research report State Supreme Court Diversity Across the Country, Courts Fail to Reflect the Racial, Ethnic, and Gender Diversity of the Communities They Serve By Laila Robbins and Alicia Bannon With Malia Reddick PUBLISHED JULY 23, 2019 Brennan Center for Justice at New York University School of Law Table of Contents Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 ABOUT THE BRENNAN CENTER FOR JUSTICE Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 I. Nationwide, State Supreme Court Benches Are Overwhelmingly White and Male . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 II. A History of Discrimination and Structural Hurdles . . . . . . . . . . . . . . . . . 9 III. Does a State’s Method of Judicial Selection Impact Diversity? . . . . . 13 1. How Justices First Reach the Bench. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2. How Justices Fare After They Reach the Bench . . . . . . . . . . . . . . . . . . . 17 IV. A Closer Look at Electoral Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 1. Fundraising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2. Spending by Interest Groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 3. Additional Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 The Brennan Center for Justice at NYU School of Law is a nonpartisan law and policy institute that works to reform, revitalize — and when necessary defend — our country’s systems of democracy and justice. The Brennan Center is dedicated to protecting the rule of law and the values of constitutional democracy. We focus on voting rights, campaign finance reform, fair courts, ending mass incarceration, and preserving our liberties while also maintaining our national security. Part think tank, part advocacy group, part cuttingedge communications hub, we start with rigorous research. We craft innovative policies. And we fight for them — in Congress and the states, in the courts, and in the court of public opinion. Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 ABOUT THE BRENNAN CENTER’S Endnotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 D E M O C RACY P RO G RA M The Brennan Center’s Democracy Program encourages broad citizen participation by promoting voting and campaign finance reform. We work to secure fair courts and to advance a First Amendment jurisprudence that puts the rights of citizens — not special interests — at the center of our democracy. We collaborate with grassroots groups, advocacy organizations, and government officials to eliminate the obstacles to an effective democracy. © 2019. This paper is covered by the Creative Commons Attribution-NonCommercial-NoDerivs license. It may be reproduced in its entirety as long as the Brennan Center for Justice at NYU School of Law is credited, a link to the Center’s web pages is provided, and no charge is imposed. The paper may not be reproduced in part or in altered form, or if a fee is charged, without the Center’s permission. Please let the Center know if you reprint. Brennan Center for Justice State Supreme Court Diversity Foreword C ourts have tremendous power. Both state and federal judges make decisions that affect virtually every facet of our lives, from voting rights to educational equity, from disability rights to immigrant justice, from abortion access to safety in the workplace. And having a bench that reflects and represents diverse perspectives is core to achieving a system of justice that lives up to its name. Yet the lack of diversity in our nation’s courts is at crisis level. President Donald Trump’s nominees to the federal courts are the least diverse in decades. As of early June 2019, none of President Trump’s 46 appellate court nominees are African American, none are Latinx, and only nine are women. Currently, the Seventh Circuit Court of Appeals has no judges of color, while the Fifth Circuit has no active Latinx judges. In the Eighth Circuit, President Trump nominated four white men, despite there being only one woman and one person of color on a bench with 17 active and senior judges. Similarly, President Trump’s district court nominees are overwhelmingly white and male. This is a troubling deviation from nearly a century of nominations. Presidents Jimmy Carter and Barack Obama made diversity on the federal bench a priority — and now, President Trump is the first Republican president since Herbert Hoover to nominate fewer women and people of color to the federal bench than the previous Republican president. As Justice Thurgood Marshall said, we condemn the courts to “one-sided justice” when we deprive the legal process of “differing viewpoints and perspectives on a given problem.” And this whitewashing of the courts comes at a time when the legal profession has more female attorneys and attorneys of color than ever before. Representation matters not only for the paths of future lawyers, but also for the communities who depend on our judicial system to affirm their lived experiences and recognize injustice from the perspective of many — not the isolated perspective of one. Extensive research has also shown that including a broad range of viewpoints in the judiciary enriches deliberations, fosters better-informed decisions, and enhances public confidence in our system of justice. But the lack of judicial diversity is not exclusive to the federal judiciary. State Supreme Court Diversity vividly shows that state supreme courts, which are generally the final word on state law, do not reflect the diversity of the communities they are supposed to serve. White men compose a disproportionate majority of state high court judges, while women and people of color are dramatically underrepresented. By some measures, over the past 20 years state high courts have become less reflective of an increasingly diverse America. This not only narrows the pipeline of potential nominees to the federal bench but also threatens the legitimacy of state courts in the eyes of the communities whose rights they are charged to protect. Most people who interact with the legal system do so in state courts. And as the federal courts become increasingly inhospitable to civil rights, state courts are poised to become even more important venues for litigation. This report should spark critical dialogue at a time when those in our nation’s highest offices seek to erode the rule of law and silence communities of color. Diversity on the bench is not a panacea. Differing perspectives do not themselves ensure that courts will recognize and protect our civil and human rights. But promoting decision-making that includes voices from our nation’s diverse communities will help foster an equitable system of justice. All people in the United States deserve a system of justice that realizes its name. But today, at both the state and federal levels, courts lack the rich experience of our diverse communities. Building diverse and representative state and federal benches should be an urgent priority for advocates and lawmakers. We must demand more from leaders who have the power to build benches as vibrantly diverse as our nation. Vanita Gupta President and Chief Executive Officer, The Leadership Conference on Civil and Human Rights State Supreme Court Diversity Brennan Center for Justice 1 Introduction A mid growing recognition of disparities in America’s justice system, this report highlights a critical but under-scrutinized problem: the lack of racial, ethnic, and gender diversity on state supreme court benches across the United States. State supreme courts, which sit atop state judiciaries, do not typically garner the same attention as the U.S. Supreme Court, but they hold substantial power. As a whole, state courts hear 95 percent of all cases filed in the United States.1 State supreme courts generally provide the final word in interpreting state law and set precedents that bind more than 23,000 lower state court judges.2 In recent years, state supreme courts have reversed multi-million dollar verdicts in commercial disputes,3 struck down restrictive abortion laws,4 and ordered hundreds of millions of dollars in additional funding for education5 — all as matters of state law. Drawing on nearly 60 years of data, we looked at who has been empowered to don a robe and sit on these powerful courts. A few numbers begin to tell the story: ƒƒTwenty-four states currently have an all-white supreme court bench, including eight states in which people of color are at least a quarter of the state’s general population.6 ƒƒOnly 15 percent of state supreme court seats nationwide are held by individuals who are Black, Asian, Latino, or Native American — though nearly 40 percent of the nation’s population are people of color.7 ƒƒSince at least 1960, the earliest year for which we were able to obtain comprehensive data, 13 states have never seated a person of color as a justice, and six more states have only had one justice of color. More than a third of all states — 18 in total — have never seated a Black justice. ƒƒWomen hold only 36 percent of state supreme court seats. Currently, 17 states have only one female justice on their supreme court bench. By some measures, state supreme courts are less reflective of the United States’ diversity than they were a generation ago. The gap between the proportion of people of color on the supreme court bench and their representation in the U.S. population was higher in 2017 (the most current year for which we have available population data) than it was over two decades ago, in 1996. This deficit of diversity among judges threatens the legitimacy of the judiciary in the eyes of the communities it serves. As former Ohio Supreme Court Justice 2 Brennan Center for Justice Yvette McGee Brown observed, “The public’s perception of justice suffers . . . when the only people of color in a courthouse are in handcuffs.”8 This is particularly so in light of the vast racial disparities in the American criminal justice system,9 where 1 in 3 Black men are incarcerated in their lifetimes, compared with 1 in 17 white men.10 A 2015 National Center for State Courts survey of public confidence in state courts found a “massive racial gap” in trust in the fairness of the courts, revealing a “deep distrust of courts among African Americans.”11 An absence of judicial diversity also limits the perspectives available to inform judicial deliberations, undermining state courts’ ability to develop a legal jurisprudence for an increasingly diverse America.12 Research shows that judicial diversity enriches judicial decision-making,13 promotes public confidence in the judiciary,14 and establishes role models across demographic groups.15 Many factors drive this, including a long history of racial and gender discrimination and inequities in access to law schools and the legal bar (see Part II, “A History of Discrimination and Structural Hurdles”). Women and people of color continue to be underrepresented in the legal profession, and prior research has found that structural barriers — including implicit and explicit biases, disparities in access to mentoring, and unequal work assignments — impact their advancement into leadership positions in the law, which can in turn impact who reaches the bench.16 Using new data, this report breaks ground in analyzing how another factor may impact diversity on the bench: a state’s method of judicial selection. Our analysis reveals that judicial elections have rarely been a path for people of color to reach the supreme court bench. This is true even in states that use contested elections to choose their justices. In these states, interim appointments, which occur when a seat opens in the middle of a justice’s term, have been the principal path to the bench for justices of color — but not for white justices. (Our findings are limited to state supreme courts, and dynamics may differ in lower courts.) And as detailed in Parts III and IV, we find racial disparities in virtually every element of state supreme court elections, from who wins, to how frequently incumbent justices are challenged, to how much money candidates raise, to who is supported by special interest groups. Diversity on state supreme courts is under-studied and under-scrutinized, in part due to a lack of data: few states State Supreme Court Diversity collect this information or make it publicly available. This report helps fill that gap, documenting missing diversity in the makeup of state high court benches across the country. However, understanding the starkly inadequate number of people of color and women on state supreme court benches is only a first step. We hope these findings will add urgency to efforts to build and strengthen pipe- lines to law school and the bench for underrepresented communities, encourage reforms to make both judicial elections and appointments more open to a diverse set of candidates, and inform discussions about how states should choose their justices in the first instance. Building a more inclusive judiciary should be a priority — not just for our elected officials and law schools, but for all of us. Data Sources For a full description of our data sources, see Appendix 1. Most states do not collect and/or make available records of state supreme court justices’ race, ethnicity, or gender, making it difficult to compile such data. Our analysis relies heavily on a dataset compiled and generously shared by Greg Goelzhauser, a political scientist at Utah State University, that includes information on how each justice who joined a state supreme court between 1960 and State Supreme Court Diversity 2014 first reached the bench, the gender of each justice, and whether publicly available information indicated that the justice was a person of color (individuals who are Black, Asian American, or Latino). We supplemented the Goelzhauser dataset to include justices who were first named or elected to the bench between 2015 and 2019 and to include Native American justices when information was available. We also added the last year each justice sat on the bench, making it possible to create snapshots of the composition of state supreme courts over time. We also created a dataset comprised of all supreme court candidates who stood for election between 2000 and 2016, whether or not they won or lost. Race, ethnicity, and gender data was derived from secondary sources with hallmarks of credibili- ty, including biographical statements, obituaries, newspaper articles, and listed membership in affinity organizations. Fundraising data comes from the National Institute on Money in Politics. Estimates of television ad spending by outside groups was provided to the Brennan Center by Kantar Media/CMAG. Brennan Center for Justice 3 I. Nationwide, State Supreme Court Benches Are Overwhelmingly White and Male A cross the country, state supreme courts fail to reflect the diversity of the communities they serve. The disparities are stark: People of color compose nearly 40 percent of the U.S. population, but hold only 15 percent of state supreme court seats as of May 15, 2019. Women make up roughly half the U.S. population but hold only 36 percent of state supreme court seats. White men make up less than a third of the U.S. population, but they constitute 56 percent of today’s supreme court justices (Figure 1).17 Figure 1: Demographic Composition of State Supreme Courts as Compared to National Population National population (2017) State high courts (2019) 56% 30% 31% 29% 20% 19% 8% White Men White Women Men of Color 7% Women of Color Population data from the U.S. Census Bureau 2013–2017 American Community Survey 5-Year Estimates. State high court data collected by the Brennan Center as of May 15, 2019. White men and white women refer to non-Hispanic/Latino whites. These trends are borne out across the country. Only five states — California, Connecticut, Minnesota, North Carolina, and Oregon — have a supreme court bench where the percentage of people of color is higher than their representation in the state’s population as a whole.18 (See Appendix 2 for a state-by-state breakdown.) Twenty-four 4 Brennan Center for Justice states do not have a single person of color on their state supreme courts, as reflected in Figure 2. (Supreme court benches vary in size from five to nine justices.) This includes eight states in which people of color are at least a quarter of the state population. State Supreme Court Diversity Figure 2: States with All-White Supreme Courts, May 2019 WA MT ME ND MN OR ID NY WI SD CA AZ PA IA NE UT IL CO MA CTRI MI WY NV VT NH KS OK NM MO WV KY MD DE VA NC TN AR SC MS TX OH IN NJ AL GA LA AK FL HI State has no justices of color on its high court We use “white” to refer to non-Hispanic/Latino whites. Figure 3: General Population Demographics in the 24 States With All-White State Supreme Courts State Percentage People of Color State Percentage People of Color Nevada 51% Utah 21% Alaska 39% Idaho 18% Delaware 38% Wisconsin 18% Alabama 34% South Dakota 18% Rhode Island 27% Wyoming 16% Arkansas 27% Kentucky 15% Tennessee 26% North Dakota 15% Michigan 25% Montana 14% Kansas 24% New Hampshire 9% Pennsylvania 23% West Virginia 8% Indiana 21% Vermont 7% Nebraska 21% Maine 7% For full table see Appendix 2. Population demographics from U.S. Census Bureau, Population Estimates Program, July 2018. Supreme court diversity data as of May 2019. People of color includes individuals who are Black, Asian, Latino, Native American, or multiracial. State Supreme Court Diversity Brennan Center for Justice 5 Likewise, while there is regional variation with respect to racial and ethnic diversity within the United States,19 every region has a substantial gap between the proportion of people of color on the supreme court bench and the proportion of people of color in the general population (Figure 4). To calculate this gap, we used a methodology similar to one used in a recent study commissioned by the American Constitution Society, calculating how far the proportion of justices of color on each region’s supreme courts falls short of what would be predicted based on the representation of people of color in the region’s population.20 As of 2017, the most recent year for which population estimates are available from the U.S. Census Bureau’s American Community Survey, the South had the largest diversity gap: there were 72 percent fewer justices of color on supreme court benches in the South than would be predicted by the representation of people of color in the general population. By contrast, the Northeast had the lowest diversity gap at 46 percent. Unfortunately, data limitations mean that we cannot provide a more detailed description of racial and ethnic diversity on the bench, such as the proportion of justices who identify as Latino, or address the representation of other historically marginalized communities, such as Figure 4: Gap Between Population Diversity and Court Diversity by Region, 2017 72% 69% 66% 46% Northeast West Midwest South These figures represent each region’s diversity gap: what percentage fewer justices of color are on the state supreme court bench than would be predicted base on the proportion of people of color in the general population. Higher percentages indicate a greater racial diversity gap. Figure 5: Gender Diversity on State Supreme Courts, May 2019 WA MT ME ND MN OR ID NY WI SD MI WY UT CA AZ PA IA NE NV IL CO KS OK NM TX OH IN MO WV KY MA CTRI NJ MD DE VA NC TN AR SC MS AK VT NH AL GA LA FL HI 10% 20% 30% 40% 50% 60% 70% Share of women on state supreme court 6 Brennan Center for Justice State Supreme Court Diversity justices who identify as LGBTQ. More research and better data are required to provide a more complete picture.21 States across the country also have a lack of gender diversity on their state supreme court benches, as reflected in Figure 5. These disparities are particularly severe for women of color. Thirty-three states do not have a single female justice of color. You are more likely to encounter a white justice named John, David, or James on today’s high court benches (31 combined) than you are to meet a female justice of color (24 in total). Female justices are also concentrated in just a handful of states. In 17 states, there is currently only one woman on the supreme court bench. In 38 states, women make up less than half of the supreme court bench. Among white female justices, 23 percent sit in only five states. The lack of diversity on today’s supreme court benches is even more striking in historical context. Since at least 1960, the earliest year for which we were able to obtain comprehensive data, more than a quarter of all states (13) have never had a person of color serve on their high court, and six states have only seated one justice of color. More than a third of all states — 18 in total — have never seated a Black justice.22 And over the past 20 years, two states (Louisiana and Rhode Island) have not had any new female justices join their high court benches, and seven states have only had one.23 Troublingly, by some measures, state supreme courts have actually become less reflective of an increasingly diverse American population in recent years. While there are more sitting state supreme