The lack of diversity on state courts is well established. In 2005, blacks and women made up about 8 percent and 27 percent of the nation's state supreme court justices, respectively (Reference Hurwitz and LanierHurwitz & Lanier 2008). Not as well recognized is the long process of diversification that took place in the states or the fact that this process is ongoing. According to data provided by the American Judicature Society, 31 states did not have a black justice on their supreme courts in 2009 and two did not have a female justice. More surprising, perhaps, is that as of 2009, 19 states had yet to seat their first black justices.Footnote 1 Although every state has seated its first female justice, the process of gender diversification spanned from 1922–2002.
Why do some states diversify their high courts sooner than others? And how has the diversification process differed for blacks and women? Understanding the determinants of judicial diversity is an important task for several reasons. First, the presence of blacks and women on the bench may influence judicial outcomes (e.g., Reference BaldezBaldez et al. 2006; Reference SongerSonger et al. 1994; Reference WelchWelch 1988). Even the inclusion of black and women judges on panels may affect how other judges decide cases (e.g., Reference BoydBoyd et al. 2010; Reference Cox and MilesCox & Miles 2008; Reference PeresiePeresie 2005). More generally, representation may enhance trust in institutions among political minorities, while increasing democratic engagement (e.g., Reference Bobo and GilliamBobo & Gilliam 1990; Reference HansenHansen 1997; Reference VerbaVerba et al. 1997).
To explore the diversification of state supreme courts, I collected information on the first black and female justices seated in each state. The theoretical contention is that political pressures and institutions affect the process of diversification. The results suggest that selection mechanisms, institutions affecting turnover, and the appointment of political minorities to the U.S. Supreme Court are associated with the length of time until states seat their first black and female justices. Although a large literature exists on the effect of selection institutions on diversity in state courts (e.g., Reference AlozieAlozie 1990; Reference Bratton and SpillBratton & Spill 2002; Reference Hurwitz and LanierHurwitz & Lanier 2003), this work typically focuses on the proportion of positions occupied by blacks or women. By focusing on the diversification of these courts, the results contribute to the institutional literature on this topic while providing new information about the relative level of judicial diversity in the states.
The First Black and Female Justices
Blacks and women have endured long struggles for greater representation and equality. The Reconstruction and Civil Rights eras led to black suffrage and enhanced legal rights (Reference KlarmanKlarman 2004), which resulted in more blacks being elected to Congress (Reference Freedman and JonesFreedman & Jones 2007), state legislatures (Reference Grofman and HandleyGrofman & Handley 1991; Reference King-Meadows and SchallerKing-Meadows & Schaller 2006), and local offices (Reference AndrewsAndrews 1997; Reference Karnig and WelchKarnig & Welch 1980). Blacks encountered great difficulties breaking into the legal profession on account of long-standing political, social, and educational barriers (Reference SmithSmith 1999). And notwithstanding recent increases in the number of black lawyers, “blacks have had little success breaking into the upper echelons of the elite bar” (Reference Wilkins and GulatiWilkins & Gulati 1996: 497). As noted previously, this is particularly evident looking at the lack of racial diversity on state supreme courts.
Jonathan Jasper Wright became the first black justice to sit on a state supreme court when the South Carolina state legislature appointed him in 1870. Wright, who had been educated in Pennsylvania and New York, became a prominent member of South Carolina's political scene during Reconstruction (see Reference WoodyWoody 1933). He was the first African American admitted to the Pennsylvania bar and one of the first three African Americans admitted to the South Carolina bar. Wright also served in the South Carolina state senate prior to being appointed to the state's supreme court.
More than 90 years passed before Michigan became the second state to seat its first black justice, when Governor John Swainson appointed Otis M. Smith to the bench in 1961. Robert M. Duncan, who was elected to the bench by Ohio voters in 1969, was the only other black state supreme court justice to take office during the 1960s. Diversification continued sporadically over the ensuing decades, as evidenced in Figure 1, which plots the growth in the number of states that seated their first black supreme court justices from 1960–2009. As of 2009, 19 states had yet to seat their first black justices.
Figure 1. States seating their first black and female justices.
