Risk Assessments in Cook County Ineffective
by Jayson Hawkins
Risk assessments have been championed as a tool to help remove bias from criminal justice decisions. While there have been improvements in some areas, overall performance has fallen short of many expectations. Cook County, Illinois, began using the Public Safety Assessment (“PSA”) four years ago as a means of reducing its pretrial jail population, but data show it has consistently overestimated the actual risk of releasing incarcerated individuals who have yet to be convicted of a crime.
The PSA was intended to gauge which individuals were more likely to commit new crimes, especially violent ones, prior to trial as well as the risk of them failing to appear at scheduled court dates. The tool does this by classifying each person into a low, moderate, or high risk category. This recommendation is then taken into consideration by a judge who determines if a defendant should he freed and what conditions should be imposed upon that freedom.
Judges at Cook County’s Central Bond Court decide the fates of about 90 individuals a day, typically in under a minute for each case. Although other factors are considered, PSA scores weigh heavily on the decisions. Almost every person rated as high risk faces jail time unless they can post bail, except roughly a third of these cases, which are given no opportunity at bail whatsoever.
For those rated low risk, about 75 percent are set free at their bond hearing.
Criticism has come after statistics from October 2017 to December 2018 show that among those freed before trial who were ranked high risk for violence, 99 percent did not incur any additional violent charges. This percentage was essentially no different than those in moderate or low categories. This has brought into question the value of risk assessment scores altogether. As a New York Times op-ed observed, “If these tools were calibrated to be as accurate as possible, then they would simply predict that every person is unlikely to commit a violent crime while on pretrial release.”
Cook County had the nation’s biggest jail population in 2015, 90 percent of which were pretrial detainees. It started applying the PSA that year to reduce that number and, within months, saw a 15 percent increase in low-risk prisoners freed without a cash bond; however, that gain was countered by even fewer high-risk defendants being released. Use of the PSA thus made virtually no difference in the overall jail population.
Seeking a more effective solution, the circuit court ordered judges to no longer demand cash bond in most cases and place “the least restrictive conditions” on defendants to assure they would appear in court and not threaten public safety. After this order was enacted in September 2017, the jail population dipped almost 20 percent. Over 97 percent of low-risk defendants are now released, yet fewer than half deemed high risk gain any measure of freedom before trial. That might be acceptable if PSA scores reflected real dangers, but data have shown risk categories bear little relation to rates of rearrest for violent crime.
Universal tools like the PSA often fail to reflect regional differences. For example, the PSA cites a 55 percent rate of rearrest among high-risk individuals released before trial, but from October 2017 to December 2018 that number was only 24 percent in Cook County. The rate of those with new violent crimes was less than a tenth of that used by the PSA. Although assessments could theoretically be recalibrated for local conditions, few areas have access to accurate data.
One thing seems certain in Cook County: The more defendants released pretrial, the more new crimes will be committed. The question is, does the 1 percent who commit them make it worth holding the other 99?