court justices of color today than there were two decades ago, proportionally, state high courts have become even less reflective of the nation’s changing demographics. From 1996 to 2017, the gap between the proportion of people of color on state high court benches as compared with the overall national population increased. In 1996, there were 63 percent fewer justices of color on state high court benches than would be predicted based on their representation in the general population. In 2017, the most recent year for which population estimates are available from the U.S. Census Bureau’s American Community Survey, the gap was 66 percent.24 Several individual states’ supreme courts have also become less racially diverse in absolute terms in recent years, a trend that is most pronounced in states that use contested judicial elections to choose their justices.25 For instance, Michigan’s high court bench was more racially diverse in 1962 than in May 2019.26 As early as 1986, two of seven justices on Michigan’s supreme court were people of color, while today, the state’s supreme court bench is Figure 6: States With No Justices of Color Since at Least 1960 WA MT ME ND MN OR ID NY WI SD MI WY UT CA AZ PA IA NE NV IL CO VT NH KS OK NM MO WV KY NJ MD DE VA NC TN AR SC MS TX OH IN MA CTRI AL GA LA AK FL HI State high court has not had a justice of color since at least 1960 See Appendix 2 for the corresponding proportion of each state’s general population composed of people of color. State Supreme Court Diversity Brennan Center for Justice 7 all white. And in Pennsylvania, two of seven justices were people of color in 1988, including the nation’s first Black female state supreme court justice, Juanita Kidd Stout. Today, Pennsylvania’s high court is all white. Similarly, at the start of 2000, two of nine seats on Alabama’s state high court were held by Black justices. That November, both were ousted in an election. For the past 18 years, Alabama’s high court bench has been all white, even though people of color currently make up a third of the state’s population. One potential explanation for the reduction in racial diversity on the bench in many states that use contested elections is that during the past two decades, supreme court elections have become increasingly costly and politicized.27 It is possible that under these circumstances, it has been more challenging for candidates of color to wage successful campaigns. We closely examine racial disparities in judicial elections in Part IV of this report, finding that from 2000 to 2016, on average, candidates of color lost more often, raised less money, and were challenged more frequently than white candidates in supreme court elections. Unfortunately, due to data limitations, we cannot determine whether these racial disparities are more pronounced than in the pre-2000 era, indicating an area warranting further study. Figure 7: States that Have Never Had a Black Justice WA MT ME ND MN OR ID NY WI SD MI WY UT CA AZ PA IA NE NV IL CO KS OK NM TX OH IN MO WV KY MA CTRI NJ MD DE VA NC TN AR SC MS AK VT NH AL GA LA FL HI State high court has never had a Black justice 8 Brennan Center for Justice State Supreme Court Diversity II. A History of Discrimination and Structural Hurdles T he lack of diversity in state judiciaries has deep roots, as women and racial minorities were excluded, both formally and informally, from practicing law and serving in the judiciary for much of American history.28 In 1873, for example, the U.S. Supreme Court ruled that Illinois could deny law licenses to women.29 In 1890, the California Supreme Court denied a Chinese immigrant named Hong Yen Chang the right to practice law in the state because he was barred from becoming a citizen under the Chinese Exclusion Act.30 And African Americans were excluded from Texas’s state law school until as late as 1950, when the U.S. Supreme Court struck down the discriminatory practice. At the time the lawsuit was filed, there were no law schools in the entire state that admitted African Americans.31 Not surprisingly, state supreme courts long remained the province of white men. The first Black supreme court justice, Jonathan Jasper Wright, was appointed to the South Carolina Supreme Court in 1870 during the shortlived Reconstruction era. No other Black justice reached the high court bench in any state until 91 years later, in 1961.32 The first woman to sit on a state supreme court bench was Florence Ellinwood Allen, in 1922; another woman did not reach a state supreme court bench until 1959.33 The first Black woman to reach a state supreme court bench was Juanita Kidd Stout in Pennsylvania in 1988.34 The persistence of racial and gender disparities on today’s state supreme courts reflects myriad and complex factors. In Part III, this report provides new evidence about how a state’s method of selecting judges may impact diversity on the bench, showing that elections have rarely served as a path to the bench for people of color. But prior research and accounts from judges and judicial candidates suggest a host of factors that contribute to the lack of diversity on today’s supreme court benches — of which the method of selection is only one. Among other factors, there continues to be a lack of racial and gender diversity among people with the professional experiences that have historically served as a pathway to the judiciary — and an emphasis in judicial selection on career patterns that have traditionally been the province of white men.35 Women, for example, have outnumbered men in law schools since 201636 and as early as 1985 made up 40 percent of enrolled law students.37 Yet the number of women holding high-prestige leadership positions within the legal profession has not reflected these trends. As of 2018, women make up 36 percent of active attorneys.38 But according to a 2018 report from the American Bar Association’s Commission on Women in the Profession, they are only 23 percent of law firm partners, 26 percent of Fortune 500 general counsels, and 32 percent of law school deans.39 Notably, several empirical studies have State Supreme Court Diversity found that having a larger pool of female attorneys in a state does not have a statistically significant relationship to the proportion of women serving on state supreme courts, underscoring that having a diverse pool of lawyers does not necessarily translate into a diverse judiciary.40 The judicial pipeline narrows even earlier for people of color, who continue to be dramatically underrepresented in the legal profession as a whole, making up an estimated 15 percent of lawyers nationwide, according to the American Bar Association.41 People of color are further underrepresented in many leadership positions that have often served as a path to the bench. For example, while people of color made up about 20 percent of first-year law students as early as 199342 (and were 31 percent of first year law students in 201843), in 2018, only 9 percent of all law firm partners were people of color.44 Women and people of color are also underrepresented among judges on state trial and intermediate appellate courts.45 Why does the leadership pipeline narrow? Prior research suggests that implicit bias, harassment, and discrimination have been among the hurdles that women and people of color face in advancing in the legal profession, along with disparities in access to legal networks and mentorship.46 For example, a recent survey of women and lawyers of color working at law firms and as in-house counsel, commissioned by the ABA’s Commission on Women in the Profession and the Minority Corporate Counsel Association, found that women and people of color reported higher levels of bias than white men with respect to equal opportunities for everything from the quality of assignments, to networking opportunities, to performance evaluations and promotions.47 And in a 2014 study, law firm partners asked to review a legal memorandum found more errors and gave it a lower rating when they were told it was authored by a Black (as opposed to white) man, comporting with social science literature finding “leniency bias” favoring “in-groups” in employment settings.48 Of course, even though women and people of color Brennan Center for Justice 9 Figure 8: Racial Diversity on State Supreme Courts, Snapshots from 1989 to 2019 1989 WA MT ME ND MN OR ID NY WI SD IL CO CA AZ KS NM OH IN MD DE WV MO VA KY NC TN OK NJ PA IA NE UT MA CTRI MI WY NV VT NH AR SC MS AL GA LA TX AK FL HI 0% 10% 20% 30% 40% 50% 60% 70% 1999 WA MT ME ND MN OR ID NY WI SD MI WY UT CA AZ PA IA NE NV IL CO VT NH KS OK NM MO WV KY NJ MD DE VA NC TN AR SC MS TX OH IN MA CTRI AL GA LA AK FL HI 0% 10% 20% 30% 40% 50% 60% Share of People of Color on State Supreme Court The maps for 1989, 1999, and 2009 divide the number of justices of color who sat on a state’s high court bench in a given year by the total number of justices who sat on that high court bench that year. In some instances, the number of justices who sat on a state’s bench in a given year may exceed the number of seats on a state’s high court bench, such as when a justice retires and a replacement is named 10 Brennan Center for Justice State Supreme Court Diversity 2009 WA MT ME ND MN OR ID NY WI SD PA IA NE IL UT CO CA AZ MA CTRI MI WY NV VT NH KS MD DE WV MO VA KY NC TN OK NM OH IN NJ AR SC GA AL MS LA TX AK FL HI 0% 10% 20% 30% 40% 50% 60% Share of people of color on state supreme court 0% 10% 20% 30% 40% 50% 60% 70% 2019 WA MT ME ND MN OR ID NY WI SD MI WY UT CA AZ PA IA NE NV IL CO VT NH KS OK NM MO WV KY NJ MD DE VA NC TN AR SC MS TX OH IN MA CTRI AL GA LA AK FL HI 0% 10% 20% 30% 40% 50% 60% Share of People of Color on State Supreme Court the same year. Given data limitations, we were only able to examine data at the aggregate year-level, rather than by individual month; as a result, these percentages reflect aggregate state-level diversity over a given year. The 2019 map reflects the state high court bench as of May 15, 2019. State Supreme Court Diversity Brennan Center for Justice 11 are underrepresented among leadership positions within the legal profession, in every state there are many women and people of color who could serve with distinction to fill state supreme court seats. Research suggests several other factors that may also deter such candidates from seeking judicial positions, including low salaries49 and a lack of formalized mechanisms for recruiting candidates from diverse backgrounds.50 Political dynamics may also affect the extent to which a given state’s high court is likely to reflect the diversity of the communities it serves. For instance, several studies have found a relationship between a state’s dominant political ideology and racial and gender diversity on the bench, with states with more “liberal” ideologies more likely to have diverse courts when controlling for other factors.51 Likewise, studies suggest that the number of justices on a court and the frequency with which court seats open have a relationship with judicial diversity.52 12 Brennan Center for Justice State Supreme Court Diversity III. Does a State’s Method of Judicial Selection Impact Diversity? T his report provides new evidence regarding another factor that may affect diversity on the supreme court bench: a state’s method of choosing judges. To select supreme court justices, most states use either competitive elections or appointments by the governor after vetting by an independent nominating commission, often known as “merit selection.”53 (These appointments are often followed by periodic retention elections where a justice stands for an up-or-down vote.54) We provide new evidence that, at least at the supreme court level, the vast majority of justices of color first reached the bench via judicial appointments. Judicial elections have rarely been a path to the supreme court bench for people of color, and they have often been a path off the bench for incumbents of color. The question of whether judicial elections or appointments yield more diverse benches has loomed large in reform debates over how to respond to the growing politicization of supreme court elections, including whether states should do away with elections altogether. (In a 2018 report, the Brennan Center urged states to end supreme court elections, replacing them with a publicly accountable appointment process with input from a diverse and independent nominating commission.55) In particular, some minority bar associations and civil rights groups have previously expressed skepticism about ending judicial elections because of concerns about the potential impact on judicial diversity.56 Historically, studies aiming to assess whether a state’s method of judicial selection affects diversity on the bench have yielded mixed findings.57 One of the hurdles to assessing the impact of judicial selection methods on diversity has been the complexity of states’ judicial selection systems. For example, while 22 states use contested elections as part of their system for choosing justices, many of these states routinely rely upon interim appointments, where justices are temporarily appointed when a position opens in the middle of a term — and then enjoy an incumbent advantage in future elections.58 In fact, prior scholarship has found that approximately half of all state supreme court justices in contested election states first reached the bench via interim appointments.59 Thus, simply treating a State Supreme Court Diversity state as having an “elected” or “appointed” supreme court may miss critical nuance regarding the method by which individual judges actually reach the bench. To overcome this issue, we looked at how individual justices first reached the supreme court bench over a nearly 60-year period (1960–2018) — including whether justices in election states first reached the bench via an interim appointment.60 We also looked at how justices fared once on the bench. This more granular analysis tells a powerful story: people of color have consistently made up a higher proportion of appointed, as compared with elected, first-time supreme court justices. Incumbent justices of color have also disproportionately been challenged and lost elections once on the bench, as compared with incumbent white justices. By contrast, by most measures, women have fared similarly under both elective and appointive methods (without controlling for race). Our findings are consistent with some prior academic studies that also looked at the method by which individual justices reached the bench and sought to control for other factors that could explain racial or gender disparities in order to isolate the effect of the selection method. These studies found that appointive methods were more likely than elections to place people of color onto the supreme court bench, and that the relationship was statistically significant. No such relationship existed for women.61 Brennan Center for Justice 13 What About Lower Courts? >> This report’s analysis is limited to diversity on state supreme courts and does not address how judicial selection may affect the racial or gender composition of lower court benches. Indeed, at least one prior study that looked at a sample of lower court judges found that appointments produced more racial diversity at the supreme court level, while at the lower court level, the method of selection was not related to racial diversity on the bench. These findings suggest that dynamics may differ based on court level.62 >> State supreme courts differ from lower courts on many dimensions: there are far fewer judges, as an institution they are precdent-setting and have more power to shape a state’s 1. How Justices First Reach the Bench legal and policy landscape, and they are typically higher-prestige positions. In states that use elections, state supreme court candidates’ experiences also frequently differ from those of lower court candidates in important respects. For example, state supreme court candidates usually run in statewide elections, while lower court judges run in smaller jurisdictions. State Figure 10: People of Color as a Percentage of All Initially Appointed or Elected Justices, by Time Period Supreme court elections have rarely been a path to the bench for people of color — a fact that is driven both by a candidate pool that lacks racial diversity and by disparities in success rates for white candidates and candidates of color. Initially appointed justices – % people of color Initially elected justices – % people of color Elections have rarely been a path to the bench for people of color From 1960 to 2018, a total of only 17 justices of color first reached the bench through an election, comprising 4 percent of initially elected justices. During the same time period, 141 justices of color were appointed to the bench, comprising 12 percent of all appointed justices (Figure 9). Only 6 women of color first reached a state supreme court bench through an election, making up 1 percent of all initially elected justices. Women of color made up 3 percent Figure 9: People of Color as a Proportion of All Initially Elected and Initially Appointed Justices, 1960–2018 Justices of Color White Justices 4% 12% Initially Elected 96% 14 Brennan Center for Justice Initially Appointed 88% supreme courts have also seen a proliferation of million-dollar elections, while most lower court elections are far less costly. For all these reasons, our findings may not be generalizable to all court levels; we encourage further research on diversity on lower court benches. 18% 13% 5% 6% 5% 0% 1960–79 1980–99 2000–18 of all appointments. By contrast, female justices have been first elected and first appointed to the bench in roughly equivalent proportions (without controlling for race).63 Judicial appointments, as compared with elections, have also been the principal selection method responsible for the increase in racial diversity in state supreme courts over recent decades. The proportion of people of color among initially appointed justices has increased over time. By contrast, the proportion of people of color among initially elected justices has declined. As reflected in Figure 10, from 2000 through 2018, only 5 percent of justices first elected to the bench were people of color — a lower percentage than in 1980 through 1999, when 6 percent of justices first elected to the bench were people of color. During the same period, the proportion of people of color among justices first appointed to the State Supreme Court Diversity Figure 11: People of Color as a Percentage of All Initially Elected and Initially Appointed Justices, 1961–2018 Initially appointed justices — % of people of color (raw numbers in parentheses) 29% (10) 31% (4) Initially elected justices — % of people of color (raw numbers in parentheses) 21% (8) 18% (6) 9% (4) 9% (1) 18% (7) 17% (3) 18% (8) 13% (7) 11% (2) 19% (6) 5% (2) 1961–62 6% (1) 67–68 73–74 79–80 85–86 bench grew from 13 percent to 18 percent. Remarkably, more justices of color were first elected to the bench in the 1990s than in the 2010s. And from 2003 to 2015, not a single person of color first reached the bench through an election. By contrast, there has not been a single cycle since 1963–64 where there were no justices of color among initially appointed justices. People of color have also made up a higher proportion of initially appointed justices as compared to initially elected justices during every election cycle since 1965–66, with the exception of 2017–18 (Figure 11). While in absolute numbers the 2017–18 cycle saw more people of color appointed than elected, there was both an unusually high percentage of initially elected justices of color and a low percentage of appointed justices of color relative to other recent cycles. (We analyze the 2017–18 election cycle in greater detail in the box, “A Closer Look at the 2017–18 State Supreme Court Elections.”) When we limit our analysis to the 22 states that currently provide for contested elections to fill seats on their high courts, the results are even more striking (Figure 12). Since 1960 (the earliest year for which we have comprehensive data), more than half of all white justices in these states first reached the bench via an election (as compared to an interim appointment).64 Yet among justices of color in these same states, only one in five first reached the bench via an election.65 Since 1960, only 10 of the 22 states that currently use contested elections have State Supreme Court Diversity 91–92 97–98 2003–04 09–10 15–16 ever had a justice of color first reach the bench via an election. Eighteen of these 22 states have had a justice of color first reach the bench via an interim appointment.66 Racial disparities in candidate pools and rates of electoral success Why have so few people of color reached the bench via an election? Disparities in both the composition of the candidate pool and in rates of success are both part of the story. For nearly two decades, the proportion of supreme court candidates who are people of color (including both open seat candidates, challengers, and incumbent justices) has remained flat. As reflected in Figure 13, the percentage of candidates of color in 2017–18 (12.7 percent) was lower than the corresponding percentage in 2001–02 (12.9 percent).67 Candidates of color also have been less successful than white candidates in winning elections (Figure 14). In order to reach the bench via an election, a candidate must either challenge and defeat an incumbent justice or win an open seat. In both types of races, candidates of color have lost more often than white candidates. From 2000 to 2016, only five candidates of color won state supreme court elections as nonincumbents, while 46 candidates of color lost. The flat candidate pool and lower success rates among people of color may also be interrelated, as racial disparities in success rates may dissuade potential judicial candidates from running for supreme court seats. Brennan Center for Justice 15 Figure 12: Contested Election States: People of Color as a Percentage of All Initially Elected or Appointed Justices, 1960–2018 Initially elected justices — % people of color Initially appointed justices — % people of color 50% 40% 30% 20% l) Illi no is M ich ig an M Te on xa ta s( na Cr im in al *L ) ou isi an a No Ore go rth Ca n ro lin Ne a w M ex ico M in ne so ta io ivi (C Oh as Te x a ns ylv an W ia as hi ng t M iss on iss ip pi Ne va d Ge a or gi W isc a on W s es t V in irg in ia Ke nt uc ky Id No rth aho Da ko ta ab am Pe n Al Ar ka ns a s 10% * Data for Louisiana is incomplete. Figure 13: Supreme Court Candidate Demographics, 2000–2016 Candidates: Percentage White Population: Percentage People of Color Population: Percentage White Candidates: Percentage People of Color 90.2% 88.1% 69.1% 2000 88.3% 89.5% 87.1% 88.4% 88.1% 87.6% 88.1% 67.9% 66.7% 65.4% 63.7% 62.8% 61.9% 61.1% 60.6% 36.3% 37.2% 38.1% 38.9% 39.4% 13.8% 11.6% 12.8% 13.1% 12.7% 32.1% 30.9% 9.8% 90.4% 12.9% 2001–02 10.4% 2003–04 33.3% 34.6% 12.5% 10.5% 2005–06 2007–08 2009–10 2011–12 2013–14 2015–16 2017–18 Population data from the U.S. Census Bureau. Yearly data was unavailable for 2001-02; the dotted line reflects a linear estimate, which connects the data point from 2000 to that from 2003-04. 