Authors: Brett Curry and Banks Miller
Publication: Justice System Journal 42: 252-270
Original source: jsj_r_r_submitted5.21.20v2.docx
Conversion note: this Markdown was generated by text extraction from the hosted file. Review equations, tables, notes, and citations before treating it as a final web copy.
The Influence of Diversity and Qualifications in Presidential Nominations: The Case of United States Attorneys
In this project, we employ original data on the future positions of United States Attorneys (USAs) to assess the influence of formal qualifications in structuring the post-service careers of a group of elite political figures. As compared to white male USAs, we find that presidents are significantly more likely to nominate USAs who are females or minorities to positions on the federal bench and in the upper federal bureaucracy. In other words, and after controlling for a number of additional individual-level differences, we find the same professional credential—service as a USA—to be more advantageous in the context of nomination to these positions for traditionally underrepresented individuals than for white men. We also find that more fine-grained qualifications influence the likelihood of promotion; in some instances, those individual qualifications also operate differently across traditional and nontraditional nominees.
Keywords: U.S. Attorneys; presidential selection; representative bureaucracy;
In the fall of 1994, President Bill Clinton nominated Vicki Miles-LaGrange to a federal district judgeship in Oklahoma’s Western District. Upon her confirmation, she became the first African American and the second female to serve as a federal judge in that state. During his second term, Clinton recommended Henry Solano, a Latino, to the Senate as his nominee to serve as Solicitor at the Department of Labor. Aside from the race and gender diversity Miles-LaGrange and Solano brought to the bench and bureaucracy, respectively, these two presidential nominations were otherwise routine. There was, however, a notable link between them: both Miles-LaGrange and Solano had served as United States Attorneys (USAs), the chief prosecutors in the federal system.
USAs, though sometimes overlooked by scholars of law and politics, are vital actors in executing criminal justice policy. But their subsequent occupations are often no less consequential. Like Miles-LaGrange, some are promoted to the federal courts. Others, including Henry Solano, are elevated to higher positions within the executive branch. Yet, despite the fact that the position of USA is a common pipeline for higher positions in government that require presidential nomination, we know nothing systematic about the dynamics that may structure those post-service careers. For instance, what role—if any—do variations in individual characteristics such as performance, ideology, or educational pedigree play in explaining these USA promotions? Does the relative importance of those individual characteristics operate differently for members of groups that have been traditionally underrepresented in the halls of power as compared to white men? More fundamentally, if the acknowledged elite status of USAs collectively sets them apart, activating a heuristic that overwhelms the operation of individual-level characteristics that scholars have found to be important in other employment contexts, is that disproportionately beneficial to nontraditional or traditional candidates when it comes to these upward positions? Finally, are the answers to questions of this sort contingent upon the partisanship of the appointing president or consistent across different types of opportunities?
These are salient issues for scholars of American politics and law. They illuminate the intersection of opportunity structures and ambition when it comes to presidential efforts to staff the public sector, and these opportunities can have profound implications. For instance, in 1997 Eric Holder vacated the position of USA for the District of Columbia to become Deputy Attorney General. Had that promotion not occurred, it seems reasonable to wonder whether he would have gone on to become the nation’s first African American Attorney General just over a decade later (Johnson 2009). More fundamentally, the demographic composition of the public sphere has consequences for public policy and even institutional legitimacy (e.g., Riccucci, Van Ryzin, and Lavent 2014; Krislov 2012; Scherer and Curry 2010; Mansbridge 1999).
In this paper, we offer the first systematic look at whether underrepresented USAs are treated differently than white men when it comes to the president’s nominations to higher positions in the federal government. We do this in two main ways, both of which focus on the centrality of qualifications. First, we consider the potential influence of a qualification that has no individual variation in our data—one’s service as a USA. USAs are select members of the legal profession who, apart from their individual-level characteristics, all possess an elite marker of quality and status by virtue of their prosecutorial service (e.g., Perry 1998). Virtually all of them go on to meaningful post-service careers; in fact, the New York Times once referred to the USA position as “a steppingstone for law’s best and brightest” (Weiser 2009). The public prominence of USAs, both as federal prosecutors and, for many, subsequent careers in higher echelons of the federal government, make investigating their future employment interesting in itself. Possessing this professional credential may be especially important for women and minorities, given that the career success of traditionally underrepresented populations has historically been more dependent on formal, objective factors such as prior experience than their white male counterparts (Daley 1996). However, this possibility has yet to be empirically tested.
Our approach also affords us the ability to examine the notion of qualifications from a more individualistic vantage point. In other words, above and beyond their shared experience as USAs, are there individual-level qualifications that help structure these governmental promotions that run through the White House? Our access to data on USA case filings (which we treat as evidence of prosecutorial effort or performance), the ability to capture the ideological leanings of USAs and compare them to the preferences of appointing presidents, information about each USA’s educational background, and the size of the prosecutorial operations under their supervision permit us to control for formal individual-level variation across USAs. We consider the relationship of these variables to promotion decisions generally as well as whether they might operate differentially across traditional (white male) and nontraditional nominees.
We find that USAs who are members of traditionally underrepresented groups are significantly more likely to be nominated to federal executive and judicial positions than are white males. This suggests service as a USA is an important path for the advancement of these groups into federal governmental positions that are contingent on presidential selection. It also indicates that a lack of attention to diversity when selecting USAs may spill beyond the realm of federal prosecution and have deleterious implications for the diversity of the federal government more broadly. A further exploration of this result shows that Democratic administrations are particularly likely to nominate nontraditional USAs to the bench (though not the upper-echelons of the federal bureaucracy). Moving beyond the general USA credential, we find some evidence to suggest that prosecutorial effort and ideological proximity to the president help structure these promotions in some instances.
We begin with a brief explanation of the work USAs perform, describe the types of post-service careers they commonly pursue, and explain our focus on a subset of those careers. We then review scholarship from the field of industrial relations to inform additional aspects of our theory and, in turn, motivate our hypotheses. After explaining our data and describing our research design, we articulate those hypotheses. Next, we present our results, discuss them, and conclude by offering potential implications and possibilities for future research.
United States Attorneys and their Careers
United States Attorneys are the federal government’s chief prosecutors, and they have existed nearly as long as the nation itself. The position dates to the Judiciary Act of 1789, which afforded each federal judicial district an officer to prosecute violations of civil and criminal law and defend the United States in civil actions. Today there are more than 90 USAs, each assigned to a federal judicial district. They are nominated by the president to a four-year term and require Senate confirmation. Because they are political appointees, USAs commonly leave office upon a change of administration. Some do so before their terms conclude (e.g., Scott 2007).
