THEORETICAL GROUNDING: RACIAL DIVERSITY AMONG JUDGES

There is presently a widespread normative commitment to racial diversity and representation across the United States from private industry to government institutions, including those comprising the carceral system. However, scholarship in this area, which usually focuses on Black and white officials’ behavior, often lacks consistent estimates of a relationship between individual government officials’ racial identities and their behavior. With respect to judges and police officers, for instance, some studies point to differences in Black and white judges’ sentencing behavior (Cohen and Yang Reference Cohen and Yang2019; Scherer Reference Scherer2004; Welch, Combs, and Gruhl Reference Welch, Combs and Gruhl1988). Other studies highlight a lack of difference in judicial behavior based on judges’ racial identities (Abrams, Bertrand, and Mullainathan Reference Abrams, Bertrand and Mullainathan2012; Schanzenbach Reference Schanzenbach2005; Spohn Reference Spohn1990), while others, still, find that Black judges are especially punitive (Steffensmeier and Britt Reference Steffensmeier and Britt2001). The findings on the role of police officer race in policing outcomes are similarly mixed (see, Ba et al. Reference Ba, Knox, Mummolo and Rivera2021; Eckhouse Reference Eckhouse2018; Legewie and Fagan Reference Legewie and Fagan2016; Sharp Reference Sharp2014). I argue that increasing the representation of racial-minority-group members in the judiciary decreases disparities in felony sentencing, not because of these judges’ behavior, but because their presence alters their peers’ behavior.

Analyses of the effect of diversity (of any kind) on institutional outcomes must conceptualize it accurately. Differences in behavior based on personal characteristics is one component of identity-based diversity, but evaluations of this component, alone, cannot capture the full impact of shifts in diversity on institutional outcomes. An individual cannot be “diverse.” Individuals contribute to a group’s diversity, but a large group with one or two non-majority-group members is still relatively homogenous, for example. Broadening the conception of what diversity is allows us to better understand how it might influence institutional outcomes. The variety of identities represented in an institution are a contextual feature of the institution, and the context in which individuals work is important for explaining individuals’ behavior and institutional outcomes.

Culture and norms also influence institutional context, and racial inequality is an enduring feature of American society. White and non-white Americans have different experiences, including their interactions with the state. This is true across a range of settings, including education, social services, and the criminal legal system. Those who are not white also remain under-represented among institutional officials. While co-racial descriptive representation sometimes results in substantive representation, the actions of individual officials (in legislatures, bureaucracies, or courts) are not enough to address deep-seated inequality. Additionally, attempts to understand the effect of racial diversity among officials on institutional outcomes that focus solely on the behavior of non-majority-group-member officials may fail to capture all the ways that racial diversity can influence institutional outcomes. If, for example, criminal legal officials of different races have different preferences, those preferences may have a weaker influence on their on-the-job behavior than the culture of their institution, which has been shown to influence judicial behavior (Abrams et al. Reference Abrams, Galbiati, Henry and Philippe2022). And institutional culture will be influenced by internal and societal power dynamics that typically favor majority-group member preferences (Legewie and Fagan Reference Legewie and Fagan2016; Ward, Farrell, and Rousseau Reference Ward, Farrell and Rousseau2009).

As members of minority racial groups gain representation in homogenous institutions dominated by majority-racial-group members, the altered context can influence the decision making of individuals from all racial groups. In these instances, we may expect institutional members’ decisions to reflect resulting shifts in the institution’s power balance and societal norm commitments. Scholars have identified a number of reasons to explain why white governmental officials’ (especially, those working in institutions that are products and producers of racial inequalities) decisions might be biased against non-white individuals, including implicit bias and political threat (Greenwald, Oakes, and Hoffman Reference Greenwald, Oakes and Hoffman2003; Legewie and Fagan Reference Legewie and Fagan2016; Ward, Farrell, and Rousseau Reference Ward, Farrell and Rousseau2009). Non-white government officials working in white-dominated institutions may also face pressures to make decisions that are generally more favorable to majority-group members (Allport Reference Allport1955; Steffensmeier and Britt Reference Steffensmeier and Britt2001).