In many ways, blacks and women faced similar struggles securing public office. Like blacks, women faced severe educational, political, and social barriers to entering the legal profession (Reference DrachmanDrachman 1998). Women were increasingly elected to public office after ratification of the 19th Amendment, which granted women suffrage (Reference CarrollCarroll 2007; Reference WasniewskiWasniewski 2007). Florence Ellinwood Allen, for example, became the first female state supreme court justice following her election by Ohio voters in 1922—two years after ratification of the 19th Amendment. This was just one in a list of firsts for Allen, who was also the country's first female assistant prosecutor, the first woman to sit on a court of general jurisdiction in Ohio, and the first woman to serve on an Article III court following her appointment to the Sixth Circuit Court of Appeals by President Roosevelt in 1934. Allen also nearly became the first woman appointed to the U.S. Supreme Court; President Truman considered nominating her for one of two vacancies in 1949 before “ultimately decid[ing] the time was not yet ripe” (Reference GinsburgGinsburg 2004: 805).
After Allen's election in 1922, another 37 years passed before Hawaii and Massachusetts seated the next first female justices with the appointments of Rhoda Valentine Lewis and Jennie Loitman Barron in 1959. Only two other states seated female justices during the 1960s. Figure 1 plots the increase in the number of states that seated their first female justices over time. Although the number of states having seated their first black and female justices remains similar throughout the 1960s and early 1970s, there is a sharp increase in the number of states seating their first female justices around 1980. South Dakota was the last state to seat its first female justice in 2002.
Explaining Diversification
Why do some states diversify their supreme courts sooner than others? In this section, I contend that political pressures and institutions shape the diversification of state supreme courts. As is common in studies of policy adoption in the states (e.g., Reference Berry and BerryBerry & Berry 1990), I highlight both internal and external pressures on states to diversify their high courts. Existing explanations for variation in the level of diversity on state courts typically center on the influence of judicial selection mechanisms. There is little consensus, however, about how these institutions affect judicial diversity. Examining the time until states seat their first black and female justices allows for a fresh look at the question of whether selection institutions affect judicial diversity. I also consider the importance of judicial institutions that affect turnover. Although the existing literature pays comparatively little attention to these institutions, they may speed diversification by creating more opportunities to seat political minorities. Lastly, I explore the potential influence of political minorities being seated on the U.S. Supreme Court and neighboring state high courts.
Internal Pressures
Selection Institutions
A large literature is devoted to explaining the determinants of judicial diversity in the states. Much of the debate centers on unraveling the effect of judicial selection mechanisms (Reference AlozieAlozie 1990, Reference Alozie1996, Reference Alozie1988; Reference Bratton and SpillBratton & Spill 2002; Reference DuboisDubois 1983; Reference Flango and DucatFlango & Ducat 1979; Reference Glick and EmmertGlick & Emmert 1987; Graham Reference Graham1990a, Reference Graham1990b; Reference Hurwitz and LanierHurwitz & Lanier 2001, Reference Hurwitz and Lanier2003, Reference Hurwitz and Lanier2008; Reference Martin and PyleMartin & Pyle 2002; Reference WilliamsWilliams 2007). The results are mixed, but much of this work finds little if any relationship between selection institutions and diversity. Until recently, however, the work in this area relied primarily on descriptive statistics to draw inferences about whether institutions influence the level of diversity on state courts. Nonetheless, more methodologically advanced treatments continue to yield mixed results. Reference Bratton and SpillBratton and Spill (2002), for example, find that women are more likely to be seated on state supreme courts under an appointment system—especially if the court is all male. In the most comprehensive multivariate analysis to date, Reference Hurwitz and LanierHurwitz and Lanier (2003) find mixed results regarding the effect of selection institutions on the number of black and female judges on state appellate courts in 1985 and 1999. States with a merit selection system, for example, had fewer black justices on average in 1985 and more in 1999. For the most part, however, institutions appeared to play little role in determining the number of political minorities on state supreme courts in those years.
Much of the existing literature examines whether selection institutions have a direct effect on judicial diversity. However, given the conventional wisdom that liberals are more likely to value representative institutions (see, e.g., Reference Bratton and SpillBratton & Spill 2002), it is reasonable to expect the effect of selection systems to be conditioned by ideology (see also Reference Hurwitz and LanierHurwitz & Lanier 2003). For example, we might expect states that utilize judicial elections to yield their first black and female justices sooner, as citizen liberalism increases. Even if liberal voters are not more likely to seat a black or female justice all else equal, blacks and women tend to be more liberal on average, which suggests that they will fare better generally as citizen liberalism increases. Furthermore, judicial experience helps propel political minorities to state supreme courts (Reference Martin and PyleMartin & Pyle 2002), and they are more likely to obtain this experience as state liberalism increases for the same reason.