USAs lead offices across the country that vary in size and are staffed by assistant USAs with civil service protection (Lochner 2002). Criminal cases represent the bulk of USA activity, and they wield substantial discretion over them (Perry 1998, 139; Whitford 2002; Eisenstein 1978). The USA position is a common point of departure for careers in both the public and private sectors (Eisenstein 1978; Eisinger 2017; Barkow 2009) and, while the range of future employment possibilities is theoretically endless, the most common paths include private legal practice, elected office, nomination to the federal judiciary, or appointment to another position in the executive branch. We focus on the career path that involves promotion within the U.S. government, because it permits a focus on the factors that one particular actor—the president—finds particularly determinative. This encompasses two basic types of positions—movement within the executive branch and federal judicial positions. Aside from private practice, these represent the two most common initial post-tenure moves by USAs in our data. However, before turning to that data, we consider several empirical and theoretical points that help frame our study.
Racial and Gender Diversity in the Federal Courts and Bureaucracy
Since the administration of Jimmy Carter, numerous presidents—particularly Democratic ones—have made racial and gender diversity central components of their judicial selection strategies (Goldman 1997; Scherer 2005; Goldman, Schiavoni, and Slotnick 2009). In every administration prior to Carter’s, well above 90% of district court appointees were white men (Wheeler 2009). That percentage dropped precipitously under Carter to just 68% and plummeted to 38% during Barack Obama’s presidency (Slotnick, Schiavoni, and Goldman 2017). These statistics signal a clear pattern—in recent decades, the federal judiciary has become far more diverse in terms of race and gender.
Trends related to diversity within the executive bureaucracy have tracked those in the federal judiciary. Many of these appointees are lawyers by trade and, as noted, this is a common career option for USAs. Like judicial candidacies, these positions are dependent on the president’s support. Just as the judiciary’s diversification began in earnest with the Carter presidency, so too did that of its executive branch appointments—20% of Carter’s senior executive appointees were nontraditional (King and Riddlesperger 2015). Likewise, recent presidential cabinets have grown more diverse in terms of gender, with a number of notable strides coming in the Clinton, George W. Bush, and Obama administrations (Borelli 2010). By 2013, Anestaki et al. (2019) report that more than 44% of political appointees were women, nearly 14% were African American, and 6.6% were Latinx.
Increased diversity in these appointive positions has important potential implications. Presidents are uniquely equipped to influence the diversity of federal judicial and bureaucratic institutions, and Democratic presidents in particular have prioritized such diversity in recent decades (e.g., Anestaki et al. 2019; Naff and Crum 2000; Scherer 2005; Goldman 1997).
Theories of representative bureaucracy suggest the administrative state will be more accountable to the public if it reflects the demographic characteristics of those it serves (e.g., Krislov 2012; Krislov and Rosenbloom 1981; Selden 1997). Indeed, numerous studies have shown that better representation of traditionally underrepresented groups can promote the interests and improve the fortunes of those groups (e.g., Sowa and Selden 2003; Dolan 2002). This passive representation’s symbolic nature has “important effects on the attitudes and behaviors of underrepresented groups toward the bureaucracy and the political system generally (Lim 2006, 194); it is therefore an “important predictor of inclusiveness” (Clark et al. 2013, 78).
Scholars of the courts have said much the same. Theories of descriptive representation posit that institutions such as the courts increase their legitimacy with the public as their race and gender composition comes to mirror that broader society (e.g., Scherer and Curry 2010). As then-candidate Bill Clinton phrased it, the federal courts must “look like America” in order for the judiciary to be seen by the public as broadly legitimate (Mannies 1992). Attorney General Griffin Bell summarized President Carter’s prioritization of diversity in judicial appointments similarly: “…[a] national judiciary should resemble its national demographic constituency. Therefore, large groups which have been denied extensive representation in government should now be given a greater degree of representation” (Bell, quoted in Neff 1981, 150). All this begs a question: how can this diversity be facilitated? One potential route by which underrepresented groups can move into important policymaking positions in the federal bureaucracy and on the federal bench is by gaining a widely recognized credential for competence, such as service as a USA. Indeed, it may be that presidents use the position of USA strategically, to help advance the careers of those they hope to install in these positions.
Service as a USA represents a highly attractive attribute for presidential nominees (e.g., Weiser 2009). For example, as Christopher Kang, a key adviser in the Obama administration on judicial appointments remarked about Ed Stanton, a USA stationed in Memphis, “he’s a U.S. Attorney, you couldn’t ask for anything better than that in terms of the typical judicial nominee” (qtd. in Slotnick, Schiavoni, and Goldman 2017, 384). With all this in mind, in light of their historically underrepresented status on the federal bench and presidential efforts in recent decades to rectify that underrepresentation, not to mention the scholarship on career progression to which we pivot in the next section, past service as a USA may be a uniquely valuable qualification for nontraditional nominees to the federal courts and executive bureaucracy. This leads to hypothesis 1:
Hypothesis 1: All else equal, we expect that presidents will be more likely to nominate USAs from traditionally underrepresented groups for positions in the judiciary and the executive branch as compared to white men.
Career Progression, Qualifications, and Individual Characteristics
USAs are bureaucrats (e.g., Perry 1998; Whitford 2002) and bureaucrats, like employees generally, understand that their prospects for promotion are likely to be influenced by some combination of their formal qualifications and prior performance. That may be particularly true for USAs, whose connection to the president is recognized as being of a principal-agent variety (e.g., Miller and Curry 2019; Whitford 2002; Perry 1998). And the president is more than the federal prosecutor’s current principal—when it comes to aspirations to judicial or higher executive service, he is also indispensable in determining whether that USA will be promoted.
At a general level, there is substantial evidence that the processes relevant to promotion differ between white men and their minority counterparts (see Smith 2005; Wilson et al. 1999). As an example, Baldi and McBrier’s (1997, 492) work with a sample of organizations concludes that employers utilize different criteria when assessing the “promotability” of white and nonwhite workers. The primary way in which those criteria differ, they conclude, is that decision makers are more likely to rely upon formal qualifications and performance metrics when assessing nontraditional job candidates than they are when evaluating white men.
Similar findings characterize examinations of the differences between employer assessments of women and men (McGuire and Reskin 1993; Olson and Becker 1983; Wolf and Fligstein 1979). According to Eagly and Carli (2003, 819), “For women to be considered as instrumentally competent as men, perceivers must be given clear evidence of women’s greater ability or superior performance…” This is largely consistent with Lazarus and Steigerwalt’s (2018) finding in the legislative context that the phenomenon of “gendered vulnerability” leads women in Congress to introduce more legislation, secure more government spending for their constituents, and be more responsive to voters than their male counterparts. As Gill and Jensen (2020, 122) put it in their examination of gender’s relevance to appointments on the European Court of Justice (ECJ), “female candidates are expected to demonstrate partisan neutrality or policymaking expertise, while male candidates are assumed to have these traits.” Though they find such gender-based norms operate uniquely across common and civil law traditions, Gill and Jensen’s conclude these double standards likely contribute to the underrepresentation of women on the ECJ (Gill and Jensen 2020, 137).