Social psychological studies of the effect of group diversity on group and individual judgment and decision making suggest some pathways through which racial diversity in government institutions might affect individuals’ decisions and, therefore, group outcomes. First, the introduction of individuals that increase a group’s diversity may change the type of information exchange that occurs, thereby influencing the knowledge and considerations individuals rely on when making decisions (Jehn, Northcraft, and Neale Reference Jehn, Northcraft and Neale1999; Sommers Reference Sommers2006). The general process here is that as racial diversity in a group increases, the content of group conversations will include information related to race and racism that leads individuals in the group to more positively evaluate minority-racial-group members, such that future decisions will be more favorable to members of that racial minority group (Sommers Reference Sommers2006). This pathway might similarly influence institution members who are in the majority and the minority racial group. However, there are additional ways that racial diversity within a group can impact individuals’ behavior, and these pathways may differ with the race of the individual in question.

Officials who are members of a majority racial group, especially in jobs that require them to make decisions or judgements about others may make decisions that are less biased against minority racial groups as they gain more colleagues who are minority-racial-group members for three reasons. First, seeing and interacting with a more racially diverse set of colleagues may make majority-racial-group institution members desire not to appear discriminatory against out-group members or biased in favor of in-group members in their behavior. Second, it could remind majority-group-member officials of their own desire to avoid prejudiced behavior (Antonio et al. Reference Antonio, Chang, Hakuta, Kenny, Levin and Milem2004; Hans and Vidmar Reference Hans, Vidmar, Kerr and Bray1982; Sommers Reference Sommers2006). Third, a racially diverse group of colleagues may change the extent to which race is a salient consideration in their decision making.

Additionally, scholars have found that while inter-group contact can often lead to negative outcomes, including “exclusionary attitudes” and discrimination toward out-group members (Forbes Reference Forbes1997), repeated, long-term, and positive inter-group contact may lead to more positive outcomes, including decreasing out-group prejudice (Allport Reference Allport1955; Enos Reference Enos2014; Selvanathan et al. Reference Selvanathan, Techakesari, Tropp and Barlow2018). Officials have repeated contact with their colleagues and these interactions are likely positive, at least when compared with the negative sentiments they are more likely to associate with civilian interactions, whether these interactions are with defendants, new drivers at the Department of Motor Vehicles, or benefits applicants. Overall, we might expect these positive experiences with peer racial-out-group members to improve majority-racial-group member officials’ perceptions and treatment of minority-racial-group members.

For non-white institution members, the pressures of tokenism may subside as more minority-racial-group members join an institution and their under-representation within that institution becomes less extreme. When there are low levels of minority-group representation in an institution, minority-group members of the institution may experience tokenism, and feel pressure to conform to the majority-groups’ typical behavior (Asch Reference Asch1956; Steffensmeier and Britt Reference Steffensmeier and Britt2001). The pressures associated with tokenism occur due to the scrutiny and visibility that come from being a member of an extreme minority group (Cook and Glass Reference Cook and Glass2015; Kanter Reference Kanter1977; Sommers Reference Sommers2006; Yoder Reference Yoder1991). As the pressures alleviate, minority-racial-group members may feel more able to exhibit behavior that is favorable to members of their racial group, more broadly. They may also prioritize their own or their in-group’s well-being over the commitments of a racially homogenous majority-racial-group-member-led institution. However, the shifts in racial diversity in many U.S. institutions may not be large enough to alleviate these pressures, thus preventing any change in minority-racial-group-member officials’ behavior as they gain additional minority-racial-group-member colleagues.

I have presented many pathways through which racial diversity in institutions may influence individuals’ behavior, but, taken together, they all suggest that this feature of group context will lead majority- and, potentially, minority-racial-group members to behave in ways that make institutional outcomes less punitive and less disparate. Importantly, all of the presented pathways are interactive in nature. In order for racial diversity to influence decision making in institutions, institution members must be able to see and interact with those colleagues whose presence increases diversity within the institution.