16 Brennan Center for Justice State Supreme Court Diversity Figure 14: Success Rates Among Challengers and Open Seat Candidates in Supreme Court Elections, 2000–2016 Candidates of Color 36% White Candidates 19% 14% 3% Challengers Open Seat Candidates 2. How Justices Fare After They Reach the Bench Regardless of whether a state provides for appointments or elections to initially seat justices, the vast majority of states require sitting justices to stand for periodic election. Thirty-eight states provide for elections for subsequent terms on the supreme court bench, including 20 states that provide for contested elections,68 and 16 states where justices are first appointed and then must stand for a periodic up-or-down retention election. (Two more states provide for retention elections after an initial contested election.) Likewise, justices who first reach the bench via an interim appointment must later stand in a contested election to keep their seats.69 These elections have frequently been a path off the bench for incumbent justices of color. Analyzing every contestable supreme court election from 2000 to 2016, we found that incumbent justices of color were challenged more often and won less often than white incumbents. Male incumbents of color faced challengers the most often, and white men were challenged the least often of any demographic group (Figure 15).70 And while incumbents of all races have higher success rates than challengers or open seat candidates in supreme court elections, incumbent justices of color have lost State Supreme Court Diversity more often than white incumbents in states that use contested elections (Figure 16). These figures are driven by men of color, who have the lowest incumbent success rate of any demographic group (77 percent). Similarly, while interim appointments have been the most common path to the bench for justices of color in states that provide for contested elections, appointed justices of color have been disproportionately likely to later lose their seats when forced to stand in an election. In contested election states, among justices who first reached the bench between 2000 and 2017 via an interim appointment and who later stood for election, 25 percent of justices of color were ousted in an election, compared with 12 percent of white justices. There is also anecdotal evidence that justices of color who were initially appointed may choose not to run in elections for additional terms out of concern that they will face disadvantages at the polls. For example, former Texas Court of Criminal Appeals Judge Elsa Alcala, who was first appointed by Gov. Rick Perry (R), expressed concerns along these lines after initial straw polls for her 2012 reelection campaign showed her 10 points below her opponent, Ken Law, despite her incumbency status and party support: “I thought to myself that if the choice was between the name ‘Ken Law’ versus the name ‘Elsa Alcala’ in front of an uneducated voter, that I was going to lose,”71 she said in an interview with the Texas Tribune. While Alcala did run for and win reelection after her opponent was forced off the ballot, she chose not to stand for reelection again in 2018 and said that her decision was influenced in part by concern that her Hispanic surname would be seen as a liability and help draw a Republican primary challenger.72 Not surprisingly given these dynamics, justices of color have served on the bench for fewer years on average than have white justices. This disparity is most pronounced in states that require incumbent justices to compete in contested elections (Figure 17).73 Figure 15: Proportion of Incumbents Who Faced a Challenger, 2000–2016 White Male Incumbents 58% White Female Incumbents 65% Female Incumbents of Color 71% Male Incumbents of Color 77% Data excludes retention elections. Brennan Center for Justice 17 Figure 16: Incumbent Success Rates, 2000–2016 Incumbents of Color 82% White Incumbents 90% Figure 17: Average Term Lengths (Years) on State Supreme Courts by Method of Reselection, 1985–2018 White Justices Justices of Color 11.7 Data excludes retention elections. 11.1 11.1 10.7 9.2 7.5 Contested elections for additional terms Retention elections for additional terms No elections for additional terms* *Includes states that provide for reappointment and states where justices have life tenure. 18 Brennan Center for Justice State Supreme Court Diversity A Closer Look at the 2017–18 State Supreme Court Elections >> In 2017–18, four justices of color (including three women of color) first reached the supreme court bench via an election, making up 31 percent of all initially elected justices that cycle. Seven justices of color first reached the bench via judicial appointments, making up 13 percent of all first-time appointed justices. This marks 2017–18 as only the third election cycle that saw more than one person of color first elected to the bench (1991–92 and 1983–84 being the other two). It is also the first time since at least 1960 that a higher proportion of first-time elected justices were people of color, as compared with first-time appointed justices. >> Does this indicate a new trend, with elections becoming a more regular path to the bench for people of color, or is it an anomaly best explained by the idiosyncrasies of particular races? It is difficult to draw clear conclusions from a single election cycle, particularly when the numbers involved are so small. The last time there was a similar electoral “wave” was in 1991–92, when three justices of color first reached the bench via an election. The years that followed did not see any sustained increase in first-time elected justices of color. >> Only future election cycles will provide a definitive answer, but there are several reasons to be cautious in concluding that 2017–18 was the start of a long-term shift: First, by several other measures people of color did not perform unusually well in 2017–18 supreme court elections. Overall, supreme court candidates of color had lower ƒƒ success rates in 2017–18 supreme court elections than in prior non-presidential election cycles. In 2017–18, 53 percent of all candidates of color (incumbents, challengers, and open seat candidates) won their races, compared with 56 percent of white candidates. This success rate for candidates of color was the lowest of any non-presidential election cycle since our data begins in 2000.74 number of candidates of color who won (8) was lower than that of the two prior cycles (9 and 10). >> Second, several of the first-time elected justices of color in 2018 had wins that suggest idiosyncratic circumstances or the effects of an election that in some states saw a Democratic wave. In North Carolina, for example, a Republican incumbent justice faced both a Republican and Democratic challenger in the general election, splitting the vote and handing the Democratic candidate of color a win with less than 50 percent of the vote. In New Mexico, every statewide office up for election in 2018 was won by the Democratic candidate, and the supreme court candidate of color who won in 2018 had previously lost a race for supreme court justice in 2016. In Texas, a Latina candidate who won both a Republican primary and the general election for a seat on the Texas Court of Criminal Appeals, Michelle Slaughter, did not have a Hispanic-sounding surname, avoiding the “surname challenge” that many observers within Texas have suggested disadvantages Latino candidates in typically low-information judicial elections.75 >> Finally, we also saw an unusually low proportion of people of color among appointed justices during the 2017–18 cycle, likely due in part to the political composition of governors’ seats, where Republicans held the highest number of seats since at least 1939, when available data begins.76 Historically, a higher proportion of justices of color ƒƒ have been appointed by Democratic governors than Republican governors: of the justices of color appointed by governors between 2000 and 2016, 67 percent were appointed by Democratic governors, and 33 percent were appointed by Republican governors. In comparison, of all justices appointed by governors between 2000 and 2016, 42 percent were appointed by Democratic governors, and 57 percent were appointed by Republican governors (a small number were appointed by third party or independent governors). Partisan differences in appointment rates were even ƒƒ Looking at all supreme court election winners, not just ƒƒ those who first reached the bench via an election, both the proportion and absolute number of winners who were people of color was not exceptional relative to prior cycles. In all, 12.1 percent of winners in 2017–18 were people of color (8 people), compared with 11.8 percent in 2015–16, and 13.3 percent in 2013–14. And the total State Supreme Court Diversity more pronounced during 2017–18, where Democratic governors were responsible for 83 percent of gubernatorial appointments of justices of color.77 By contrast, of white justices appointed by governors in 2017–18, 77 percent were appointed by Republican governors, while 23 percent were appointed by Democratic governors. Brennan Center for Justice 19 IV. A Closer Look at Electoral Dynamics O ur analysis makes clear that candidates of color and white candidates have had disparate experiences in supreme court elections, and that elections have only rarely been a path to the supreme court bench for people of color. In this section, we take a closer look at supreme court election dynamics. Focusing on elections from 2000 to 2016 — a time period in which, as the Brennan Center has documented in previous reports, supreme court elections have become increasingly high cost and politicized— we found numerous racial disparities, including in fundraising and support by outside groups.78 Together, these findings tell a more detailed story about the specific ways candidates of color may experience running for judicial office differently than white candidates. 1. Fundraising While raising more money does not mechanically translate into winning more votes, supreme court elections are increasingly multimillion-dollar campaigns that require substantial fundraising efforts from judicial candidates.79 Prior studies of supreme court elections have found that both challengers and initially appointed incumbents gained electoral advantages when they spent more money in their campaigns.80 Analyzing every supreme court election from 2000 through 2016, we found that candidates of color raised less money than their white counterparts when compar- ing challengers and open seat candidates (Figure 18).81 Among incumbents, the story was more complex. As reflected in Figure 19, although female incumbents of color were slightly out-fundraised by white female incumbents and white male incumbents, male incumbents of color raised more than any other incumbent demographic group. One likely explanation is that male incumbents of color have faced challengers more often than other demographic groups, comporting with prior research finding that incumbents raise more funds when they face well-resourced challengers.82 Our data indicates that male incumbents of color who faced opposition raised more funds than their white male, white female, and female of color counterparts, providing support for this explanation.83 We also found that women of color have often faced the greatest fundraising disparities. Between 2000 and 2016, women of color raised fewer funds than any other demographic group in both nonpartisan and partisan supreme court elections. Similarly, women of color competing as open seat candidates or incumbents raised less than their counterparts in any other demographic group.84 White women, by contrast, have generally had a very different Reporting Statistical Relationships >> Some of our analyses in this section rely upon statistical regressions to evaluate the relationship between different variables. In conducting regression analyses, social scientists control for factors that can influence a perceived relationship between two variables, such as temporal or geographic variation. After controlling for such factors, social scientists traditionally report results as statistically significant if they have a “p value” below 20 Brennan Center for Justice 5 percent — indicating that there is at least a 95 percent chance the reported relationship is attributable to the factor(s) tested for, rather than due to random chance. Social scientists have conventionally held that a p value below 5 percent is sufficiently robust to have confidence in the regression results. >> In instances where our regressions yielded p values below 5 percent, we report those findings as statistical- ly significant in the report text and provide the corresponding p value(s) in endnotes. Our full regressions are available in Appendix 5. We also report relationships with p values below 15 percent, with the corresponding p value in endnotes, though we do not characterize such relationships as statistically significant in text. A p value of 15 percent indicates that there is at least an 85 percent chance the reported relationship is attributable to the factor(s) tested for, rather than due to random chance. While our 15 percent threshold is above the conventional level of statistical significance used by social scientists, we included these analyses because given the small number of observations and limited time scope, our results are suggestive for future social science research and may be informative for policymakers and advocates. State Supreme Court Diversity fundraising experience. In contested elections, being a white woman was positively and statistically significantly related to the amount of funds raised.85 Similarly, we found that in contested partisan elections, women of color had less “purchasing power” for the money they raised — in other words, their vote shares86 did not benefit from additional funds to the same extent as other candidates.87 For candidates who raised an equivalent amount of funds, we found that being a woman of color reduced a candidate’s expected vote share by 4 percent.88 We found a similar (though smaller) relationship in contested elections for male candidates of color. For candidates who raised an equivalent amount of funds, we found that being a male candidate of color reduced a candidate’s expected vote share by 1 percent.89 Public Financing and Diversity >> Prior research and observations by candidates have suggested that public financing can increase the viability of otherwise qualified candidates who lack access to networks of wealthy donors — and can open the door for candidates who may have otherwise been discouraged from running in the first place.90 Multiple studies have also found that after establishing public financing, jurisdictions experienced increases in the racial and gender diversity of candidates for elected office.91 In the context of state supreme court elections, our data indicates substantial racial disparities in fundraising. Prior studies have also suggested that people of color perceive fundraising as a barrier to entry for competing in judicial elections.92 Figure 18: Average Non-Incumbent Fundraising, 2000–2016 Candidates of Color White Candidates $584,288 $337,265 $314,403 $154,815 >> During the time period covered by our data, four states provided for public financing of state supreme court elections: Wisconsin (2011), North Carolina (2004–2012), West Virginia (2012–present), and New Mexico (2008–present). Although the numbers involved were small,93 we found that among the candidates who were eligible, people of color opted into public financing more often than their white counterparts — a finding that is consistent with prior research that found women and people of color participate in public financing programs at higher rates than white male candidates.94 Women overall did not participate at higher rates than men, but the two female candidates of color who were given the opportunity chose to opt into public financing. Figure 20: Opt-In Rate Among Candidates Eligible for Public Financing, 2000–2016 Challengers Open Seat Candidates All monetary amounts are adjusted to 2016 dollars. Figure 19: Average Incumbent Fundraising, 2000–2016 Candidates of Color 83% White Candidates 61% Women of Color $422,505 White Men $430,543 White Women $463,742 Men of Color $650,122 All monetary amounts are adjusted to 2016 dollars. Table does not include retention elections. State Supreme Court Diversity Brennan Center for Justice 21 2. Spending by Interest Groups We also looked at outside spending by interest groups, which have become increasingly important players in state supreme court elections. In 2015–16, political action committees, social welfare organizations (also known as 501(c)(4)s), and other non-party groups spent a record $27.8 million on supreme court elections nationwide, making up 40 percent of all supreme court election spending.95 Because this outside spending is a relatively new phenomenon, and because these groups have tended to concentrate their attention on only a handful of races each cycle, we have only limited data to draw from in assessing any racial disparities. In total, between 2004 and 2016, groups took out negative television ads against 3 candidates of color and 61 white candidates, and positive ads supporting 5 candidates of color and 82 white candidates. Nevertheless, controlling for other factors,96 we found some indication that candidates of color may be less likely to attract positive ads from outside groups than white candidates. In the most common circumstances (i.e., with relevant factors held at their average values), groups ran ads that were positive in tone in contested elections 12 percent of the time. This frequency decreased to 4 percent of the time when male candidates of color were involved. Groups aired positive ads for white male candidates 12 percent of the time, and for white female candidates 18 percent of the time.97 When groups went on the attack, they also spent more on average opposing candidates of color relative to white candidates (Figure 21). However, this relationship did not approach statistical significance, and disparities Figure 21: Average Spending by Groups on TV Ads, 2004–2016 Candidates of Color Targeted White Candidates Targeted $641,990 $459,701 $377,186 Negative Ads $407,079 Positive Ads Average spending is by unique candidate. When applicable, spending was analyzed by unique candidate over multiple campaigns, rather than by unique candidate and by unique campaign. All monetary amounts are adjusted to 2016 dollars. 22 Brennan Center for Justice were largely fueled by outside group spending on a single election.98 Our corresponding data on the success rates of white candidates and candidates of color who faced group-run attack ads is similarly limited in scope. However, within this small sample, all three candidates of color who faced negative group-run ads from 2004 to 2016 lost their races.99 By contrast, half of all white candidates who faced negative group-run ads during this period (32 of 64) won their election. 