 |
| Jeffrey Kremers |
On any given day, there are more than 730,000 people in jail in the United States, two-thirds of whom are there pretrial, not yet convicted of a crime. It is a system that runs contrary to our nation's core promise of liberty — but it doesn't have to be like this.
One of the major problems in this pretrial phase is the reliance on bail in determining release conditions for people charged with, but not yet convicted of a crime. While courts have historically set bail according to the seriousness of a person's charge and concerns regarding public or victim safety, they do not always have the information necessary to accurately assess pretrial risk.
So instead of looking at a defendant's likelihood of committing another crime or failing to appear at court, bail is instead tied to the charge: the more serious the charge, the higher the bail. The result has been the dramatic over-incarceration of low-risk individuals at great public expense, and the release of people who by any objective measure of risk should be detained.
Relying on this charge-based methodology for making release decisions has significant harmful consequences. Since the focus is on the charge and not the individual, people with money, regardless of how risky or dangerous, can get out.
But people without money, who may have given every indication they will return to court, and by definition have less ability to flee, often end up being held at great public and personal expense. There is a growing body of research that shows pretrial incarceration, even for people ultimately acquitted, can lead to loss of jobs, housing, benefits and childcare, and also higher rates of recidivism.
Here's the urgent and ugly reality of America's pretrial system: The burden unfairly falls predominantly on members of poor minority communities who are least able to afford bail. This means the pretrial population is disproportionately comprised of people of color, contributing even more to racial disparities in our jails.
However, many jurisdictions are making progress in fixing their pretrial systems and reducing unjust incarceration. One commonly used reform is reliance on pretrial risk assessment to help set bail.
In July, a group of researchers published an essay in the New York Times taking aim at pretrial risk assessment.[1] They reached the conclusion that risk assessment fails to deliver on necessary improvements to the pretrial system.
While the authors' intentions are in the right place, their critique only looked at part of the picture.
Research shows that when jurisdictions implement broad reforms — which can include, but are not limited to, risk assessment — they are consistently successful in reducing the number of people held before trial. One needs to look no further than Mecklenburg County, North Carolina, where researchers this year found the implementation of risk assessment and related reform led to a significant drop in pretrial jail bookings without an increase in crime.[2]
Advocates are right to argue that we must address the public's deep-seated misunderstandings about public safety. An inflated sense of people's risk seized on by the bail industry and perpetuated by alarmist media stories and tough-on-crime politicians continues to drive over-incarceration.
In the limited role risk assessments have to play in pretrial reform, they should work to undo these misconceptions. To that end, risk assessment developers must consider making it clearer to judges and the public what risk scores actually mean in terms of a person's likelihood of future arrest for crime of violence — which, as the opinion authors point out, is almost guaranteed to be low, even for the highest risk defendants.
Like every judge, I have released people accused of crimes on bail who go on to commit serious violent offenses. These situations are obviously tragedies for the victim and the community. Yet in each of these sad circumstances, I had little or no evidence-based measurement of the defendant's risk of reoffending. Usually I had only the defendant's local record and the charge.
Thankfully, such events are quite rare in the context of the thousands of people released on bail each day. Despite this rarity, the typical response to such tragedies is to set higher bail in the future.
It's a decision not based on evidence of risk. Each time I made such a decision in my own court, I failed to understand the harm done to individuals who did not have enough money to be released, and who, if released, would not have reoffended.
Risk assessment was developed as a resource meant to serve a system plagued by a chronic lack of information and ruled by subjective — often biased — decision-making. It should never be mistaken either for a root cause of the system's ills or a silver bullet. Instead, research shows that, if applied together with policies aimed at the safe and monitored reduction in pretrial jail populations, risk assessment can be a powerful tool for reform.
The true key to reducing unjust incarceration is jurisdictions committing to the hard work of actual policy change beyond the option of a risk assessment: This includes limiting the categories of offenses eligible for jail time, emphasizing a presumption of release, and strengthening due process protections through individualized hearings, access to robust representation and speedy trials.
Only then will eliminating unjust and unnecessary pretrial detention be within reach. Only then can we truly claim to value liberty.
Jeffrey Kremers was a judge for the Milwaukee County Circuit Court in Wisconsin from 1992 to 2018.
"Perspectives" is a regular feature written by guest authors on access to justice issues. To pitch article ideas, email expertanalysis@law360.com.
The opinions expressed are those of the author(s) and do not necessarily reflect the views Portfolio Media Inc., or any of its affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.
[1] https://www.nytimes.com/2019/07/17/opinion/pretrial-ai.html.
[2] https://www.mdrc.org/sites/default/files/PSA_Mecklenburg_Brief1.pdf.