The components that make up pretrial risk assessment tools (RATs) only tell one piece of the story. How pretrial service agencies, courtrooms, and jails implement RATs on the ground in local jurisdictions, and how RATS interact with local laws and court rules, determine much of the impact that RATs have on an accused individual and overarching levels of bias and decarceration.
Regardless of RAT developers’ intent or policy surrounding RATs, the ultimate effects that RATs have on individuals comes down to how judges, magistrates, and other criminal legal system officials choose to use the tool’s results. It’s not just about what the tools say, but how they are used.
Pretrial RATs fit into each jurisdiction’s criminal legal system differently, depending on the laws and practices that govern the pretrial process. Some states require the use of RATs as part of pretrial decision-making, some laws permit but do not require RATs, some court systems recommend or set internal rules governing RATs, and other counties have decided to pilot or adopt them independently.
RATs generate predictions on risk levels for “failure to appear” and “new arrest”, along with predictions for “violence”, which then combine with local policy and decision-making frameworks to recommend specific pretrial outcomes and supervision levels. The recommendations often include specific levels of bail or the use of electronic monitoring as well.
Explore how tools are used in our Policies and Practices of Implementation section, which details how pretrial RATs interact with local laws, bail, and e-carceration.
There is also a lot of variation between jurisdictions in terms of which agencies conduct the assessments, which accused individuals are assessed, who gets to see the results of the assessment, and how the RAT results are actually used. Learn more in the RATs in Action section that examines transparency, tool use and misuse, and the assessed and assessors.
Judges and magistrates still make the ultimate decisions regarding pretrial freedom, detention, and supervision conditions. Implementing a RAT does not erase judges’ and magistrates’ racial biases, and in one study the use of a RAT actually increased bias against the poor.1Jennifer L. Skeem, Nicholas Scurich, and John Monahan: Impact of Risk Assessment on Judges’ Fairness in Sentencing Relatively Poor Defendants, Virginia Public Law and Legal Theory Research Paper No. 2019-02
Proponents of RATs claim that these tools have the power to reduce both pretrial jail populations and racial disparities in pretrial detention.2Jon Kleinberg, Himabindu Lakkaraju, Jure Leskovec, Jens Ludwig, Sendhil Mullainathan: Human Decisions and Machine Predictions, The National Bureau of Economic Research But independent studies have shown mixed results.
Learn more about the impact RATs may or may not have on pretrial jail populations and racial demographics in the Do They Impact Pretrial Populations? section.
This section will explore how pretrial RATs impact jail populations and pretrial outcomes, based largely on our interviews with jurisdictions across the country.