Criminal trial courts are a useful lens through which to study the processes outlined above. First, trial court judges hear and decide cases alone, usually in their own courtrooms. However, they see and interact with colleagues in the hallways, elevators, and parking garages. Some criminal trial courts even assign judges to working groups in which judges can discuss the work of the court and receive communication from upper-level officials. Second, the work judges do is the type we would expect to be susceptible to shifts in institutional context related to race. Judges make judgements, and these are the types of individual decisions where social psychologists have found that individuals are susceptible to group diversity. In fact, many of these findings come from studies of juries, including the role of jury diversity on individual juror’s pre-deliberation decisions (Sommers Reference Sommers2006). Additionally, trial court judges make decisions in a relatively high-discretion, high-information environment, the type where the representative bureaucracy literature suggests passive representation may have substantive effects (see discussion in Baumgartner et al. Reference Baumgartner, Bell, Beyer, Boldrin, Doyle, Govan and Halpert2021, 862).

I derive the following expectations about the relationship between racial diversity and sentencing in trial courts based on the theory presented above. In general, I expect increases in racial diversity among judges to lead to decreases in punitiveness and decreases in race-based sentencing disparities. Focusing on Black and white judges, due to theoretical relevance and the nature of the data, which I discuss below, I expect increases in Black judicial representation will lead to decreases in the probability of incarceration and sentence length.

As the court becomes more racially diverse, I expect white judges to become more lenient toward Black defendants due to the processes described above. Such a behavioral shift will decrease punitiveness, overall, since the vast majority of criminal defendants are Black. If white judges become more lenient in their sentencing of Black defendants, this will also decrease the Black–white gap in sentencing, bringing Black defendants’ sentences closer in line with those of white defendants. It is possible that white judges will adjust their sentencing of white defendants as well, rendering slightly more punitive sentences to these defendants, which would also decrease disparities in sentencing. Black judges may similarly become less likely to incarcerate Black defendants as they gain Black colleagues. However, it is unclear, how many Black colleagues would be necessary in order to alleviate pressures of tokenism Black judges may experience in situations where they are extremely underrepresented. Further, it is also possible that harsher sentences for Black defendants might reflect Black judges’ true preferences, such that gaining Black colleagues might have no effect on or reinforce that preference (see, Forman Jr Reference Forman2017). Finally, because the theory implies at least some interaction among colleagues, I only expect to find a relationship between racial diversity among judges and judges’ sentencing decisions when judges work in the same location as those judges whose presence increases racial diversity on the bench.

While much of the research on racial diversity among judges has focused on how individual judge’s racial identities influence their decisions, some studies have conceived of racial diversity more broadly, lending support to the present argument. Studies of three-judge U.S. Courts of Appeals panels show that one judge’s characteristics can affect how the other two judges vote and, ultimately, the panel’s final decision (Cox and Miles Reference Cox and Miles2008a; Reference Cox and Miles2008b; Kastellec Reference Kastellec2013). The evidence from federal district courts is weaker. Ward, Farrell, and Rousseau (Reference Ward, Farrell and Rousseau2009) find that increased Black prosecutorial, though not judicial representation, is associated with decreases in sentencing disparities between Black and white defendants in federal judicial districts while Schanzenbach (Reference Schanzenbach2005) finds weak relationships between racial representation in judicial districts and judges’ sentencing.

Previous studies in this area represent important advances in the study of the relationship between institutional context and outcomes. However, they do not investigate the process by which shifts in individual judges’ decisions lead to these institutional outcomes. Additionally, the analyses that follow use data from a large lower-level state trial court, the type of court where the vast majority of criminal cases are heard in the United States highlighting the potential scope and importance of this relationship.

THE CIRCUIT COURT OF COOK COUNTY, CRIMINAL DIVISION

The data used in this article come from the Circuit Court of Cook County in Illinois, one of the largest unified court systems in the United States. The Circuit is composed of six municipal districts, with the First Municipal District covering the entire city of Chicago. A chief judge presides over the Circuit, which includes nonjudicial offices and the judiciary. The Criminal Division is responsible for hearing most of the county’s felony cases.