3. Additional Factors Existing research also suggests other factors that may disadvantage people of color in supreme court elections, potentially contributing to the disparate success rates documented in this report. First, supreme court elections are typically low-information races. Research suggests that when voters know little about the candidates they are voting for, they may be more prone to draw on implicit biases and stereotypes to inform their decisions.100 One manifestation of this dynamic is in the “surname challenge” reported by candidates of color in some states, where having a surname associated with a particular racial or ethnic group can make it harder to win judicial races.101 Two quantitative studies that looked at judicial elections in Cook County, Illinois, for example, found that having an Irish surname increased a candidate’s likelihood of success.102 Illustrating how some surnames can be seen as a barrier in judicial races, one Asian American judge who sat in Cook County, Sandra Otaka, recounted in a 2005 report by the Lawyers’ Committee for Civil Rights Under Law that she was “told to put an apostrophe after my O because if I did that, I would have a greater chance at winning county-wide.”103 Another potential impediment for candidates of color is racially polarized voting, which occurs when voters of color and white voters have different candidate preferences.104 Racially polarized voting poses a particular concern for supreme court diversity because nearly all states elect their supreme court justices statewide, rather than via districts, meaning that minority-preferred candidates may be shut out of state high court seats entirely or face higher hurdles to stay competitive. A 2012 supreme court election in Washington State illustrates how the “surname challenge” can contribute to racially polarized voting dynamics, making it harder for some candidates of color to compete in supreme court elections. There, the state’s first Mexican American supreme court justice, Justice Steven González, who had originally been appointed to the bench, was challenged by a little-known attorney, Bruce Danielson. Danielson did not raise or spend any money on the race, while Justice González spent over $300,000. Despite Danielson’s lack of fundraising and González’s status as an incumbent, State Supreme Court Diversity Danielson won over 70 percent of the vote in some counties in Central Washington. While González ultimately held onto his seat, a study by a political scientist at the University of Washington found racially polarized voting in these regions that crossed party and ideological lines.105 The study concluded that “racial voting bias distorted the González-Danielson race in certain Washington counties.”106 After the election, Justice González reflected, “Frankly[,] I want voters to know the candidate they’re voting for and vote because of the candidate’s qualifications[,] not because of their reaction to a last name.”107 Other studies, often conducted in the context of litigation, have likewise identified racially Disparities in Prior Judicial Experience We also examined the proportion of state supreme court candidates from 2000 to 2016 who had prior judicial experience. Perhaps reflecting higher thresholds to entry, we found that among supreme court candidates, a far higher percentage of candidates of color had prior judicial experience than did white candidates, a disparity that was most pronounced for female candidates of color.A Female candidates as a whole were also more likely to have judicial experience than their male counterparts. polarized voting in lower court judicial elections in California,108 Texas,109 and Louisiana.110 Finally, several supreme court elections involving candidates of color have seen examples of racial appeals, which can have the effect of prompting and eliciting racial biases.111 (See box, “Racial Appeals in Supreme Court Elections.”) While we lack comprehensive data on the prevalence of such appeals, one possible proxy is the prevalence of attack ads that address criminal justice themes. Such ads frequently feature images of people of color as criminal defendants — sometimes juxtaposed against an image of the judicial candidate. Looking only at television ads run by groups in supreme court elections from 2004 to 2016, we found that 99.6 percent of negative ad spots targeting candidates of color involved criminal justice themes, where candidates were attacked as soft on crime. By contrast, only 50.1 percent of negative ad spots targeting white candidates had a criminal justice theme. These figures reflect only a small number of advertisements, but suggest potential racial disparities in the tenor and tone of negative ads in supreme court elections. Figure 22: Candidates with Prior Judicial Experience, 2000–2016 Female Candidates 79% Male Candidates 67% Candidates of Color 92% White Candidates 71% Female Candidates of Color 94% All Other Candidates 72% Note: Data only relevant to candidates in contestable elections, as all retention election candidates have prior experience. A  This finding is also consistent with research on judicial qualifications among federal judicial nominees. See Roger E. Hartley, “Senate Delay of Minority Nominees to the Federal Judiciary: A Look at Race, Gender, and Experience,” Judicature 84 (2001): 195; Rorie L. Spill Solberg and Kathleen A. Bratton, “Diversifying the Federal Bench: Presidential Patterns,” The Justice System Journal 26 (2005): 124. State Supreme Court Diversity Brennan Center for Justice 23 Racial Appeals in Supreme Court Elections Racial appeals have been factors in at least some supreme court elections involving candidates of color. While we lack comprehensive data on their prevalence, we highlight some notable examples. >> In North Carolina, civil rights attorney Anita Earls, who is biracial and whose father is Black, challenged and defeated incumbent Republican Justice Barbara Jackson in 2018 for a seat on the North Carolina Supreme Court.112 During the campaign, the state Republican Party funded a mailing and website opposing Earls’s candidacy that featured an image of Earls with a former client whom she represented in a civil rights challenge, in which Earls’s skin was “several shades darker than in real life,” with the headline “Dangerous Anita Earls.”113 >> During the same North Carolina race, North Carolina Republican Party Executive Director Dallas Woodhouse falsely claimed that Earls had gotten three defendants of color “off death row,” using Twitter to post photos of the defendants.114 Woodhouse later acknowledged that Earls had not represented any of the defendants. He said he was referring to her support of the state’s Racial Justice Act in 2009, under which the defendants’ sentences were later commuted.115 >> In Michigan, incumbent Republican Justice Robert P. Young and Democratic challenger Denise Langford-Morris, who are both Black, were each subject to attack ads involving racial appeals in their 2010 election.116 The Law Enforcement Alliance of America spent over $393,000 on a television ad attacking Langford-Morris, claiming she was “soft on crime for rappers.”117 The ad stated that “for a rapper facing a gun charge while on probation, Langford-Morris let him walk free.” The “rapper” referenced was Eminem, who is white — but the ad never referenced him by name, instead using the racially evocative term “rapper.”118 [Announcer]: Judge Denise Langford Morris has been soft on crime for rappers, lawyers, and child pornographers. 24 Brennan Center for Justice State Supreme Court Diversity by asking viewers to “Call Bob Young, the comedy >> On the other side, the Michigan Democratic Party ran a STSUPCT/MI MIDP NOconcluded LAUGHING MATTER judge, and tell him we’re not laughing anymore.” The only television ad featuring an image of Steve Urkel, a character Brand: DEMOCRATIC ST CNTRL CMT (B339) apparent similarity between Justice Young and Steve Urkel is from the sitcom Family Matters, juxtaposed with an image PARENT of Parent: UNKNOWN both are Black, male, and wear glasses. Ultimately, Justice Justice Young. The ad asked, “What if Urkel grew up and 10/27/2010that Aired: - 11/01/2010 CreativeCourt?” Id: 9133307 Young won reelection to the bench. became the comedy judge on the Michigan Supreme It [Announcer]: What if Urkel grew up and became the comedy judge on the [Announcer]: What if Urkel grew up and became the comedy judge on the... Michigan Supreme Court? Bob Young, but it's no laughing matter. Bob Young gutted the ...Michigan MichiganSupreme Court? Bob Young, but it’s no laughing matter. Bob Young gutted the Michigan... Consumer Protection Act. Bob Young ruled that Michigan citizens can't protect the environment. ...Consumer Protection Act. Bob Young rules that Michigan citizens can’t protect the environment. that particular case because “they were trying to find someone >> In Wisconsin, the first — and only — person of color to sit who looked as close to me as possible, so that they could on the Wisconsin Supreme Court, Louis Butler, was ousted in emphasize the fact that a Black justice was on the ballot.”120 2008 after his opponent ran an ad claiming that Butler “found a Notably, the ad juxtaposed Butler’s face with that of the loophole” to allow a Black defendant to commit another crime. criminal defendant. Outside special interest groups aired The ad was contrary to the facts — Butler was a public defender 119 Bob Young even close and off the line we case. can't walk our beaches anymore. Call thesides comedy judge, and tell60 him similar ads, alleging, “Butler with criminals percent of in the case, nottried thetojudge, heshore ultimately lostsothe of our Great Lakes, Bob Young, 121 Later, Justice Butler suggested that his opponent highlighted the time.” we're not laughing anymore. [Announcer]: Louis Butler worked to put... [PFB]: MICHIGAN DEMOCRATIC STATE CENTRAL COMMITTEE ...criminals on the street, like Reuben ...with learning disabilities. Butler found Lee Mitchell who raped an 11-year-old a loophole, Mitchell went on to molest... girl... Copyright 2010 TNS Media Intelligence/CMAG www.PoliticsOnTV.com ...another child. Can Wisconsin families feel safe with Louis Butler on the Supreme Court?... State Supreme Court Diversity 1-866-559-CMAG ...[Paid for by] Gableman for Supreme Court Brennan Center for Justice 25 Conclusion O ur courts are tasked with ensuring that everyone receives equal justice under the law. Diversity on the bench is critical to this task, promoting both the appearance and reality of a fair and impartial justice system. On this measure, state supreme courts overwhelmingly fall short. Alarmingly, we find that since 1960, a quarter of all states have never had a person of color on their supreme court — and that 24 state supreme courts are currently all white. Meanwhile, white men are vastly overrepresented on state supreme courts, making up a third of the population but 56 percent of today’s supreme court bench. And though the United States has become more racially diverse, we find that state supreme courts are lagging further and further behind the demographics of the communities they are supposed to serve. While the lack of diversity on today’s supreme court benches reflects complex causes, this report provides new evidence regarding one key factor: supreme court elections have overwhelmingly yielded white benches. We find that judicial appointments have disproportionately been the path to the bench for people of color, both in absolute numbers and proportionally. Most strikingly, this holds true even in states that provide for contested elections, where justices of color — but not white justices — have most often reached the bench via interim appointments, which occur when a seat opens in the middle of a judge’s term. We also found racial disparities in virtually every aspect of how candidates experience supreme court elections. Though candidates of color are more likely to have prior judicial experience, they raise fewer funds and win less often as challengers and candidates for open seats than their white counterparts. As incumbents, they are challenged more frequently and lose more often than white incumbents. Bias and racial appeals have also been factors in at least some supreme court elections involving people of color, distorting electoral results. Likely reflecting these dynamics, the proportion of people of color in the overall supreme court candidate pool has been virtually unchanged for two decades. These findings point to judicial elections as one key inflection point for addressing diversity on the bench. Fundraising disparities, for example, may make it harder for many candidates of color to win supreme court seats and discourage them from competing in the first place. Public financing is one proven mechanism for opening the door to otherwise qualified candidates who 26 Brennan Center for Justice lack access to networks of wealthy donors. Anecdotal evidence also suggests that a lack of information about judicial candidates may elicit and exacerbate racial biases among voters, suggesting the importance of investing in voter guides and public education. The findings in this report should also help allay concerns that abolishing state supreme court elections and moving to a judicial appointment system would harm judicial diversity. (This report does not examine lower court elections, which may have different dynamics.) Nor should appointment systems escape scrutiny, however. Neither elections nor appointments have generally produced courts that are reflective of the diversity of the communities they serve, and both systems can impose hurdles for diverse candidates. For states that use appointment systems, there is substantial research suggesting ways that judicial nominating commissions and governors can encourage diversity among the applicant pool for judgeships through their recruitment processes, as well as well-developed best practices for how to mitigate implicit biases in interviewing and evaluating potential judges.122 Our findings also underscore the need for more research and better data on judicial diversity, so that policymakers, advocates, and judges can better understand what hurdles currently exist. A diverse bench is crucial in securing an impartial and fair system of justice — and in achieving judicial legitimacy in the eyes of the communities our judiciary is intended to serve. But as this report makes clear, state high courts across the country are falling short in this critical task. We hope this report will prompt action and reform to address these vast inequities. State Supreme Court Diversity Appendix 1: Data Sources and Methodology T he Brennan Center created three datasets for this report: all candidates (winning and losing) who ran in a state supreme court election between 2000 and 2016, all state supreme court justices who served between 1959 and 2018 (including whether they were first elected or appointed to the bench and the years they served on the state high court), and a snapshot of every state supreme court as of May 15, 2019.123 For each dataset, we identified individuals’ race and gender based on several credible secondary sources. Greg Goelzhauser, a political scientist at Utah State University, generously shared a dataset he created, which includes information on how each state supreme court justice who joined a state supreme court between 1960 and 2014 first reached the bench, the gender of each justice, and whether each justice was white, Black, or a “person of color” (Black, Latino, and/or Asian American).124 We extended the dataset to include Native American justices, where data was available. Goelzhauser drew the names of most justices who reached the bench from 1990 through 2012 from a list compiled by Adam Bonica and Michael Woodruff (2015); to fill in the remaining years, Goelzhauser extended the list through 2014 and back through 1960, and coded each justice’s race and gender from biographical statements, “archived newspaper articles, law review articles, bar journals, obituaries, remembrances, law firm web pages, and biographical databases accessed through LexisNexis.”125 We expanded on this data by adding every justice who joined the bench between 2015 and 2018, every justice sitting on a high court bench as of May 15, 2019, every candidate who lost a supreme court election from 2000 through 2016, and every candidate who won reelection to the bench from 2000 through 2016. To classify these individuals’ race and gender, we referred to a number of sources, including individuals’ self-description in biographies, court-level statistics collected by the National Center for State Courts, listings in the Directory of Minority Judges, data collected by the American Bar Association, individuals’ affiliation with affinity bar associations, and news articles. In instances in which these sources did not indicate that a justice was a person of color, two researchers independently coded individuals’ race based on an analysis of available photographs. It is important to note that our figures are estimates, as we collected judges’ characteristics from publicly available sources. For every justice who joined the bench between 1960 and 2018, we also identified the last year they sat on the bench, as well as the membership of each state’s supreme court in 1959. This enabled us to determine the racial and gender composition of each state’s supreme court by year. State Supreme Court Diversity The year coded as the year each justice was named to the bench designates the year a nominee was named or the year they were elected to the bench — which can, in some instances, differ from the year they first sat on the bench. For instance, a justice who won an election in November of 2018 but did not begin sitting on the bench until January of 2019 would be coded as named to the bench in 2018. This creates some data limitations: the snapshots we provide of high court diversity over time are estimates, given that a justice may have been named to a court the year before they actually began sitting on the court. In all, our 2000–2016 dataset includes 1,069 individuals for whom we were able to identify race and gender. We were unable to identify the race of 45 (non-winning) candidates. For our 1959–2018 dataset, race and gender are identified for all 1,676 instances in which a justice joined the bench in that time period. It is important to note the many ways in which our data on diversity is limited. First, our full dataset only classifies individuals as people of color or white, so we are not able to discern how different racial or ethnic groups are represented in the judiciary, or whether different racial or ethnic groups are disproportionately affected by the dynamics identified in this report. Second, we do not capture other important demographic information about justices or judicial candidates, such as whether a person identifies as LGBTQ. We were limited by a lack of available data; according to a study by Lambda Legal, 49 states “do not formally collect data on sexual orientation and gender identity as part of a judicial application and reporting process.”126 For our analysis of 2000–2016 elections, a substantial portion of our candidate-level judicial election data comes from the National Institute on Money in Politics (NIMP). We used NIMP data to identify which candidates won or lost their elections and the amount of monetary contributions each candidate raised. For independent expenditures, we relied upon a historical dataset of television advertisements provided to the Brennan Center by Kantar Media/CMAG, which covers 2004 to 2016. Brennan Center for Justice 27 Appendix 2: Court and Population Demographics for All States General Population Demographics Compared to Demographics on State Supreme Courts State General Population – % People of Color General population – % White Supreme Court – Number of Justices Supreme Court – % Justices of Color Nevada 51% 49% 7 0% Alaska 39% 61% 5 0% Delaware 38% 62% 5 0% Alabama 34% 66% 9 0% Rhode Island 27% 73% 5 0% Arkansas 27% 73% 7 0% Tennessee 26% 74% 5 0% Michigan 25% 75% 7 0% Kansas 24% 76% 7 0% Pennsylvania 23% 77% 7 0% Indiana 21% 79% 5 0% Nebraska 21% 79% 7 0% Utah 21% 79% 5 0% Idaho 18% 82% 5 0% Wisconsin 18% 82% 7 0% South Dakota 18% 82% 5 0% Wyoming 16% 84% 5 0% Kentucky 15% 85% 7 0% North Dakota 15% 85% 5 0% Montana 14% 86% 7 0% New Hampshire 9% 91% 5 0% West Virginia 8% 92% 5 0% Vermont 7% 93% 5 0% Maine 7% 93% 7 0% Texas (Supreme Court) 58% 42% 9 11% Texas (Court of Criminal Appeals) 58% 42% 9 11% Mississippi 43% 57% 9 11% New Jersey 45% 55% 7 14% 28 Brennan Center for Justice State Supreme Court Diversity State General Population – % People of Color General population – % White Supreme Court – Number of Justices Supreme Court – % Justices of Color Louisiana 41% 59% 7 14% Illinois 39% 61% 7 14% Oklahoma (Supreme Court) 34% 66% 9 14% Massachusetts 28% 72% 7 14% Ohio 21% 79% 7 14% Missouri 20% 80% 7 14% Iowa 14% 86% 7 14% South Carolina 36% 64% 5 20% Oklahoma (Court of Criminal Appeals) 34% 66% 5 20% Georgia 47% 53% 9 22% Washington 31% 69% 9 22% Arizona 45% 55% 7 29% Virginia 38% 62% 7 29% Colorado 32% 68% 7 29% Oregon 24% 76% 7 29% Minnesota 20% 80% 7 29% Hawaii 78% 22% 5 40% New Mexico 62% 38% 5 40% Maryland 49% 51% 7 43% Florida 46% 54% 7 43% New York 45% 55% 7 43% North Carolina 37% 63% 7 43% Connecticut 33% 67% 7 43% California 63% 37% 7 71% Population demographics from U.S. Census Bureau, Population Estimates Program, July 2018. Supreme court diversity data as of May 2019. People of color includes individuals who are Black, Asian, Latino, Native American, or multiracial. White is limited to nonHispanic/Latino whites. State Supreme Court Diversity Brennan Center for Justice 29 Appendix 3: Comparing Elections and Appointments Percentage of People of Color Among Initially Elected and Initially Appointed Justices (1960-2018) State Initially Elected Justices – % People of Color” Initially Appointed Justices – % People of Color Arkansas 0% 29% Alabama 0% 18% Pennsylvania 0% 17% Washington 0% 15% Mississippi 0% 14% Nevada 0% 11% Georgia 0% 9% Wisconsin 0% 7% West Virginia 0% 7% Kentucky 0% 6% Idaho 0% 0% North Dakota 0% 0% Ohio 3% 18% Texas (Civil) 4% 20% Illinois 4% 8% Michigan 5% 32% Montana 6% 0% Texas (Criminal) 7% 21% Louisiana 7% N/A Oregon 9% 8% North Carolina 11% 22% New Mexico 27% 19% Minnesota 50% 8% Chart is limited to states that currently use contested elections to fill open seats on their supreme court. 