Smith (2005, 1158) interprets this vein of research to show that, for women and minorities, “promotion decisions are less likely to be influenced by [informal] sponsorship ties and, instead, minorities’ and women’s formal credentials are placed under greater scrutiny compared to those of their white male counterparts.” As Wilson and his colleagues (1999) see it, nontraditional candidates for promotion tend to have fewer opportunities to demonstrate more “informal” characteristics such as loyalty, judgment, and leadership potential than do white males. As a consequence, the mobility of nontraditional candidates is typically restricted to a more “formal and deterministic route” (Wilson 1997) in which more objective signs of loyalty and performance are paramount. Smith (2005, 1162-1163) describes it this way:
…[T]he formal performance and work commitment investments of women and minorities are placed under tighter scrutiny compared with those of their White male counterparts. This means that in the absence of information about workers’ loyalty and devotion to the job (information that is more difficult to gain from women and minorities), superiors place a comparatively high premium on “objective” factors for promotion, namely, performance and … credentials.
Haire and Moyer’s (2015) overview of judicial backgrounds on the U.S. Courts of Appeals, which emphasizes the frequency of prosecutorial and/or judicial experience among judges from historically underrepresented groups, is broadly consistent with these conclusions. For instance, they note prosecutorial experience is common among African American judges on the Courts of Appeals (2015; 20-21), that nearly all such Latinx judges possessed trial court experience (25), and that women nominated to those courts since the Carter administration have been likely to possess prior judicial experience (40).
The extent to which such patterns characterize presidential nomination decisions more generally remains less conclusive. However, with this as background, it seems plausible to predict that race- and gender-based differences could exist in patterns of promotion among USAs. But how might we operationalize measures of performance and credentials in the context of federal prosecutors and their “formal qualifications” for nomination to subsequent judicial or bureaucratic positions? As previously indicated, our investigation is two-pronged. First, we leverage the fact that all USAs share a common credential. Second, we examine specific aspects of background and performance on which there is individual variation. Extant literature on USAs suggests two such constructs—the individual’s ideological compatibility with the promotor and the USA’s performance in office as captured by objective case processing statistics (Miller and Curry 2018). Here, the pertinent indicator of ideological compatibility should compare the ideological proximity of a USA to the ideological position of the president who is formally responsible for their nomination to the bureaucracy or the judiciary (Miller and Curry 2019). Quantifying one’s formal performance or “effort” as a USA could be captured in several ways, but we focus on performance as measured by case filings because it is less likely than alternative measures to be affected by the preferences of other actors in the criminal justice system such as judges or juries (Shermer and Johnson 2010, 402; Richman 2003).
We also investigate two other possible sources of individual variation between USAs being considered for promotion: the prestige of an individual’s law school education and the size of the federal prosecutor’s office managed by the individual. Office size is a key indicator of USA autonomy according to Eisenstein (1978), with larger USA offices typically seen as more prestigious. Indeed, it may be that effective management of larger offices is seen as a predictor of future performance managing similarly complex organizations or the handling of a complex docket. It is also plausible that more well-respected USA candidates are placed in larger offices, thus creating a sort of selection effect whereby those most likely to be promoted are first groomed with service in large offices. Attendance at an elite law school might also predict, as a formal credential, the likelihood of promotion; prior studies have controlled for the possibility in related contexts (e.g., Haire 2001). The top ten schools for placements on the federal bench include nine of the fifteen highest ranked law schools (the one outlier is Florida). These four factors—ideological distance, case filings, office size, and law school attended—comprise our measures of formal qualifications. All else equal, we expect that presidents will be more likely to nominate USAs with more attractive formal individual qualifications. Based on the above discussion, this leads to Hypotheses 2a – 2d:
Hypothesis 2a: Presidents will be more likely to nominate USAs who are highly compatible with them ideologically.
Hypothesis 2b: Presidents will be more likely to nominate USAs whose effort, as measured by case filings, is high.
Hypothesis 2c: Presidents will be more likely to nominate USAs who oversee larger offices.
Hypothesis 2d: Presidents will be more likely to nominate USAs who have attended elite law schools.
Data and Methods
We set out to code the initial positions of USAs after completing their terms as prosecutors. We undertook searches to gather career data on the USAs who served during 1986-2015. That yielded information on 93% of all USAs who served during this period; to the best of our knowledge, this represents the most comprehensive collection of that data to date. We utilized a number of sources, including the Federal Judicial Center’s Biographical Directory, biographical and autobiographical works, news stories, and even obituaries, in compiling this information.
We identified individuals as a federal judge if they were nominated to serve as a district or court of appeals judge. We coded those who were put forward for promotion within the executive branch (federal bureaucracy). Turning to our independent variables of interest, we created a variable called white male (1 if white and male; 0 otherwise). We captured USA ideology with the approach developed by Giles, Hettinger, and Peppers (2001) in conjunction with lower federal court judges. Those scores are scaled from -1 (very liberal) to +1 (very conservative). As with lower court judges, the norms of senatorial courtesy are strong when it comes to the appointment of USAs (Nelson and Ostrander 2016). Therefore, when both of a state’s senators belong to the president’s party when a USA was nominated, we average their NOMINATE ideology scores; when one senator is of the president’s party we use that senator’s score; and when no senators of the president’s party are in place, we use the president’s score for USA ideology. We measure the difference between the USA’s ideology and the president’s (ideological distance). In general, we expect former USAs who are in closer ideological alignment with the president to be among those most likely tapped for these two appointive positions. Since executive actors who prove to be non-congruent with the administration can be easily removed but judicial “mistakes” are guaranteed lifetime tenure, the importance of ideology may be especially important in the context of judicial nominations.
We measure USA performance as the average total case filings per year for USAs (average filings). There is ample evidence these sorts of quantitative measures of performance are important to administrations wishing to monitor USA performance (e.g., Chesney 2007; Eisenstein 2007; Iglesias with Seay; Miller and Curry 2018; 2019). We do not have direct measures of office size for all years in our data, so we rely on a proxy measure. Our proxy is the average number of courtroom hours worked by attorneys in an office (average workhours). Lastly, we include a dummy variable equal to one if a USA attended one of the law schools regularly ranked in the top fifteen by U.S. News over the period of our study (top 15).
In the regressions that follow, we control for a number of additional factors that can affect USA performance. These control variables include whether the USA serves in a district bordering Mexico (border district; coded 1 if yes and 0 otherwise), whether the USA left office during a transition in the partisanship of the presidency (transition year; coded 1 if yes and 0 otherwise), and the year in which the USA left office (year leaving). More details on these variables and the reasons for their inclusion are available in the appendix.
Table 1 contains descriptive statistics for the included variables. In addition, Table 2 contains demographic information on the USAs in our data. The majority of USAs are white men, comprising 78% of all of those who served in our period of study. It is clear in Table 2 that, although underrepresented groups are 22% of the USAs, they are more likely to move into both high level federal bureaucratic positions (39% of the total) and be nominated to the federal courts (35% of the total) than are white men. Among the groups that comprise our underrepresented category, the majority are women. The totals for each subgroup add up to more than the total for the underrepresented category because of intersectionality between women and the racial and ethnic groupings. It is notable that each of the subgroups of the underrepresented category is more prevalent in both high-level federal bureaucratic positions and in the federal courts than their representation in the population of USAs suggests.
