THE BLOG

May the Odds Be (Never) in Minorities' Favor? Breaking Down the Risk-Based Sentencing Divide

08/22/2014 12:30 pm ET | Updated Oct 22, 2014
Yellow Dog Productions via Getty Images

Attorney General Eric Holder recently stirred up a small hornet's nest in the world of sentencing reform when he called for more research on risk-based sentencing, seemingly decrying the move towards evidence-based reforms. Holder expressed concern that the growing use of aggregate data to anticipate outcomes at sentencing "may exacerbate unwarranted and unjust disparities that are already far too common in our criminal justice system and in our society."

The attorney general is right to be concerned. Risk assessment tools depend upon several factors, many of which are proxies for structural inequities disproportionately affecting racial and ethnic minorities. These include criminal history, drug and alcohol abuse, employment history and education, connection to others with criminal histories, mental health, and financial situations. These factors are weighted differently depending on which risk assessment tool is used, but prior correctional contact (arrests, convictions, incarcerations) tends to be weighted more heavily given its highly predictive value. Today, blacks comprise 31 percent of those arrested for drug law violations despite making up only 13 percent of U.S. population and using drugs at similar rates as other races. Moreover, black men are six times more likely to be incarcerated than white males; similarly Latino men 2.5 times more likely to be incarcerated than white males. It is clear that contact with the criminal justice system skews against minorities.

But the structural inequities go further. In searching for a job, black and Latino applicants with no criminal record fare no better than white applicants just released from prison. One in 9 black children born today has an incarcerated parent, as compared to 1 in 57 white children. Children of incarcerated parents tend to suffer from learning and behavioral problems at higher rates than peers whose parents are not incarcerated. By 2009, one in every three black children and one in every four Latino children lived in poverty. Clearly, the odds are not in minorities' favor to "succeed" in any risk assessment applied at sentencing.

Given these realities, it may be surprising that Holder's cautious words - that use of these tools should be studied further before being adopted full stop - received such vehement responses. According Judge Richard Kopf, Holder's perspective is akin to an ostrich burying his head in the sand: "[I]f we believe that public safety is or should be a central goal of our criminal justice system we ought not to ignore the truth..." offered in empirical data. Others concur.

Indeed, it is a "powerful current" that Holder is swimming against when he speaks out against aggregate data use at the front end of the criminal justice system. Policymakers are successfully applying evidence-based programs and practices to different areas of the criminal justice system. From CompStat data collection used to allocate police resources to predictive models used to assist a parole board's decision to release an inmate to probation supervision programs, the use of statistical methods on large datasets of criminal offending rates to estimate criminal justice outcomes are everywhere.

But evidence-based practices can be useful in one arena and not another. For example, Kentucky had great success in incorporating risk assessment tools at the pretrial stage to determine whether a person will appear in court. That assessment does not answer the separate question presented at sentencing: for example, whether a sentence or two or four years better meets the purposes of punishment. At best, risk assessment tools answer a different, but related question: whether an offender is likely to reoffend.

The U.S. criminal justice system is built on the premise that one size does not fit when meting out justice. An individualized sentencing practice is key to a fair and just sentence. It may be true that evidence-based programming improves certain aspects of the system. But, sentencing practices may not be one of those areas. In light of this reality, the U.S. Sentencing Commission's decision to seek further information on the use of this practice at sentencing is a wise one. Figuring out whether the risks are worth the costs is just the next step.