The first thing you need to know about Patrick Bayer is that he's an economist, a social scientist and part of a discipline that relies heavily on all sorts of data about human behavior and financial matters. That's part of the reason economics is sometimes called the dismal science.
Bayer began his career studying urban economics. So for him, that work also often included examining residential segregation, school choice and competition, social interactions and the effects of different neighborhoods on people's lives. Today, Bayer is also a professor of economics at Duke University, where his most recent research has gone deep on the effects of discrimination in mortgage lending and housing markets.
It's serious, academic stuff, not included in publications on your local newsstand. And it is important stuff. For instance, this year Bayer co-authored a journal article called, “The Vulnerability of Minority Homeowners in the Housing Boom and Bust,” published in the American Economic Journal. He's also been published in several other peer-reviewed journals with economic in the title.
Bayer is a guy who knows his stuff. And back in 2012, he co-wrote a study that did get a lot of attention in newspapers across the country because it was the first of its kind — ever. He decided to take a look at what effect, if any, all-white juries had on actual trials over a 10-year-period. The study Bayer and his research partners, Shamena Anwar and Randi Hjalmarsson, published, "The Impact of Jury Race in Criminal Trials" arrived at an answer: a lot.
With the Supreme Court's Monday decision to make way for a new trial in a case involving a black man on death row for killing an elderly white woman, heard by an all-white Georgia jury which prosecutors intentionally formed, The Fix thought a deeper look at the phenomenon of all-white juries might be in order. How does this happen even in diverse areas of modern America? What does this do to justice and trial decisions? Bayer helped us sort this out. It's really worth a read. Be sure you take a look at Bayer's final answer, which puts the "dismal" in dismal science.
What follows is a Q&A conducted via email, edited only for clarity and length.
THE FIX: How would you describe your study's major findings?
BAYER: Let me first provide a little background on the study and the context in which it was conducted.
Our analysis was designed to examine the impact of the racial composition of juries on conviction rates for white and black defendants. The study was based on data from Lake and Sarasota counties in Florida [between 2000 and 2010], where we were able to acquire information on the characteristics of not only the seated jury but also the pool of potential jurors. In order to identify causal effects (rather than just correlation), our analysis examines how trial outcomes change as the result of random variation in the pool of potential jurors called for jury duty for each trial.
The population of Lake and Sarasota counties in Florida is approximately 4 to 5 percent black [Editor's note: these figures have grown since the period Bayer studied], non-capital trials are decided by six-person juries, and the typical jury pool has 25-30 members. As a result, 36 percent of jury pools in our study had no black members and so, by construction [when a jury has been formed], there are no black jurors. The other 64 percent of jury pools included a small number of black members, resulting in black jurors being seated in some (but certainly not all) of these trials.
The results of our study indicated that racial composition of the jury has a large effect on conviction rates. In cases with no black members of the jury pool, black defendants were convicted 81 percent of the time, while white defendants were convicted 66 percent of the time. When the jury pool included at least one black person, the conviction rates were instead nearly identical: 71 percent for black defendants, 73 percent for whites. This large shift in conviction rates occurred even though jury selection still led to all-white juries in most of the cases in which there were black members of the jury pool.
Our study was the first to establish a strong causal link between the racial composition of real-world juries and conviction rates for both white and black defendants. In addition to what the results say about the effect of juror race, they also imply that there is a great deal of arbitrariness in trial outcomes — randomness in who happens to be called for jury duty for a given trial has a substantial effect on the outcome. The extent of this arbitrariness and the large role that race plays in decision-making raises serious questions about the basic fairness of jury trials as they are currently conducted in these jurisdictions.
THE FIX: It's sometimes valuable to establish the basic facts. What's wrong with seating an all-white jury? Does that fundamentally hamper equal justice? And how do prosecutors even accomplish this?
BAYER: In the United States, jury systems are generally designed to be representative of the eligible local population. There are two broad issues related to jury selection and composition that affect the fundamental fairness of jury trials in this kind of system.
The first is related to the system of peremptory challenges, which allow attorneys on each side to strike a number of potential jurors during pretrial jury selection without cause or justification. [This] results in juries that are unrepresentative of the local population. While the Supreme Court has prohibited the use of race as a rationale for using peremptory challenges, numerous studies have shown that black members of the jury pool are systematically more likely to be excluded from juries in many contexts. It is generally not especially difficult for attorneys to provide a non-race related explanation (if needed) to justify the use of a peremptory challenge, even if the juror's race is at least part of the basis for the attorney's decision.
A second and broader issue is whether juries that are representative of the local population can impartially decide cases when the defendant (or victim) is a minority member of the local population. In our study, for example, the vast majority of juries have no blacks members not because the attorneys are seating white jurors disproportionately — in fact, white and black jurors are seated at roughly the same rate — but because the local population is only 4-5 percent black.
In both instances, the lack of inclusion of minority members of a local population on juries raises concerns about whether such juries can actually reach unbiased decisions. Concerns are heightened by results like the ones from our study, which imply that the racial composition of juries does, in fact, play a large role in trial outcomes.