[Table 1 here]
[Table 2 here]
Tables 3-6 contain additional descriptive information about the USAs included in our data. Table 3 provides a demographic breakdown of USAs by their appointing president; it shows that the percentage of white males was highest during the Reagan presidency and lowest during Clinton’s. Clinton’s USAs also contain the largest female and African American cohorts, while W. Bush’s comprise the largest share of Asian and Latinx USAs during this period. Tables 4 and 5 capture the percentage of USAs nominated to the bureaucracy and judiciary, respectively, under each of these five presidents. The figures in Table 4 include all USAs, whereas Table 5 is limited to nontraditional USAs and, in some cases, a comparison of these percentages is illuminating. Reagan was comparative likely to nominate USAs to both the bureaucracy (13%) and the judiciary (11%) and, while each of his (three) nontraditional USAs was put forward to serve in the bureaucracy, he tapped none for the federal courts. To varying degrees, with the exception of George W. Bush, each of the remaining four presidents exhibit percentages in Table 5 that are equal to or higher than those in Table 4. In other words, presidents of both parties appear to nominate nontraditional candidates for these positions at higher rates than is true of their baselines for USAs generally.
Finally, Table 6 conveys the frequency with which the same president appoints an individual to both the USA position and nominates that person to one of our two positions of interest. In other words, Table 6 captures the percentage of USAs appointed by Obama that Obama would go on to nominate to serve in the federal bureaucracy or judiciary. One immediate takeaway from Table 6 is that the only president whose USA nominees to the bureaucracy and courts are more likely than not to have been initially appointed USA by another president is the only chief executive in our data to succeed a predecessor of the same party—George H.W. Bush, a single-term president who tapped a high proportion of co-partisan Ronald Reagan’s USAs for these promotions. The second important takeaway is that each of the remaining four presidents is far more likely to nominate USAs appointed by others for bureaucratic positions than for judicial ones.
[Tables 3-6 here]
Given our interest in two distinct kinds of promotions, both controlled by presidential discretion, we estimate a series of models for movement into the federal bureaucracy and for nomination to the federal bench. For each dependent variable, we first estimate a simple model without interactions and then proceed by interacting the variable white male with each of the formal qualification factors we highlight above to better gauge whether these variables have differential effects for white male and underrepresented groups. In the analysis section that follows we present substantive effects generated from these regression models (which are included in the appendix).
Analysis
Table 7 presents results from the simple regression models without interaction effects (Table A1 in the appendix). Substantive effects in the table are calculated by shifting dichotomous variables from zero to one and continuous variables from the 10th to 90th percentile of the data—95% confidence intervals are in brackets. Only four variables have statistically significant effects, highlighted in gray in the table. USAs from underrepresented groups are 9 percentage points more likely to be considered for a high-level position in the federal bureaucracy than are white male USAs. Similarly, USAs from underrepresented groups are 12 percentage points more likely to be nominated to the federal courts than are white male USAs. Ideological distance is also a potent predictor of nomination to the federal bench, as a voluminous literature in judicial politics would predict given the lifetime tenure of federal judges. Indeed, increasing ideological distance from the 10th to the 90th percentile reduces the likelihood of nomination by 15 percentage points, slightly more than our indicator for demographic characteristics. However, ideological distance is not a factor for promotion of USAs into the federal bureaucracy. That anyone promoted to a politically appointed position in the bureaucracy serves at the pleasure of the president likely explains the difference.
[Table 7 here]
None of the other formal indicators are important predictors of the likelihood of elevation to either position, except that USAs who manage larger offices are slightly more likely to be promoted into the bureaucracy. Using the proxy for office size of courtroom hours worked, as the number of hours worked increases from the 10th to the 90th percentile the likelihood of promotion increases by two percentage points. This is a relatively small effect compared to those for demographic characteristics and ideological distance. Nevertheless, it makes some intuitive sense that presidents would seek to move USAs with a demonstrated capability in managing large numbers of people into positions requiring similar skills in the federal bureaucracy. The insignificance of effort measured by case filings and educational pedigree captured by law school prestige suggests that having the credential of USA serves a sort of leveling or equalizing effect for movement into coveted positions in the federal government. In other words, among those with the USA qualification, most individual-level distinctions that might be expected to exert independent influence are unimportant to promotion. Lastly, it is not the case that the partisanship of the president is a significant predictor of the likelihood of a USA being elevated to either position, an issue to which we return later.
Given the continuous nature of most of the formal qualification factors (ideological distance, filings, and office size) we present several graphs of the interaction effects. Results for the interaction models that generated these effects are available in Tables A2 and A3 in the appendix. Here our approach is straightforward: we interact white male with each of the formal qualification factors to determine whether the importance of each factor varies depending on the demographic status of the USA. With the important exceptions referenced below, the general takeaway from this exercise is that there is little evidence these formal qualification factors are treated differently based on the demographic characteristics of the USA. Figure 1 presents results for the white male and ideological distance interaction for both the bureaucracy and the judiciary. Focusing on the bureaucracy, underrepresented groups are more likely to be promoted at low levels of ideological difference than are white men. However, at approximately the mid-point for ideological distance (0.4) the difference between underrepresented USAs and white male USAs is difficult to distinguish statistically. It is notable that white male USAs seems to become slightly more likely to be promoted into a bureaucratic position as they become more ideologically distant from the president, while the opposite is true for underrepresented groups. For nomination to the federal judiciary, the right panel of Figure 1 gives a good sense of the importance of ideological distance as both underrepresented USAs and white male USAs see their chances for such a nomination decline precipitously as distance increases. Again, once ideological distance reaches the mid-point in the data (0.4) the two groups have statistically indistinguishable chances for nomination.
[Figure 1 here]
Next, we turn to the effect of USA effort as measured by case filings in Figure 2. Here there is no evidence for differential treatment of effort by demographic characteristics for promotion to the federal bureaucracy. The left panel of Figure 2 shows that increasing effort has no influence on the likelihood of promotion and there is no discernable difference in how increased effort affects underrepresented groups and white men. In the right panel of Figure 2 we focus on movement to the federal bench. Here there is strong evidence for differential treatment; increasing effort seems to greatly increase the chances for underrepresented USAs. At the median level of effort, approximately 1,200 filings, underrepresented USAs are about twice as likely as white male USAs to be nominated.
