For more than a century, judicial selection methods have been a regular topic of political debate and scholarly analysis (Shugerman Reference Shugerman2012; Marcin Reference Marcin2015; Kritzer Reference Kritzer2020; Gibson and Nelson Reference Gibson and Nelson2021). These discussions often focus on how each method of selection affects judicial independence, judicial accountability, the quality of judges, judicial decision making, or diversity among state judges. Much of this rhetoric, often originating from judges and legal organizations, asserts that judicial appointments promote the selection of skilled, independent judges (Phillips Reference Phillips2009; Schneider Reference Schneider2010). Justice Sandra Day O’Connor and former Arizona chief justice Ruth V. McGregor (Reference O’Connor and McGregor2011-Reference O’Connor and McGregor2012) argued that appointment systems have “proved” themselves “as an answer to the threats posed by judicial elections” able to “maintain or restore the impartiality of our judges, minimize partisan politics in judicial selection” and “produce a well-qualified judiciary” among other benefits. Meanwhile, other observers argue that judicial elections promote accountability and citizen interest in the courts while also choosing judges similar in quality to their appointed counterparts (Bonneau and Hall Reference Bonneau and Hall2009). The result of these debates has been gradual, widespread reform of states’ judicial selection systems with various methods adopted to suit states’ own preferences. Policy debates about the ideal system to pick judges continue, as do arguments about whether judicial independence or accountability is associated with appointments or judicial elections as their backers claim.
Whatever the goals of policymakers who establish a state’s method of choosing judges, there are political and practical realities that stand in their way (Shugerman Reference Shugerman2012). Politicians are unlikely to approach judicial appointments in an apolitical fashion that values merit exclusively. Citizens tasked with choosing judges may be uninformed, respond to name recognition, or receive ballots with no alternative candidates. In addition, many judges do not reach office initially via the regular process adopted by their states. Instead, they receive interim appointments to the bench (Gibson and Nelson Reference Gibson and Nelson2021). Bannon (Reference Bannon2016) explained in the American Bar Association’s periodical The Professional Lawyer that “in some states that provide for elections, interim appointments are a central – yet under-scrutinized – aspect of the selection process, since judges routinely step down before the end of their terms so as to provide the governor with an appointment.” In states where partisan or nonpartisan elections are the regular method of selection this means that the voices of the people are hushed in favor of elites’ preferences, limiting popular influence on the composition of the courts.
Gibson and Nelson (Reference Gibson and Nelson2021) explain that it is vital to distinguish between “formal” methods of judicial selection for full terms and “informal” methods used to fill vacancies that arise when a judge resigns, retires, or dies during an unfinished term. Because interim vacancies are common, the formal method of selection adopted by a state often provides limited insight regarding how many of its judges reach the bench (Holmes and Emrey Reference Holmes and Emrey2006; Gibson and Nelson Reference Gibson and Nelson2021). In 46 states, interim vacancies can be filled by the governor via appointments (Berry and Lisk Reference Berry and Lisk2017). Interim appointments happen regardless of the formal method of judicial selection used by a state. This has potential consequences for the quality and independence of judges because the appointment power shifts from its formal specification to an informal process with the governor as the final player. Despite their frequency and importance, interim judicial appointments are rarely the subject of scholarly analysis.
As stated above, interim judicial appointments empower elites – including in states typically identified as using partisan or nonpartisan elections to pick or retain judges. Accordingly, it is important to understand the criteria used by governors and their partners to select judges for interim appointments. Advocates of judicial appointments emphasize that they promote judicial competence and independence. However, governors approach judicial selection with political goals in mind (Vining and Wilhelm Reference Vining and Wilhelm2011; Goelzhauser Reference Goelzhauser2018b). If political influence is prominent in commission-based selection processes, it would be contrary to the arguments of many reform advocates (Schneider Reference Schneider2010). We investigate the impact of political considerations and qualifications on interim judicial appointments. We expect that governors’ decisions regarding interim judicial appointments are influenced by their political attachments to applicants to the detriment of qualifications. We assess this relationship using data from Georgia’s mixed selection system (Goelzhauser Reference Goelzhauser2018a) where governors appoint judges in the event of interim vacancies. We give particular attention to the effects of ideological proximity, campaign donations to the governor, and professional background.
We analyze data including both successful and unsuccessful applicants for judicial appointments throughout the state court system, a rarity in this literature. This allows us to explore why some prospective judges are appointed while others are not. In addition, it broadens the scope of our analysis beyond the courts of last resort that are usually featured in state courts research. Our empirical analyses of interim judicial appointments in Georgia from 1991 to 2014 indicate that political attachments have a substantial impact on interim appointments while only judicial experience and law school quality have consistent effects among qualifications. Governors use informal means of filling vacancies to put likeminded judges on the bench – even where nonpartisan nominating commissions exist or the voting public might have preferred a different candidate. Applicants are more likely to be appointed if they are similar ideologically to the governor and donated to his campaigns. The influence of donations is notable, providing systematic evidence that donors are advantaged in the appointment process. The impact of political factors undercuts the effects of racial background, suggesting that judicial diversity is less of a concern than ideology or past support. These results have significant implications for the ongoing debate regarding judicial selection methods, the quality of state judges, and whether judicial selection can be isolated from state politics.
