Editor's note: Shamena Anwar is an assistant professor of economics and public policy at Carnegie Mellon University; Patrick Bayer is a professor of economics at Duke University; and Randi Hjalmarsson is an associate professor in economics at Queen Mary, University of London.
(CNN) -- The Sixth Amendment right to a trial by an impartial jury is the bedrock of our criminal justice system. Yet the promise of impartiality is called into question when defendants face juries that include few, if any, members of their race.
The small percentage of black people in the U.S. population, less than 13%, and in some cases, their systematic exclusion from juries, means that black defendants routinely face all-white juries in many states and counties.
Concerns about jury representation go hand-in-hand with the sense that the racial makeup of juries might make a big difference for conviction rates in criminal trials. Surprisingly, we know very little about this.
Because every trial has unique circumstances, it is impossible to say for certain how the jury composition affected any particular decision. Instead, much like the process of clinical trials designed to test new medications, it is necessary to examine a large number of cases to gauge the impact of juries.
In clinical drug trials, doctors randomly assign participants to treatment and control groups, giving some the medication and others a placebo. Random assignment ensures that those receiving the medication and placebo are statistically identical and, therefore, a comparison of their outcomes provides a clean test of the medicine's efficacy.
As it turns out, the real world provides exactly the same kind of random assignment for criminal trials -- the pool of jurors who are randomly called for jury duty that day. In our recent study, "The Impact of Jury Race in Criminal Trials," published in The Quarterly Journal of Economics, we analyzed the results of this kind of natural experiment for over 700 felony trials in Lake and Sarasota counties in Florida from 2000-2010. The study examined how conviction rates varied with the racial composition of the pool of jurors called for each trial.
The jury pools in these trials consisted of an average of 27 members from which juries of six people -- which Florida allows in felony trials -- plus an alternate, were seated. Because the eligible jury population in these counties was less than 5% black, jury pools typically included one or two black members. There was not a single black member of the jury pool in almost 40% of the trials.
The results of our study were straightforward and striking: In cases with no blacks in the jury pool, black defendants were convicted at an 81% rate and white defendants at a 66% rate. When the jury pool included at least one black member, conviction rates were almost identical: 71% for black defendants and 73% for whites. The impact of the inclusion of even a small number of blacks in the jury pool is especially remarkable given that this did not, of course, guarantee black representation on the seated jury.
The finding that the racial composition of the jury pool had such a large impact on conviction rates raises concerns about the fairness of our criminal justice system. It is deeply disturbing that the luck of the draw regarding the makeup of the jury pool, rather than the quality of the evidence, was decisive in so many cases.
The fact that the inclusion of even one or two blacks in the jury pool had such a large impact on conviction rates also seriously calls into question whether jury representation in direct proportion to the local population is enough to ensure juries are impartial. This is especially worrisome for members of minority groups that are severely underrepresented in that population.
These findings should push the Supreme Court to reexamine the issue of jury representation at the most basic level. Although the use of peremptory challenges by attorneys to exclude blacks from juries has been a major concern that should not be minimized, racially unbalanced juries come about in most cases simply by reflecting the local eligible juror population. In fact, in the trials that we studied, blacks in the jury pool were seated in roughly the same proportion as whites. But the majority of the juries were all-white simply because there were so few blacks in the jury pool.
Of course, rethinking features of the jury selection process that we take for granted -- like proportional representation of the local population -- and designing alternatives is not easy. But if the core promise of jury impartiality is to be met, the evidence that a minimal level of racial inclusion in the jury selection process makes a substantial difference simply cannot be ignored. The fundamental fairness of our criminal justice system is at stake.
The opinions expressed in this commentary are solely those of the writers.