[Figure 2 here]
In Figure 3 we present results for interactions with office size. We find little evidence in support of differential treatment by demographic group. In the left panel focusing on the federal bureaucracy, we see that those in underrepresented groups are always more likely to be promoted than white males at every level of office size. For both groups, managing a larger office is a positive factor in moving into the executive bureaucracy. In the right panel of Figure 3 we find similar results for nomination to the federal bench. Again, those USAs from underrepresented groups are more likely to be nominated at every level of office size and the depressive effect of nomination on running a large USA office is the same for white men and those from underrepresented groups. Lastly, we find no differential effect for attendance at an elite law school on the likelihood of promotion to the bureaucracy. For nomination to the bench, we find that attendance at an elite law school has no effect on the nomination chances of white males, nor does it substantially boost the likelihood that underrepresented USAs will be nominated as compared to white males. Indeed, as displayed in Table 8, each of these groups look similar. Instead, it appears that service as a USA is a potent booster of the chances of nomination for underrepresented USAs who do not attend a top 15-law school. At the very least, it is consistent with the notion that the USA credential may obviate the influence of a formal, individual-level characteristic for nontraditional candidates—precisely the group whose occupational advancement should be most susceptible to formal markers such as educational pedigree (e.g., Smith 2005)
[Figure 3 here]
[Table 8 here]
There is clear evidence in the literature on both promotion within the federal bureaucracy (e.g. Anestaki 2019) and on nominations to federal courts (e.g., Spill Solberg and Bratton 2005; Wheeler 2009; Goldman 1997) that Democratic presidents are more focused on efforts to diversify. Does this mean that they are more likely to seek out USAs from underrepresented groups to help in this effort? We explored this possibility by interacting our white male variable with whether or not the appointing/nominating president is a Democrat (Clinton and Obama in our data). Results of this exercise are displayed in Table 9. With respect to promotion into the upper-ranks of the federal bureaucracy, the evidence suggests that underrepresented USAs are more likely to be promoted in both Democratic and Republican administrations as compared to white men. The difference in the rate at which white male and underrepresented USAs are nominated is not significant for either Democratic or Republican presidents, but the difference in Republican administrations approaches traditional levels of significance (p=0.12) and is notably larger than is the difference in Democratic administrations. There is clear evidence that Democratic presidents are much more likely to nominate underrepresented USAs to the bench than they are their white male colleagues, as the 22-percentage point difference is statistically significant. Indeed, underrepresented USAs in Democratic administrations are about eight times more likely to be nominated to the bench than are white male USAs. Republican presidents nominate underrepresented USAs and white male USAs to the judiciary at approximately the same rate.
[Table 9 here]
Discussion
Our purpose in this study was to determine factors relevant to the promotion of underrepresented groups into two basic types of important federal policymaking positions. By training our focus on the careers of U.S. Attorneys, we undertook that determination in a way that held a general qualification constant. Beyond that, and relatively unique among studies of descriptive representation in governmental positions, this allowed us to control for potentially important variations in individual performance. It also permitted us to juxtapose each USA’s ideological position with that of the president—the individual ultimately responsible for deciding whether to nominate them to a judicial or higher executive position. We controlled for several additional factors and, after doing so, determined that one’s service as a USA is more advantageous in the context of nomination to these positions for traditionally underrepresented individuals than for white men.
In light of our results, it seems likely there are important unexplored aspects related to how a position like that of USA might be used by a forward-thinking administration to set up preferred candidates for further advancement. Additional avenues for future inquiry here might be to investigate whether there are differences in the rates at which presidents nominate former USAs to judicial and/or bureaucratic positions when the political opposition controls the confirmation process or, in the case of judicial nominees, whether presidents are more likely to nominate former USAs when filling vacancies in states whose U.S. senators are political opponents. A final extension of this line of research might involve assessments of the movement of Assistant USAs to these positions—in contrast to politically-appointed USAs, these government attorneys have civil service protection. But it is possible that the AUSA qualification may also be of special value for members of traditionally underrepresented groups; indeed, one could posit this position’s formally apolitical nature could be of special utility in some instances.
As is true of any study, ours has limitations. Perhaps most evident among them is its inability to tap the career path preferences of individual USAs. It is conceivable that some USAs presidents might wish to tap for judicial or bureaucratic positions have idiosyncratic reasons for refusing those appointments. Unfortunately, these declinations are unobservable. In other words, the full causal story incorporates two actors—the president who must nominate and the USA who must accept that nomination. Absent that information, we must focus on the president as the key gatekeeper. Even so, it is worth noting that some general evidence exists to suggest that, at least collectively, USAs appear to prioritize federal judgeships and executive promotions ahead of other alternatives (Miller and Curry 2019, 177-178; see also Eisenstein 1978).
A related question stems from evidence that women (Williams 2008; Noonan and Corcoran 2004) and minority (e.g., Lempert et al. 2000) attorneys are generally more attracted to government work than are their white, male counterparts. According to Lempert et al. (2000, 427), in every decade they examine “the proportion of minority students taking first jobs in government has substantially exceeded the numbers of whites taking government jobs.” Here, we think it important to remember that our study’s population of interest – USAs – is, by definition, comprised of individuals who have a penchant for governmental service and have self-selected into it. Though we cannot know for sure, it seems likely to us that the individual preferences of those we examine are unlikely to match the magnitude of the demographic trends observed in the broader legal profession.
Conclusion
Scholars of both the bureaucracy and the courts have long been concerned with representativeness because representativeness is fundamental for democratic governance (e.g. Mansbridge 1999). Yet social inequalities persist in the federal bureaucracy (e.g. Anestaki et al. 2019, 198) and in the federal courts (e.g. Gryski et al. 1994). For those concerned about the diversity of governmental institutions, the USA path appears to be a viable pipeline by which members of traditionally underrepresented groups might increase their share of positions in those institutions.
By the same token, the potential for the USA position to be a throughway to more opportunities for nontraditional candidates in higher governmental positions is dependent upon establishing a threshold level of diversity among USAs in the first place. As illustrated in Table 9, underrepresented USAs are more likely—about 10 percentage points more likely—to be nominated for promotion to an executive position under a Republican president than are white male USAs. This suggests that the lack of diversity amongst USAs may be especially consequential for those types of promotion opportunities under Republican administrations. If that is so, recent developments paint an ominous picture. Nearly a year into President Donald Trump’s administration, only one of 46 nominees put forward for U.S. Attorney was a woman and the lack of racial or ethnic diversity was similarly noteworthy among recent administrations, of both parties (Biskupic 2017).
Many advocates of representative bureaucracy undoubtedly hoped this dip would prove ephemeral. At this writing, however, about 87% of the Trump administration’s nominees for the position of U.S. Attorney have been white males: 92% of have been men, and 92% have been white. President Trump has nominated just 11 nontraditional USAs—seven women (four whites, two Latinas, and one Asian American) and four men (two Asian Americans and two African Americans). In light of our results here, the implications for this are evident. To whatever degree this or a future Republican administration might wish to tap partisans with the USA credential for judicial or executive positions, it will be severely limited in its ability to diversity the ranks of government. Given both the Democratic party’s more demographically diverse political coalition and the more ideologically liberal bent of the legal profession writ large (e.g., Bonica, Chilton, and Sen 2016), the implications of this lack of diversity amongst USAs are likely to be a particularly significant challenge to any Republican efforts to increase the representation of women and minorities in these higher positions.
