How a Spreadsheet Could Change the Criminal-Justice System

A lack of data instills trial-court judges with enormous, largely unrestrained sentencing power.

An illustration of a graph as jail bars
Shutterstock / The Atlantic

About the authors: Pierre H. Bergeron is a judge with the Ohio Court of Appeals. Michael P. Donnelly is a justice with the Ohio Supreme Court.

Judges have various restrictions on what they can say publicly, and for that reason, you don’t often hear our voices in contemporary public-policy debates. But as momentum builds to address deep inequities in our criminal-justice system, we feel it’s important to highlight a problem lurking in the background that could jeopardize these efforts: Many court systems lack basic data about themselves, including about their criminal-sentencing decisions. This means that when a judge considers a sentence for a criminal defendant, he or she has no way to evaluate it against others handed down for similar crimes in the same state, or even the same county.

Most people agree, in theory, that a court should treat similarly situated defendants equivalently in terms of their sentences. But this is not happening in practice. Criminal sentences for both violent and nonviolent cases vary widely from court to court and from defendant to defendant. The failure to gather and analyze sentencing information leaves our system vulnerable to the vagaries of explicit and implicit bias and instills our trial-court judges with enormous, largely unrestrained sentencing power.

Courts need to understand the scope and specifics of the problem that they are trying to solve.  Otherwise, they may adopt half-measures that make people feel good in the moment, but really leave broader issues and systemic inequities untouched.

A lack of data collection and analysis is a nationwide problem. Many states, including Ohio, where we serve, do not have reliable statewide numbers on the criminal sentences they impose. The states that do compile statistics have significant gaps. The problem extends beyond sentencing—many states also can’t measure, for instance, what the average bail rate is for various offenses, or even the effectiveness of the bail system.

All of this may strike one as inconceivable: How does a court system lack basic statistics in this technological day and age? The answer varies by state, but typically, antiquated IT infrastructure in state courts, no uniform requirements on compiling numbers, and a lack of coordination across jurisdictions precludes gathering meaningful numbers and demographics. And, in many corners, institutional interests are aligned to resist transparency out of a fear of what might show up.

None of these reasons should surprise anyone in the judiciary. Nearly 20 years ago, a blue-ribbon panel commissioned by Ohio’s Supreme Court concluded that the state should compile sentencing and other related data. The state legislature, desiring sentences to be consistent, fair, and not racially disparate, likewise urged similar collection. Despite the recognized need for a sentencing database, the political will to make it happen never materialized, and we, as a state, missed a golden opportunity.

For states that are starting to gather statistics, they are finding troubling, but not surprising, results. The Massachusetts Supreme Judicial Court commissioned an analysis of statewide numbers to evaluate racial disparities. Plagued by many data challenges, this effort took several years. The recently published report showed what many of us know: People of color are vastly overrepresented in the criminal-justice system as defendants; they receive longer sentences than white defendants; and they are typically charged with more serious offenses to begin with (a leverage tool to force plea agreements). When judges see reports that show this is happening in their own courts, they must ask themselves hard questions about their own complicity in these results.

In 2016, investigative reporters with the Sarasota Herald-Tribune conducted a comparison study that confirmed racial disparities in Florida’s criminal-justice system. One of the examples from their study examined two cases involving armed robbery. The same judge sentenced a white defendant to two years, but a Black defendant to 26 years—for essentially the same offense. These two individuals were almost the same age, both had a single prior misdemeanor, and they were rated the same based on Florida’s sentencing guidelines. When judges have virtually unchecked discretion, and they lack ready access to sentencing data, these discrepancies are bound to continue happening.

This problem manifested itself in the case of Susan Gwynne, a 55-year-old Ohio woman with no felony criminal history who faced multiple nonviolent-property-theft charges for stealing nursing-home residents’ personal property. After entering a negotiated plea agreement in which a number of her charges were dismissed, Gwynne arrived at her sentencing hearing not knowing whether she would receive probation (and thus no prison time) or a lengthy prison sentence, because either of these scenarios fell within the range of possible penalties authorized by law.

The judge, imbued with nearly unfettered discretion, sentenced her to serve 65 years in a state prison, effectively ensuring that she will die in prison for committing a series of nonviolent thefts. She received a longer sentence than many rapists and murderers, who may also have the opportunity to be released from prison in 10 or 15 years (at the low end). If Gwynne’s attorney had the relevant information, he could have demonstrated on the record that her sentence was astronomically out of proportion to similar theft-related cases, and such a record would enable appellate courts to conduct meaningful scrutiny of the sentence.

Examples like these, unfortunately, can be found in every state. And they foster an already boiling mistrust of the judicial system, particularly among people of color who intuitively know that they receive more severe sentences than similarly situated white defendants.

Although data challenges are pervasive and a key barrier to criminal-justice reform, they can be solved. In response to the statewide analysis conducted by the Sarasota Herald-Tribune, Florida legislators passed groundbreaking legislation to standardize the way the state gathers and shares information. The state has already missed some deadlines, but the legislation as designed would render Florida one of the most transparent states in the country from a criminal-data perspective.

Other states are also starting to create comprehensive databases so that informed criminal sentences are accessible to all stakeholders—judges, prosecuting and defense attorneys, defendants, and policy makers. Currently, the nonprofit Measures for Justice has compiled statistics for 16 states. Ohio, led by Chief Justice Maureen O’Connor, is in the process of developing a felony-sentencing database with the objective of making information accessible, shareable, and reportable. It can’t come soon enough. Indeed, if every state acted quickly, we could solve this issue in short order, and then move on to implement lasting criminal-justice reform that would end mass incarceration.

The goal of these efforts is not to eliminate judicial discretion (judges aren’t robots, after all) but to provide sound analysis to inform judges in the exercise of that discretion. If everyone has complete access to information, the prosecutor can make an informed sentencing recommendation, the defense counsel can use the data to make his or her case, and the judge can feel secure in knowing that the sentence imposed fits well within the range from other courts around the state. If the sentence deviates up or down, the judge can give a reason on the record, providing greater transparency in the process. Objective measures that are comparable, consistent, and reliable can better ensure the equalized application of justice.

For an illustrative example of how an evidence-based, informed criminal-justice system could work, consider the high-profile sentencing of Paul Manafort, Donald Trump’s former campaign manager. Manafort’s 47-month sentence for tax and bank fraud sparked significant outcry, with many people wondering why his sentence was so “light.” It turns out that his lawyers in the federal court system had the numbers most of our state court systems do not. While the government advocated for a 19-year sentence, Manafort’s lawyers were able to show that on multiple prior occasions, similarly situated defendants in less visible cases received much shorter sentences. Unable to distinguish these cases, the judge felt constrained to keep his sentence in conformity with this past practice, regardless of the defendant’s notoriety. Having such data available to attorneys in our state systems could prevent disparate sentences from occurring while providing appellate courts with the ability to correct sentences that are grossly disproportionate to the offenses involved.

The judicial system relies on the trust of our citizenry; public confidence is its lifeblood. We must act in deliberate and real ways to create change in our courts. And that requires working with all stakeholders—including the community, legislators, and law enforcement. No one, including judges, can sit back and pretend that the problem of inequality is too intractable or the result of someone else’s decisions any longer. Collecting and utilizing sentencing data will help build a better, more equitable justice system.