Bar News - October 18, 2017
NH Joins National Pretrial Justice Reform Initiative
By: Kristen Senz
New Hampshire will soon join a national pretrial justice reform campaign as part of a plan to advance efforts at the local, county and state level to use more data-driven approaches to setting bail in criminal cases.
The state’s Interbranch Criminal and Juvenile Justice Council (ICJJC) voted last month to make New Hampshire the fourth jurisdiction to join 3DaysCount (along with Connecticut, Washington State and Guam). An initiative of the Pretrial Justice Institute (PJI), 3DaysCount aims to help 20 states by 2020 to reduce unnecessary arrests, replace money bail with risk-based decision-making, and restrict incarceration to those who truly pose unmanageable risks if released.
The campaign’s name refers to the devastating effects that as few as three days in jail can have on a low-risk person’s life, by disconnecting them from employment and relationships. According to PJI, research shows that just three days in jail can also leave low-risk defendants less likely to appear in court and more likely to commit new crimes, the exact opposite outcomes sought by the state’s bail statute.
For New Hampshire, joining the campaign means PJI will provide support during a Grafton County pilot project using a research-based risk assessment tool, help in analyzing the state’s current jail population and pretrial processes, and assistance to state officials as they explore other avenues for reform, said NH Superior Court Chief Justice Tina Nadeau, who chairs the ICJJC.
Enfield Police Chief Richard Crate, a member of the ICJJC, originally reached out to PJI because he wanted a better way for his police officers to decide whether to arrest or release a defendant on a summons. In the pilot project, which is set to launch within the next few months, local officials in the 2nd Circuit (Lebanon, Dorchester, Enfield, Canaan, Grafton, Orange, Hanover, Orford and Lyme) will begin using an actuarial risk-assessment tool – a statistical model that relies on data from thousands of criminal cases.
The tool estimates the likelihood, based on specific risk factors within the model, that a particular person will appear for future court hearings or commit additional crimes during release. That information is meant to be used in conjunction with observations and other factors to make better decisions about bail, says Rachel Sottile Logvin, vice president of PJI.
“This is not a replacement for judicial discretion,” says Logvin. “I can’t say that enough. These actuarial tools are intended to be an additional source of information, but there are certain other critical factors, as well as the statute, that need to be taken into account by the judge.”
The Grafton County pilot project will use the latest version of the Virginia Pretrial Risk Assessment (VPRAI) tool, which was first developed about 15 years ago. It uses eight primary risk factors: active community supervision; charge is a felony drug, theft or fraud; pending charges; criminal history; two or more failures to appear; two or more violent convictions; unemployed at time of arrest; and history of drug abuse.
After reviewing a number of assessment tools, the ICJJC preferred a newer, open source tool developed by the Laura and John Arnold Foundation. Known as Public Safety Assessment (PSA), that tool is still in testing but is expected to be made available to states soon. NH Circuit Court Chief Justice Edwin Kelly, who is helping to lead the pilot project, said PSA, unlike VPRAI, does not involve interviews with the subjects and uses more data and different, more readily available data points, such as age.
“Even if we switch to the Arnold Foundation instrument six months down the road, we will have gotten police departments and judges and others used to looking at a risk-assessment tool,” he said at the ICJJC meeting.
Both VPRAI and PSA were developed using data from multiple jurisdictions. After implementation, VPRAI was locally validated in Virginia, to ensure the model didn’t contain built-in biases that would surface through disproportionate representations of certain population segments in the model, compared with the actual jail population in Virginia. It remains unclear whether any risk assessment tool selected for statewide use in New Hampshire would eventually be validated against local inmate data. Nadeau said officials are debating this question and are continuing to collect information.
According to PJI: “There are many jurisdictions that ‘borrow’ a risk tool that has been validated in another jurisdiction, with the intent of ultimately validating it. For example, many jurisdictions outside of Virginia use the Virginia tool, and many outside of Ohio use the Ohio tool. The rationale behind such an approach is that you have to start somewhere and there is enough commonality among the various empirically-derived tools, especially those developed through studying multiple jurisdictions… to suggest that any one of them would likely have some degree of accuracy in assessing the probability of success in most jurisdictions.”
On the county level, Rockingham and Hillsborough counties are exploring the development of pretrial services programs similar to the ones in Strafford and Sullivan counties, which provide a framework for supervising defendants during pretrial release.
Nationally, efforts to move criminal justice systems away from relying heavily on cash (or surety) bail, which results in poorer defendants remaining locked up because they can’t post low cash bail amounts, have been gaining momentum over the past few years. PJI, which has been around for about four decades, has emerged as a leader in this area.
“I think more people are becoming aware of what’s going on across the country with bail reform,” Nadeau said at a recent ICJJC meeting. “The goal is to be able to use objective factors, as objective as possible, as part of our decision-making.”
Nadeau said the ICJJC, with support from PJI, is exploring an array of options designed to improve pretrial processes, including the risk assessment tool, notifying defendants by text of upcoming court dates, and training for judges and others within the criminal justice system. Ultimately, she said, the group hopes to enact cost-effective reforms on which all stakeholders can agree.
As shifts in pretrial populations in local jails across the country are examined and addressed, the country as a whole could begin to see its prison population shrink, reversing the trend toward mass incarceration that has characterized the past three decades in American criminal justice, says Logvin.
“As we work to better manage our pretrial populations and decision-making, we’re going to see major reductions in our prison population,” she said. “That’s what the field is predicting.”