Judicial selection and interim appointments in the states
Among the most enduring questions among scholars and reformers interested in judicial selection is whether certain methods of selecting judges are best suited to fostering judicial independence and maintaining a highly qualified state bench. If the judiciary drifts into the political realm, critics argue that it may lose public confidence, attract inferior judges, stifle judicial diversity, disfavor minority rights, or lead judges to abandon impartiality for electoral advantage (Gill Reference Gill2013; Hume Reference Hume2013).
Reformers have responded to concerns about the balance between judicial independence and accountability by changing states’ procedures for recruiting, vetting, and choosing judges (Gill Reference Gill2013; Marcin Reference Marcin2015; Kritzer Reference Kritzer2020). The long-term trend has been for states to adopt methods of selection said to reduce the role of politics. Although partisan elections were once the primary alternative to appointment by political elites, only eight states now use them for appellate courts and eleven states do so for lower courts. The widespread adoption of nonpartisan elections or merit selection as states’ formal selection systems is intended to strengthen judges’ independence, weaken their links to the political system, and aid the recruitment of skilled judges. The informal process for interim appointments tends to follow the merit selection model, using a nominating commission to vet applicants before the governor selects a judge. These procedures are intended to elevate the goal of filling the state bench with qualified, independent judges rather than individuals who secure judgeships due to pandering, patronage, or political connections.
Advocates of commission-based selection argue that having candidates vetted by a purportedly nonpartisan judicial nominating commission (JNC) and then appointed by the governor is more likely to produce qualified, independent judges. By contrast, judicial elections or unconstrained appointments are characterized by their detractors as encouraging biased judicial decisions, demagoguery, or the ascension of less qualified judges (e.g., O’Connor and McGregor Reference O’Connor and McGregor2011-Reference O’Connor and McGregor2012). However, there have been many criticisms of the claim that merit selection is divorced from partisan politics. For example, former chief justice Clifford Taylor of the Michigan Supreme Court called merit selection “an attractive ruse” (Taylor Reference Taylor2010). Ultimately, few empirical studies assess the claim that a commission-based appointment process squelches political influence. The limited previous research tends to focus on state high courts (Holmes and Emrey Reference Holmes and Emrey2006; McLeod Reference McLeod2012). Goelzhauser (Reference Goelzhauser2018b) provides an assessment of this relationship in Nebraska’s merit selection system, identifying persistent political influences contrary to the claims of reform advocates. Despite decades of empirical research on judicial selection systems and their consequences, scholars still have little understanding of how well institutional mechanisms developed to assure judicial independence and competence actually guard courts from politics and elevate the best and brightest to the state bench.
The impact of interim judicial appointments is substantial (Berry and Lisk Reference Berry and Lisk2017).Footnote 1 Holmes and Emrey (Reference Holmes and Emrey2006) found that more than half of state supreme court judges are initially appointed to the bench. More recent research confirms this trend while also noting that roughly half of state high court justices were initially appointed, even in states where judicial elections are the formal selection mechanism (Berry and Lisk Reference Berry and Lisk2017; Gibson and Nelson Reference Gibson and Nelson2021). Judges who receive an interim appointment garner benefits from doing so. In states with retention elections, judges who take the bench through an interim appointment run for full terms without an opponent. In states with competitive elections, appointees enjoy the incumbency advantage if they seek reelection (Bonneau and Hall Reference Bonneau and Hall2009). Interim appointments allow governors to staff the courts independent of voters’ judgments even where competitive judicial elections occur, and approximately one-third of their appointees do not face electoral opposition once in office (Berry and Lisk Reference Berry and Lisk2017).
Many high-profile controversies have drawn attention to the influence of politics in the interim appointments of state court judges – including in states where a nominating commission is part of the process.Footnote 2 For example, Governor Roy Barnes of Georgia – where the formal method of selection is nonpartisan elections – was criticized in statewide media for using interim appointments to place numerous campaign donors and personal associates on the state courts during his tenure (Judd Reference Judd2002). In 2020, a Georgia state supreme court election was cancelled when the incumbent timed his retirement announcement so that Governor Brian Kemp could appoint a conservative judge rather than allow a competitive election in which former Democratic Congressman John Barrow had already launched a bid (Millhiser Reference Millhiser2020). Events like these underscore that interim appointments are susceptible to political influence despite their distance from voters, legislative debate, or a formal advice-and-consent process.