Criminal Division judges hear cases in three courthouses: the George N. Leighton Criminal Court Building in Chicago and the Skokie and Bridgeview Courthouses in suburban Cook County.Footnote ² Felonies originating (occurring) in the city of Chicago are heard in the city’s Leighton Criminal Building, and felonies originating outside of the city limits are heard by a Criminal Division judge in either Skokie or Bridgeview. The Criminal Division judges in the suburbs share courthouses with the other judges of their municipal districts who hear many different types of cases. The Division’s presiding judge and approximately 30 trial judges have courtrooms in Chicago’s Leighton Building, and there are usually six to seven Criminal Division trial judges with courtrooms in the Skokie and Bridgeview courthouses.

The distinction between the Chicago and suburban courthouses highlights a key point of the theoretical argument presented in the previous section. In order for racial diversity among judges to impact individual judges’ behavior, judges must see and, potentially, interact with the colleagues whose presence increases racial diversity on the bench. Judges sharing courthouses with judges who hear different, noncriminal, matters or those who work in courthouses alone will not have the opportunity to respond to shifts in racial diversity among their divisional colleagues, nor will defendants standing before those judges experience any related changes in sentencing behaviors.

The theory developed here can help us understand judicial behavior in criminal trial courts serving locations in which members of a minority group are over-represented among potential defendants and underrepresented among the judiciary, which is the case in much of the United States. However, the theory is likely most applicable to courts serving relatively large, densely populated areas where many judges have courtrooms in a single courthouse. This is common in counties with large cities like Chicago, including Philadelphia County, PA; Harris County, TX; and Los Angeles County, CA and in counties with medium- and smaller-sized cities like Middlesex County, NJ and Macomb County, MI, for example. Courts in these types of locations handle a significant portion of felony cases in the United States. According to the FBI’s Uniformed Crime Reporting over 40% of arrests occur in cities with at least 50 thousand residents, and these figures do not even account for the arrests in those cities’ surrounding counties that will be heard in the same court systems (FBI 2019). To provide more convincing evidence of the argument’s scope, I conducted analyses of court data from Harris County, TX (included in the Supplementary Material), and the results support those presented below.

DATA

The Cook County, IL, Criminal Division data used for this project include just over 431,000 felony cases heard from 1995 to 2013 in which the defendant was not found not guilty (Harris Reference Harris2023).Footnote ³ According to the 2013 Annual Report of the Illinois Courts, only 3% of felony defendants were found not guilty, suggesting the data used in this analysis capture nearly the universe of felony cases.Footnote ⁴ Cases missing due to a not guilty verdict are likely not missing at random and could introduce bias into analyses. However, judges only sentence defendants in cases in which the defendant pleads guilty or is convicted.

Each case is identified by a case number, and includes the following information: initiation date; defendant’s last name, date of birth, race, and sex; judge’s name; description of charge; felony class of the convicted charge; final disposition; length of minimum and maximum sentence; courthouse and number of the courtroom where the case was heard. The defendant in the vast majority of these cases—just under 76%—was Black, while white and Latinx defendants were charged in 11 and 12% of cases, respectively. As shown in Table 1, the defendants in these cases are mostly Black, male, and young, and the defendants in 46% of these cases were sentenced to incarceration for some period.

Table 1.

Case Characteristics across Locations

Note: Mean values with standard deviations in parentheses.

Felony charges are charges for which the defendant is eligible for at least 1 year of incarceration. Felonies are categorized into classes based on severity, which are then used to guide sentencing decisions. All felony charges—except those in the two most serious felony classes, which were dropped from the data—used in these analyses are eligible for probation in lieu of incarceration.Footnote ⁵

Overall, incarceration sentences have decreased over time (Figure 1). Additionally, the frequency of incarceration varies with defendant race. Cases with Black defendants end with incarceration sentences far more frequently than those with either Latinx or white defendants, and cases with Latinx defendants end with incarceration sentences slightly more frequently than those with white defendants.

Figure 1.