H1:Election System Hypothesis: States with an election system will seat their first black and female justices sooner as citizen liberalism increases.
The use of nominating commissions may also influence when states seat their first black and female justices. There is substantial variation across states in the makeup of nominating commissions and the factors they consider when selecting nominees. As a general matter, however, the governor, legislature, state bar, or some combination of these actors appoints commissioners. Most commissions are made up of lawyers and non-lawyers, and many are bipartisan. Nominating commissions were created to infuse merit into the selection process. Although merit selection tends to reward judicial experience, which has been difficult for minorities to obtain, some commissions expressly value diversity (Reference CaufieldCaufield 2007). As a result, it is reasonable to expect states that utilize nominating commissions to seat their first black and female justices sooner.
H2:Nominating Commission Hypothesis: States with nominating commissions will seat their first black and female justices sooner.
Institutions Affecting Turnover
In addition to selection systems, several other judicial institutions may play a role in determining when states seat their first black and female justices. The number of seats on a court is a consistent predictor of greater gender and racial diversity (Reference Bratton and SpillBratton & Spill 2002; Reference Hurwitz and LanierHurwitz & Lanier 2003; Reference Solberg and BrattonSolberg & Bratton 2005). One possible explanation for this result is that courts with more seats are less prestigious, which reduces the competition for office (e.g., Reference Hurwitz and LanierHurwitz & Lanier 2003). A simpler explanation is that more seats means more opportunities to seat political minorities. Similarly, other institutions that increase turnover may lead states to seat their first black and female justices sooner. Two prominent judicial institutions that affect turnover at the state level are term length and mandatory retirement rules. As term length increases, there should be less turnover, thereby reducing the number of opportunities for states to seat their first political minorities. Conversely, there should be more turnover in states that force judges off the bench at a certain age, thereby increasing the number of opportunities for states to seat their first political minorities.
H3: Seats Hypothesis: States with more seats on their high courts will seat their first black and female justices sooner.
H4: Term Length Hypothesis: States where justices enjoy longer terms will be slower to seat their first black and female justices.
H5: Mandatory Retirement Hypothesis: States that require justices to retire at a certain age will seat their first black and female justices sooner.
External Pressures
Supreme Court Appointments
Aside from state-level institutional features, states may also be more likely to seat their first political minorities when appointments to the U.S. Supreme Court generate heightened attention to the importance of judicial diversity. Thurgood Marshall became the first black justice appointed to the U.S. Supreme Court in 1967, and Sandra Day O'Connor became the first female appointed in 1981. Subsequently, Clarence Thomas became the second black justice in 1991, and Ruth Bader Ginsburg became the second female justice in 1993. In 2009, Sonia Sotomayor became the third female and first Hispanic nominated to the Supreme Court. Each of these appointments were salient political events that brought diversity to the forefront of debates over judicial selection (see, e.g., Reference AbrahamAbraham 2007; Reference OverbyOverby et al. 1992; Reference LiptakLiptak 2009). As a result, similar to the notion of vertical diffusion in the policy adoption literature (e.g., Reference AllenAllen et al. 2004; Reference Daley and GarandDaley & Garand 2005; Reference Shipan and VoldenShipan & Volden 2006), it is reasonable to expect that these federal appointments generate increased pressure to diversify state courts, and that states will learn from and imitate the federal government's efforts to diversify its highest court.
There is anecdotal evidence that the appointment of political minorities to the U.S. Supreme Court brings attention to the issue of institutional diversity in the states. For example, shortly after President Obama nominated Sonia Sotomayor to the U.S. Supreme Court, the Atlanta Journal-Constitution published a piece lamenting the absence of Hispanics on Georgia's state courts, noting: “There are no Hispanics now, nor have there ever been any, on the Supreme Court of Georgia or on the Georgia Court of Appeals” (Reference DiguetteDiguette 2009). Although the nomination of minorities to the Supreme Court clearly puts pressure on political actors in appointment systems, it is reasonable to expect them to affect decisions in election states as well, where most vacancies are filled by government officials through interim appointments (Reference Holmes and EmreyHolmes & Emrey 2006). The news story from Georgia shows how this pressure manifests itself in election states, when the author adds: “Gov. Sonny Perdue must appoint someone to fill the Supreme Court seat soon to be vacated by Chief Justice Leah Ward Sears. Perhaps a Hispanic will make the governor's short list” (Reference DiguetteDiguette 2009).