Previous empirical studies of political influence and judicial qualifications in commission-based selection are limited. Goelzhauser (Reference Goelzhauser2018b) explained that such studies are rare partially due to the scarcity of systematic data available to scholars. Despite filing public records requests with all states using merit selection, only one state provided information to Goelzhauser (Reference Goelzhauser2018b) regarding the applicants to fill judicial vacancies or the shortlists developed by the nomination commission. Using Nebraska’s data from 2000 to 2016 (N = 980 applications), Goelzhauser (Reference Goelzhauser2018b) found evidence that party affiliation, applicants’ campaign donations and professional background influenced commissions and governors – though not always in similar ways. Governors were especially responsive to applicants’ party affiliation and donations to the opposing political party. Goelzhauser (Reference Goelzhauser2018b) also found evidence that the applicant’s gender correlates with commission recommendations, with female applicants at a significant disadvantage. Unfortunately, these data lacked information regarding the race or ethnicity of applicants. This assessment of merit selection in Nebraska provides novel evidence regarding the collision of politics and commission-based appointments. We extend this literature to interim appointments in a mixed selection system (Goelzhauser Reference Goelzhauser2018a) where they occur frequently even though the state’s formal method of selection is competitive popular elections. Interim appointments in elected states undercut popular influence that would prevail in the event of regular turnover.
Political elites and interim judicial appointments
Whatever the formal means of judicial selection is in a state, the informal use of interim appointments to fill judicial vacancies gives political elites opportunities for outsized impact (Gibson and Nelson Reference Gibson and Nelson2021). Governors are central to this process in most states (Berry and Lisk Reference Berry and Lisk2017), even if these judges must later face a popular election. In thirty states, governors participate in the selection of judicial nominating commission members.Footnote 3 In three states, governors have sole authority to select these commissioners. These institutional powers give most governors a substantial role in picking judges even while commission-based selection is perceived as the least politicized method of choosing judges. While this process seems to constrain governors, it is likely that a state’s chief executive influences each stage prior to the appointment. Both potential applicants and commissioners are likely to consider the preferences of the governor given that he holds the appointment power (Watson and Downing Reference Watson and Downing1969). As Tarr (Reference Tarr2011) explains, judicial nominating commissions “assist the appointing authority, and so it should take into account her needs and predilections.”Footnote 4
Governors use their appointment power to staff both judicial nominating commissions and the bench with political allies. Political influence on nominating commissions was detected by scholars decades ago, with several discovering evidence that partisanship influenced the choice of commission members (Watson and Downing Reference Watson and Downing1969; Henschen, Moog, and Davis Reference Henschen, Moog and Davis1990) and which applicants advanced to the shortlist presented to the governor (Ashman and Alfini Reference Ashman and Alfini1974; Goelzhauser Reference Goelzhauser2018b).
Governors overwhelmingly pick judges from their own party (Dubois Reference Dubois1985; Holmes and Emrey Reference Holmes and Emrey2006; McLeod Reference McLeod2012), much like American presidents tend to staff the federal courts with their allies (Goldman Reference Goldman1997). This is expected and understandable given the aspiration of political elites to steer the courts toward their own ideological predispositions and reward co-partisans with desirable government jobs. Cross-party appointments are relatively rare, though they are more common in states with merit selection (McLeod Reference McLeod2012). Unfortunately, we have little information regarding how judicial appointments are affected by applicants’ qualifications or political cues, such as campaign donations (cf., Goelzhauser Reference Goelzhauser2018b) and the ideological proximity between the governor and each applicant.
Theory and hypotheses
We argue that, where political elites have the opportunity to inject their own preferences into the selection of state judges, they will do so. We expect to observe this behavior despite claims that appointed judges are relatively isolated from politics. The widespread use of interim appointments to fill judicial vacancies assures that most governors can exert such influence regardless of the formal selection process enshrined in state law. The relevant actors in this process, including members of the nominating commission, are aware of the governor’s preferences regarding the backgrounds and predispositions of judicial appointees (Watson and Downing Reference Watson and Downing1969). As a result, we theorize that the political attachment between the governor and judicial applicants influences the selection of a state court judge. We posit that our political attachment theory captures an important aspect of interim appointments that is often overlooked, especially for judges below the court of last resort.
There are several pathways by which a governor is likely to influence the selection of judicial candidates and appointees. First, the governor participates in the selection of nominating commission members in thirty states where they are convened.3 Governors are likely to select allies for these positions (Tarr Reference Tarr2011), and we expect that these commissioners will behave in ways consistent with the interests and preferences of the governor.
Second, whether appointed by a governor or not, commissioners’ perceptions of acceptable candidates are likely influenced by the partisan and political leanings of the governor (Tarr Reference Tarr2011). Governors usually pick judges from their own political party (Goelzhauser Reference Goelzhauser2018b), and commissioners will be aware of this tendency. Even in states where governors are bound by law to select a judge from the JNC’s shortlist, we do not expect the commissioners to shirk the governor by offering a slate of candidates who she would find unacceptable.
Finally, the governor’s political attachment to a judicial candidate is likely to influence her inclusion on a shortlist and likelihood of appointment. This may be due to direct influence by the governor (Watson and Downing Reference Watson and Downing1969) or commissioners recognizing the types of candidates favored by the governor (Sheldon and Maule Reference Sheldon and Maule1997; Tarr Reference Tarr2011; Goelzhauser Reference Goelzhauser2018b). With these conditions in mind, Judge Peter Olszewski (Reference Olszewski2004) argued that merit selection merely “substitutes committee politics for electoral politics” with “[t]he appearance of expertise and non-partisanship…. a facade.”