Criminal Division Annual Incarceration Rate

There were a total of 89 Criminal Division judges during this period, with approximately 38 judges serving in the Division each year—six to seven in the suburban courthouses and the rest in Chicago’s Leighton Building. Each judge has their own courtroom. As shown in Table 2, each judge’s courtroom hears an average of roughly 772 cases per year and Black and non-Black judges hear cases with similar characteristics (Table 3).Footnote ⁶ Black judges do preside over cases involving more Black defendants and fewer older defendants. However, if Black judges were more lenient than white judges, the main concern with respect to case characteristics would be that this was because Black judges received cases more likely to qualify for a lenient sentence, which does not appear to be the case. The minimum cases per judge per year in Chicago and Bridgeview was only one, likely because, over the course of the study period, some courtrooms no longer heard felony cases.Footnote ⁷

Table 2.

Cases per Judge per Year

Table 3.

Case Characteristics by Judge Race

Note: Mean values with standard deviations in parentheses.

I used judges’ names to merge the case data with biographical information from local legal organizations, Chicago Appleseed and Chicago Council of Lawyers. I also conducted Internet searches to identify judges’ race and gender when necessary and possible. The vast majority of judges, for whom I could identify race and gender, are white (79%) and male (79%). However, the racial composition of the Criminal Division bench has fluctuated over time.

I measure racial diversity on the bench as the percent of Black judges in the Criminal Division each year.Footnote ⁸ Throughout this period, only approximately 3% of judges could be identified as having a racial or ethnic identity other than Black or white. The percent of judges on the bench who are Black each year, shown in Figure 2, increased sharply after 1996, after which it decreased and then steadily rose from 2000 to 2004. Black judges’ representation in the Division decreased again in 2005, then experienced mild fluctuations throughout the rest of the study period. Both Black and white judges sentence Black defendants to incarceration more frequently than white defendants, and white judges are somewhat more punitive than Black judges, overall. Black judges issue incarceration sentences in 45% of cases with Black defendants and 32% of cases with white defendants, and white judges issue incarceration sentences in 50% of cases with Black defendants and 34% of cases with white defendants.

Figure 2.

Percent of Criminal Division Judges Who Are Black and White

JUDICIAL DIVERSITY AND SENTENCING

This section describes the empirical approach to evaluating the relationship between judges’ colleagues’ characteristics and judges’ sentencing decisions. I use the data described above to estimate the following linear probability and ordinary least squares regression models:

(1) $$ \begin{array}{l}\hskip-0.5em Sentence={\beta}_0+{\beta}_1(\%Black\hskip2.77695pt judges)\\ {}\hskip4.4em +{\beta}_2(Black\hskip2.77695pt defendant)+{\beta}_4Z+\zeta +\varepsilon, & \end{array} $$

(2) $$ \begin{array}{l}\hskip-0.1em Sentenc{e}_{ij}={\beta}_0+{\beta}_1(\%Black\hskip2.77695pt judge{s}_{ij})\\ {}\hskip4.5em +{\beta}_2(Black\hskip2.77695pt defendan{t}_{ij})+{\beta}_4{Z}_{ij}+{W}_j+\zeta +\varepsilon, & \end{array} $$

(3) $$ \begin{array}{l}\hskip-0.5em Sentenc{e}_{ij}={\beta}_0+{\beta}_1(\%Black\hskip2.77695pt judge{s}_{ij})\\ {}\hskip4.7em +{\beta}_2(Black\hskip2.77695pt defendan{t}_{ij})\\ {}\hskip4.6em +{\beta}_3(\%Black\hskip2.77695pt judge{s}_{ij}\times Black\hskip2.77695pt defendan{t}_{ij})\\ {}\hskip4.6em +{\beta}_4{Z}_{ij}+{W}_j+\zeta +\varepsilon, \end{array} $$

where i and j represent a case and a judge, respectively. The dependent variable in most of the models is dichotomous, taking a value of 1 if a defendant in a case is incarcerated to prison or jail and 0 otherwise. In a few of the models, the dependent variable is sentence length, measured in days. The main independent variable is the percent of Criminal Division judges who are Black each year, measured either across all three courthouses comprising the Division, or just in Chicago’s Leighton Building. Each model also accounts for the race of the defendant and includes an indicator for whether the defendant in a case is Black (1, 0). Equation 1 is a pooled model, whereas Equation 2 and Equation 3 include judge-level fixed effects ( $ {W}_j $ ) to account for unobserved factors affecting an individual judge’s decisions and the likelihood that their decisions across cases may be related. All models include standard errors clustered at the judge level to account for correlation between an individual judge’s decisions and a time trend ( $ \zeta $ ) to address broader trends that may impact the court’s work.Footnote ⁹