H6: Supreme Court Appointments Hypothesis: States are more likely to seat their first black and female justices sooner when appointments to the U.S. Supreme Court generate heightened attention to the issue of judicial diversity.
Neighboring State Diffusion
The state politics literature is replete with theoretical arguments contending that policies diffuse across neighboring states (e.g., Reference Berry and BerryBerry & Berry 1990, Reference Berry and Berry1992; Reference Mooney and LeeMooney & Lee 1995). Although the process of seating judges is quite different than the policy adoption process, it is reasonable to think that states could be influenced by their neighbors to seat minority judges—especially when these states have yet to seat their first black or female justices. News of a state seating its first black or female justice may spread across state lines, generating pressure in neighboring states to examine the level of diversity on their own courts.Footnote 2
H7: Neighboring State Diffusion Hypothesis: States are more likely to seat their first black and female justices sooner as the percentage of neighboring states having seated their first black and female justices increases.
Data and Measurement
I collected information on the first black and female justice seated by each state from a variety of sources, including state supreme court histories, law review articles, and newspaper archives.Footnote 3 To model the time until these events occur, I estimate a series of duration models (Reference Box-Steffensmeier and JonesBox-Steffensmeier & Jones 2004). The risk set, which identifies instances where states are at risk of experiencing the event of interest, consists of any opening on a state supreme court.Footnote 4 The outcome variables take values of 0 for openings prior to the first black or female justice being seated. A 1 is scored when a state seats its first black or female justice, and that state is subsequently dropped from the analysis.
Explanatory Variables
Differences in selection mechanisms are captured using indicator variables for partisan elections (Partisan Selection) and nonpartisan elections (Nonpartisan Selection). Citizen Liberalism is captured using Reference BerryBerry et al.'s (1998) measure of citizen ideology. Nominating Commission is an indicator variable scored 1 when the appointing authority chooses from a list of candidates supplied by a nominating commission and 0 otherwise. Seats counts the number of seats on a state's supreme court. Term Length scores the number of years before justices face reelection or reappointment. Mandatory Retirement is an indicator variable scored 1 for states that require judges to retire at a certain age and 0 otherwise. To capture salient federal appointments, I include indicator variables for each of the three black or female justices nominated during the sample period—Sandra Day O'Connor (O'Connor), Clarence Thomas (Thomas), and Ruth Bader Ginsburg (Ginsburg). These variables are scored 1 in the two years following the nomination and 0 otherwise. To capture any diffusion effects, I score the percentage of neighboring states that have seated their first black or female justices (Neighboring States).
Control Variables
I also include a set of control variables in each model. To capture the size of the eligibility pool, I include the percentage of black or female lawyers in a state (Black Attorneys and Female Attorneys). I also include the percentage of blacks or females of the total voting age population (Black Voting Age and Female Voting Age).Footnote 5 Openings scores the number of openings in a given year. It seems reasonable that political minorities may be more likely to break onto a state supreme court as the number of openings in a year increases. South is an indicator variable scored 1 for the eleven states of the Confederacy and 0 otherwise. I also account for whether a state already seated its first black or female justice. A state may be more likely to seat its first black justice, for example, if it already seated its first female justice. Thus, I include Female Seated in the first black justice models and Black Seated in the first female justice models, both of which are indicators scored 1 for openings following the first female and black justices being seated and 0 otherwise.
Estimation and Results
Tables 1 and 2 present results from duration models on the time until states seat their first black and female justices. The models were fit with logit. Both models were estimated with cubic polynomials to account for temporal dependence (Reference Carter and SignorinoCarter & Signorino 2010). Standard errors are clustered by state to correct for the non-independence of observations within states. The estimation period spans from 1970–2005 for the first black justice model and 1970–2002 for the first female justice model, the year the last state seated its first female justice.Footnote 6 Michigan is not included in the first black justice model because it seated its first black justice before 1970.Footnote 7 Similarly, Arizona, Hawaii, Massachusetts, North Carolina, and Ohio are excluded from the first female justice model because they seated their first female justices before 1970.
Table 1. Duration Analysis of the Time until States Seat Their First Black Justices
Note: Standard errors are clustered by state. ***p < .01; **p < .05; *p < .10. Cubic polynomials were included in both models, but are excluded from the table.