In the immediate term, the representative bureaucracy perspective predicts that this lack of diversity amongst USAs will have its own consequences for the legitimacy of governmental institutions and the decisions they reach (e.g., Lim 2006; Sowa and Selden 2003). As Vanita Gupta, who led the Civil Rights Division at the Department of Justice during the Obama administration, put it, “Having diversity both in gender and race among the crop of U.S. attorneys is vital. They need to have the trust of the community. [Any lack of diversity is] going to have an impact on how the Justice Department is viewed (qtd. in Biskupic 2017).
Anestaki, Aikaterini, Meghna Sabharwal, Kenneth Connelly, and N. Joseph Cayer. 2019. “Race and Gender Representation in Presidential Appointments, SES, and GS Levels, During Clinton, Bush, and Obama Administrations.” Administration & Society 51: 197-228.
Baldi, Stephane and Debra Branch McBrier. 1997. “Do the Determinants of Promotion Differ for Blacks and Whites? Evidence from the U.S. Labor Market.” Work and Occupations 24: 478-497.
Barkow, Rachel. 2009. “Institutional Design and the Policing of Prosecutors: Lessons from Administrative Law.” Stanford Law Review 61: 869-921.
Bell, Lauren Cohen. 2002. “Senatorial Discourtesy: The Senate’s Use of Delay to Shape the Federal Judiciary.” Political Research Quarterly 55: 589-607.
Biskupic, Joan. 2017. “Trump Has Nominated 42 People for U.S. Attorney. Only one is a Woman.” CNN, 17 September.
Bonica, Adam, Adam Chilton, and Maya Sen. 2016. “The Political Ideologies of American Lawyers.” Journal of Legal Analysis 8: 277-335.
Borelli, Maryanne. 2010. “The Contemporary Presidency: Gender Desegregation and Gender Integration in the President’s Cabinet, 1933-2010.” Presidential Studies Quarterly 40: 734-749.
Chesney, Robert M. 2007. “Federal Prosecution of Terrorism-Related Offenses: Conviction and Sentencing Data in Light of the ‘Soft-Sentence’ and ‘Data-Reliability’ Critiques.” Lewis & Clark Law Review 11: 851-901.
Clark Jr., Ronald C., Holona LeAnne Ochs, and Michael Frazier. 2013. “Representative Bureaucracy: The Politics of Access to Policy-Making Positions in the Federal Executive Service.” Public Personnel Management 42: 75-89.
Daley, Dennis M. 1996. “Paths of Glory and the Glass Ceiling: Differing Patterns of Career Advancement Among Women and Minority Federal Employees.” Public Administration Quarterly 20: 143-162.
Dolan, Julie. 2002. “Representative Bureaucracy in the Federal Executive: Gender and Spending Priorities.” Journal of Public Administration Research & Theory 12: 353-375.
Eagly, Alice H. and Linda L. Carli. 2003. “The Female Leadership Advantage: An Evaluation of the Evidence.” The Leadership Quarterly 14: 807-834.
Eisenstein, James. 1978. Counsel for the United States: U.S. Attorneys in the Political and Legal Systems. Baltimore: Johns Hopkins University Press.
Eisenstein, James. 2007. “The U.S. Attorney Firings of 2006: Main Justice’s Centralization Efforts in Historical Context.” Seattle University Law Review 31: 219-263.
Eisinger, Jesse. 2017. The Chickenshit Club: Why the Justice Department Fails to Prosecute Executives. New York: Simon & Schuster.
Giles, Micheal W., Virginia A. Hettinger, and Todd C. Peppers. 2001. “Picking Federal Judges: A Note on Policy and Partisan Selection Agendas.” Political Research Quarterly 54: 623-641.
Gill, Rebecca D. and Christian Jensen. 2020. “Where are the Women? Legal Traditions and Descriptive Representation on the European Court of Justice.” Politics, Groups, and Identities 8: 122-142.
Gryski, Gerard, Gary Zuk, and Deborah J. Barrow. 1994. “A Bench that Looks Like America? Representation of African Americans and Latinos on the Federal Courts.” Journal of Politics 56: 1076-1086.
Goldman, Sheldon. 1997. Picking Federal Judges. New Haven: Yale University Press.
Goldman, Sheldon, Sara Schiavoni, and Elliot Slotnick. 2009. “W. Bush’s Judicial Legacy: Mission Accomplished.” Judicature 92: 258-289.
Haire, Susan B. 2001. “Rating the Ratings of the American Bar Association Standing Committee on Federal Judiciary.” Justice System Journal 22: 1-17.
Haire, Susan B. and Laura P. Moyer. 2015. Diversity Matters: Judicial Policy Making in the U.S. Courts of Appeals. Charlottesville, VA: University of Virginia Press.
Iglesias, David, with Davin Seay. 2008. In Justice: Inside the Scandal that Rocked the Bush Administration. Hoboken, NJ: John Wiley & Sons.
Johnson, Carrie. 2009. “Holder Confirmed as Nation’s First Black Attorney General. He Overcame Objections by Some in GOP.” Washington Post, 3 Feb.
King, James D. and James W. Riddlesperger Jr. 2015. “Diversity and Presidential Cabinet Appointments.” Social Science Quarterly 96: 93-103.
Krislov, Samuel. 2012. Representative Bureaucracy. New Orleans: Quid Pro Books.
Krislov, Samuel and David H. Rosenbloom. 1981. Representative Bureaucracy and the American Political System. New York: Praeger.
Lazarus, Jeffrey and Amy Steigerwalt. 2018. Gendered Vulnerability: How Women Work Harder to Stay in Office. Ann Arbor: University of Michigan Press.
Lempert, Richard O., David L. Chambers, and Terry K. Adams. 2000. “Michigan’s Minority Graduates in Practice: The River Runs Through Law School.” Law & Social Inquiry 25: 395-505.
Lim, Hong Hai. 2006. “Representative Bureaucracy: Rethinking Substantive Effects and Active Representation.” Public Administration Review 66: 193-204.
Lochner, Todd. 2002. “Strategic Behavior and Prosecutorial Agenda Setting in the United States Attorneys’ Offices: The Role of U.S. Attorneys and their Assistants.” Justice System Journal 23: 271-294.
Mannies, Jo. 1992. “Clinton Goes on Attack Here; Bush Blasted on Thomas, Lack of Strategy on Jobs.” St. Louis Post-Dispatch, 1 Aug.
Mansbridge, Jane. 1999. “Should Blacks Represent Blacks and Women Represent Women? A Contingent ‘Yes.’” Journal of Politics 61: 621-657.
McGuire, Gail M. and Barbara F. Reskin. 1993. “Authority Hierarchies at Work: The Impacts of Race and Sex.” Gender & Society 7: 487-506.
Miller, Banks and Brett Curry. 2019. U.S. Attorneys, Political Control, and Career Ambition. New York: Oxford University Press.