The percent of Criminal Division judges who are Black, shown in Figure 2, increased through 2004 after which it decreased then leveled off. It is important to note that while defendants’ cases are randomly assigned to judges in Chicago’s Leighton Courthouse (though not in the suburban courthouses), racial diversity on the bench is not randomly assigned across cases. However, the random assignment of defendants’ cases to judges should ensure that case assignment is not systematically related to defendant characteristics or any individual judge’s sensitivity to the level of racial diversity among their colleagues.

Z is a set of defendant, case, and judge characteristics. These include defendant race (binary indicators for Latinx and “other race” in addition to Black), age (in years), and gender (a binary indicator for whether the defendant is recorded as “female” in court records). It accounts for the seriousness of the conviction class, measured ordinally from least to most serious of seven classes (three misdemeanor classes and four felony classes). Z also includes binary indicators for whether a judge is up for retention at the time they hear a case and, in some of the pooled models includes, binary indicators for whether a judge is a Democrat, judge gender, judge race, and whether the judge was on the bench the previous year (to account for turnover among the judges).

Equation 1 tests the expectation that increases in Black judicial representation should decrease judges’ punitiveness in criminal sentencing. $ {\beta}_1 $ represents the change in the probability that a case will end with an incarceration sentence when Black judicial representation increases by 1 percentage point. Negative values on $ {\beta}_1 $ mean that as Black judges gain more representation on the court, sentencing becomes more lenient. Equation 2 is similar to Equation 1, but it incorporates judge-level fixed effects ( $ {W}_j $ ). Equation 3 evaluates whether judges’ responses to racial diversity within their group of colleagues varies with defendant race. This model interacts the indicator for when the defendant in a case is Black with the main independent variable, such that $ {\beta}_1 $ represents the change in the probability of an incarceration sentence (or the change in sentence length) associated with a 1-percentage point increase in Black judicial representation in cases with white defendants while $ {\beta}_3 $ represents the additional impact on sentencing for cases with Black defendants.

DISCUSSION AND CONCLUSION

Diversity with respect to background, ideology, gender, and race can greatly alter any environment, and the analyses presented above show that there is much at stake when considering racial diversity among trial court judges. First, the impact of such diversity extends beyond the behavior of any one judge whose identity might make the bench more racially diverse. Second, racial diversity among members of the judiciary has the potential to do more than instill a sense of institutional legitimacy among those most likely to appear in criminal courts as defendants (Scherer and Curry Reference Scherer and Curry2010); it may have the ability to impact the lives and freedom of individuals who interact with the institution directly and reduce incarceration. The results show that increasing the number of judges who look like the majority of defendants could increase the likelihood that defendants of different races receive more similar sentences and reduce incarceration. Finally, these findings have important implications for the struggle to address mass incarceration, disparate outcomes for non-white people throughout the carceral system, and demands for and implications of diversity (along multiple dimensions) within institutions.

This article moves beyond previous scholarship evaluating the relationship between racial diversity among judges and judicial decision making by providing an individual judge-level analysis of this relationship, understanding that each judge’s response will be different and conditioned by their own identity. Though the analyses cannot definitively determine the mechanisms, they do provide some support that positive, repeated contact with Black colleagues and/or the increased salience of racism in the criminal legal system drives white judges’ responses to increases in racial diversity among their colleagues.

