before the pandemic were arrested for a violent crime within 1 year of their release, compared to 9% among the cohort released when the COVID-19 mitigation strategies were implemented. Because most of those returned to prison are returned because of new arrests or convictions, and arrest rates fell during COVID-19, so too did the 1-year rate of return to prison. The magnitude of decrease in these recidivism rates were similar when those released to urban versus rural counties in Illinois were compared during the pre-COVID and COVID-19 mitigation periods.



Timing of Recidivism and Implications for Mandatory Supervised Release (MSR)

It is also important to consider the timing of arrests following release from prison, since being arrested while on MSR constitutes a violation of supervision, which can result in a return to prison. Further, being arrested for specific types of crimes (e.g., domestic violence offenses) while on MSR constitutes a technical violation that requires return to prison. Research conducted by Loyola’s Center for Criminal Justice found that a substantial portion of those returned to prison for technical violations of MSR were in fact returned for arrests for violent crimes.

Periods of MSR supervision were recently shortened for many crimes as part of the Safety, Accountability, Fairness and Equity-Today (SAFE-T) Act. For all felony classes except first degree murder, offenses involving domestic battery or violations of orders of protection, and sex offenses, the MSR periods were cut in half. Thus, the length of MSR for Class X offenses (excluding sex offenses) went from 3 to 1.5 years, for Class 1 to 2 felonies (excluding sex and domestic violence offenses) MSR length went from 2 to 1 year, and for Class 3 to 4 felonies (again, excluding domestic violence and sex offenses) MSR length went from 1 year to 6 months. These shortened lengths of MSR became effect January 1, 2022.

What impact is this policy likely to have on prison readmissions for technical violations? Figure 4 examines the timing of post-release violent crime arrests among those exiting IDOC in 2018 and 2019, after serving prison terms for various classes of felonies. While these individuals were released prior to the new policy, this retrospective analysis gives a sense of the potential impact of the change in MSR lengths, by showing (1) what percentage were arrested for violent crimes within an average of 4 years, and (2) how the timing of those arrests relates to the old and new MSR lengths. For example, among those released after serving a sentence for a Class 3 or 4 felony (excluding the domestic violence and sex offenses not subject to the change in MSR length), 70% had no violent crime arrests at all during the period examined, and 18% were not arrested for a violent offense until more than a year after release (i.e., beyond the MSR length in place prior to the SAFE-T Act changes). Thus, 88% had no violent crime arrests while on MSR, and the change to MSR lengths would not have had any impact on their risk of being returned to prison as technical violators. Conversely, 6% were arrested within 6 months of their release—which meant that they would have been under MSR under both the old and the new policies, so that again, the change would not have affected their risk of being returned to prison as technical MSR violators. But another 6% were arrested for a violent crime in the period between 7 and 12 months of release, and so were still on MSR under the old policy, but would not have been if the new policy had been in effect. This group was at risk of return to prison under the old policy, but would not have been under the new.

Similar patterns were found among the other felony classes, with most of those released from prison not being subsequently arrested for a violent crime, or not being arrested within the MSR period under either the old or new policies. But the change in MSR lengths would have affected 8% of those released after serving prison sentences for Class X felonies, and 9% of those released after serving prison sentences for Class 1 or 2 felonies: if these groups had been serving the shortened MSR periods now in effect, they would not have been subject to return to prison as technical MSR violators, and could only have been imprisoned following conviction for the offenses for which they were arrested (if it were a felony-level violent crime).



Methodology

Case-level information was obtained from the Illinois Department of Corrections (IDOC) and the Illinois State Police/Illinois Criminal Justice Information Authority for all individuals released from prison in Illinois between January 2018 and June 2021, including demographic characteristics, the nature of the crimes for which they were sentenced to prison, length of time incarcerated, and criminal history record information (CHRI). Cases included in the recidivism analyses were limited to those in which individuals were released from prison after having completed their original prison sentence. Exits following a return to prison for a technical violation of mandatory supervised release, or MSR, were excluded. Also excluded were those who were “released” but immediately returned to prison (i.e., “gate violators” who could not be released to MSR because they had no approved place to live). Finally, when an individual had more than 1 exit from prison that met the selection criteria during the period examined, only the first exit was included in the analyses. The detailed CHRI data allowed for the identification of any post-release arrest and the date of the arrest. Arrests were divided into three groups: arrests in which none of the arrest charges included a violent crime (i.e., arrest for non-violent offenses only), arrests for violent offenses other than domestic violence (i.e., arrests for any violent crime other than a domestic violence offense), and arrests for domestic violence offenses (i.e., arrests on charges that included domestic battery or violation of an order of protection). Any individual arrested for a domestic violence offense was placed into that category, regardless of whether there were other violent or non-violent arrest charges as well. Those placed in the group arrested for non-violent offenses had no arrest charges for any violent crime during the post-prison follow-up period.