Table 2. Duration Analysis of the Time until States Seat Their First Female Justices
Note: Standard errors are clustered by state. ***p < .01; **p < .05; *p < .10. Cubic polynomials were included in both models, but are excluded from the table.
In both tables, Model 1 is a full-sample model that excludes the diffusion variable; Model 2 includes the diffusion variable, but excludes information from Alaska and Hawaii, which are not bordered by another state. Overall, the results are quite similar across models. As a result, I will discuss the results from Model 1 for both the first black and first female justice models unless otherwise noted. The overall model fit in both instances is quite good. Both the first black justice (χ2(20) = 126.33, p < .001) and first female justice (χ2(20) = 83.18, p < .001) models offer improvements over the null models. Moreover, the areas under the ROC curve—which convey the percentage of correct classifications from a random draw of 0,1 pairs on the outcome variables—are.91 and.85 for the first black and female justice models, respectively.Footnote 8
The estimated coefficients for Partisan Selection and Nonpartisan Selection are not of substantive interest since they represent effects only for the counterfactual scenario where Citizen Liberalism = 0 because of the interaction terms. The estimated coefficient for Citizen Liberalism gives its effect for states with appointment systems (i.e., when Partisan Selection and Nonpartisan Selection = 0). 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. Substantively, a one standard deviation increase in Citizen Liberalism increased the odds of the first black and female justices being seated by a factor of about 2.0 and 2.2, respectively, holding the other explanatory variables constant.
The key results concerning the influence of selection systems are the estimated coefficients for the interaction terms. The interaction term between Partisan Elections and Citizen Liberalism is positive and significant in the first black justice model, but is statistically indistinguishable from 0 in the first female justice model. For the first black justice model, this result suggests that the likelihood of a state seating its first black justice increases on average in states with Partisan Selection as Citizen Liberalism increases. A similar increase in Citizen Liberalism in states with Nonpartisan Selection, however, has no effect on average on the likelihood of the first black or female justice being seated. Overall, these results are intriguing. Selection systems seem to matter, but only for the seating of a state's first black justice, and only at higher levels of citizen liberalism.
States that utilize a Nominating Commission seat their first black justices sooner than states that do not. Substantively, the use of a nominating commission increases the odds of the first black justice being seated by a factor of about 4.5, holding the other explanatory variables constant. These commissions had no effect, however, on the likelihood of a state seating its first female justice. Although nominating commissions sometimes expressly value diversity in their consideration of candidates, there has been some question about whether they have actually increased diversity on the bench. The result from the first black justice model suggests that nominating commissions have been at least somewhat successful in increasing diversity, although commissions did not appear to have an effect on the likelihood of a state seating its first female justice.
Turning to institutions that affect turnover, the number of Seats on a state's supreme court is positively associated with the timing of states seating their first black justices. Substantively, an additional seat increases the odds of a state seating its first black justice by a factor of about 2.1, holding the other explanatory variables constant. The effect of the number of seats on the likelihood of a state seating its first female justice is not statistically distinguishable from 0. However, the estimated coefficient falls just short of conventional significance (p = .10, one-tailed). In Model 2, the estimated coefficient on the number of seats is statistically distinguishable from 0. Substantively, an additional seat increases the odds of a state seating its first female justice by a factor of about 1.4, holding the other explanatory variables constant.
The estimated coefficient on Term Length is statistically indistinguishable from 0 in the first black justice model. Counterintuitively, however, states with longer terms seem to have seated their first female justices sooner. Substantively, each additional year of term length increased the odds of a state seating its first female justice by about 1.0, holding the other explanatory variables constant. One possible explanation for this result may be that seats with longer terms attract more attention from interest groups, thereby increasing the pressure on states to seat their first female justices.Footnote 9 The final institution affecting turnover—Mandatory Retirement —had no impact on the timing of states seating their first female justices. A retirement rule does, however, affect the timing of states seating their first black justices. Substantively, such a rule increases the odds of a state seating its first female justice by about 3.0, holding the other explanatory variables constant. Overall, the results from both models suggest that institutions affecting turnover have a substantial impact on the timing of states seating their first black and female justices.