If social pressure to exhibit behavior that is acceptable to an increasingly diverse group drives white judges’ behavior, then we might expect them to sentence less disparately only, or especially, when that diversity is most noticeable—when Black judges’ have courtrooms nearby. In this situation, white judges may have no choice but to see Black colleagues frequently in the hallway, for example, and, therefore, be reminded that these colleagues are nearby and potentially observing them. The results of a floormate analysis (included in the Supplementary Material) do not suggest that this is the case. Rather, white judges appear to become more punitive toward Black defendants when they have many Black floormates.

Results from the main analyses, however, support the idea that judges respond to racial diversity among colleagues at the courthouse level, suggesting that shifts in the racial identities represented in the judiciary alter the context in which judges work and how they approach their jobs. For white judges, this could be because, over time, as they are exposed to more Black colleagues and have the opportunity to interact with them, they either update their personally held beliefs about race or the role of racism in the criminal legal system becomes more salient to them. However, further disentangling the nuances of this relationship requires additional research. Observations of court operations could be especially helpful to identify how and where judges interact. It is possible that some interactions that cannot be measured with the type of data used here might create the type of social pressure that would lead white judges to alter their sentencing behavior. For example, on some courts, judges are assigned to teams in which they discuss the court’s work and receive communication from the presiding judge. Interviews with judges and court staff may also illuminate how judges interact with their colleagues, the types of relationships they develop, and how those interactions and relationships influence their work.

This article presents evidence from analyses of state trial court data. State trial court judges handle the vast majority of felony cases in the United States, and the factors most likely to contribute to the observed relationship between racial diversity among judges and sentencing are, by no means, unique to Cook County, IL. The theory and potential mechanisms underlying the relationship suggest that a similar relationship between racial (or ethnic) diversity among judges and disparity reduction in criminal sentencing could emerge in settings where there are large racial disparities in sentencing, and members of the racial group most frequently incarcerated are relatively underrepresented among the judiciary. These conditions are met in many places throughout the United States, and analyses of data from Harris County, TX (included in the Supplementary Material) yield similar results to those reported here.

Black people are more frequently incarcerated than white people in every state in the United States.Footnote ¹³ The judiciary remains mostly white in every state except for Hawaii and also the District of Columbia. The Black–white incarceration disparity will obviously vary within a state, as will Black judicial representation. However, counties where this disparity is greatest and where Black judges are more likely to have and seek seats are those where populations are usually most racially diverse. These might be counties that are more urban rather than rural, for example, and that tend to be more densely populated and have higher crime rates. Such a description applies to very many urban (and suburban) areas in the in the United States.

Throughout this article, I have focused on the Black–white sentencing disparity and on Black and white judges, which deserve close attention considering Black Americans’ increased likelihood of contact with all segments of the criminal legal system, and Black Americans’ general lack of representation on the judiciary. But future research might consider the relationship between Latinx judicial representation and sentencing outcomes for Latinx defendants, especially in settings where there are disparities in sentencing between Latinx and white individuals, Latinx judges remain in the minority, and Latinx judges are likely to have and seek seats on the bench.

The Latinx–white incarceration disparity is not as stark as the Black–white disparity, but Latinx individuals are still more frequently incarcerated than white individuals throughout the United States. This is true in New Mexico, Arizona, and Colorado, three states which also have significant representation of non-white people on the Bench.Footnote ¹⁴ These are prime examples of states with criminal courts that might produce results similar to those presented here, but for the sentencing of Latinx defendants.

Beyond what they suggest about judicial decision making and criminal legal outcomes, a major contribution of the findings is how they inform our conception of racial diversity in institutions. This article reinforces the importance of acknowledging that diversity is a group-level characteristic, a definitionally and analytically important distinction. An individual cannot be “diverse,” and reducing evaluations of racial diversity to comparisons of non-white and white individuals’ behavior reinforces the belief that white individuals’ behavior represents an accepted baseline while failing to capture all the ways racial diversity might influence institutional outcomes. In fact, this article shows that, in trial courts, racial diversity is most influential on white judges’ behavior. These findings suggest a new focus for efforts to increase racial diversity in institutions, which have typically focused on inclusion and representation—the potential to shift institutional power balances and alter behavior among majority-group members.