By ERIC A. HOWALD
Of the Keizertimes
In recent months, several accused criminals who call Keizer home have been released back into the community on forced release as a result of overcrowding at the Marion County Jail (MCJ).
Some of the accused have taken their regained freedom as license to reoffend, others appear to be adhering to the conditions of their release when jail overcrowding causes them to be released.
When news of the forced releases hits social media, area residents cry foul and suggest that police and courts are not living up to the roles the community have given them. Others call for more primitive forms of capital punishment. Neither position takes into account how such decisions are made or why.
In large part, a risk assessment score, determined by a computer algorithm and based on numerous factors gleaned from other residents of jails and prisons and those under community monitoring, carries the burden of determining who gets released when overcrowding occurs. Often, the decisions the computer reaches appear illogical to a layperson, but it does work, said Commander Tad Larson, who oversees the jail for the Marion County Sheriff’s Office.
“We’ve had people with medium or high risk scores end up out on forced release and still fulfill their court obligations, even when it ended up with them in prison,” Larson said. “Does it look right? Did it feel right? Were the neighbors upset? You bet, but the risk assessment process is working.”
Understanding where risk assessment fits into decisions at MCJ requires understanding some of the realities on the ground. First, jail is not prison. Jails house arrested individuals who are awaiting court proceedings and those who are sentenced to less than a year after conviction in a court. Prisons are where those who are convicted and sentenced to more than a year of detention go.
As of July 2018, MCJ had processed more than 8,000 individuals since the beginning of the year, and that number is expected to double by the end of the year. The vast majority were booked into the jail, had their fingerprints taken, were assigned a court date for arraignment on the charges they are facing and then released back to the streets. Others, who pose a greater risk, are held at the jail. The average stay for those held at the jail this year is 17 days.
MCJ has four cell blocks, but only three in operation due to lack of funding. The three operating blocks have 415 beds that are almost always full. In addition, there may be up to 60 inmates flowing in and out of the jail – e.g. making their way to court appearances or other facilities – on a daily basis.
At intake, a risk assessment tool, which crunches probabilities of re-offending based on factors such as age, gender, age of first arrest, current criminal charges, type of crime (personal vs. property), previous convictions and custody cycles, spits out a risk assessment score for each inmate at the jail. The database compares the new inmates to the performance of 300,000 past inmates based on their demographics and criminal history and looks for signs that might predict future behavior.
The one area which is conspicuous by its absence in the data used to assess risk is what happens when the suspect has victims known to him who he can return to torment once released. The answer is there, but it’s hidden. Baked into the charges a suspect faces are the circumstances of their crime. If a violent altercation is part of a domestic abuse incident, the risk scores take that into account.
When the jail can no longer hold additional bodies, the ones with the lowest scores are put back on the street under an agreement to appear in court at a future date (For an example of how this works, and sometimes doesn’t, see related story Third time page A5).
The intake process is just as much a factor for the jail’s administrators when they are looking at who to release, Larson said. If a suspect was arrested while making threats to individuals or known victims, the officer who takes the suspect to the jail can request a public safety override. If approved by the jail administrators, the override guarantees the suspect stays in jail until an arraignment, which is typically held as soon as possible. Larson said the jail tries to grant all of the requests for public safety overrides.
Overrides rely on good communication when officers from local police departments and other arresting agencies make the hand-off at MCJ.
“Somebody has to ask the question. And it has to be a credible threat to the public,” Larson said.
At the Keizer Police Department, there are no hard-and-fast policies for when to request such overrides, officers are asked to use their own discretion.
While the risk scores are a large element of determining who stays and who goes, there are human components.
“The risk score doesn’t override common sense. It’s an unbiased tool and scores [inmates] among their peers,” added Commander Jeff Wood, who oversees community corrections programs that provide alternatives to detention.
In an country where any dispute can end up in court, the risk assessment score acts a bulwark against perceived biases. Moreover, the risk assessment tool has a calming influence on the shifts in public and political priorities. When the community attention turned to obliterating the production and use of methamphetamine 15 years ago, it became difficult to release individuals arrested for trespassing while under the influence – even if that was the worst of the crimes they committed while high.
“The same thing happened with gang crimes,” Larson said. “Every agency was arresting people with gang involvement and the call from the community was to release other people and not those charged with gang crimes. What if one inmate is a member of a gang and some other guy is charged with murder, which one do you want out on release?”
Another component of intake is classification, which determines where an inmate goes in the jail. As much as possible, administrators try to keep cooperative inmates, unruly ones and those perceived as suffering from medical issues separated. Blending them can have unintended consequences.
“If we keep a low risk offender and put him in jail with another offender who scores higher, all he might gain is learning how to commit a crime better or he might be encouraged not to go to court,” Larson said.
That’s another human calculation and, occasionally, another reason for putting someone out on forced release when the crimes they are charged with offend the sensibilities of an average person.
Amid all the other considerations, timeliness is also important. Not only is it the right of the accused to be moved into and through the court system quickly, but allowing some individuals to return to society in short order ensures their support systems stay in place and that reduces the risk of re-offending.
“For every cycle of arrest and detention we put an inmate through, the risk becomes greater. We also have to consider we might be taking away some of the stabilizing factors that they might lose if they are sitting in jail. A job may not wait for them. They may lose their address, or children. As more stabilizing factors diminish, the less likely it is they will appear in court,” Larson said.