February 2018 OPM - Criminal Justice Policy & Planning Division CT recidivism rates, 2014 cohort Highlights During 2014, 11,245 men and women left prison facilities in the state having completed, or partially completed, a prison sentence. Over the next 3 years: 60% were re-arrested, 53% were returned to prison, for at least one single day, 45% were convicted for a new offense, and 34% returned to prison to begin a new term of incarceration. Offenders could be returned to prison for a variety of reasons including new criminal charges or technical violations of their community supervision. Overall, women who were released from prison in 2014 recidivated at moderately lower rates than men, regardless of the type of recidivism (see page 2). Overall recidivism rates – 2014 release cohort The Criminal Justice Policy & Planning Division (CJPPD) at CT OPM is required to report on recidivism among Connecticut prisoners on an annual basis. OPM calculates four separate measures of recidivism for offenders leaving prison. They are: 1) new arrests 2) new convictions 3) returns-to-prison for any reason, and 4) returns-to-prison to begin a new sentence of incarceration. OPM typically produces analysis based on the 3-year period following an offender’s release from prison. In earlier studies, CJPPD has reported on recidivism rates among prisoners released in 2004, 2005, 2008 and 2011. This report tracks offenders who left prison in 2014. 3-year recidivsm rates for offenders leaving prison in 2014 70% New arrest rate 60% Return to prison rate 60% New Conviction Rate 50% 53% New sentence rate 40% 45% 30% 34% 20% 10% 0% General recidivism rates have been dropping at a moderate pace over the last decade. Between 2011 and 2014, recidivism rates declined for every category of recidivism with the exception of return-to-prison rates for men. When segmented by offender age and prior sentence history, the 3year return to prison rate varied between 15% and over-80% for different sub-groups in the 2014release cohort. CT OPM – February 2016 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Many offenders returned to prison more than once during the 36month study window. Among the 6,026 offenders who were returned to prison within three years, less than a third were admitted only once (see page 3). Months since 2014 release 6 months 12 months 24 months 36 months New arrest rate 19% 34% 51% 60% New conviction Return-to-prison New sentence rate rate rate 6% 21% 4% 16% 33% 11% 35% 47% 25% 45% 53% 34% For its analysis, OPM relies on data provided by the Court Support Services Division (CSSD) in the Judicial Branch and the CT DOC MIS Unit. OPM has followed the same research methodology for almost a decade. The recidivism rates shown here aggregate the first recidivism event, over time, for each recidivating offender in the cohort. Because the overall recidivism rate changes so incrementally, OPM has chosen to evaluate prisoner-cohorts spaced at three year intervals. The data used in this report ran through December 31, 2017. Over the next two years, OPM anticipates that it will produce two further analyses of recidivism among the 2014 cohort. In the past, OPM’s off-year analysis has focused on recidivism among specific subgroups of offenders that present special challenges to the prison system. Page 1 of 4 Recidivism and CT prisons How is recidivism measured? Connecticut is one of six states in the nation with a unified prison system. In unified systems, both prisons and jails are operated by the state, and the incarcerated population is composed of both sentenced prisoners and pre-trial detainees. In most states, county-level government operates the jails. The recidivism rates reported here are restricted to sentenced prisoners leaving prison in 2014. CT’s prison population has been declining for a decade and the sentenced component of that population has been shrinking at an even faster rate. The number of sentenced prisoners currently incarcerated, is at its lowest level since early-1993. With fewer prisoners, there are fewer sentenced offenders leaving prison each year. As a result, the size of the prisoner-cohorts OPM uses in its analysis has been declining. The 2014release cohort includes 11,245 men and women, 10.6% less that the 2011-release cohort that included 12,579 male and female offenders. CT's Sentenced prison population 16,000 14,998 14,745 14,000 13,876 13,578 12,955 12,000 12,104 12,417 12,063 11,601 10,748 10,000 9,871 8,000 6,000 4,000 2,000 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 In addition to contracting, the composition of the sentenced population has also changed significantly. Legislation that raised the age at which youths could be treated as adults in the criminal justice system dramatically reduced the number of young offenders in state prisons. While the overall size of the 2014 cohort