Miller, Banks and Brett Curry. 2018. “Haphazard, Systematic, or Both? An Empirical Investigation of the United States Attorney Firings in 2006.” Journal of Law and Courts 6: 379-403.
Naff, Katherine C. and John Crum. 2000. “The President and Representative Bureaucracy: Rhetoric and Reality.” Public Administration Review 60: 98-110.
Neff, Allen. 1981. The United States District Judge Nominating Commissions: Their Members, Procedures and Candidates. Chicago: American Judicature Society.
Nelson, Michael J. and Ian Ostrander. 2016. “Keeping Appointments: The Politics of Confirming United States Attorneys.” Justice System Journal 37: 211-231.
Noonan, Mary C. and Mary E. Corcoran. 2004. “The Mommy Track and Partnership: Temporary Delay or Dead End?” The Annals of the American Academy of Political and Social Science 596: 130-150.
Olson, Craig A. and Brian E. Becker. 1983. “Sex Discrimination in the Promotion Process.” ILR Review 36: 624-641.
Perry, H.W. 1998. “United States Attorneys—Whom Shall They Serve?” Law & Contemporary Problems 61: 129-148.
Riccucci, Norma M., Gregg G. Van Ryzin, and Cecilia F. Lavena. 2014. “Representative Bureaucracy in Policing: Does It Increase Perceived Legitimacy?” Journal of Public Administration Research and Theory 24: 537-551.
Richman, Daniel C. 2003. “Prosecutors and Their Agents, Agents and Their Prosecutors.” Columbia Law Review 103: 749-832.
Rutkis, Denis Steven. 2013. “Role of Home State Senators in the Selection of Lower Federal Court Judges.” Washington, DC: Congressional Research Service.
Scherer, Nancy. 2005. Scoring Points: Politicians, Activists, and the Lower Federal Court Appointment Process. Palo Alto, CA: Stanford University Press.
Scherer, Nancy and Brett Curry. 2010. “Does Descriptive Representation Enhance Institutional Legitimacy? The Case of the Federal Courts.” Journal of Politics 72: 90-104.
Scott, Kevin M. 2007. “U.S. Attorneys Who Have Served Less Than Full Four-year Terms, 1981-2006.” Washington, D.C.: Congressional Research Service.
Selden, Sally C. 1997. The Promise of a Representative Bureaucracy: Diversity and Responsiveness in a Government Agency. Armonk, NY: M.E. Sharpe.
Shermer, Lauren O’Neill and Brian D. Johnson. 2010. “Criminal Prosecutions: Examining Prosecutorial Discretion and Charge Reductions in U.S. Federal District Courts.” Justice Quarterly 27: 394-430.
Slotnick, Elliot, Sara Schiavoni, and Sheldon Goldman. 2017. “Obama’s Judicial Legacy: The Final Chapter.” Journal of Law and Courts 5: 363-422.
Smith, Ryan A. 2005. “Do the Determinants of Promotion Differ for White Men Versus Women and Minorities? An Exploration of Intersectionalism Through Sponsored and Contest Mobility Processes.” American Behavioral Scientist 48: 1157-1181.
Sowa, Jessica E. and Sally Coleman Selden. 2003. “Administrative Discretion and Active Representation: An Expansion of the Theory of Representative Bureaucracy.” Public Administration Review 63: 700-710.
Spill Solberg, Rorie and Kathleen A. Bratton. 2005. “Diversifying the Federal Bench: Presidential Patterns. Justice System Journal 26: 119-133.
Weiser, Benjamin. 2009. “A Steppingstone for Law’s Best and Brightest.” New York Times, 29 Jan.
Wheeler, Russell. 2009. The Changing Face of the Federal Judiciary. Washington, D.C.: Brookings Institution Press.
Whitford, Andrew B. 2002. “Bureaucratic Discretion, Agency Structure, and Democratic Responsiveness: The Case of the United States Attorneys.” Journal of Public Administration Research and Theory 12: 3-27.
Williams, Margaret S. 2008. “Ambition, Gender, and the Judiciary.” Political Research Quarterly 61: 68-78.
Wilson, George. 1997. “Pathways to Power: Racial Differences in the Determinants of Job Authority.” Social Problems 44: 38-54.
Wilson, George, Ian Sakura-Lemessy, and Jonathan P. West. 1999. “Reaching the Top: Racial Differences in Mobility Paths to Upper-Tier Occupations.” Work and Occupations 26: 165-186.
Wolf, Wendy C. and Neil D. Fligstein. 1979. “Sexual Stratification: Differences in Power in the Work Setting.” Social Forces 58: 94-107.
Brackets represent 95% confidence intervals. Results for regressions available in Tables A1. Highlighted cells are statistically significant (p < 0.05).
We include three control variables in the models and we describe our reasons for including them more fully here. First, we include a dummy variable indicating whether a district borders Mexico. Starting in the Clinton administration, immigration cases began to make up an ever-larger portion of the caseload for USAs. This increase was particularly acute in the border districts, where immigration cases can number in the tens of thousands and dominate the case filings. Therefore, border districts have higher case filings than other districts by virtue of geography. Second, we include a dummy variable that captures changes in the partisanship of the presidency. During our period of study, this includes the years 1993, 2001, and 2009. Prior work indicates that USAs who leave office in these transitional years predominantly move into private practice given the ideological disjunction between who appointed them to the position of USA and who currently occupies the White House (Miller and Curry 2019). Finally, we include a variable measuring the year a USA leaves her position to help account for any over-time trends in the likelihood of presidents to nominate USAs to the positions of interest.
Table A1 below provides regression results for the model used to generate the substantive results displayed in Table 7. Table A2 displays regression results for interactive models on movement to the federal bureaucracy, while Table A3 shows results for interactive models on movement to the federal bench. Given the dichotomous nature of the dependent variables, we estimate logit models; in each, standard errors are clustered by judicial district.
[Table A1 here]
[Table A2 here]
[Table A3 here]