There is no diffusion effect from Neighboring States in the female justice model. However, in the first black justice model, states were actually less likely to seat their first black justice given an opening as the percentage of neighboring states who had seated their first black justice increased. Substantively, a standard deviation increase in the percentage of neighboring states having seated their first black justice decreased the odds of a state seating its first black justice by about 0.4, holding the other explanatory variables constant. Because of large geographic variation in black populations (e.g., smaller populations in much of the Northeast, Northwest, and Great Plains regions), I also fit a model interacting Neighboring States and Black Voting Age. The estimated coefficient for the interaction term was positive and significant (p < .10, two-tailed), which means that the diffusion effect is strongest in states with low black populations, but grows weaker on average as those populations increase. This suggests that the counterintuitive result may reflect the fact that states that have not yet seated their first black justices tend to be grouped together geographically and have small black populations.
The appointments of certain black and female justices to the U.S. Supreme Court are associated with seating decisions at the state level as expected. The odds of a state seating its first black justice increased by a factor of about 4.6 after the appointment of Clarence Thomas (Thomas) to the U.S. Supreme Court, holding the other explanatory variables constant. Furthermore, the odds of a state seating its first black justice increased by a factor of about 11.2 after the appointment of Ruth Bader Ginsburg (Ginsburg). In the first female justice model, the appointment of Sandra Day O'Connor (O'Connor)—the first female to sit on the U.S. Supreme Court—is positively associated with the seating of the first female justices on state supreme courts. Substantively, the odds of a state seating its first female justice increased by a factor of about 2.9 following Justice O'Connor's appointment. Overall, these results offer support for the idea that the appointment of political minorities to the U.S. Supreme Court heightens interest in diversifying courts at the state level. Because of the necessarily blunt nature of these measures and the possibility of confounding influences, however, more work should be devoted to determining the extent to which national events affect state-level decisions to diversify courts.
Turning to the control variables, states with higher proportions of attorneys who are black (Black Attorneys) are more likely to seat their first black supreme court justices sooner. Substantively, a 1 standard deviation increase in the percentage of attorneys in a state who are black increases the odds of a state seating its first black justice by a factor of about 1.4, holding the other explanatory variables constant. For those who value institutional diversity, this result speaks to the importance of educating more black attorneys and doing more to place blacks in positions that can serve as stepping stones to the highest judicial offices.
A 1 standard deviation increase in the size of the Black Voting Age population increases the odds of a state seating its first black justice by about 7.8, holding the other explanatory variables constant. This result may also help to explain why many states have yet to seat their first black supreme court justices: many of the states that have not seated their first black justices have relatively small black populations (e.g., Idaho, New Hampshire, Wyoming, and Vermont). Neither the percentage of attorneys who are women (Women Attorneys) or the size of the female voting age population (Women Voting Age) affected the likelihood of a state seating its female justice.
The first black and female justices were no more likely to be seated in years when states had multiple Openings. The odds of a state seating its first black justice increased by a factor of about 6.3, holding the other explanatory variables constant, when that state had previously seated its First Female justice. Similarly, the odds of a state seating its first female justice increased by a factor of about 3.0, holding the other explanatory variables constant, when that state had previously seated its First Black justice. Southern states (South) were no faster or slower to seat their first black or female justices.
Lastly, it is instructive to consider the underlying hazard rates. As noted previously, each of the duration models includes a cubic polynomial to account for temporal dependence. Figures 2 and 3 plot the probabilities of seating the first black and female justices respectively as a function of time (note that the probabilities on the y-axes are different for each model). Figure 2 shows that the baseline probability of a state seating its first black justice increased or held steady from 1970 until the late 1970s, then dropped steadily through about 1990, at which point it flattened out until taking a sharp upward turn around 2000. In contrast, the hazard for the first female justice model—plotted in Figure 3—looks quite different. Following 1970, there is a slow but steady increase in the probability of a state seating its first female justice until around 1990, at which point there is a sharp upward turn until 2002 when the final state seated its first female justice.
Figure 2. Hazard rate (first black justice).
Figure 3. Hazard rate (first female justice).
The baseline hazard rate is instructive insofar as it provides insights regarding the impact of omitted variables (see Reference BeckBeck 2010). The ultimate goal is to produce a model accounting for the diversification of state supreme courts without duration dependence. Figures 2 and 3 suggest that future work would benefit from more fully considering how the process of diversifying courts has differed for blacks and women over time. Although blacks and women have both encountered formidable barriers securing public office, there is reason to expect that institutional and political pressures may have affected the underlying processes differently for these groups over time.