30 Brennan Center for Justice State Supreme Court Diversity Appendix 4: Average Fundraising in Supreme Court Elections Mean Fundraising by Candidate Status White Women White Men Women of Color Men of Color Challenger $549,151 $282,780 $185,086 $120,220 Incumbent $463,742 $430,543 $422,505 $650,122 Open Seat $715,414 $530,057 $165,283 $349,490 White Women White Men Women of Color Men of Color Partisan $934,477 $642,303 $331,802 $680,083 Nonpartisan $285,028 $196,086 $186,494 $216,954 Mean Fundraising by Election Type Mean Fundraising by Incumbents Who Faced Opposition White Women White Men Women of Color Men of Color $665,320 $705,077 $554,159 $826,729 State Supreme Court Diversity Brennan Center for Justice 31 Appendix 5: Literature Review and Regressions This section provides a discussion of the congruence between our findings and similar past research, lists the variables used in our analyses and a description of their coding, and provides our regression tables. Literature Discussion In addition to exploring the impact of a candidate’s gender and race/ethnicity on campaign fundraising, TV advertising, and outcomes in state supreme court elections, we also controlled for factors that have been identified as relevant in past research. Only two studies have explored the factors that influence the ability of candidates in judicial races to raise campaign funds. One study examined fundraising in state supreme court races from 1990 to 2000 (Bonneau);127 the other looked at candidate spending on contested races for seats on intermediate appellate courts from 2000 to 2009 (Frederick and Streb).128 Although our study indicates that white women are prolific fundraisers in supreme court campaigns, neither prior analysis reported a relationship between the level of campaign financing and the gender or race/ethnicity of the candidate. Consistent with past research on appellate court candidates, as well as with research on candidates for other elective offices, our findings demonstrate that supreme court candidates who are incumbents and who are “quality” candidates (as indicated by having held prior judicial office) are more successful fundraisers than their opponents. According to our study, open-seat candidates also enjoy a fundraising advantage over challengers to incumbents. When it comes to the relevance of the electoral context, our findings differ from past work in important respects. While Bonneau129 and Frederick and Streb130 reported that appellate court candidates in nonpartisan elections outraised and outspent candidates in partisan contests, we discovered no such dynamic. (Note that scholars differ in their coding of election types in Michigan and Ohio. In both states, candidates stand in nonpartisan general elections; however, in Michigan, supreme court candidates are nominated at political party conventions, and in Ohio, candidates compete in partisan primaries. Because of the strong partisan overtones in supreme court races in these states, we coded both states as having partisan elections, while others code them as nonpartisan.) We also established that candidates in district- or circuit-based races were more successful fundraisers than candidates in statewide races, though no such relationship was found in the earlier studies. We found some evidence that the larger institutional context is relevant to fundraising capacity. As with Bonneau,131 our study suggests that supreme court candidates raise more in odd-year elections than in even-year elections. At least two possibilities are in play here: candi- 32 Brennan Center for Justice dates may be more successful fundraisers because they are not competing for funds with candidates for federal and other statewide offices, or this result may be driven by the fact that two of the states that consistently see high-dollar judicial elections (Pennsylvania and Wisconsin) hold these elections in off-years. Regarding television advertising in supreme court elections, Hall132 has done important work. She examined various aspects of television advertising in contested races between 2002 and 2008. One of her key findings was that nonpartisan elections were more likely than partisan elections to feature TV ads. We found, to the contrary, that a partisan election increases the likelihood that an outside group will air a positive ad in a contested race, but that election type is unrelated to the likelihood of a group airing a negative ad in such a race. In addition, while Hall found that both positive and negative ads were aired most often in contested open-seat races, we found that being an incumbent in a contested race increases the likelihood that a group will air a negative ad but that no such relationship exists for being a candidate in a contested openseat race. Consistent with Hall,133 however, we found that both being a candidate in a contested open-seat race and being an incumbent in a contested race increases the likelihood that a group will air a positive ad. (Note that Hall did not distinguish between group-aired and partyaired ads or code contrast ads as positive or negative, as we did.) List of Variables Dependent Variable Fundraising = Log of total fundraising by candidate in 2016 dollars Vote Share = Percentage of votes the candidate received Group Ad = 1 if at least one group ad was aired re: a candidate in a particular election cycle; 0 if otherwise Independent Variables Woman = 1 if the candidate is a woman; 0 if otherwise Minority = 1 if the candidate is a racial/ethnic minority; 0 if otherwise Minority x Woman = 1 if the candidate is a minority woman; 0 if otherwise Woman x Fundraising = Interaction term to test the effects of being a woman on the value (in terms of vote share) of fundraising Minority x Fundraising = Interaction term to test the effects of being a minority on the value (in terms of vote share) of fundraising Minority x Woman x Fundraising = Interaction term to test the State Supreme Court Diversity tion in the state in that year effects of being a minority woman on the value (in terms of vote share) of fundraising Partisan Election = 1 if the election is partisan (including Michigan and Ohio elections); 0 if otherwise Prior Experience = 1 if the candidate has prior judicial experience; 0 if otherwise Off-year Election = 1 if the election takes place in an odd-numbered year; 0 if otherwise Incumbent = 1 if the candidate is a sitting justice (not coded for retention elections); 0 if otherwise Presidential Election Year = 1 if the election takes place in a presidential election year; 0 if otherwise Unopposed = 1 if the candidate is unopposed in all election stages; 0 if otherwise District-based = 1 if the candidate is running in a district, rather than statewide; 0 if otherwise Opposed Twice = 1 if the candidate is opposed in at least two election stages (e.g., primary and general; primary and primary run-off); 0 if otherwise Multimember = 1 if the candidate is running in a multimember race (e.g., Mississippi, Pennsylvania) rather than a single-seat race; 0 if otherwise Number of Seats = number of supreme court seats up for elec- Regression Tables Table 1: Multivariate Regression – Fundraising in Contestable Elections Coefficient Std. Error t Prob > t (Constant) 6.974 .455 15.342 .000 Woman .684 .290 2.359 .019 Minority -.002 .469 -.005 .996 Minority x Woman -.321 .794 -.405 .686 Prior Experience 3.396 .318 10.691 .000 Incumbent 1.383 .362 3.815 .000 Open Seat 1.331 .313 4.257 .000 Unopposed -5.605 .414 -13.528 .000 Opposed Twice 1.493 .320 4.669 .000 Number of Seats .023 .112 .201 .841 Partisan Election .329 .263 1.252 .211 Off-year Election .794 .509 1.561 .119 Presidential Election Year .241 .250 .964 .335 District-based .947 .336 2.821 .005 Multimember .452 .481 .941 .347 N = 883 F(14, 821) = 35.066 Prob. F > 0.00 = .000 Adjusted R2 = 0.364 State Supreme Court Diversity Brennan Center for Justice 33 Table 2: Multivariate Regression – Fundraising in Contested Elections Coefficient Std. Error t Prob > t (Constant) 7.484 .432 17.305 .000 Woman .660 .279 2.364 .018 Minority .026 .445 .058 .954 Minority x Woman -.519 .753 -.689 .491 Prior Judicial Experience 3.449 .286 12.067 .000 Incumbent 1.494 .326 4.581 .000 Open Seat 1.187 .280 4.235 .000 Opposed Twice 1.511 .286 5.290 .000 Number of Seats -.094 .109 -.865 .387 Partisan Election .047 .248 .188 .851 Off-year Election .676 .464 1.458 .145 Presidential Election Year .141 .244 .577 .564 District-based 1.017 .329 3.091 .002 Multimember .593 .431 1.375 .169 N = 768 F(13, 707) = 30.603 Prob. F > 0.00 = .000 Adjusted R2 = 0.348 Table 3: Multivariate Regression – Candidate Vote Share in Contested Elections Coefficient Std. Error t Prob > t (Constant) 10.180 3.257 3.126 .002 Woman 5.728 7.765 .738 .461 Minority 11.526 7.453 1.546 .123 Minority x Woman -18.184 23.607 -.770 .441 Fundraising 2.247 .269 8.348 .000 Woman x Fundraising -.209 .617 -.339 .735 Minority x Fundraising -1.140 .610 -1.869 .062 Minority x Woman x Fundraising 1.095 1.951 .561 .575 34 Brennan Center for Justice State Supreme Court Diversity Prior Experience 5.832 1.329 4.388 .000 Incumbent 12.732 1.415 8.997 .000 Open Seat -1.616 1.212 -1.333 .183 Opposed Twice 6.361 1.208 5.266 .000 Number of Seats .026 .458 .058 .954 Partisan Election -1.435 1.095 -1.311 .190 Off-year Election -6.706 1.910 -3.510 .000 Presidential Election Year -3.040 1.058 -2.874 .004 District-based -1.025 1.498 -.684 .494 Multimember -20.742 1.842 -11.258 .000 N = 768 F(17, 627) = 41.817 Prob. F > 0.00 = .000 Adjusted R2 = 0.519 Table 4: Multivariate Regression – Candidate Vote Share in Contested Partisan Elections Coefficient Std. Error t Prob > t (Constant) 14.871 4.047 3.675 .000 Woman -16.084 11.666 -1.379 .169 Minority 7.051 7.277 .969 .333 Minority x Woman 44.390 29.121 1.524 .128 Fundraising 2.316 .320 7.239 .000 Woman x Fundraising 1.411 .882 1.601 .110 Minority x Fundraising -.589 .595 -.991 .323 Minority x Woman x Fundraising -4.024 2.370 -1.698 .091 Prior Experience 2.406 1.920 1.253 .211 Incumbent 10.083 1.848 5.457 .000 Open Seat -1.376 1.645 -.837 .403 Opposed Twice 8.685 1.597 5.438 .000 Number of Seats -1.069 .588 -1.817 .070 Off-year Election -8.743 2.309 -3.787 .000 Presidential Election Year -2.272 1.368 -1.661 .098 State Supreme Court Diversity Brennan Center for Justice 35 District-based -8.195 2.146 -3.819 .000 Multimember -23.126 1.841 -12.563 .000 N = 397 F(16, 322) = 32.224 Prob. F > 0.00 = .000 Adjusted R2 = 0.596 Table 5: Logistic Regression – Likelihood of Groups Airing Positive Ads in Contested Elections Coefficient Std. Error t Prob. > Wald Woman .484 .283 2.924 .087 Minority -1.143 .760 2.258 .133 Minority x Woman -.219 1.096 .040 .842 Partisan Election .814 .275 8.751 .003 Odd-year Election .062 .488 .016 .900 Presidential Election Year -.481 .276 3.028 .082 Number of Seats -.259 .130 3.984 .046 District-based .425 .362 1.378 .240 Multimember -.057 .461 .015 .902 Incumbent 1.454 .326 19.845 .000 Open Seat .802 .339 5.608 .018 Constant -2.008 .519 14.955 .000 N = 549 -2 Log likelihood = 404.758 Cox & Snell R2 = 0.081 Table 6: Logistic Regression – Likelihood of Groups Airing Negative Ads in Contested Elections Coefficient Std. Error t Prob. > Wald Woman .349 .338 1.063 .302 Minority -.566 .772 .537 .464 Minority x Woman -.977 1.323 .546 .460 Partisan Election -.038 .358 .012 .914 Odd-year Election .392 .472 .692 .406 Presidential Election Year -.414 .335 1.526 .217 36 Brennan Center for Justice State Supreme Court Diversity Number of Seats -.724 .166 19.051 .000 District-based .597 .428 1.948 .163 Multimember .676 .508 1.769 .183 Incumbent .637 .377 2.861 .091 Open Seat -.082 .386 .045 .832 Constant -.474 .532 .795 .372 N = 520 -2 Log likelihood = 301.673 Cox & Snell R2= 0.070 State Supreme Court Diversity Brennan Center for Justice 37 Endnotes 1  Alicia Bannon, Rethinking Judicial Selection in State Courts, Brennan Center for Justice, 2016, 1, https://www.brennancenter.org/ publication/rethinking-judicial-selection-state-courts. 2  “Number of Trial Court Judges,” National Center for State Courts, https://www.ncsc.org/microsites/sco/home/List-Of-Tables.aspx; National Center for State Courts, “Number of Appellate Court Judges,” https://www.ncsc.org/microsites/sco/home/List-Of-Tables. aspx. This reflects figures self-reported by states as of 2016 or 2017; because not all states provided information or complete information, this is a minimum estimate for the number of lower state court justices. 3 Joyce Mazero, “The Contract Is The Contract: Reinforced By Recent Texas Supreme Court Decision,” Forbes, March 25, 2019, https:// www.forbes.com/sites/joycemazero/2019/03/25/the-contractis-the-contract-reinforced-by-recent-texas-supreme-court-decision/#351850bc70aa 4  Dan Margolies and Celia Llopis-Jepsen, “Kansas Supreme Court Rules State Constitution Protects Right to Abortion,” NPR, April 26, 2019, https://www.npr.org/2019/04/26/717449336/kansas-supreme-court-rules-state-constitution-protects-right-to-abortion; Tony Leys and Stephen Gruber-Miller, “Iowa Supreme Court rejects 72-hour abortion waiting period requirement, says women have right to abortion,” Des Moines Register, June 29, 2018, https://www. desmoinesregister.com/story/news/health/2018/06/29/abortion-iowa-supreme-court-planned-parenthood-72-hour-waiting-period-american-civil-liberties-union/745068002/. 5  Dion Lefler, Hunter Woodall, Katy Bergen, and Suzanne Tobias, “Kansas school funding still inadequate, Supreme Court says,” The Kansas City Star, June 25, 2018, https://www.kansascity.com/news/ politics-government/article213797099.html. 6 In this report, we define an all-white bench as one composed of only non-Hispanic/Latino white justices. We use the term “people of color” to refer to individuals who, per the U.S. Census Bureau’s classification system, are not non-Hispanic/Latino whites (including individuals who are Black, Asian, Latino, Native American, Alaskan Native, Pacific Islander, and multiracial). 7  Oklahoma and Texas have two state high courts (one civil and one criminal). For these states, if either high court had a justice of color we coded the state as a whole as having a justice of color on its high court. 8  Yvette McGee Brown, “Foreword,” Building a Diverse Bench: A Guide for Judicial Nominating Commissioners, Brennan Center for Justice, 2016, 1, https://www.brennancenter.org/sites/default/files/ publications/Building_Diverse_Bench.pdf; see also, Nancy Scherer and Brett Curry, “Does Descriptive Race Representation Enhance Institutional Legitimacy? The Case of the U.S. Courts,” The Journal of Politics 72 (2010): http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.333.6739&rep=rep1&type=pdf (finding that “greater descriptive representation for blacks causes increased legitimacy for the institution among African Americans”). 9  Elizabeth Hinton, LeShae Henderson, and Cindy Reed, An Unjust Burden: The Disparate Treatment of Black Americans in the Criminal Justice System, Vera Institute of Justice, 2018, https://storage. googleapis.com/vera-web-assets/downloads/Publications/for-therecord-unjust-burden/legacy_downloads/for-the-record-unjust-burden-racial-disparities.pdf. 10  Ibid, 1. 11  For example, 32 percent of African American respondents, compared to 57 percent of overall respondents, when asked how well the assertion “[p]rovide equal justice to all” describes state courts, responded “well or very well.” Similarly, 42 percent of African American respondents, compared to 60 percent of respondents overall, when 38 Brennan Center for Justice asked how well the description “fair and impartial” characterizes state courts, responded “well or very well.” (The question was “How well does each of the following describe state courts?” and the answer was “provide equal justice to all” and “fair and impartial.”) See “The State of State Courts: 2015 Poll,” National Center for State Courts, 2015, 10, https://www.ncsc.org/~/media/Files/PDF/Topics/ Public%20Trust%20and%20Confidence/SoSC_2015_Presentation. ashx. See also “Analysis of National Survey of Registered Voters,” Gerstein Bocian Agne Strategies to the National Center for State Courts, November 17, 2015, https://www.ncsc.org/~/media/Files/PDF/ Topics/Public%20Trust%20and%20Confidence/SoSC_2015_Survey%20Analysis.ashx. 12  See, e.g., Sherrilyn A. Ifill, “Judging the Judges: Racial Diversity, Impartiality and Representation on State Trial Courts,” Boston College Law Review 39 (1998): 99, https://lawdigitalcommons.bc.edu/cgi/viewcontent.cgi?referer=https://www.google. com/&httpsredir=1&article=2086&context=bclr. 13  Empirical research has found that having diverse viewpoints on the bench can affect judicial decision-making. See, e.g., Adam B. Cox and Thomas J. Miles, “Judging the Voting Rights Act,” Columbia Law Review 108 (2008): 4, 49 (examining all published federal cases decided under section 2 of the Voting Rights Act since 1982, and finding that in the context of voting for liability under section 2 of the Voting Rights Act, “when a white judge sits on a panel with at least one African-American judge, she becomes roughly 20 percentage points more likely to find a section 2 violation,” a finding that crossed party lines); Jonathan P. Kastellec, “Racial Diversity and Judicial Influence on Appellate Courts,” American Journal of Political Science 57 (2013): 167 (analyzing three-judge panels of the Court of Appeals, and finding that “[r]andomly assigning a black counterjudge—a black judge sitting with two nonblack judges—to a three-judge panel of the Courts of Appeals nearly ensures that the panel will vote in favor of an affirmative action program,” an effect that “preliminary evidence…suggests… is due either to a presence or deliberation effect”) ; Jennifer L. Peresie, Note, “Female Judges Matter: Gender and Collegial Decisionmaking in the Federal Appellate Courts,” Yale Law Journal 114 (2005): 1761 (examining 556 federal appellate cases involving gender-coded cases and finding that “the presence of a female judge significantly increased the probability that a male judge supported the plaintiff”). Judges themselves have also reflected on how diversity on the bench enriches judicial deliberations. See, e.g., Harry T. Edwards, “Race and the Judiciary,” Yale Law and Policy Review 20 (2002): 329 (“And in a judicial environment in which collegial deliberations are fostered, diversity among the judges makes for better informed discussion. It provides for constant input from judges who have seen different kinds of problems in their pre-judicial careers, and have sometimes seen the same problems from different angles.”); James Andrew Wynn, Jr. & Eli Paul Mazur, “Judicial Diversity: Where Independence and Accountability Meet,” Albany Law Review 67 (2004): 789 (“However, it is generally difficult for a homogenous judiciary of affluent white men to understand and explain the socially diverse realities of poverty, race, and gender. For instance, a recent study of one federal circuit reveals that female judges are more likely than male judges to observe, report, and intervene when instances of gender-related incivility are directed at women.”); Sandra Day O’Connor, “Thurgood Marshall: The Influence of a Raconteur,” Stanford Law Review 44 (1992): 1217-1218, 1220 (“Although all of us come to the Court with our own personal histories and experiences, Justice Marshall brought a special perspective… At oral arguments and conference meetings, in opinions and dissents, Justice Marshall imparted not only his legal acumen but also his life experiences, constantly pushing and prodding us to respond not only to the persuasiveness of legal argument but also to the power of moral truth…Occasionally, at Conference meetings, I still catch myself looking expectantly for his raised brow and his twinkling eye, hoping to hear, just once more, another story that would, by and by, perhaps change the way I see the world.”). State Supreme Court Diversity 14  See, e.g., Scherer and Curry, “Does Descriptive Race Representation Enhance Institutional Legitimacy? The Case of the U.S. Courts” (“Using a novel experimental design centered on the federal courts, we find that greater descriptive representation for blacks causes increased legitimacy for the institution among African Americans”); see also Katherine Tate, Black Faces in the Mirror: African Americans and Their Representatives in the U.S. Congress (Princeton: Princeton University Press, 2002), 151 (“Blacks who believed Blacks to be numerically strong in Congress had more positive views of Congress and greater trust than those Blacks who felt Blacks were numerically weak.”); Adrian D. Pantoja and Gary M. Segura, “Does Ethnicity Matter? Descriptive Representation in Legislatures and Political Alienation Among Latinos,” Social Science Quarterly 84 (2003): 441 (finding that “The presence of Latino representatives in the state assembly, state senate, and/or U.S. House is associated with lower levels of political alienation among Latino constituents.”); Gabriel R. Sanchez and Jason L. Morin, “The Effect of Descriptive Representation on Latinos’ Views of Government and of Themselves,” Social Science Quarterly 92 (2011): 483 (finding “a meaningful relationship between descriptive representation and the political attitudes of Latino citizens”). Not all studies have found evidence of such a relationship, however. One study found no “systematic” evidence that increased descriptive representation through Black judicial officials “ameliorates the suspicions of the black public regarding the fairness of the state courts.” L. Marvin Overby, Robert D. Brown, John M. Bruce, Charles E. Smith, Jr., and John W. Winkle III, “Race, Political Empowerment, and Minority Perceptions of Judicial Fairness,” Social Science Quarterly 86 (2005): 444. 