There are multiple ways for commissioners and governors to identify the political allies of the state’s chief executive. One indicator of shared political predispositions is the ideology of the potential judge. Despite claims that commission-based selection dampens the role of politics in judicial selection, these considerations may remain important in interim appointments. Commissioners and governors are likely to favor candidates whose ideological leanings align with the governor’s preferences. Governors, like presidents, recognize that the judges they appoint represent important opportunities to affect the law and politics of their states. Judges tend to serve longer terms than either legislators or governors, and are usually retained if they seek reelection or reappointment.
Contributions to the governor’s political campaigns are a more direct manifestation of a candidate’s political attachment to the governor. These donations provide an overt signal of support for the governor. They may also set certain applicants apart from their competition when their qualifications are similar. In short, governors may be inclined to choose candidates because of their political and financial support. Our expectations regarding the impact of political attachment on judicial selection motivate two key hypotheses regarding actions by judicial nominating commissions and governors:
Hypothesis 1 The greater the ideological similarity between a judicial applicant and the governor, the more likely she will be appointed to the bench.
Hypothesis 2 Judicial applicants who contributed to the governor’s political campaign are more likely to be appointed to a judgeship than applicants who did not.
In addition to the political attachments described above, we also expect that several personal characteristics and professional qualifications influence selection for judgeships. Several studies confirm that judicial appointments are influenced by the race and gender of potential judges. Martin and Pyle (Reference Martin and Pyle2002) reported that appointment systems were related to the selection of African-American state supreme court judges serving in 1999. In addition, Bratton and Spill (Reference Bratton and Spill2002) and Holmes and Emrey (Reference Holmes and Emrey2006) determined that appointment systems facilitate the selection of nontraditional judges. The latter study concluded that interim appointments are a key pathway for nonwhite high court justices to be appointed, especially while a Democrat is governor. Whether they prefer to diversify the judiciary per se or appeal to constituent groups by appointing their members, governors have incentives to choose judges who are women or identify as racial or ethnic minorities.Footnote 5
There is also substantial evidence that regionalism affects the judicial selection process in the states (Canon Reference Canon1972; Dubois Reference Dubois1983; Glick and Emmert Reference Glick and Emmert1987). We consider whether judges received their legal education in-state. We perceive this as an indication that a judicial applicant is more likely to have professional and social ties to the state’s elites. Numerous studies of judicial selection conclude that legal qualifications influence the process (Glick and Emmert Reference Glick and Emmert1987; Reddick Reference Reddick2002). Two key determinants of qualifications are professional experience and education. Applicants with greater experience and a more prestigious educational background are likely to have an advantage over their competitors with lesser experience or training.
Data and research design
To determine the relationship between political attachment and judicial applicants’ success in the appointment process, we investigate this phenomenon using an original data set of applicants for interim appointments in Georgia from 1991 to 2014 filed with the Georgia Judicial Nominating Commission. This covers the gubernatorial administrations of Zell Miller to Nathan Deal, and includes 4,835 applications for interim appointments to both trial and appellate courts. These data were provided by request from the privately held records of former chairmen of the Georgia Judicial Nominating Commission. These data include the applicants’ characteristics, their legal experience, and the outcome of their application. Unfortunately, information about applicants who made shortlists was discarded. As Goelzhauser (Reference Goelzhauser2018b, 158) noted, the practice of discarding such data or exempting it from public disclosure is common throughout the United States. These data provide a rare opportunity to examine the impact of political attachments between governors and those selected for the bench throughout the state court system. This is the first study to analyze a diverse set of interim appointments in a mixed selection state with systematic data.
Interim vacancies in Georgia are filled by gubernatorial appointments after vetting by the Judicial Nominating Commission. We focus on interim appointments in Georgia to examine the influence of applicants’ political attachments and qualifications where the informal method of selection permits the governor to appoint judges rather than rely on the state’s formal method of selection – nonpartisan judicial elections. Georgia has used nonpartisan elections since 1984. Most Georgia jurists initially come to the bench via appointment (Bullock and Gaddie Reference Bullock and Gaddie2018).
Georgia’s sociopolitical context reflects broader American political trends in important ways. The time period we examine included episodes of both Democratic and Republican control of state government, as well as a gradual polarization between the parties (Shock Reference Shock2020).Footnote 6 In addition, the state underwent a demographic change, including urbanization (Nord Reference Nord2018) and an emergent minority population – especially among Hispanics (Hood and Bullock Reference Hood and Bullock2021). These trends have continued, contributing to narrow wins in statewide elections for Governor Brian Kemp (R) in 2018, President Joe Biden (D) in 2020, and both Senator Jon Ossoff (D) and Senator Raphael Warnock (D) in January 2021 runoff elections (Bullock Reference Bullock, Bullock and Rozell2022).
