Ask any of the highly-paid professional anti-bail lobbyists lurking in the halls of the Colorado state house what’s wrong with the current cash bail bonds system and you will invariably get the answer that it’s a racist system that punishes minorities. Irony of ironies then, that a July study conducted by the University of Northern Colorado (UNC) discovered that the risk assessment algorithm promoted by anti-bail zealots and used by a number of Colorado counties is itself biased against African Americans.
It seems the algorithm, known as CPAT - short for the Colorado Pretrial Assessment Tool and - is prone to concluding that blacks are more likely to miss a court date than whites. Even though statistics show that’s not the case. Defenders of the CPAT algorithm claim the UNC study was flawed. But the people behind the study believe otherwise. They say their study indicates that the CPAT algorithm uses aspects of poverty to incorrectly suggest that a person represents a flight risk or is a danger to others.
For instance, CPAT determinations are based on 12 questions including ‘Do you have a home phone or cell phone?’ and ‘Do you own or rent your place of residence?’ On the face of it, those questions might seem like a good way to separate the risky people from the more settled, trustworthy types. But in reality, what they represent is a poverty test. Poor people are far more likely to rent than own. They are also more likely to have a landline than a cell phone.
Another question used to help make CPAT risk determinations is ‘At what age were you first arrested?’ That’s another question that might seem like a good way to find out if someone is a career criminal and a danger to society. But statistics show that young blacks are far more likely to get arrested than whites simply because policing of poor black neighborhoods is typically much more aggressive than it is in more affluent white neighborhoods.
Once it was established that the CPAT test was actually a type of means testing which was slanted to give poor people a higher threat assessment than well-to-do people, the next question became: “Who are these poor people it is discriminating against?” More often than not, at least in big cities like Denver, they are African Americans.
For a variety of complex reasons, 25% of African Americans live below the poverty line, as opposed to 7% of whites. That is especially true as we said, in big cities, including Denver, which uses the CPAT test to make risk assessments of defendants. According to the UNC study what happens is that an African American is arrested. He or she is subjected to the CPAT risk assessment test. They answer the test truthfully and the test concludes they are a flight risk or a danger to society. But what the test is actually concluding is that they are poor.
(For the record, if a poor white person is arrested and subjected to the CPAT algorithm, they too will be unfairly judged a risk to society. But because a much smaller percentage of whites live below the poverty line in the United States today, the CPAT test is skewed to discriminate against black Americans.)
To drive the point home of how inaccurate and twisted the CPAT test is, the study followed some 3,400 defendants after they were assessed by CPAT and released. It discovered that CPAT placed more than half of them into the highest risk categories based on the slanted questions. Yet 98% of those deemed to represent the highest risks (mostly blacks) stayed out of trouble while awaiting their court appearance. At the same time nearly 50% of those with more money who were deemed less risky (mostly whites), got into some sort of trouble after release. These results strongly suggest that CPAT has very little to do with risk assessment, but is instead little more than a classic manifestation of systemic racism.
Before we wrap up we thought we’d demonstrate an example of CPAT bail bonding reform in action.
Samuel Bouknight is an African American resident of Denver who was arrested for driving under the influence in October of 2018. He was taken in and subjected to the CPAT assessment to determine if he was a flight risk or danger to his fellow citizens. Bouknight was assessed at the highest level of risk. Because of this, the judge set a $750 bond which Bouknight, because he had limited financial means, could not pay. So he languished in jail for 59 days.
When he was finally released the few clients he had for his photography business had gone elsewhere and Bouknight had to spend time and what little money he had going across town 3 times a week to take court-ordered sobriety tests. According to Bouknight, the biased CPAT assessment, which had concluded he was a risk because he was both black and poor, essentially ruined him. Oh, and for the record, despite being ranked at the highest risk level by the CPAT algorithm he kept all his court dates and stayed out of trouble.
The current crusade against the bail bondsman must stop. Not only are cities and towns in Jefferson County, Arapahoe County, and Denver County buckling under the cost of having to keep tabs on so many people released on their own recognizance, not only are truly dangerous individuals being put back on the streets, but people who are guilty of little more than being black are having their lives ruined because of biased assessment tests like CPAT.