The last time government officials in Kentucky made headlines, it wasn’t for a flattering reason. Rowan County Clerk Kim Davis incited controversy after refusing to issue marriage licenses to same-sex couples in 2015, defying a U.S. Supreme Court ruling and order. But two hours south of Davis in Somerset, Circuit Court Judge David Tapp believes the state’s court system is a national leader worth boasting about when it comes to one innovation.
“Aside from bourbon and basketball, this is one of the few things we really do a good job on,” Tapp told TakePart.
Tapp is referring to the state’s pretrial release system, which since 1976 has embraced groundbreaking risk assessment tools to help judges decide which defendants can safely be released back to their homes instead of sitting in jail to wait for trial. In July 2013, the state switched to its fourth and current risk assessment tool—Tapp says it is the best so far at helping courts figure out which defendants can be released before their trial without posing a threat to the public. While 62 percent of criminal defendants nationally are released prior to their trial, Kentucky releases 73 percent of its pretrial defendants. Of those released pretrial, 88 percent show up for their court dates, and 90 percent of that group have not been rearrested.
The tool—which was developed by the Laura and John Arnold Foundation and made available to 29 local jurisdictions across the country at no cost—is also being rolled out across Arizona and New Jersey. Similar tools peddled by private companies are licensed to state and local governments for a fee. Kentucky was the first state to use the foundation's tool and has long been considered an innovator in pretrial assessment. In 1976 the state abolished commercial bail—eliminating the lobbying power of private bail bond companies that maintain a foothold in other states and are considered a roadblock to reform. Illinois, Oregon, and Washington, D.C., have also abolished commercial bail.
Detaining someone prior to trial is a big deal, in both human and fiscal costs. The dangers associated with jailing a low-risk defendant were seen most sharply in the suicide of Kalief Browder in 2015, who as a teen spent three years in New York City’s Rikers Island jail awaiting trial because he and his family could not afford his bail of $3,000. He was arrested for allegedly stealing a backpack, a charge he denied. After spending two of those years in solitary confinement and being released, Browder struggled to acclimate to life outside jail and took his life at 22.
Beyond the psychological toll that incarceration can take, research has found that spending even a few days behind bars increases the likelihood that someone will commit another crime in the short and long term. That means detaining low- and moderate-risk defendants—people who are supposed to be presumed innocent until proved guilty—before their trials can cause them to miss shifts at work or leave children unattended or otherwise make it harder for them to return to their normal lives.
“It only takes a couple days to turn a low-risk person into a higher-risk person,” said Cherise Fanno Burdeen, executive director of the Pretrial Justice Institute.
Allowing defendants to go home before their trials is also a boon to taxpayers. In Kentucky, the cost of pretrial supervision averages $1.90 per defendant per day, depending on the location and level of monitoring. Jail costs average $10 to $30 per inmate per day, depending on the jurisdiction.
To figure out who should return to the community before their trial, researchers at the Laura and John Arnold Foundation studied 1.5 million cases from 300 jurisdictions across the country to learn more about what risk factors most strongly correlate with an individual’s return to court. Prior to the creation of the foundation’s Public Safety Assessment tool, many jurisdictions relied on risk assessment surveys that required time-intensive interviews with each defendant involving a litany of questions about criminal history, employment, and other variables believed to predict the likelihood that a person will return to court. The PSA whittled down the list of questions to just nine predictive factors that could be answered without an in-person interview—such as previous failures to appear in court, age, current charge, and criminal history—making the tool easier to use.
“Tools like this have the potential to help reduce disparities in the justice system, simply because there are so many low-risk, nonviolent defendants sitting in jail,” said Matt Alsdorf, vice president of criminal justice at the Laura and John Arnold Foundation. “This tool can help judges and prosecutors identify who those defendants are and get them out.”
Not all risk assessment tools are created equal. Some, particularly proprietary tools used in the sentencing phase, have faced criticism for exacerbating existing racial bias in the criminal justice system. A ProPublica investigation into the COMPAS risk assessment tool developed by a for-profit company called Northpointe found black defendants routinely received higher risk scores than white defendants. Northpointe, like other companies that have created proprietary tools, doesn’t release data on all the risk factors that are used to predict criminal behavior—making it difficult to fully assess their effectiveness and potential for bias and drawing criticism from advocates and others. According to Alsdorf, the foundation has invited an independent third party to conduct an 18-month analysis of its PSA tool to assess its impact on “race, ethnicity, and neutrality.”
On July 13, the Wisconsin Supreme Court ruled that judges using COMPAS during the sentencing phase must receive written warnings not to rely solely on the tool’s score after a defendant’s lawsuit argued that Northpointe’s secret scoring system was unfair and a threat to his due process rights.
Tapp calls the PSA a “huge time saver” that helps him quickly assess whether or not individuals might pose a danger to the community if he releases them.
“Instead of taking a whole lot more time to look at those factors individually and do my own imprecise calculation, the [PSA] has done that for me,” said Tapp. “We want to detain people who are scary, who are going to hurt somebody—but most of what we deal with is not those people. The vast majority of defendants I see come back as low or moderate risk.”
In spite of the state’s successful adoption of the tool, both Tapp and Alsdorf emphasized that it is not a silver bullet but a supplement intended to ease a complicated process.
“Risk assessment tools can’t supplant anybody’s discretion,” said Tapp. “It doesn’t always work; I still have defendants occasionally not show up and others who violate the conditions of their release.”
David Mink, a 36-year-old resident of Somerset, was released before his October 2014 trial by Tapp after the PSA established him as a low risk. Mink had pleaded not guilty to one count of property theft and two counts of firearm theft. It was his first offense. Mink, who was struggling with opioid addiction after being prescribed oxycodone to treat several herniated discs in his spine, found himself with few restrictions and struggled to stay out of trouble until his court date. When a court officer called him in to take a drug test, he failed.
“The problem with pretrial release is there’s no structure,” Mink told TakePart. “You’re just turned loose. If you’re an addict, you’re going to go back to it.”
Mink found himself before Tapp again after failing the drug test and spent 30 days in jail. Tapp diverted him to a drug treatment program, and he has been clean for 25 months after being treated with Vivitrol, an opiate blocker. “I just wish I’d have gotten thrown in drug court a lot sooner,” said Mink.
In spite of pretrial release slip-ups like Mink’s, Tapp advocates for the PSA’s use.
“It’s a huge improvement over a normal or old-fashioned bail bond system,” said Tapp. “This is definitely the way forward for every state.”