BJS report examines 10-year recidivism rates of state prisoners

As stated in a recent Bill Otis’s post on Wednesday, the Bureau of Justice Statistics (BJS) recently released a report that examines 10-year recidivism patterns of prisoners released in 2008 from 24 states. The study has a lengthy follow-up period of ten years, which is useful in tracking recidivism rates over time. A similar report was released on the topic in July 2021 that included data from 34 states, but the follow-up period was only five years (2012-2017). Similarly, BJS also released a report in May 2018 that examined recidivism rates of offenders from 30 states with a nine-year follow-up period. Extending the follow-up period allows researchers to examine recidivism patterns over longer periods of time, which is one of the main benefits of the current study.

The most recently released recidivism study per BJS sampled 73,600 prisoners, tracking their recidivism patterns for ten years (from 2008 through 2018). Similar to reports in the past, source data are from prisoner records reported by state departments of corrections to BJS’s National Corrections Reporting Program and criminal-history records from the FBI’s Interstate Identification Index and state repositories via the International Justice and Public Safety Network. Criminal history records were combined with arrest data and prison admissions/release data to provide information about offending patterns and recidivism.

Among the sample of offenders, 81.9% were arrested at least once during the ten years following release, including arrests for new crimes as well as arrests for violations of parole. When violations of parole were excluded from analysis, the re-arrest rate was not much lower, at 80.5%. Age also appeared to impact recidivism, as those who were 24 or younger at the time of release were more likely to be re-arrested during the ten-year period (90%) compared with those who were age 40 and older at the time of release (75%). These findings are generally similar to the aggregate findings from 2014 data showing nine-year re-arrest rates of 83%, with those younger than 24 having higher re-arrest rates (90%) than those 40 and older (43%).

Recidivism rates are slightly higher than those presented in a previous BJS report released in July 2021, which showed that 71% of prisoners were re-arrested within five years following their release. The July 2021 report examined 34 states (10 more states than the current study), but only used a five-year follow-up period, examining data from 2012-2017. The September 2021 BJS report examined here reveals the most recent estimates of recidivism (2008-2018), with a lengthier follow-up period of ten years. It is hard to directly compare reports to each other because there are differences in the states represented, however the differences in recidivism rates across the reports clearly show why it is useful to use lengthy follow-up periods when examining recidivism.

Per the most recent estimates, prisoners with fewer prior arrests were less likely to be arrested during the ten-year period following their release. The average number of prior arrests among released prisoners was nine (for any type of offense). Those with four or fewer prior offenses re-offended at a rate of 67.2%, those with five to nine offenses re-offended at a rate of 80.9%, and those with ten or more offenses re-offended at a rate of 89.0%. Age at time of release also appeared to have an impact on re-offending rates. At the end of the ten-year period the cumulative arrest percentage was 90% for those who were under age 18 at the time of release, compared with 41% of those who were 40 or older at the time of release.

Regarding violent offenses, 40% of the sampled offenders were re-arrested for a violent crime during the ten-year follow-up period. Assaults accounted for 31% of violent re-offenses, while robberies accounted for 7%, and homicides accounted for 1%. Among those who were initially imprisoned for a violent offense, 44% were re-arrested for a violent offense within ten years, compared with 40% of those initially arrested for a property offense, and 35% of those who were initially imprisoned for a drug offense.

22 of the 24 states provided information on re-arrests leading to a conviction that occurred within the follow-up period. Overall, 69% had an arrest that led to a conviction during the ten-year follow-up period. This rate was higher for younger prisoners, with 79% of prisoners age 24 or younger receiving arrests leading to a conviction during the ten-year period, compared to 59% of those age 40 or older.

Prisoners who had served less than the 15-month median were more likely to be re-arrested during the follow-up period (81%) than those who served more than the median time frame (76%). Among those serving 61 months or more, 65% were re-arrested during the ten-year follow-up period. Among violent offenders, released prisoners whose time served was less than the 29-month median were more likely to be re-arrested within ten years (78%) when compared with those serving sentences longer than the median (66%). Similarly, drug offenders who served less than the 13-month median were slightly more likely to be re-arrested (79%), compared with 75% of those who served more than the median.

While people may be most likely to recidivate among the first few years following release, it appears that this is not the case for everyone. When extending the follow-up period to ten years, estimates for aggregate recidivism rates tend to increase. However, because these findings are usually reported in the aggregate, differences at the state-level remain unclear. Future research should attempt to look at records from individual states and compare them to each other to gain a better understanding of how recidivism varies across different states and how it fluctuates over time.