The Georgia Judicial Nominating Commission was established in 1972 by Governor Jimmy Carter via executive order. Subsequent governors have periodically renewed the JNC through further executive orders. The current order describes it as a “[n]on-partisan, non-political, judicial nominating commission composed of outstanding lawyers and laypersons…” (Georgia Executive Order No. 02.07.19.01 2019). The commission makes recommendations to the governor to fill vacancies on all courts of record, offering to the governor a slate of candidates it deems qualified. Accordingly, it assists with the evaluation of candidates for Georgia’s high court, court of appeals, superior courts, and state courts. The governor unilaterally appoints members of Georgia’s JNC who serve at his pleasure.
Data provided by the commission chairmen include several characteristics of the candidates relevant to the evaluation process. These include identifying information about each applicant (full name and city) and their Georgia Bar number. This information was used to code limited demographic characteristics (whether they are African American and their gender) and information about their legal background (the name of the law school they attended and their year of admission to the Georgia State Bar) by hand. Because Georgia does not require party registration, we are unable to replicate Goelzhauser’s (Reference Goelzhauser2018b) use of state voter records to identify applicants’ partisan loyalties. Instead, we rely on recent advances in measurement to examine the impact of ideology in a subset of the data.
We examine the entire population of applicants for the effects of diversity, legal background, and merit. These factors are acknowledged as appropriate considerations for merit selection by Georgia’s JNC and are consistent with previous research findings. We then examine the influence of additional variables that capture political ideology and prior campaign contributions. We use a merged subset of observations from the Georgia JNC that were combined with information from Adam Bonica’s Database on Ideology, Money in Politics, and Elections (DIME) (Bonica Reference Bonica2016). The DIME data contain a massive amount of information on individual political contributors, contributions, and the candidates who received these donations. Bonica used these data to develop a common-space measure of political ideology by mapping the positions of donors and their recipients (Bonica Reference Adam2013; Bonica Reference Adam2014). The measure, called a campaign finance score or CFScore, is a continuous operationalization of ideology where the most negative values represent the most liberal ideal points and the most positive represent the most conservative ones.
DIME was particularly useful for this project because it contains donation records and CFScores for individual contributors who are not elected officials in contrast to many existing measures of ideology. The vast majority of applicants to the Georgia JNC were practicing attorneys in the state without experience in elected office. Thus, inferring partisan affiliation or ideology for these otherwise private individuals presents a challenge that DIME can overcome.
Applicants considered by the Georgia JNC were linked to corresponding DIME contributor records using biographical data. DIME includes over two million contributor records for Georgia between 1980 and 2014; thus, covering donations well before the beginning of our JNC applicant data in 1991. Bringing DIME together with the JNC data required clear merging criteria to combine them because there are no common identification numbers shared between these datasets to facilitate a straightforward merge. Instead, an applicant’s first name, middle initial, last name, suffix, and home city were used to merge the two data sources.
These merge criteria performed fairly well. A subset of 2,002 applications of the total 4,835 applications (41.41 percent) submitted to the JNC during our time period were merged with DIME. Several variables within the DIME data relate to a contributor’s occupation; we used them to assess whether the merged applicants were, in fact, attorneys.Footnote 7 Using these fields, only 71 individual contribution records of 5,191 successfully merged from the Georgia DIME data (1.37 percent) had information that indicated they were engaged in a different field of work (e.g., timber farmer, car dealer); thus, they were excluded from the subset. Overall, 95.4 percent of the JNC applications merged with DIME had at least one contribution, as well as information confirming they were an attorney, available either in their occupation fields or via a simple web search. The remaining 4.6 percent either had vague career fields that did not prove whether the individual was an attorney (e.g., business man, vice president) or their career information was left blank in commission records. While this merged subset of observations is necessarily imperfect due to a lack of common identifiers, it permits an analysis of a political process usually shielded from scrutiny.
Finally, we utilize imputed ideological distance scores to address the 2,833 applications which did not have identified past campaign contribution activity. Specifically, we calculate and use the median CFScore for all attorneys in the city where the applicant resides. Utilizing the DIME data, we identified all contributions during the study period made by Georgia residents whose occupation field indicates they are an attorney (Bonica Reference Bonica2016).Footnote 8 From this, we utilize the median score for attorneys from the Georgia Judicial Nominating Commission applicant’s home city. This measure allows us to operationalize a heuristic a governor may rely on when there are no obvious external indicators of political ideology, such as past campaign contributions or party registration. A governor or the JNC is likely to know a legal community’s culture by reputation, and apply that perception to an individual applicant from that community when they lack other knowledge. Further, a governor might be more likely to draw geographically from constituencies that mirror his or her own ideology in order to reward them. While this is an inexact measure of this dynamic, it facilitates a preliminary examination of political effects among a larger set of applicants without the precondition that each of them made or received a campaign donation.