15  A number of judges, advocates, and other observers have discussed how diversity on the bench can establish role models and contradict stereotypes that certain groups cannot obtain judicial positions. See, e.g., Improving Diversity on the State Courts: a Report from the Bench, Lawyers’ Committee for Civil Rights Under Law, Justice at Stake Campaign, and Center for Justice, Law and Society at George Mason University, 2009, 6, http://www.justiceatstake.org/ media/cms/DiversityReport2009_4F739E0E55910.pdf (“A diverse bench also provides new role models for current and future law students and younger members of the bar, who in turn may be encouraged to seek the bench.”); Sheryl J. Willert, Building a Diverse Court: A Guide to Recruitment and Retention, Washington State Minority and Justice Commission (2010), https://www.courts.wa.gov/committee/ pdf/WaMJCCompleteManual0403.pdf (“The opportunity to observe persons of color working in all areas of the court system also provides role models for young people, graphically demonstrating that career opportunities in the courts are open to everyone.”); Edwards, “Race and the Judiciary,” 328-329 (describing a meeting with the Black Law Students Association in 1970 at the University of Michigan where students emphasized: “We do not need you on any picket lines. We need you to be a role model. We want you to be as good as any professor in the law school. We want students to subscribe to your classes in the same numbers as they do other classes. If you are respected, we will be respected.”); Answering the Call for a More Diverse Judiciary: A Review of State Judicial Selection Models and Their Impact on Diversity, Lawyers’ Committee for Civil Rights Under Law, 2005, 2, https://www.opensocietyfoundations.org/sites/default/files/answering_20050923.pdf (emphasizing the importance of “role models for minority youth”). In the context of politics and elections, several empirical studies have also found that enhanced representation fosters increased political engagement. See, e.g., David E. Campbell and Christina Wolbrecht, “See Jane Run: Women Politicians as Role Models for Adolescents,” The Journal of Politics 68 (2006): 244 (finding that “The more politics is infused with visible female role models, the more adolescent girls report an expectation of being involved in politics.”); Lonna Rae Atkeson, “Not All Cues Are Created Equal: The Conditional Impact of Female Candidates on Political Engagement,” The Journal of Politics (2003): 1040 (finding “overwhelming support for the hypothesis that women citizens in states with competitive and visible female candidates increase their political engagement.”); Kira Sanbonmatsu, “Electing Women of Color: The Role of Campaign Trainings,” Journal of Women, Politics & Policy 36 (2015): 145 State Supreme Court Diversity (qualitatively finding that among steering committee members of the Diversity Initiative, a campaign training program for women of color, that a perceived barrier facing women of color was the “lack of female role models of the same racial/ethnic background.”). 16  See Part II for research outlining structural barriers that can impede advancement in the legal profession; see also Elaine Martin and Barry Pyle, “Gender and Racial Diversification of State Supreme Courts,” Women & Politics 24 (2002): 41-42 (“Historically, the white male lawyers tapped for judgeships have followed a relatively narrow political-legal career path. There is substantial evidence that non-traditional lawyers do not follow the same career patterns that white men lawyers do. This may be, as many have suggested, because women and African American lawyers do not have the same range of opportunities as white men lawyers do. Certainly, women lawyers are clustered in the lower prestige ranks of their professions, such as public sector employment — making less money, having less opportunity for advancement, and often subjected to varying degrees of gender bias.”). 17  “Sex by Age, Table B01001,” 2013-2017 American Community Survey 5-Year Estimates, U.S. Census Bureau, https://factfinder. census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_17_5YR_B01001&prodType=table; “Sex by Age (White Alone, Not Hispanic or Latino), Table B01001H,” 2013-2017 American Community Survey 5-Year Estimates, U.S. Census Bureau, available at https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_17_5YR_B01001H&prodType=table. 18  Five states have a higher proportion of justices of color on the bench than in their overall population, and one state (IA) has the same proportion of justices of color on the bench as in the general population. In an additional two states, the percentage of justices of color is no more than 10 percent less than what researchers would expect based on the diversity of the general population. All other states have larger gaps in representation. 19  According to the U.S. Census Bureau’s American Community Survey, people of color make up 34 percent of the general population in the Northeast, 24 percent in the Midwest, 42 percent in the South, and 49 percent in the West. 20  We followed the methodology laid out in Tracey E. George and Albert H. Yoon, The Gavel Gap: Who Sits in Judgment on State Courts?, American Constitution Society, 2016, http://gavelgap.org/pdf/gavel-gap-report.pdf; “The Gavel Gap: Methodology,” American Constitution Society, http://gavelgap.org/methodology.html. We divided the gap between the percentage share of people of color on the bench and their percentage share in the region’s overall population by the percentage share of people of color in the region’s overall population: (fraction of judges who are people of color minus fraction of general population who are people of color) divided by fraction of general population who are people of color). We then multiplied by 100 to yield a percentage. The resulting positive or negative percentage shows what percentage fewer (or more) people of color a region’s high court benches have than researchers would expect from the region’s overall population demographics. In this case, our calculations were as follows: in the West in 2017, people of color were 49 percent of the overall population and 16 percent of high court justices (0.164835164835165-0.49199213259)/(0.49199213259) = -.66496 x 100 = -66.5 percent. In the Northeast in 2017, people of color were 34 percent of the general population and 18 percent of state high court justices (0.181818181818182-0.33920622362)/(0.33920622362)= -.46399 x 100 = -46.4 percent. In the South in 2017, people of color were 42 percent of the overall population and 12 percent of high court justices (.12- 0.42319399466)/(0.42319399466)= -.71644 x 100 = -71.6 percent. In the Midwest in 2017, people of color were 24 percent of the overall population and 7 percent of high court justices (0.0731707317073171-0.23733696695)/(0.23733696695) = -.6917 x 100 = -69.2 percent. To calculate the proportion of justices on the bench in 2017 who were people of color, we divided the number of justices of color who sat on a state’s high court bench in 2017 by the total number of justices who sat on that high court bench that year. In some instances, the number of justices who sat on a state’s bench in Brennan Center for Justice 39 a given year may exceed the number of seats on a state’s high court bench, such as when a justice retires and a replacement is named the same year. Given data limitations, we were only able to examine data at the aggregate year-level, rather than by individual month; as a result, these percentages are approximations and reflect aggregate state-level diversity over 2017. 21  For best practices on collecting data on judicial diversity, see Yuvraj Joshi, Diversity Counts: Why States Should Measure the Diversity of Their Judges and How They Can Do It, Lambda Legal and the American Constitution Society, 2017, https://www.lambdalegal.org/sites/ default/files/legal-docs/downloads/20170607_diversity-counts.pdf. 22  This figure includes some states with comparatively low Black populations. However, the fact that these states have never had a Black justice in their entire history is still troubling and notable given the lengthy time period. 23  Delaware, Florida, Hawaii, Idaho, Indiana, Iowa and South Carolina each only had one female justice first reach the bench between 1998 and 2018. 24  We followed a methodology similar to what was laid out in George and Yoon, The Gavel Gap: Who Sits in Judgment on State Courts?; “The Gavel Gap: Methodology,” American Constitution Society, http://gavelgap.org/methodology.html. We divided the gap between the percentage share of people of color on the bench and their percentage share in the state’s overall population by the percentage share of people of color in the state’s overall population: (fraction of judges who are people of color minus fraction of general population who are people of color) divided by fraction of general population who are people of color). We then multiplied by 100 to yield a percentage. The resulting positive or negative percentage shows what percentage fewer (or more) people of color a state’s high court bench has than researchers would expect from the state’s overall population demographics. In this case, our calculations were as follows: in 1996, 10.05% of state high court justices were people of color, and 26.84% of the population were people of color. Calculation: ((.1005-.268417)/ (.268417)) = -0.62558 (x 100) = -62.6%. In 2017, 13.33% of justices were people of color, and 39.4% of the overall population were people of color. Calculation: ((.1333-.394)/(.394)) = -.662 (x 100) = -66.2%. Population estimates from the U.S. Census Bureau: “Resident Population Estimates of the United States by Sex, Race, and Hispanic Origin: April 1, 1990 to July 1, 1999, with Short-Term Projection to November 1, 2000,” Population Estimates Program, U.S. Census Bureau, published January 2, 2001, available at https://www.census. gov/population/estimates/nation/intfile3-1.txt; “Sex by Age (White Alone, Not Hispanic or Latino), Table B01001H,” 2013-2017 American Community Survey 5-Year Estimates, U.S. Census Bureau, available at https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_17_5YR_B01001H&prodType=table. 25  Of the state supreme courts that use contested elections, 74 percent have had a higher proportion of justices of color on the bench at some time in the past, as compared with today (as of 2018). In comparison, of the state supreme courts that do not use contested elections, only 31 percent have had a higher proportion of justices of color at some time in the past. 26  In 1962, one justice on Michigan’s high court was a person of color. As of May 2019, Michigan’s high court bench is all-white. 27  Alicia Bannon, Who Pays for Judicial Races? The Politics of Judicial Elections 2015-16, Brennan Center for Justice, 2017, https://www. brennancenter.org/publication/politics-judicial-elections. 28  Greg Goelzhauser, Choosing State Supreme Court Justices: Merit Selection and the Consequences of Institutional Reform (Philadelphia: Temple University Press, 2016), 86-94; Ifill, “Judging the Judges: Racial Diversity, Impartiality and Representation on State Trial Courts,” 126; Cynthia Grant Bowman, “Women in the Legal Profession from the 1920s to the 1970s: What Can We Learn From Their Experience About Law and Social Change?,” Maine Law Review 61 (2009). 29  Bradwell v. Illinois, 83 U.S. 130 (1873). 30  In re Hong Yen Chang, 84 Cal. 163 (1890). 40 Brennan Center for Justice 31  Sweatt v. Painter, 339 U.S. 629 (1950). 32  Gregory L. Acquaviva and John D. Castiglione, “Judicial Diversity on State Supreme Courts,” Seton Hall Law Review (2010): 1210-1211; Goelzhauser, Choosing State Supreme Court Justices: Merit Selection and the Consequences of Institutional Reform, 96-97. 33  Michael J. Gabrail, “Hon. Florence Ellinwood Allen, U.S. Court of Appeals for the Sixth Circuit: A Judge of Many Firsts,” The Federal Lawyer, August 2016, 51, http://www.fedbar.org/PDFs/Past-Judicial-Profiles/Sixth-Circuit_1/Ellinwood-Allen-Hon-Florence.aspx; Goelzhauser, Choosing State Supreme Court Justices: Merit Selection and the Consequences of Institutional Reform, 93 (“Thirty-seven years passed following Allen’s 1922 election before Rhoda Valentine Lewis (Hawaii) and Jeannie Loitman Barron (Massachusetts) became the next female state supreme court justices seated in 1959”). Justice Sandra Day O’Connor became the first female justice on the U.S. Supreme Court in 1981: see Linda Greenhouse, “Justice O’Connor Seated on Nation’s Highest Court,” The New York Times, September 26, 1981, https://www.nytimes.com/1981/09/26/us/justice-o-connor-seated-on-nation-s-high-court.html. 34  Robert Mcg. Thomas Jr., “J.K. Stout, Pioneering Judge in Pennsylvania, Is Dead at 79,” New York Times, August 24, 1998, https:// www.nytimes.com/1998/08/24/us/j-k-stout-pioneering-judge-inpennsylvania-is-dead-at-79.html. 35  Martin and Pyle, “Gender and Racial Diversification of State Supreme Courts,” 41-42 (“Historically, the white male lawyers tapped for judgeships have followed a relatively narrow political-legal career path. There is substantial evidence that non-traditional lawyers do not follow the same career patterns that white men lawyers do. This may be, as many have suggested, because women and African American lawyers do not have the same range of opportunities as white men lawyers do. Certainly, women lawyers are clustered in the lower prestige ranks of their professions, such as public sector employment — making less money, having less opportunity for advancement, and often subjected to varying degrees of gender bias.”). 36  Stephanie Francis Ward, “Women outnumber men in law schools for first time, newly updated data show,” American Bar Association Journal, December 19, 2016, http://www.abajournal.com/ news/article/women_outnumber_men_in_law_schools_for_first_ time_newly_updated_data_show; “Law School Rankings by Female Enrollment (2017),” Enjuris, https://www.enjuris.com/students/ law-school-female-enrollment-2017.html; “Law School Rankings by Female Enrollment (2018),” Enjuris, https://www.enjuris.com/students/law-school-female-enrollment-2018.html. 37  “First Year and Total J.D. Enrollment by Gender, 1947-2011,” American Bar Association, https://www.americanbar.org/content/ dam/aba/administrative/legal_education_and_admissions_to_the_ bar/statistics/jd_enrollment_1yr_total_gender.authcheckdam.pdf. 38  These figures reflect demographic data on resident and active attorneys provided to the American Bar Association from state bar associations or licensing agencies in individual states. The demographic numbers reflect the “best available data” provided to the ABA. “ABA National Lawyer Population Survey: 10-Year Trend in Lawyer Demographics,” American Bar Association, 2018, https://www.americanbar.org/content/dam/aba/administrative/market_research/ National_Lawyer_Population_Demographics_2008-2018.pdf. 39  A Current Glance at Women in the Law: January 2018, American Bar Association, 2018, 2-4, https://www.americanbar.org/content/ dam/aba/administrative/women/a-current-glance-at-women-inthe-law-jan-2018.pdf. 40  The size of the attorney pool has been found to have a relationship with racial diversity, but not gender diversity, on the bench. See Malia Reddick, Michael J. Nelson, and Rachel Paine Caufield, “Examining diversity on state courts: How does the judicial selection environment advance — and inhibit — judicial diversity?,” American Judicature Society (2009): 6 (“[S]tates with a higher percentage of minority attorneys were more likely to have minority judges serving on all three levels of courts,” but “the percentage of women attorneys in the state was not tied to women attaining seats on the court of last State Supreme Court Diversity resort”); Mark Hurwitz and Drew Noble Lanier, “Explaining Judicial Diversity: The Differential Ability of Women and Minorities to Attain Seats on State Supreme and Appellate Courts” State Politics & Policy Quarterly 3 (2003): 339-340, 342 (“With respect to demographics, the candidate pool [of attorneys of color] has a rather strong, positive coefficient in both years... [but] the candidate pool [of female attorneys] does not influence the number of women judges.”). 41  These figures reflect demographic data on resident and active attorneys provided to the American Bar Association from state bar associations or licensing agencies in individual states. The demographic numbers reflect the “best available data” provided to the ABA. In some states, responders are able to choose more than one category for race/ethnicity, as a result, the “race/ethnicity percentages may total to more than 100%.” “ABA National Lawyer Population Survey: 10-Year Trend in Lawyer Demographics.” 42  “Statistics Archive: Longitudinal Charts, First-Year & Total JD Minority,” American Bar Association, https://www.americanbar.org/ groups/legal_education/resources/statistics/statistics-archives/. 43  “Statistics: 2018 1L Enrollment by Gender & Race/Ethnicity (Aggregate),” American Bar Association, https://www.americanbar. org/groups/legal_education/resources/statistics/. 44  2018 Vault/MCCA Law Firm Diversity Survey, 2018, 5, https:// www.mcca.com/wp-content/uploads/2018/11/2018-Vault-MCCA-Law-Firm-Diversity-Survey-Report.pdf. 45  George and Yoon, The Gavel Gap: Who Sits in Judgment on State Courts?, 8-9. 46  Some causes identified in prior scholarship include disparate legal social networks, divergent interorganizational mobility, disparities in formal and informal mentoring opportunities, and perceived “affirmative action stigma.” See Christopher I. Rider, Adina D. Sterling, and David Tan, “Career Mobility and Racial Diversity in Law Firms,” in Diversity in Practice: Race, Gender, and Class in Legal and Professional Careers (Cambridge: Cambridge University Press, 2016): 358-361 (The authors examined the unexpected dissolution of six large law firms, and found that white lawyers “were more likely to regain employment in the largest, highest-grossing, and most prestigious US law firms” relative to attorneys of color, supporting the theory that “the differences in career attainment between white lawyers and lawyers of other races can be attributed to variation in opportunities to change employers.” The authors also found that Black attorneys “face mobility constraints due at least in part to the intra-organizational structure of work relationships.”); Monique R. Payne-Pikus, John Hagan and Robert L. Nelson, “Experiencing Discrimination: Race and Retention in America’s Largest Law Firms,” Law and Society Review 44 (2010): 567 (“Turning to the institutional discrimination theory of contact and mentoring, there is a [] large difference in partner contacts and mentoring for African American and white associates.”); Veronica Root, “Retaining Color,” University of Michigan Journal of Law Reform 47 (2014): 610-611 (One “cause for the retention rate differentials appears to be a result of ‘perceived affirmative action-stigma.’ The perception that all Black or Hispanic associates have benefited from affirmative action in both law school admissions and the firm hiring process solidifies the idea that whites are more qualified than blacks and Hispanics.”). For an explanation of how implicit biases and discrimination can impact advancement in the legal profession, see Nicole E. Negowetti, “Implicit Bias and the Legal Profession’s Diversity Crisis: A Call for Self-Reflection,” Nevada Law Journal 15 (2015): 941-949; Root, “Retaining Color,” 606-607 (explaining that one factor that inhibits retention of attorneys of color is “second generation bias,” including assignment systems that lack “standardization or systematic checks to ensure that all similarly situated associates are receiving the same type of work.”); Luis J. Diaz and Patrick C. Dunican Jr., “Ending the Revolving Door Syndrome in Law,” Seton Hall Law Review (2011): 977 (citing multiple studies that found perceived bias against female attorneys and attorneys of color in law firm assignment distribution); From Visible Invisibility to Visibly Successful: Success Strategies for Law Firms and Women of Color in Law Firms, American Bar Association Commission on Women in the Profession, 2008 (finding that 44% of female attorneys of color “re- State Supreme Court Diversity ported being passed over for desirable work assignments” compared to 2% of white male attorneys, and that 31% of female attorneys of color “reported getting at least one unfair performance evaluation” compared to less than 1% of white male attorneys). 47  Joan C. Williams, Marina Multhaup, Su Li, and Rachel Korn, You Can’t Change What You Can’t See: Interrupting Racial & Gender Bias in the Legal Profession, American Bar Association Commission on Women in the Profession and the Minority Corporate Counsel Association, 2018, 7-10, https://www.americanbar.org/content/dam/aba/ administrative/women/Updated%20Bias%20Interrupters.pdf. 48  Debra Cassens Weiss, “Partners in Study Gave Legal Memo a Lower Rating When Told Author Wasn’t White,” ABA Journal, April 21, 2014, http://www.abajournal.com/news/article/hypothetical_legal_memo_demonstrates_unconscious_biases; Diaz and Dunican Jr., “Ending the Revolving Door Syndrome in Law,” 986 (describing “leniency bias,” where “[c]oldly objective judgment is reserved for out-groups…The higher the group’s status, the more convincing the demonstration of incompetence will have to be”). 49  Ciara Torres-Spelliscy, Monique Chase, and Emma Greenman, Improving Judicial Diversity, Brennan Center for Justice, 2010, 32, http://www.brennancenter.org/sites/default/files/legacy/Improving_Judicial_Diversity_2010.pdf (Judicial nominating commissioner Von G. Keetch (UT) was concerned that “we speak with candidates all the time who flatly say that they cannot afford to become a judge.”). 50  Improving Diversity on the State Courts: A Report from the Bench, 27 (“[N]early all of the judges mentioned (often without prompting) that political leaders, more senior judges, or other influential individuals had actively recruited them. However, these recruitment efforts were frequently not conducted through any formalized recruiting process.”). 51  Kathleen Bratton and Rorie Spill, “Existing Diversity and Judicial Selection: The Role of the Appointment Method in Establishing Gender Diversity in State Supreme Courts,” Social Science Quarterly 83 (2002): 515 (“We also add to current research by demonstrating more directly the political influences on the selection of women to state high courts: relatively liberal states are particularly likely to have gender-diverse courts.”); Hurwitz and Noble Lanier, “Explaining Judicial Diversity: The Differential Ability of Women and Minorities to Attain Seats on State Supreme and Appellate Courts,” 340 (“Among the political variables in the supreme court models, elite ideology is a positive influence on minority representation in 1985…suggesting that minorities benefit when state policymakers are more liberal.”); Greg Goelzhauser, “Diversifying State Supreme Courts,” Law & Society Review 45 (2011): 771 (“The estimated coefficient for Citizen Liberalism is positive and significant in both the first black and first female justice models. This means that states that utilize appointment systems seated their first black and female justices sooner on average as Citizen Liberalism increased.”). 52  Goelzhauser, “Diversifying State Supreme Courts,” 776 (“States are more likely to seat their first black and female justices sooner when institutional rules result in more frequent openings.”); Hurwitz and Noble Lanier, “Explaining Judicial Diversity: The Differential Ability of Women and Minorities to Attain Seats on State Supreme and Appellate Courts,” 339 (“[C]ourt size remains a positive influence on minority judicial representation at both levels in both time periods. Clearly, minority judges benefit from having more judicial seats available.”); Nicholas O. Alozie, “Selection Methods and the Recruitment of Women to State Courts of Last Resort,” Social Science Quarterly 77 (1996): 122 (“Compared to the specification for percentage of women on the court, this outcome suggests that while size of court does not influence the degree of women’s representation, it helps determine whether or not women secure a seat on the court.”). 53  “Judicial Selection: An Interactive Map,” Brennan Center for Justice, http://judicialselectionmap.brennancenter.org/?court=Supreme. 54  A small number of states provide for appointment by the governor without the input of a commission or for legislative appointment. 55  Alicia Bannon, Choosing State Judges: A Plan for Reform, Brennan Center for Justice 41 Brennan Center for Justice, 2018, 2, https://www.brennancenter.org/ sites/default/files/publications/2018_09_JudicialSelection.pdf. 56  John F. Kowal, Judicial Selection for the 21st Century, Brennan Center for Justice, 2016, 15, https://www.brennancenter.org/sites/ default/files/publications/Judicial_Selection_21st_Century.pdf. 57  Reddick, Nelson, and Paine Caufield, “Examining diversity on state courts: How does the judicial selection environment advance — and inhibit — judicial diversity?,” 6 (“Appointive methods were more likely than popular elections to place minority judges on courts of last resort … In terms of selection methods, merit selection and gubernatorial appointment systems did not place more, or fewer, women on state high courts or trial courts than did popular elections. However, after taking the larger political, institutional, and contextual environment into account, we found that merit selection placed significantly fewer women on intermediate appellate courts than did partisan or nonpartisan elections.”); Martin and Pyle, “Gender and Racial Diversification of State Supreme Courts,” 48 (“It is clear that appointment processes advantage African-Americans, male and female… only one African-American male and no African-American females in elective states initially reached the bench through election. White female fare somewhat better than white males); Bratton and Spill, “Existing Diversity and Judicial Selection: The Role of the Appointment Method in Establishing Gender Diversity in State Supreme Courts,” 504 (“Women are significantly more likely to be selected to a state high court when initially appointed, and this effect is particularly pronounced when the governor is Democratic. When an appointment system is used, women are much more likely to be appointed to an all-male court than to a gender-diverse court.”); Nicholas O. Alozie, “Black Representation on State Judiciaries,” Social Science Quarterly 69 (1988): 984-985 (“The important finding is that methods of judicial selection do not significantly contribute to explaining the percentage of black judges in the state…Rather, the overwhelming factor is the percentage of black lawyers among all lawyers in the state.”); Nicholas O. Alozie, “Distribution of Women and Minority Judges: The Effects of Judicial Selection Methods,” Social Science Quarterly 71 (1990): 318 (“Though the legislative election method grants women better access to states’ judicial offices, the relationship is quite weak. In general, the data lead to the more fundamental observation that judicial selection methods do not seem to be the major agents some analysts think they are.”); Hurwitz and Noble Lanier, “Diversity in State and Federal Appellate Courts: Change and Continuity Across 20 Years,” 47 (“Based on our prior research and the findings we report here, we assert that the selection system, including the merit system, does not relate to judicial diversity.”); “Study Finds Appointive Process Yields More Women and Minority Judges,” Modern Courts 2 (1986): 1 (“The success of women and minorities in achieving judicial office depends in large measure upon the method of selection, according to a study released today by the Fund for Modern Courts.”); Diane M. Johnsen, “Building a Bench: A Close Look At State Appellate Courts Constructed by The Respective Methods of Judicial Selection,” San Diego Law Review 53 (2016): 870 (“[A]mong the statistically significant differences between elected and appointed judges in election states is that a significantly greater portion of the appointed judges are non-white.… By contrast, there is virtually no distinction between the proportion of women among judges first appointed and first elected to the bench in election states.”). 58  In a few states, such as Arkansas and Louisiana, justices who reach the bench via interim appointments are barred from running for the seat in a subsequent election. 59  Lisa M. Holmes and Jolly A. Emrey, “Court Diversification: Staffing the State Courts of Last Resort Through Interim Appointments,” The Justice System Journal 27 (2006): 1 (from 1964 to 2004, 52 percent of supreme court justices reached the bench as interim appointments in state that use elections); Reddick, Nelson, and Paine Caufield, “Examining diversity on state courts: How does the judicial selection environment advance — and inhibit — judicial diversity?,” 2 (“[A]pproximately 45 percent of judges serving in 2008 in states with partisan or nonpartisan elections were initially appointed.”) 60  Our data also includes New Mexico, which uses a hybrid system 42 Brennan Center for Justice by which justices are first appointed to the bench via a merit selection system, and then must compete in contested partisan elections. 61  Reddick, Nelson, and Paine Caufield, “Examining diversity on state courts: How does the judicial selection environment advance — and inhibit — judicial diversity?,” 6; Johnsen, “Building a Bench: A Close Look At State Appellate Courts Constructed by The Respective Methods of Judicial Selection,” 870; Goelzhauser, Choosing State Supreme Court Judges: Merit Selection and the Consequences of Institutional Reform, 104-105 (though Goelzhauser did not find that merit selection or appointments placed significantly more female judges on the state high court bench than did elections, he did find that “[a]ppointment systems are more likely on average to seat political minorities, which is a category that combines female, black, Hispanic, and Asian American justices.”). 62  See, for instance, Reddick, Nelson, and Paine Caufield, “Examining diversity on state courts: How does the judicial selection environment advance — and inhibit — judicial diversity?,” 6. 63  Overall, 79 of 471 (16.8%) initially elected justices were female, and 210 of 1202 (17.5%) of initially appointed justices were female. 64  These figures include New Mexico, which has a hybrid system by which justices are first appointed to the bench and then must stand in a contested partisan election. 65  In states that provide for contested elections, 52 percent of white justices first reached the bench via an election, while 22 percent of justices of color first reached the bench via an election. 66  Looking at the seven states formerly covered by Section 5 of the Voting Rights Act that currently use contested elections for filling seats on their state high court bench (AL, GA, LA, MI, MS, NC, TX), three have not had a justice of color first reach the bench via an election since at least 1960 (AL, GA, MS) and one had just one justice of color first elected to the bench (MI). Comparatively, six of the seven states formerly covered that use contested elections have had multiple justices of color appointed to the bench since 1960 (AL, GA, MI, MS, NC, TX). 67  When we limited out analysis to challengers and open seat candidates, the proportion of candidates of color fluctuated by cycle but remained low. When available, data comes from the U.S. Census Bureau’s 1-Year American Community Survey Estimates. “ACS Demographic and Housing Estimates, Table DP05,” 2017 American Community Survey 1-Year Estimates, U.S. Census Bureau, available at https:// factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?pid=ACS_17_1YR_DP05&prodType=table; “ACS Demographic and Housing Estimates, Table DP05,” 2016 American Community Survey 1-Year Estimates, U.S. Census Bureau, available at https:// factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?pid=ACS_16_1YR_DP05&prodType=table; “ACS Demographic and Housing Estimates, Table DP05,” 2014 American Community Survey 1-Year Estimates, U.S. Census Bureau, available at https:// factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?pid=ACS_14_1YR_DP05&prodType=table; “ACS Demographic and Housing Estimates, Table DP05,” 2012 American Community Survey 1-Year Estimates, U.S. Census Bureau, available at https:// factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?pid=ACS_12_1YR_DP05&prodType=table; “ACS Demographic and Housing Estimates, Table DP05,” 2010 American Community Survey 1-Year Estimates, U.S. Census Bureau, available at https:// factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?pid=ACS_12_1YR_DP05&prodType=table; “ACS Demographic and Housing Estimates: 2008, Table DP05,” 2008 American Community Survey 1-Year Estimates, U.S. Census Bureau, available at https:// factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?pid=ACS_08_1YR_DP5&prodType=table; “ACS Demographic and Housing Estimates: 2006,” 2006 American Community Survey, U.S. Census Bureau, https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_06_EST_DP5&prodType=table; “Resident Population by Sex, Race, and Hispanic Origin Status: 2000 to 2003, Table No. 13,” Statistical Abstract of the United States: 2004-2005, U.S. Census Bureau, available at https://www2. State Supreme Court Diversity census.gov/library/publications/2004/compendia/statab/124ed/ tables/pop.pdf?#; “Profile of General Demographic Characteristics: 2000,” Census 2000 Summary File 1 (SF 1) 100-Percent Data, U.S. Census Bureau, available at https://factfinder.census.gov/faces/ tableservices/jsf/pages/productview.xhtml?pid=DEC_00_SF1_ DP1&prodType=table. Estimates for the years 2001-2002 and 2004 were not available. 68  This figure includes Montana and New Mexico, which have hybrid systems. In Montana, a sitting justice may seek additional terms through participating in an election where multiple candidates may vie for the seat — but if no candidates register to challenger the incumbent judge, she stands in an unopposed yes/no retention election. In New Mexico, a sitting justice must compete in a partisan election during the first general election after her appointment, and if successful, serves the remainder of the seat’s eight-year term. A justice approaching the end of that full term may stand in an unopposed yes/no retention vote. She may stand for additional eight-year terms in the same retention process. “Judicial Selection: An Interactive Map.” 69  Louisiana and Illinois are exceptions to this general framework for interim appointments. In Louisiana and Illinois, when a seat on the state high court becomes open in the middle of a justice’s term, the remaining justices of the court appoint a judicial candidate to the empty seat on the bench. In Louisiana, within a year, the governor calls a special election to fill the seat. The judge appointed in the interim by the supreme court may not run in the special election. In Illinois, the appointed justice holds office until the next general election more than 60 months after she was appointed, unless the remainder of the seat’s term runs out before then. During the election, multiple candidates may vie for the seat in a partisan election. 70  These findings are consistent with those of a 2015 study by the Center for American Progress, which examined reelection rates for incumbent state supreme court justices in contested races from 2000 to 2015 and found that white justices had a 90% reelection rate, compared with an 80% reelection rate for Black justices, and a 66.7% reelection rate for Latino justices. See Michele L. Jawando and Billy Corriher, More Money, More Problems: Fleeting Victories for Diversity on the Bench, Center for American Progress, 2015, 8, https://cdn. americanprogress.org/wp-content/uploads/2015/10/26055523/ JudicialDiversity-reportB.pdf. 71  Emma Platoff and Alexa Ura, “In Texas Republican judicial primaries, do Hispanic-sounding surnames spell loss?,” The Texas Tribune, March 1, 2018, https://www.texastribune.org/2018/03/01/ texas-highest-courts-have-few-hispanic-judges-some-attribute-surname-c/. 72  Alexa Ura, “Trial begins in case targeting Texas’ statewide elections of judges,” The Texas Tribune, February 12, 2018, https://www. texastribune.org/2018/02/12/lawsuit-puts-texas-statewide-elections-judges-trial/. 73  We limited our analysis to justices who were first named to the bench between 1985 and 2018, in order to avoid including states whose method of judicial selection changed. (In 1988, New Mexico was the last state to change its high court judicial selection system to a hybrid system. Because partisan elections continued to be used in New Mexico for additional terms to justices, we did not restrict the data to 1989 onwards.) We also limited our analysis to justices whose term ended between 1985 and February 2019, to avoid attributing shorter term lengths to justices still on the bench. 74  Non-presidential election cycles have consistently seen higher success rates for candidates of color than presidential election cycles; to compare like entities, we isolated success rates in non-presidential election cycles over time. 75  Platoff and Ura, “In Texas Republican judicial primaries, do Hispanic-sounding surnames spell loss?.” 76  In 2017, there were 33 Republican governors, 16 Democratic governors, and 1 Independent governor. No year from 1939 to 2016 had a higher number of Republican governors than 2017. In 1970, 1996, 1997, 1998, and 2016, there were 32 Republican governors. Data from 1939-2011 from Carl Klarner, “Measurement of the Partisan Balance State Supreme Court Diversity of State Government,” State Politics and Policy Quarterly 3 (2003): 309-19; Carl Klarner, “State Partisan Balance Data, 1937-2011,” Harvard Dataverse, 2013, https://dataverse.harvard.edu/dataset.xhtml?persistentId=hdl:1902.1/20403. Data from 2012-2017 from the National Conference of State Legislatures: “2012 State and Legislative Partisan Composition,” National Conference of State Legislatures,” http://www.ncsl.org/documents/statevote/LegisControl_2012.pdf; “2013 State and Legislative Partisan Composition,” National Conference of State Legislatures,” http://www.ncsl.org/documents/statevote/LegisControl_2013.pdf; “2014 State and Legislative Partisan Composition,” National Conference of State Legislatures,” http://www. ncsl.org/documents/statevote/LegisControl_2014.pdf; “2015 State and Legislative Partisan Composition,” National Conference of State Legislatures,” http://www.ncsl.org/Portals/1/Documents/Elections/ Legis_Control_2015.pdf; “2016 State and Legislative Partisan Composition,” National Conference of State Legislatures,” http://www.ncsl. org/portals/1/documents/elections/Legis_Control_2016.pdf; “2017 State and Legislative Partisan Composition,” National Conference of State Legislatures,” http://www.ncsl.org/Portals/1/Documents/ Elections/Legis_Control_2017_March_1_9%20am.pdf.). 77  Seven justices of color were appointed in 2017-18. Five were appointed by Democratic governors, 1 by a Republican governor, and 1 by the state high court. 78  Bannon, Who Pays for Judicial Races? The Politics of Judicial Elections 2015-16. 79  Ibid. 80  Chris W. Bonneau and Damon Cann, “Campaign Spending, Diminishing Marginal Returns, and Campaign Finance Restrictions in State Supreme Court Elections,” The Journal of Politics 73 (2011); Chris W. Bonneau, “Electoral Verdicts: Incumbent Defeats in State Supreme Court Elections,” American Politics Research 33 (2005). 