was 10.6% smaller than the 2011-prisoner cohort, the number of offenders under the age of 25 declined by 31%. The number of offenders under age-18 dropped by an astounding 74% between 2011 and 2014. Despite this, the number of sentenced women declined by only a meager 3.5%. Changes in the 2011 and 2014 cohorts All All - men All - women Under 25 Under 18 2011 cohort 12,579 11,235 1,344 2,712 236 CT OPM – February 2016 2014 chort 11,245 9,948 1,297 1,878 62 Change -10.6% -11.5% -3.5% -30.8% -73.7% Since younger offenders tend to recidivate at higher rates than their older counterparts, fewer young offenders might be expected to translate into lower overall recidivism rates. Recidivism and gender Regardless of the measure, women offenders recidivated at lower rates than their male counterparts. While the reasons are not entirely clear, it is well acknowledged that the circumstances and pathways into prison for most women is considerably different than it is for most male prisoners. Recidivsm rates and gender, 2014-cohort Cohort size New arrest rate 6 month 12 month 24 month 36 month New conviction rate 6 month 12 month 24 month 36 month Return-prison rate 6 month 12 month 24 month 36 month New sentence rate 6 month 12 month 24 month 36 month 2014 cohort, all 11,245 2014 female cohort 1,297 2014 male cohort 9,948 19% 34% 51% 60% 20% 32% 48% 56% 19% 34% 51% 60% 6% 16% 35% 45% 5% 14% 31% 40% 6% 16% 35% 46% 21% 33% 47% 53% 17% 26% 38% 43% 22% 34% 48% 55% 4% 11% 25% 34% 3% 8% 18% 24% 4% 12% 26% 35% On average, female prisoners in the cohort were admitted to prison at an older age than their male counterparts. For males, more than 50% of the group had been admitted to the CT DOC by the age of 21. In contrast, only 26% of women had been admitted by the age of 21. By the age of 30, 82% of the men had already been admitted to a DOC facility at least once: for women the figure was 63%. On average, male offenders first entered the DOC at the age of 24. Among female offenders the average age of first DOC entry was 29. Despite all this, the mean age of women in the 2014-cohort (35.1 years), only slightly older than the mean age for male members of the cohort. (34.5 years). Page 2 of 4 In addition to entering prison at an earlier age, the average male offender had a more extensive sentence history with the DOC. Among women, over half (51%) were completing their first or second prison sentence at the time of their 2014 release. Among men, slightly more than a third (34%) were completing their first or second prison sentence in the state. The average female offender in the cohort had served 4 prison sentences in CT at the time of their 2014 release. Men in the cohort averaged prior 6 sentences. The 2014 women’s-cohort, generally, had lower sentence rates for violent offenses compared to males. While 16% of the men had been sentenced to prison for a felony assault, only 11% of women had served a sentence for the same offenses. Similarly, 16% of men had served time for robbery compared to only 10% of women. Women exhibited higher sentence rates, than men in only two crime categories: theft/larceny and DUI offenses. Prior sentences served by 2014 cohort members Crime type Court-related Drug-related Theft/larceny DUI-related Burglary Felony Assault Robbery Weapon Threat/stalk/harass Sex crime Risk of injury Kipnap-related Homicide Murder 2014 male cohort 6,529 4,444 3,435 2,184 2,048 1,608 1,563 1,592 1,256 774 557 233 112 58 % of males 66% 45% 35% 22% 21% 16% 16% 16% 13% 8% 6% 2% 1% 1% 2014 female cohort 776 337 528 334 112 142 121 30 43 16 68 3 13 2 % of females 62% 27% 42% 27% 9% 11% 10% 2% 3% 1% 5% 0% 1% 0% Court-related: VOPs, FTAs Re Multiple returns to prison history of incarceration The recidivism data shown on page 1 aggregates data for each offender based on the first recidivism event within the three-year study window. What the data does not reveal is that many offenders returned to prison more than once during the 3-years following their 2014 release. In the following table, we observe that a significant proportion of offenders were returned to prison returned multiple times. CT OPM – February 2016 Of the 6,026 offenders who were returned to prison, within 3 years, only 32% (1,916) were readmitted just once. In fact, a larger percentage of offenders (38%) were returned to prison at least three times in the 3years following their releases. Offenders w/multiple returns Number of returns to prison within 3 years 1 2 3 4 5 6 7 8 9 10+ Total % returning once 2014 cohort, all 1916 1834 1090 641 279 125 59 30 21 31 6026 32% 2014 female cohort 225 129 109 57 29 8 4 2 1 3 567 40% 2014 male cohort 1691 1705 981 584 250 117 55 28 20 28 5459 31% The 6,026 offenders who returned to prison were admitted to the DOC on 14,784 separate occasions during the following three years. Approximately 400 offenders were readmitted ten or more times. Not surprisingly, these offenders exhibited high rates of chronic substance abuse and/or mental health issues. Recidivism: 2011 and 2014 Connecticut has been fortunate in recent years with respect to trends in the criminal justice system. According to FBI data, crime rates, for both property and violent crime, have declined dramatically. As a consequence, many fewer people are being arrested each year and fewer are being convicted and sentenced to prison. With raise-the-age legislation, CT essentially eliminated youth under the age of 18 from the prison system without a significant jump in reported crime rates. Overall recidivism rates have been declining for a decade. 