Methods and variables
We first explore the effect of ideological distance or prior donations to the sitting governor in a bivariate context. Then, we utilize three models to test our political attachment theory. We utilize logistic regression models with fixed effects for the vacancy to explain whether an applicant to the JNC will be appointed. The dependent variable takes the value of 1 if an applicant is appointed by the governor, and 0 otherwise. These data capture who applied for each vacancy and which applicant received the appointment.Footnote 9 In Model 1, we utilize the population of applications considered by the JNC. In Model 2, we focus on the subset merged with information from DIME on contributions and ideology. Finally, Model 3 includes the ideology score of the median lawyer in the applicant’s home city for those applicants who had no CFScore within DIME identified through the merging criteria. Summary statistics for our data are reported in Table 1. The summary statistics for each sample are similar. The subset of applicants merged with DIME data is most distinct, in part, because of the restricted sample size resulting from employing a fixed effects approach.Footnote 10
Table 1. Summary Statistics for Georgia Interim Judicial Appointments Data (1991–2014)
Note: Observations include applicants for Georgia judicial vacancies from 1991 to 2014. “All applicants” includes all records with complete merit and diversity data. The center column includes applicants with paired donation records in DIME. The right-hand column includes all applicants with CFScores, if available, and imputed median CFScores for lawyers in an applicant’s home city if the individual applicant’s CFScore is not in DIME. DIME, Database on Ideology, Money in Politics, and Elections; LSAT, Law School Admissions; SD, standard deviation.
We capture legal qualifications through several variables. To account for their experience as attorneys, we incorporate “Years of Experience,” the number of years that have elapsed since admission to the Georgia State Bar. The impact of experience is likely non-monotonic. Applicants with little experience as attorneys may lack the track record to be suitable for an appointment. However, the most senior attorneys may be perceived as too old to fill a judgeship given the governor’s desire to choose judges who will serve long and productive careers. Thus, “Years of Experience2” helps capture the form of this non-monotonicity.
Next, a binary variable indicates whether the applicant was a judge at the time of the application. We expect that applicants with prior judicial experience will have a substantial advantage over those who do not. Aside from merit, a current judge may also be preferred for an interim appointment because her selection creates another vacancy for the governor to fill. For example, if a superior court judge is elevated to the appellate bench, the governor can reward another ally with a superior court judgeship. The same logic applies to state court judges who are elevated to the superior court, creating vacancies in the lower trial courts.
We measure the quality and prestige of an applicant’s legal education by including the median Law School Admissions Test (LSAT) score for their law school alma mater. As a measure of quality, Law School Admissions Test scores are roughly equivalent to including other alternatives, such as the U.S. News and World Report’s law school rankings. For example, Klein and Hamilton (Reference Klein and Hamilton1998) find that the median LSAT score can explain ninety percent of the differences in rankings. These were gathered from the American Bar Association’s 509 information spreadsheet for 2011 (earliest available).Footnote 11 We expect that, as the median LSAT score increases, the probability that an applicant is appointed will increase. To control for whether the law school attended by the applicant is public or private, we use a binary indicator with a value of 1 if it was a private law school, and 0 otherwise. Since regionalism likely has an impact on whether an applicant is chosen for an interim vacancy, a variable indicates if the applicant attended a Georgia law school. We expect attorneys who were educated in the state to have an advantage over those who received their education elsewhere. These lawyers are likely to be immersed in social and political networks amenable to favorable treatment in the appointment process.
To measure the diversity of candidates, we include information related to gender and race. Accordingly, binary variables indicate whether an applicant is female or African American. We expect that, all else equal, diversity should have a positive influence on the likelihood an applicant receives an appointment. Notably, Goelzhauser (Reference Goelzhauser2018b) lacked information about the racial backgrounds of Nebraska’s pool of applicants. We provide the first systematic assessment of links between race and applicants’ success in a commission-based appointment process. Finally, to account for the context of the selection process, previous applications include the number of times an applicant has been considered by the JNC before the current application. We anticipate that attorneys who have been previously passed over for an interim appointment will be less likely to be appointed.
Models 2 and 3 utilize the merged subset of applicants with DIME contributor records and ideology. “Ideological Distance” is the distance in CFScore ideal points between the applicant for a vacancy and the sitting governor. In Model 2, these differences are associated directly with each applicant. In Model 3, we assign the imputed scores described above to the 2,773 applicants who were not successfully paired with DIME. Consistent with Hypothesis 1, we expect that as the distance between the governor and the applicant increases, the probability that person will receive an appointment will decrease. Governors likely prefer to select jurists who are ideologically similar to support their agenda and further their policy goals. The variable “Donated to Governor” takes the value of 1 when the applicant has given money to the sitting governor, and 0 otherwise. Consistent with Hypothesis 2, applicants who have donated to the governor responsible for filling the vacancy will have a substantial advantage over their competitors.Footnote 12
Results
Turning first to a bivariate analysis of our two key independent variables, “Ideological Distance” and “Donated to Governor,” we find results suggestive of both statistical and substantive impacts on the probability of receiving a final appointment. In Figure 1, we display the bivariate relationship from a logistic regression with applicant-governor ideological distance used to predict appointment. This regression includes fixed effects on the vacancy.Footnote 13 Marks along the bottom and top indicate the distribution of the dependent variable. The relationship between ideological distance and appointment is distinctly negative, conveying that applicants who are more politically dissimilar from the governor are less likely to obtain an interim appointment. The graph conveys that an applicant who is the closest to the sitting governor’s ideology has a probability of appointment of 0.172, indicative of a better than average chance of appointment. By contrast, the most ideologically distant applicant from the governor has a probability of appointment is just 0.033 (a decrease in probability of 0.139). These bivariate results suggest ideology has an extremely influential effect on a supposedly apolitical merit system.