THE FIX: Among your findings, was there anything that truly surprised you?
BAYER: The direction of the findings was consistent with our hypothesis: That defendants of each race are less likely to be convicted when the jury has more members of the same race. The magnitude of the findings, however, was really striking, implying that even a small degree of inclusion of black jurors makes a large difference for conviction rates.
Keep in mind that we are using random variation in who is called for jury duty, so the cases that face an all-white jury pool are statistically identical (i.e. have the same objective quality of evidence) to the cases that face a jury pool with a small number of black members.
THE FIX: Could you explain what it is about all-white juries, or what is it about the dynamics of an all-white jury considering the fate of a black defendant that contributes to these outcomes? Or can you not jump to those conclusions?
BAYER: Unfortunately, our study provides little direct insight into the dynamics of jury decision-making. Our findings demonstrate that (randomly) changing the composition of the jury has a large impact on conviction rates but does not tell us exactly why or how this happens.
We also need to be careful about drawing any conclusions about what conviction rates for white and black defendants should be in these counties, as we have no direct measures of the quality of the evidence in the cases that are brought to trial against defendants of each race. If prosecutors bring a similar set of cases to trial for white and black defendants, impartial conviction rates should be identical. But if, for example, prosecutors bring weaker cases to trial against black defendants knowing that they will face all-white juries the majority of the time, impartial conviction rates should be lower for black versus white defendants. Unfortunately, our study does not provide a definitive answer on this.
THE FIX: What's known about the way that all-white juries decide when a white defendant is accused of killing or somehow harming a black victim (the police officer's acquittal in Baltimore brings this to mind)? Or, the impact of an all-black jury considering a case that involves a white victim and black defendant?
BAYER: Unfortunately, we know very little about the effect of the racial composition of juries on trial outcomes. Our study was conducted in a setting in which blacks constitute a small proportion of the local population, which means that the variation is primarily between all-white juries and those with the inclusion of a very small number of black jurors. I am not aware of any study of the impact of all-black juries using real world data.
THE FIX: Your study focused on a decade of non-death penalty cases in Florida between 2000 and 2010. That's a substantial stretch of time. But, some readers will wonder, how applicable are your findings to the rest of the country?
BAYER: We certainly need a lot more research on this topic in jurisdictions throughout the country. A broader set of studies would provide more evidence on whether our results generally hold in similar circumstances or whether they are special to these jurisdictions in Florida. Additional research could also provide evidence on whether certain rules and procedures for jury trials lead to the greater diversity of seated juries and/or less arbitrary trial outcomes.
The setting that we studied is an important one in the sense that one might be most concerned about the basic inclusion of minority members of a population on juries in settings in which the minority group constitutes a small proportion of the local population and, therefore, can more readily be systematically excluded from juries. In more racially diverse jurisdictions, on the other hand, the use of peremptory challenges by attorneys on each side is more likely to cancel or balance out, resulting in more racially diverse juries. [Editor's note: Most Americans live in communities that score high on the racial segregation index — meaning all but a very small share of most Americans' neighbors are the same race.]
It is also worth pointing out that it is generally very difficult for researchers to access data on jury composition and trial outcomes. So, a wider set of studies will really only be possible if courts throughout the country show a greater interest in transparency and in conducting rigorous studies of the efficacy of the trials conducted in their jurisdiction.
THE FIX: What could be done in individual courtrooms, or in the process of building jury pools to assure greater jury diversity?
BAYER: There are a number of potential policy remedies that courts could use to address issues related to jury diversity.
First, to ensure that juries more accurately reflect the diversity of the local population, the number of peremptory challenges provided to each side could be limited and the use of such challenges could be given greater scrutiny to ensure that they are used in a way that is consistent with Batson v. Kentucky, the Supreme Court case that prohibits race as a basis for peremptory challenges.
In addition, while it is very difficult to establish whether race is a consideration in the use of a peremptory challenge on a case-by-case basis, statistical analysis of attorney behavior over a large set of cases could be used to scrutinize, for example, whether prosecutors systematically strike black potential jurors in cases with black defendants. The privilege of using peremptory challenges in a jurisdiction could be tied to the continued demonstration that such challenges are used in a race-neutral way.
Second, in settings in which minority groups constitute a small fraction of the local population, jurisdictions could design the jury summons process to increase the likelihood of including members of the minority group on juries. While I am not aware of any jurisdiction in the country that currently does this, such a step could be justified as a requirement for basic fairness in jurisdictions where statistical analysis demonstrates that the racial composition of juries has a substantial impact on trial outcomes.
The American court system currently has very few protections to assure that race is not used in the selection of juries. To successfully argue that jury selection was racially biased, defendants typically need to provide a smoking gun, such as the prosecutor notes that indicated selection was race-based the Foster v. Chapman case that was decided by the Supreme Court this week. It is extremely rare for any court in the United States to do any systematic analysis of jury selection and trial outcomes to ensure the essential fairness and efficacy of jury trials.