America's prison populations are disproportionately filled with people of color, but prosecutors' biases toward defendants' race and class may not be the primary cause for those disparities, new research from the University of Arizona suggests.
The finding, which comes from a unique study involving hundreds of prosecutors across the U.S., counters decades' worth of previous research. Those studies relied on pre-existing data, such as charges and punishments that played out in courtrooms. In a 1993 study, for example, researchers found that prosecutors in Los Angeles were 1.59 times more likely to fully prosecute an African American defendant for crack-related charges than a white defendant. That likelihood was 2.54 times greater for Hispanic defendants compared to white defendants.
The new study, led by Christopher Robertson, a professor of law and associate dean for research and innovation at the James E. Rogers College of Law, involved a controlled experiment with prosecutors, asking them to examine the same hypothetical case but changing the race and class of the defendant.
The study, administered online, provided prosecutors with police reports describing a hypothetical crime, which the researchers designed with assistance from experienced prosecutors. All details of the case were the same except for the suspect's race - either black or white - and occupation - fast-food worker or accountant - to indicate the suspect's socioeconomic status. Roughly half of the prosecutors received one version of the case; the other half received the other.
The study allowed researchers to "really isolate the prosecutor's decision-making in a way that mere observational research wouldn't allow," said Robertson, whose co-authors are Shima Baradaran Baughman of the University of Utah and Megan Wright of Penn State. The paper was published in the Journal of Empirical Legal Studies.
The outcomes the study looked for included whether prosecutors charged a felony, whether they chose to fine the defendant or seek a prison sentence, and the proposed cost of the fine or length of the sentence.
"When we put all those together, we see the same severity of charges, fines and sentences across all the conditions, whether the defendant was black, whether the defendant was white, whether the defendant had a high-class career or a low-class career," Robertson said. "Differences in the actual outcomes - in the actual behavior of the prosecutors - is what we would have expected if they were biased. But since we see no difference in the outcomes, we concluded that they were not substantially biased."
Given previous research that indicated rampant bias drives criminal justice disparities, Robertson's results may surprise many - just like they did the researchers.
"We were surprised at the bottom line," he said.
Robertson offered one possible explanation for the unexpected result.
"We conducted this study in 2017 and 2018 and prosecutors have been under a spotlight for some time," he said. "They've been training and are aware of and are working hard to not be biased in their own decision-making."
The results do not rule out race and class bias as factors in prosecutorial decision-making but suggest that policymakers committed to addressing systemic racism and classism in the legal system may be more successful seeking reforms in other areas.
"The disparities in outcomes are indisputable," Robertson said. "As we go through the criminal justice system and think about what the right reforms are, the sheer bias of the prosecutor doesn't seem to be the biggest one."
Robertson said policymakers may be better off focusing on disparities that occur before someone is even arrested, in areas such as economic development and education.
"Crime is associated with poverty, and race in America is associated with poverty, so I think some very front-end questions of social policy are really important," he said. "At the same time, I think, on the back end, to shift the focus, there's a growing consensus among people on the left and the right that our 40-year-long war on crime has been ineffectual in some ways and that we could make the criminal justice system much less severe and much less expensive and thereby reduce some of these same disparities."
Robertson also stresses that his study's results aren't the final word on prosecutor bias - a problem that still needs addressing, he said. Even after these findings, he remains a proponent of blinding prosecutors to defendants' race, a detail that is often not relevant to prosecutors after an arrest is made. Prosecutor blinding is the focus of Robertson's next research project.
Journal of Empirical Legal Studies