81  Prior scholarship examining racial disparities in fundraising has had mixed findings. A 2014 study of 2006 state legislative elections found that candidates of color raise significantly fewer campaign funds than white candidates (Laura M. Albright, “Not Simply Black and White: The Relationship Between Race/Ethnicity and Campaign Finance in State Legislative Elections,” American Political Science Association, 2014, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2475889). An earlier study found that Black candidates receive significantly fewer contributions from PACs than white candidates do (Allen Wilhite and John Theilmann, “Women, Blacks, and PAC Discrimination,” Social Science Quarterly 67 (1986)). However, two studies that specifically examined judicial races, one looking at state supreme court races from 1990 to 2000 (Chris W. Bonneau, “Campaign Fundraising in State Supreme Court Elections,” Social Science Quarterly 88 (2007): 81-82), and one looking at contested races for seats on intermediate appellate courts from 2000 to 2009 (Brian Frederick and Matthew J. Streb, “The Cost of Going for the Gavel: Individual Candidate Spending in Intermediate Appellate Court Elections,” Justice System Journal 32 (2011): 40), did not find relationships between the levels of campaign financing and the race or ethnicity of the candidate. 82  Christopher S. P. Magee, “The Incumbent Spending Puzzle,” Social Science Quarterly 93 (2012): 947-948; Tony Nguyen, “Why Women Win: Gender and Success in State Supreme Court Elections,” American Politics Research (2019): 592. 83  See Appendix 4 for full fundraising figures. 84  See Appendix 4 for full fundraising figures. 85  p = .018. See Appendix 5 for full regression tables. 86  We primarily relied on individual states’ Secretary of State websites for vote share data, and used publicly available databases to fill in data gaps. See Herbert Kritzer, “State Supreme Court Election Data,” Harvard Dataverse, 2015, https://dataverse.harvard.edu/ dataset.xhtml?persistentId=doi:10.7910/DVN/1P1JFG; “Judgepedia,” Ballotpedia, https://ballotpedia.org/Judgepedia. 87  Prior research on whether the campaign funds of male candidates have more “purchasing power” than those of female candidates Brennan Center for Justice 43 in general (including both women of color and white women) in other contexts have yielded mixed findings. See Rebekah Herrick, “Is There a Gender Gap in the Value of Campaign Resources?” American Politics Quarterly 24 (1996): 73 (“Campaign spending helps male candidates more than female candidates. For every $100,000 a male candidate spends, he receives .8% more votes than does a woman spending $100,000.”); Robert E. Hogan, “The Effects of Candidate Gender on Campaign Spending in State Legislative Elections,” Social Science Quarterly 88 (2007): 1102 (“[T]he results demonstrate that men and women receive a similar level of return for their campaign spending”); Joanne Connor Green, “The Times… Are They A-Changing? An Examination of the Impact of the Value of Campaign Resources for Women and Men Candidates for the U.S. House of Representatives,” Women & Politics 25 (2003): 19 (“[I]n the 1992-2000 period, we see a reversal. Spending by men is not significant, while women challenger spending is a significant and positive predictor of vote.”) 88  p = 0.091. See Appendix 5 for full regression tables. 89  p = .062. See Appendix 5 for full regression tables. 90  DeNora Getachew and Ava Mehta, Breaking Down Barriers: The Faces of Small Donor Public Financing, Brennan Center for Justice, 2016, 2, https://www.brennancenter.org/sites/default/files/publications/Faces_of_Public_Financing.pdf; A Civil Rights Perspective on Money in Politics, Brennan Center for Justice and Demos, 2016, https://www.brennancenter.org/analysis/civil-rights-perspective-money-politics; see also The Case for Small Donor Public Financing in New York State, Brennan Center for Justice, 2019, https://www. brennancenter.org/sites/default/files/publications/CaseforPublicFinancingNY_0.pdf; J. Mijin Cha and Miles Rapoport, Fresh Start: The Impact of Public Campaign Financing in Connecticut, Demos, 2013, https://www.demos.org/sites/default/files/publications/FreshStart_PublicFinancingCT_0.pdf. For a case study of a state’s experience with judicial public financing, see “A Profile of the Judicial Public Campaign Program, 2004-2010,” Democracy North Carolina, https:// democracync.org/wp-content/uploads/2017/05/JudicialPubFinSuccess2004-2010.pdf; Albright, “Not Simply Black and White: The Relationship Between Race/Ethnicity and Campaign Finance in State Legislative Elections”; Richard L. Fox and Jennifer L. Lawless, “If Only They’d Ask: Gender, Recruitment, and Political Ambition,” The Journal of Politics 72 (2010), 310-326, http://digitalcommons.lmu.edu/cgi/ viewcontent.cgi?article=1001&context=poli_fac; Jennifer L. Lawless and Richard L. Fox, Why Are Women Still Not Running for Public Office?, Brookings Institution Issues in Governance Studies, 2008, 13 (among surveyed lawyers, business leaders, educators, and political activists, 64 percent of women thought it was “harder for a woman to raise money for a campaign than a man”). 91  See Public Financing in California: A Model Law for the 21st Century, Center for Governmental Studies, 2011, 12, http://www. ilcampaign.org/wp-content/uploads/2014/07/CGS-Model-Hybrid-Matching-System.pdf; Steven M. Levin, “Keeping It Clean: Public Financing in American Elections,” National Civic Review 95 (2006); 2007 Report on the Maine Clean Election Act: 2007 Study Report, Has Public Funding Improved Maine Elections?, Maine Commission on Governmental Ethics and Election Practices, 2007, 17, 20, https:// www1.maine.gov/ethics/pdf/publications/2007_study_report.pdf; Angela Migally and Susan Liss, Small Donor Matching Funds: The NYC Experience, Brennan Center for Justice, 2010, http://www.brennancenter.org/sites/default/files/legacy/Small%20Donor%20Matching%20Funds-The%20NYC%20Election%20Experience.pdf. 92  Improving Diversity on the State Courts: A Report from the Bench, 17 (“[E]lected judges frequently cited campaign money, or, rather, the difficulty of raising it, as a leading issue that discourages minority candidates for judgeships.”). 93  From 2000 to 2016, a total of 44 candidates were eligible to opt into public financing (6 candidates of color and 38 white candidates). 94  Levin, “Keeping It Clean: Public Financing in American Elections”; Timothy Werner and Kenneth R. Mayer, “Public Election Funding, Competition, and Candidate Gender,” Political Science and Politics 40 (2007). 44 Brennan Center for Justice 95  Bannon, Who Pays for Judicial Races? The Politics of Judicial Elections 2015-16, 2. 96  See Appendix 5 for full list of variables controlled for. 97  p = 0.133 and p = .087. 98  Estimates of outside group spending on television ads from 2004 through 2016 were inflation adjusted to 2016 dollars. Unique candidates targeted were isolated; spending was aggregated by unique candidate targeted, and then averaged by the number of unique candidates targeted. Spending was analyzed by unique candidate over multiple campaigns, rather than by unique candidate and by unique campaign (when applicable). 99  These candidates were Louis Butler (Wisconsin); John Guidry (Louisiana); Denise Langford-Morris (Michigan). 100  See, e.g., Monika L. McDermott, “Race and Gender Cues in Low-Information Elections,” Political Research Quarterly 51 (1998): 912-13; Jim Walker, “The Politics of State Courts,” in The Judicial Branch of State Government: People, Process, and Politic, ed. Sean O. Hogan (Santa Barbara: ABC-CLIO, 2006), 171, 178. 101  See Platoff and Ura, “In Texas Republican judicial primaries, do Hispanic-sounding surnames spell loss?.” 102  Albert J. Klumpp, “Judicial Primary Elections in Cook County, Illinois: Fear the Irish Women!,” DePaul Law Review 60 (2011); “Electing Judges in Cook County: The Role of Money, Political Party, and the Voters,” The Chicago Appleseed Fund for Justice with the Assistance of the Chicago Council of Lawyers, 2003, http://chicagoappleseed.org/ wp-content/uploads/2012/08/electing_judges_in_cook_county.pdf. 103  Answering the Call for a More Diverse Judiciary: A Review of State Judicial Selections Models and their Impact on Diversity, 17 (quoting Judge Sandra Otaka). 104  See Matt A. Barreto, “Defining Racially Polarized Voting — Examples from Washington State,” 2013, http://mattbarreto.com/ papers/polarized_voting_wa.pdf. 105  Matt A. Barreto, Cameron Caldwell and Kassra A. R. Oskooii, “Dissecting Voting Patterns in the González-Danielson Supreme Court Contest in Washington State,” 3, 2012, http://mattbarreto.com/papers/gonzalez_primary2012.pdf. 106  Paula Wissel, “Racial Bias a Factor in Washington Supreme Court Election, Research Finds,” KPLU.885 (Seattle), September 26, 2012, http://www.kplu.org/post/racial-bias-factor-washington-supreme-court-electionresearch-finds; Barreto, Caldwell and Oskooii, “Dissecting Voting Patterns in the González-Danielson Supreme Court Contest in Washington State,” 1. 107  Wissel, “Racial Bias a Factor in Washington Supreme Court Election, Research Finds.” 108  In Los Angeles, a series of studies of Los Angeles County Superior Court elections in 2006, 2008, and 2010, all found evidence of racially polarized voting. See Matt A. Barreto, “Latino Candidates and Racial Block Voting in Primary and Judicial Elections: An Analysis of Voting in Los Angeles County Board Districts, 2006 Democratic Primary Election & L.A. County Superior Court Election,” B&W Consulting for The Los Angeles County Chicano Employees Association, 2007, http://mattbarreto.com/papers/laccea2006.pdf; Matt A. Barreto and Loren Collingwood, “Latino Candidates and Racial Block Voting in Primary and Judicial Elections: An Analysis of Voting in Los Angeles County Board Districts, 2008 Primary Election & L.A. County Superior Court Election,” Barreto Consulting for The Los Angeles County Chicano Employees Association, 2009, http://mattbarreto.com/papers/laccea2008.pdf; Matt A. Barreto, “Latino Candidates and Racial Block Voting in Primary and Judicial Elections: An Analysis of Voting in Los Angeles County Board Districts, 2010 Primary Election & L.A. County Superior Court Election,” Barreto Consulting for The Los Angeles County Chicano Employees Association, 2010, http://www. mattbarreto.com/papers/laccea2010.pdf. 109  A study of Harris County, Texas, found racially polarized voting from 2002 to 2010 “regardless of whether we analyze elections for county commission, county judge, U.S. Senate or Lieutenant Governor,” and concluded that “[i]n every instance, Latino voters greatly State Supreme Court Diversity prefer to vote for Latino candidates, and in every instance Latino voters demonstrate a statistically significantly higher vote for a Latino candidate than do Whites by huge margins.” (Francisco I. Pedraza and Matt A. Barreto, “Harris County Minority Vote Dilution Analysis Expert Report,” 2012, http://www.mattbarreto.com/papers/harriscounty12. pdf). The study was conducted as an expert report in Rodriguez v. Harris County, 964 F. Supp. 2d 686 (S.D. Tex. 2013), where the court also concluded that there was racially polarized voting. Id. at 777. 110  In Terrebonne Parish Branch NAACP v. Jindal, plaintiffs successfully argued that at-large elections for Louisiana’s 32nd Judicial District violate the Voting Rights Act and the U.S. Constitution. 274 F. Supp. 3d 395 (M.D. La. 2017). In ruling in favor of the plaintiffs, the trial court found evidence of racially polarized voting, explaining that the analyses of two experts “clearly show that in Terrebonne, blacks vote cohesively and non-black voters usually vote as a bloc to defeat the black-preferred candidates…Additionally, the pattern shows that black candidates receive minimal support from the white electorate. Notably, in a parish-wide election, no black candidate has ever received over one-third of the overall vote.” Id. at 435. 111 See, for instance, Tali Mendelberg, “Executing Hortons: Racial Crime in the 1988 Presidential Campaign,” Public Opinion Quarterly 61 (1997): 134-157; Nicholas A. Valentino, Vincent L. Hutchings, and Ismail K. White, “Cues that Matter: How Political Ads Prime Racial Attitudes During Campaigns,” American Political Science Review 96 (2002): 75-90. 112  Jackson also faced Republican challenger Chris Anglin, who split the Republican vote. Earls won the race with less than 50 percent of the vote. 113  Barry Yeoman, “The North Carolina GOP Is Trying Every Trick to Keep a Supreme Court Seat,” Talking Points Memo, October 18, 2018, https://talkingpointsmemo.com/feature/the-north-carolina-gop-istrying-every-trick-to-keep-a-supreme-court-seat. 114  Paul Woolverton, “NC GOP tries to connect Supreme Court candidate to convicted murderers,” The Fayetteville Observer, April 1, 2018, https://www.fayobserver.com/news/20180401/nc-gop-triesto-connect-supreme-court-candidate-to-convicted-murderers. verse-bench-guide-judicial-nominating-commissioners. 123  Court of criminal appeals candidates (Texas and Oklahoma) were not included in this analysis. 124  Goelzhauser, Choosing State Supreme Court Justices: Merit Selection and the Consequences of Institutional Reform, Appendix A (“To assemble a list of state supreme court justices seated from 1960 through 2014, I began with the list of justices compiled by Adam Bonica and Michael Woodruff (2015). Their list included most of the justices who served at some point from 1990 through 2012. Next, I extended this list through the end of 2014 and backdated it to 1960 using lists provided by state supreme courts, historical societies, various volumes of the Council of State Governments’ Book of the States, and other sources. The resulting list includes, to the best of my knowledge, every state supreme court justice seated from 1960 through 2014. The list does not include justices who served temporarily, for example, pending an ill justice returning or a vacancy being filled. The seating decision is the event that led to the individual becoming a justice rather than the actual start date…For most of the justices who had recently served or are still serving, reasonably complete biographical statements are available online through state supreme court websites or news articles. Other sources include archived newspaper articles, law review articles, bar journals, obituaries, remembrances, law firm web pages, and biographical databases accessed through LexisNexis”). Goelzhauser also differentiated whether justices reached the bench via “merit selection,” “elite appointment” or “election.” In our analysis, we do not differentiate between different appointment methods (i.e. gubernatorial appointment, merit selection, legislative selection, selection by the state high court, or interim appointments); instead, we compare all appointments to all elections. 125  Ibid. 126  Eric Lesh, Justice Out of Balance: How the Election of Judges and the Stunning Lack of Diversity on State Courts Threaten LGBT Rights, Lambda Legal, 2016, 14, https://www.lambdalegal.org/sites/ default/files/justiceoutofbalance_final_rev1_2.pdf. 127  Bonneau, “Campaign Fundraising in State Supreme Court Elections.” 115  Ibid. 128  Frederick and Streb, “The Cost of Going for the Gavel.” 116  Three other white candidates (one incumbent and two challengers) also competed in this election for two seats on the Michigan Supreme Court. 129  Bonneau, “Campaign Fundraising in State Supreme Court Elections.” 117  This figure is inflation-adjusted to 2016 dollars. 131  Bonneau, “Campaign Fundraising in State Supreme Court Elections.” 118  Christina Fuoco-Karasinski, “Eminem Gets Probation,” Rolling Stone, June 28, 2001, https://www.rollingstone.com/music/music-news/eminem-gets-probation-243439/; “Eminem gets one year of probation for gun charges,” CNN.com/Law Center, June 29, 2001, www.cnn.com/2001/LAW/06/28/eminem.sentencing/. 119  Ryan J. Foley, “Supreme Court Deadlocks in Gableman Ethics Case,” The Associated Press, July 1, 2010, https://madison.com/wsj/ news/local/govt_and_politics/supreme-court-deadlocks-in-gableman-ethics-case/article_059e2f86-8522-11df-83b6-001cc4c03286. html (“Butler had won the man a new trial from the appeals court based on a procedural error, but the decision was overturned by the Supreme Court. The man served his full prison term before being released and committing another offense.”) 130  Frederick and Streb, “The Cost of Going for the Gavel.” 132  Melinda Gann Hall, Attacking Judges: How Campaign Advertising Influences State Supreme Court Elections (Stanford, CA: Stanford University Press, 2015). 133  Ibid. 120  Jawando and Corriher, More Money, More Problems: Fleeting Victories for Diversity on the Bench, 15, https://cdn.americanprogress.org/wp-content/uploads/2015/10/26055523/JudicialDiversity-reportB.pdf. 121  Groups also labeled Justice Butler “Loophole Louie,” a nickname they claimed Butler found “affectionate.” “Loophole Louie,” Wisconsin Manufacturers and Commerce, 2008, https://www. brennancenter.org/sites/default/files/legacy/video/Judicial%20 Ads%2008/WI/boards/03-25-08%20STSUPCT_WI_WMC_LOOPHOLE_LOUIE.pdf. 122  Kate Berry, Building a Diverse Bench: A Guide for Judicial Nominating Commissioners, Brennan Center for Justice, 2016, https://www.brennancenter.org/publication/building-di- State Supreme Court Diversity Brennan Center for Justice 45 ABOUT THE AUTHORS AC K N O W L E D G M E N T S  Laila Robbins is a research and program associate in the Democracy Program at the Brennan Center for Justice, where she conducts quantitative and qualitative research and advocacy pertinent to promoting fair, diverse, and impartial courts. She graduated cum laude from Yale College with a BA in history with distinction. The Brennan Center for Justice gratefully acknowledges Carnegie Corporation of New York, Democracy Alliance Partners, The JPB Foundation, Martin and Ruth Krall, Open Society Foundations, Piper Fund, a Proteus Fund initiative, and Rockefeller Brothers Fund for their generous support of our fair courts work. We also gratefully acknowledge Arnold Ventures’ founders, Laura and John Arnold, who are donors to the Brennan Center for Justice. This is an independent Brennan Center publication; the views expressed are those of the authors and do not necessarily represent those of our supporters.  Alicia Bannon is managing director of the Brennan Center’s Democracy Program. She leads the Center’s fair courts project, where she directs research, advocacy, and litigation to promote a fair and impartial judicial system. She has authored several nationally recognized reports and articles on judicial selection, access to justice, and government dysfunction. Prior to joining the Brennan Center, Bannon was a John J. Gibbons Fellow in Public Interest and Constitutional Law at Gibbons P.C. in Newark, N.J., where she engaged in a wide range of public interest litigation. She also clerked for the Honorable Sonia Sotomayor in the Court of Appeals for the Second Circuit and the Honorable Kimba M. Wood in the Southern District of New York. She received her JD from Yale Law School and her AB from Harvard College, where she graduated summa cum laude with a degree in Social Studies.  Malia Reddick is a nationally recognized researcher and commentator in the areas of judicial selection and judicial performance evaluation. Reddick recently worked with the Quality Judges Initiative at the Institute for the Advancement of the American Legal System at the University of Denver, developing informed recommendations for selecting, evaluating, recusing, and disciplining judges who are highly qualified, impartial, and publicly accountable and who inspire public trust. Reddick also previously served as director of research and programs at the American Judicature Society. She earned a BS in public administration, summa cum laude, from Harding University, and a MA and PhD in Political Science from Michigan State University. S TAY C O N N E C T E D T O T H E BRENNAN CENTER Visit our website at www.brennancenter.org Sign up for our electronic newsletters at www.brennancenter.org/signup 46 Brennan Center for Justice The authors of this report are grateful to their Brennan Center colleagues who lent their support to the creation of this report. Wendy Weiser provided strategic guidance and vision throughout the course of this project. Dorothy Samuels provided invaluable support in honing our arguments and editing the report. Oliver Jones, Henry Costello, Jaya Aiyer, Sara Graziano, Julia O’Regan, Michael Loedel, Jeremy Gomez, Katherine Hacthoun, Gustav Wachhaus, Julia Stadlinger, Chabely Jorge, Ian Morris, Ayanna MillerSmith, Gwen Cappel-McCoy, Marco Balestri, and Julia Udell provided crucial research assistance. Peter Miller and Kevin Morris contributed important guidance in the creation and interpretation of the report’s methodology. Patrick Berry, Douglas Keith, Chisun Lee, Ian Vandewalker, and Nirali Vyas provided helpful feedback. The editorial and design assistance of Lisa Benenson, Stephen Fee, Lorraine Cademartori, Jeanne Park, Derek Rosenfeld, Josh Bell, Alden Wallace, and Yuliya Bas brought this project to publication. The authors are also indebted to Professor Greg Goelzhauser for generously sharing his dataset of state supreme court justices who reached the bench from 1960 to 2014 and to Professor Michael Nelson for his feedback regarding the report’s methodology and analyses. We also extend our gratitude to foreword author Vanita Gupta for her contribution. State Supreme Court Diversity Brennan Center for Justice at New York University School of Law 120 Broadway 17th Floor New York, NY 10271 www.brennancenter.org