Figure 1. Bivariate Logistic Regression of Ideological Distance on Appointment.
The “Donated to Governor” variable is also influential. We display the crosstabulation of donation to the sitting governor and appointment in Table 2. Among those applicants who had not donated to the governor, only 6.28 percent of them received an appointment. By contrast, a much larger 13.73 percent of applicants who donated were appointed. These descriptive analyses provide strong initial support of our hypotheses. However, to understand fully the role of political attachments in interim appointments we analyze three models to further illustrate its effects.
Table 2. Crosstabulation of Donating to Governor and Appointment
The results of our main analytical models are in Table 3. Model 1 includes the legal qualifications, diversity, and the context of the selection process for JNC applicants from 1991 to 2014. In this model, all of these factors significantly affect whether an applicant receives an appointment to fill a judicial vacancy. “Years of Experience2” is significant and negative, indicating that the effect of experience initially increases the probability of receiving an appointment, but declines past a certain point. Similarly, prior judicial experience is a strong signal of applicant quality in the appointment process. The predicted probability of receiving an appointment based on having been a judge is approximately 0.198 while those who have no prior service as a judge have a nearly 0.051 probability of appointment. Considering few applications result in an appointment, this is a large effect.
Table 3. Logistic Regression Models Predicting Georgia Interim Judicial Appointments
Note: Coefficients are from logistic regression models fitted on the appointment outcomes. Appointed = 1. Vacancy fixed effects are included in each model.
* Significant at p < 0.10
** Significant at p < 0.05
The prestige of an applicant’s legal education, as measured by the median LSAT score of the law school they attended, significantly influences whether they will be appointed. The predicted probability of being appointed ranges from approximately 0.04 at the minimum score to approximately 0.101 at the maximum score. Within Model 1, this suggests that the quality of one’s legal education matters when being considered for a judgeship. The impact of attending a Georgia law school is statistically significant. Those who attended an in-state institution are slightly more likely to receive an appointment relative to those who did not (change in probability from 0.053 to 0.069). Another indicator related to legal qualifications, the effect of the applying attorney having attended a private law school, is not significantly different from zero.
The diversity of applicants influences the interim judicial selection in Georgia insofar as African-American attorneys are more likely to receive an appointment (statistically significant at p < 0.10). African-American applicants have a slightly increased predicted probability of receiving an appointment, roughly 0.02 greater than other individuals. Thus, race appears to be a significant, but marginal, factor in making appointments. Georgia governors have incentives to appoint African-American jurists in order to appeal to constituencies of racial or ethnic minorities, and assure representation of the state’s large Black population.Footnote 14 The coefficient for the variable tracking female applicants is not statistically significant. The context of the vacancy also constrains whether an applicant is selected for the position. The larger the number of previous applications by the attorney for the vacancy significantly decreases the probability that an applicant will be chosen.
As previously explained, Model 2 and Model 3 report two different approaches to operationalizing the role of political attachments for applicants before the Georgia Judicial Nominating Commission. The former relies only on applicants with ideology scores directly matched according to our merge criteria with Bonica’s DIME. The latter includes imputed ideology scores for attorney applicants who do not have a matching record within DIME. While the impact of legal qualifications and applicant characteristics differ between the two models, ideological distance and prior donations to the governor are statistically significant in both of them. More importantly, the substantive nature of these effects conveys the same pattern and conclusion as the bivariate analysis: the more aligned with the sitting governor an applicant is, the more likely it is they will receive an interim appointment to a judgeship.
Figure 2 displays the predicted probabilities for appointment relative to ideological distance for Models 2 and 3. The shape of the effect is similar, but Model 2’s direct ideology measure displays a more drastic effect, but with less confidence. This is likely due, at least in part, to the smaller number of observations. A substantial portion of Model 2’s observations were deleted listwise, because they lacked CFScores gathered from merging the JNC data with Bonica’s DIME. Still, the results support Hypothesis 1 in either case. Increasing ideological distance from minimum to maximum is predicted to decrease the probability of appointment by nearly 0.111 in Model 2 and 0.058 in Model 3. A one standard-deviation increase from the minimum distance decreases the probability of appointment by 0.028 in Model 2 and 0.014 in Model 3. In any event, these are important effects on the probability of overall appointment from a supposedly merit-based process.
Figure 2. Predicted Probabilities of Appointment from Models 2 and 3.
Supporting Hypothesis 2, the impact of an applicant having previously donated to the sitting governor’s campaigns is positive and significant in both models. The increase in predicted probability for having donated to the governor over not donating is 0.081 in Model 2 and 0.046 in Model 3. This suggests that not only does ideological concordance with the governor increase the chances of being selected, so does directly giving money to his or her political campaigns. These connections seem to violate the intent behind the creation of a Judicial Nominating Commission: removal from the political process and making decisions solely on applicant merit.