Conclusion
When it comes to determining the time until states diversify their supreme courts, institutions matter—particularly institutions that affect turnover. States are more likely to seat their first black and female justices sooner when institutional rules result in more frequent openings. Furthermore, in the ongoing process of states seating their first black justices, the choice of selection system matters: on average, states with partisan selection systems have seated their first black justices sooner, but only in states with relatively liberal citizenries. Thus, just as the interaction of institutional design and politics matters in the context of judicial decision-making (e.g., Reference Brace and BoyeaBrace & Boyea 2008; Reference Brace and HallBrace & Hall 1997; Reference CaldaroneCaldarone et al. 2009), so too does it affect judicial selection. One remaining issue is to examine the impact of interim appointments on judicial diversity. Unfortunately, we lack systematic information on how particular judges obtain their seats. But given the important role that interim appointments play in diversifying courts (Reference Holmes and EmreyHolmes & Emrey 2006), further exploring this issue will be an important step forward in the study of judicial diversity in the states.
The results also suggest that states may be more likely to seat their first black and female justices during periods of heightened attention to the issue of judicial diversity generated by the appointment of political minorities to the U.S. Supreme Court. These results lend support to a growing literature on vertical diffusion, providing additional evidence that national forces can help shape state level outcomes. Because of the necessarily blunt nature of measuring vertical influence and the potential for confounding effects, however, more work is needed to examine the possibility that decisions at the federal level help shape judicial diversification in the states. The recent appointments of Sonia Sotomayor—the first Hispanic Supreme Court justice—and Elena Kagan offer new opportunities to examine whether appointment decisions at the federal level have spillover effects in the states.
These results are interesting from a political development perspective, but they are also relevant for enhancing our understanding of the political and institutional conditions that promote judicial diversity. This is important for several reasons. First, 19 states have yet to seat their first black justices. Moreover, groups such as Hispanics and Asian Americans are still underrepresented on state courts (Reference Hurwitz and LanierHurwitz & Lanier 2008). To the extent policy makers and voters value institutional diversity, the results suggest a normative argument for considering changes in institutional rules such as appointment systems, court size, term length, and mandatory retirement rules. The results also suggest that states with smaller numbers of black lawyers may need to focus on ways to elevate those lawyers into positions that will allow them to gain the experience necessary to obtain a seat on the state's highest court. This is a lesson that is likely to also hold true when it comes to seating members of other minority groups.
More broadly, understanding the determinants of judicial diversity is an important task because of the impact diverse courts have on political processes and public opinion. There is evidence, for example, that female (Reference BaldezBaldez et al. 2006; Reference SongerSonger et al. 1994; Reference PeresiePeresie 2005) and black (Reference SchererScherer 2005; Reference WelchWelch 1988) judges vote differently on certain issues than their male and nonblack counterparts do.Footnote 10 Moreover, the mere presence of a female (Reference BoydBoyd et al. 2010; Reference Farhang and WawroFarhang & Wawro 2004; Reference PeresiePeresie 2005) or black (Reference Cox and MilesCox & Miles 2008) judge on a panel may influence the way other judges decide cases.Footnote 11 Political minorities also bring unique perspectives to the bench that help shape the design of legal rules (Reference SiskSisk et al. 1998). Thus, diversity on the bench has important implications for judicial decision-making, the interaction of judges on collegial courts, and judicial policymaking. Descriptive representation on courts also enhances perceptions of institutional legitimacy among blacks (Reference Scherer and CurryScherer & Curry 2010).Footnote 12 More generally, representation fosters trust in institutions among political minorities and increases democratic engagement (e.g., Reference Bobo and GilliamBobo & Gilliam 1990; Reference VerbaVerba et al. 1997; Reference Voss and LublinVoss & Lublin 2001).Footnote 13 If judicial diversity matters for these reasons, it is important to understand why some states lag behind in generating that diversity.
Several important questions remain in our effort to further unravel the determinants of judicial diversity. For example, how do we explain the 115-year gap between South Carolina seating its first and second black justice, or the fact that several states have not yet seated their second black or female justices? Do different retention systems influence how long minority justices serve? How important are interim appointments to diversification? Questions also remain as to what factors will explain the seating of justices from other minority groups. Exploring these and similar questions will be important for further developing our understanding of how political institutions and pressures influence judicial diversity in the states.