3-year recidivism rates. 2005 through 2014 2005 2008 2011 2014 New arrest 66% 64% 63% 60% New conviction 54% 51% 50% 45% Return to prison 56% 54% 54% 53% New sentence 37% 37% 35% 34% In the 2014-cohort, almost two-thirds of the offenders who returned to prison did so as pre-trial detainees, on new charges. A quarter of the offenders who returned, were re-admitted to prison as remandees, having violated the technical conditions of their community supervision. Page 3 of 4 In the following table, the total 2014-release cohort has been segmented into 25 subgroups based on offender ages and their sentence histories. The columns show the age quintiles of offenders by their ages in 2014. The rows segment the offenders by the number of prison sentences the offenders had served with the DOC prior to their 2014 release. First returns to prison Offenders, 1st return Pre-trial admit 3821 Technical violation 1473 Criminal violation 458 Return absconder/escapee 196 Without prejudice 40 Return from other juris. 16 Offenders who returned 6004 % of returns 64% 25% 8% 3% 1% 0% 100% In recent years, the remand rate for parolees has attracted some scrutiny in the state. Since 2017, the CT DOC and the Board of Pardons and Paroles (BOPP) have undertaken a major reassessment of remand policies and practice. OPM will be tracking the impact of these reforms. 3-year recidivism rates, 2011 and 2014 Cohort size New arrest 6 month 12 month 24 month 36 month New conviction 6 month 12 month 24 month 36 month Return-to-prison 6 month 12 month 24 month 36 month New sentence 6 month 12 month 24 month 36 month Women 2011 2014 1,344 1,297 Men 2011 2014 11,235 9,948 20% 36% 54% 61% 20% 32% 48% 56% 21% 39% 55% 63% 19% 34% 51% 60% 9% 21% 39% 49% 5% 14% 31% 40% 8% 23% 44% 55% 6% 16% 35% 46% 18% 26% 37% 44% 17% 26% 38% 43% 21% 34% 48% 55% 22% 34% 48% 55% 5% 10% 20% 28% 3% 8% 18% 24% 4% 13% 27% 37% 4% 12% 26% 35% The top section of the table shows the percentage at which each subgroup that was returned to prison within three years. The lower table, lists the number of offenders in each subgroup. From top section we observe that the 3-year, return– to-prison rate varies widely, between 15% -for 47 to 92 year-olds who were completing their first prison sentence - to over 80% for offenders under the age of 30 who had completed 13 or more sentecnes with the DOC. Let us recall, the return-to-prison recidivism rate for the entire cohort was 53%. From the table we can observe two truisms with respect to recidivism 1) recidivism rates decline with age if we hold the sentence history constant, and 2) recidivism rates worsen across all age groups as an offender’s sentence history increases. Return-to-prison, age and sentence history Age quintiles - 2014 cohort A r A range of recidivism rates incarceration In any discussion of recidivism it is important for readers to maintain sight of the fact that an entire prisoner-release cohort, like the one observed here, is a non-homogenous population. Although entire cohorts of prisoners may behave similarly over time, and exhibit comparable outcomes, there are subgroups within the population - some quite large - that behave much better, or much worse than the average with respect to recidivism. CT OPM – February 2016 Prior sentences 15 to 25 26 to 30 31 to 37 38 to 46 Return-to-prison recidivism rate First 52% 32% 31% 26% 2nd or 3rd 64% 50% 47% 35% 4th to 6th 71% 67% 59% 43% 7nd to 12th 82% 77% 70% 59% 13 or more 89% 83% 76% 75% 47 to 93 Prior sentences 15 to 25 26 to 30 31 to 37 38 to 46 Offender distribution First 806 469 407 357 2nd or 3rd 859 590 527 418 4th to 6th 462 589 594 410 7nd to 12th 157 381 665 618 13 or more 9 36 193 456 47 to 93 15% 29% 41% 53% 65% 428 409 352 470 579 This report was produced by the Research Unit in the Criminal Justice Policy & Planning Division at the Connecticut Office of Policy and Management. 450 Capitol Avenue Hartford CT, 06106 Mike Lawlor, Under-Secretary Ivan Kuzyk – Principal author Kyle Baudoin – Planner Kendall Bobula - Research This report is available on-line at the OPM-CJPPD website. All questions, comments and suggestions should be directed to the Research Unit at CJPPD. Page 4 of 4