Models 2 and 3 report mixed results for merit’s effect in the selection process. In Model 2, the effect of “Years of Experience2” and attending a Georgia law school are not distinguishable from zero. This is a surprising result. At least within the context of this merged subset of observations, the effect of being ideologically similar to the governor or contributing funds to their campaign(s) outweighs time spent as a lawyer or being trained in the state – both of which are potentially important signals of applicant quality. This suggests that for applicants with an easily observable informational cue about their political affiliations, experience may no longer be as powerful a consideration during the selection process. Model 3 reports contrasting results, however, with “Years of Experience2” and “Georgia Law School” are statistically significant and in the expected directions. The effect of attending an in-state law school is an increase in predicted probability of 0.016, smaller than having donated to the governor or being very close to the governor ideologically. Both models contain significant and positive results for prior service as a judge and median LSAT score. Having prior judicial experience results in a predicted probability of selection of 0.292 in Model 2 and 0.151 in Model 3. As mentioned, current judges may be favored for their experience or because their elevation creates an additional vacancy on the bench for the governor to fill (or both). The increase in predicted probability of moving from the minimum median LSAT score to the maximum is 0.093 in Model 2 and 0.058 in Model 3.
Neither Model 2 nor Model 3 had statistically significant effects for Black or female applicants to the Georgia Judicial Nominating Commission. In Model 1, which did not include ideology or prior donations, the effect of being Black was predicted to be significant and positive. This suggests that concerns regarding judicial diversity are secondary in interim appointments, with their predictive power waning when political considerations enter the governor’s decision calculus. Finally, the number of previous applications was significant and negative in both models.
Discussion
Overall, the results of this examination of governors’ interim judicial appointments indicate that political attachments have a powerful influence – even where nominating commissions are established to prioritize the merit and qualifications of aspiring judges. In spite of the Georgia JNC’s role as a “[n]on-partisan, non-political” body that considers the qualifications of potential judges (Georgia Executive Order No. 02.07.19.01), ideology and donation activity affect whether an individual applicant is tapped to fill a vacancy. Our study provides additional evidence that commission-based appointment systems do not separate judges from politics despite the best intentions of reformers. Instead, ideological congruity between the eventual appointee and the governor is associated significantly with successful judgeship applications in Georgia. Prior donations to the governor are also a significant indicator of an applicant’s success when seeking a judgeship. This suggests that long-held associations with the governor bolster the chief executives’ confidence in the applicant as a potential jurist. Both our study of interim appointments in Georgia and Goelzhauser’s analysis of merit selection in Nebraska find that political factors influence governors when picking judges. This may be unsurprising given the overtly political actions of many governors in judicial selection during recent decades (Holmes and Emrey Reference Holmes and Emrey2006; Vining and Wilhelm Reference Vining and Wilhelm2011; Gibson and Nelson Reference Gibson and Nelson2021), but runs contrary to the typical arguments in favor of merit selection and nominating commissions. The inconsistent impact of our variables for professional qualifications casts doubt on claims that commission-based appointments are more suited to choosing high-quality judges than elected systems.Footnote 15
While our study is novel in examining the outcomes of interim appointments in a nonpartisan election state, it does have clear limitations. First, since the data provided by former Georgia JNC chairmen did not indicate which applicants were selected for the shortlist, our understanding of this process is incomplete. It remains unclear whether our observed effects are the result of the members of the JNC filtering certain types of candidates to the shortlist for the governor, the governor choosing from the slate of candidates, or a combination of the two. Second, Georgia’s JNC may not be representative of other states’ commission-based selection systems. In most other states that use JNCs, the governor plays a more limited role in selecting the members of the commission. Further, since the Georgia commission evaluates only applicants for interim vacancies, it may operate differently than commissions that serve as the primary pathway for selecting judges. Nonetheless, our research contributes to a growing consensus that applicants’ ideologies and prior campaign donations influence their likelihood of being appointed by a governor.
Our results and related research suggest that reformers should reconsider the extent to which they can expect to separate judicial selection from politics, even where nominating commissions are operational. Advocates of appointment systems may need to revisit their assumptions about the extent to which professional evaluators outperform the public in terms of assuring state judges are competent and independent. In addition, the distinction between formal and informal selection systems (Gibson and Nelson Reference Gibson and Nelson2021) must be taken seriously. States that set their expectations for judicial independence and accountability with an eye on their formal selection method may face a rude awakening if prospective judges routinely sidestep the expected path to the judiciary. This is a common occurrence, so its consequences must be subject to critical evaluation if we hope to understand the factors that shape judicial selection and their likely results.
Acknowledgments
Previous versions of this paper were presented at the 2017 meeting of the Midwest Political Science Association and the 2021 meeting of the Southern Political Science Association. The authors thank Jeff Segal, Ben Kassow, Shane Gleason, and the anonymous reviewers for their helpful remarks.
Competing interest
The authors declare no competing interests exist.
Data availability statement
Replication materials for this article are available at the Journal of Law and Courts’ Dataverse archive.
Open access