** is statistically significant at p < 0.05 (two-tailed); * is statistically significant at p < 0.10 (two-tailed).
** is statistically significant at p < 0.05 (two-tailed); * is statistically significant at p < 0.10 (two-tailed).
** is statistically significant at p < 0.05 (two-tailed); * is statistically significant at p < 0.10 (two-tailed).
| Variable | Mean | Std. Dev. | Min. | Max |
| Federal Bureaucracy | 0.08 | 0.28 | 0 | 1 |
| Federal Judge | 0.11 | 0.31 | 0 | 1 |
| White Male | 0.78 | 0.42 | 0 | 1 |
| Ideological Distance | 0.43 | 0.33 | 0 | 1 |
| Average Filings (thous) | 1.24 | 2.91 | 0.05 | 34.77 |
| Ct. Workhours (thous) | 11.31 | 18.11 | 0.86 | 176.18 |
| Top 15 Law School | 0.27 | 0.45 | 0 | 1 |
| Democratic President | 0.52 | 0.50 | 0 | 1 |
| Border District | 0.06 | 0.24 | 0 | 1 |
| Transition Year | 0.33 | 0.47 | 0 | 1 |
| Year Leaving | 2001 | 8.25 | 1986 | 2016 |
| Demographics | Total % of USAs | Federal Bureaucracy % | Federal Courts % | Federal Courts % | Total N |
| White Male | 78% | 62% | 62% | 64% | 356 |
| Underrepresented | 22% | 38% | 38% | 36% | 101 |
| African American | 6% | 10% | 10% | 12% | 28 |
| Asian | 1% | 3% | 3% | 2% | 6 |
| Latinx | 3% | 10% | 10% | 6% | 14 |
| Female | 15% | 15% | 15% | 24% | 67 |
| % White Male | % Female | % African Am. | % Asian | % Latinix |
| Reagan | 97% | 0% | 1% | 0% | 1% |
| H.W. Bush | 85% | 13% | 2% | 0% | 2% |
| Clinton | 64% | 24% | 14% | 1% | 3% |
| W. Bush | 77% | 15% | 4% | 4% | 4% |
| Obama | 73% | 17% | 11% | 0% | 3% |
| % Fed. Bureaucracy | % Fed. Judiciary |
| Reagan | 13% | 11% |
| H.W. Bush | 9% | 5% |
| Clinton | 7% | 7% |
| W. Bush | 7% | 14% |
| Obama | 6% | 13% |
| % Fed. Bureaucracy | % Fed. Judiciary |
| Reagan | 100% | 0% |
| H.W. Bush | 38% | 13% |
| Clinton | 11% | 16% |
| W. Bush | 6% | 12% |
| Obama | 12% | 35% |
| % Fed. Bureaucracy | % Fed. Judiciary |
| Reagan | 75% | 100% |
| H.W. Bush | 33% | 33% |
| Clinton | 50% | 89% |
| W. Bush | 64% | 95% |
| Obama | 50% | 89% |
| Federal Bureaucracy | Federal Judge |
| White Male | -0.09 [-.17, -.00] | -0.12 [-.19, -.05] |
| Ideological Dist. | 0.01 [-.10, .12] | -0.15 [-.22, -.09] |
| Avg. Filings | -0.00 [-.01, .01] | 0.01 [-.00, .03] |
| Office Size | 0.02 [.00, .03] | -0.03 [-.07, .02] |
| Top 15 Law School | -0.05 [-.12, .02] | -0.04 [-.10, .02] |
| Democratic Pres. | -0.04 [-.09, .02] | -0.06 [-.12, .00] |
| Yes | No |
| Underrepresented | 0.09 [-.02, .21] | 0.25 [.16, .34] |
| White Male | 0.08 [.03, .12] | 0.09 [.06, .12] |
| Bureaucracy | Judiciary |
| Underrepresented | 0.12 | 0.24 |
| White Male | 0.05 | 0.02 |
| Difference | 0.07 [-.03, .17] | 0.22 [.11, .33] |
| Republican President | Republican President |
| Bureaucracy | Judiciary |
| Underrepresented | 0.18 | 0.11 |
| White Male | 0.08 | 0.13 |
| Difference | 0.10 [-.02, .23] | -0.02 [-.12, .08] |
| Model 1: Bureaucracy | Model 1: Bureaucracy | Model 1: Bureaucracy | Model 2: Judiciary | Model 2: Judiciary | Model 2: Judiciary |
| Coeff. | Robust S.E. | Robust S.E. | Coeff. | Robust S.E. |
| White Male | White Male | -0.98** | 0.44 | 0.44 | -1.22** | 0.32 |
| Ideological Dist. | Ideological Dist. | 0.12 | 0.89 | 0.89 | -2.64** | 0.61 |
| Avg. Filings | Avg. Filings | -0.00 | 0.04 | 0.04 | 0.084* | 0.045 |
| Avg. Workhours | Avg. Workhours | 0.013** | 0.004 | 0.004 | -0.016 | 0.015 |
| Top 15 Law School | Top 15 Law School | -0.67 | 0.49 | 0.49 | -0.53 | 0.37 |
| Dem. Pres. Appt. | Dem. Pres. Appt. | -0.50 | 0.38 | 0.38 | -0.69* | 0.37 |
| Border District | Border District | 0.04 | 0.86 | 0.86 | -0.64 | 0.83 |
| Transition Year | Transition Year | -0.64 | 0.63 | 0.63 | -2.28** | 1.07 |
| Year Leaving | Year Leaving | -0.034 | 0.02 | 0.02 | -0.02 | 0.02 |
| Constant | Constant | 65.81 | 48.95 | 48.95 | 35.44 | 43.14 |
| N | N | 457 | 457 | 457 | 457 | 457 |
| Districts | Districts | 90 | 90 | 90 | 90 | 90 |
| Wald χ2 | Wald χ2 | 28.30 (0.00) | 28.30 (0.00) | 28.30 (0.00) | 71.51 (0.00) | 71.51 (0.00) |
| Area under ROC | Area under ROC | 0.69 | 0.69 | 0.69 | 0.83 | 0.83 |
| Model 3 | Model 4 | Model 5 | Model 6 | Model 7 |
| White Male | -2.14 (.77)** | -1.05 (.46)** | -1.18 (.50)** | -1.06 (.47)** | -0.95 (.55)* |
| Ideological Dist. | -1.84 (1.08)* | 0.12 (.89) | 0.11 (.89) | 0.13 (.89) | 0.13 (.90) |
| Avg. Filings | 0.01 (.04) | -0.046 (.19) | -0.003 (.042) | -0.003 (.042) | -0.000 (.042) |
| Avg. Workhours | 0.011 (.006)* | 0.013 (.005)** | 0.007 (.005) | 0.013 (.006)** | 0.013 (.005)** |
| Top 15 Law School | -0.64 (.49) | -0.67 (.49) | -0.70 (.49) | -0.93 (.72) | -0.67 (.49) |
| Dem. Pres. Appt. | -0.77 (.41)* | -0.50 (.38) | -0.51 (.38) | -0.51 (.38) | -0.46 (.56) |
| WM * Ideo. Dist | 3.04 (1.33)** | — | — | — | — |
| WM * Avg. Filings | — | 0.05 (.18) | — | — | — |
| WM* Avg. Workhours | — | — | 0.012 (.006)** | — | — |
| WM * Top 15 | — | — | — | 0.40 (.75) | — |
| WM * Dem. Pres. | — | — | — | — | -0.07 (.82) |
| Border District | -0.03 (.82) | 0.12 (1.03) | 0.02 (.87) | 0.05 (.85) | 0.04 (.88) |
| Transition Year | -0.53 (.59) | -0.64 (.64) | -0.64 (.64) | -0.64 (.64) | -0.64 (.63) |
| Year Leaving | -0.03 (.02) | -0.03 (.02) | -0.035 (.024) | -0.03 (.02) | -0.03 (.02) |
| Constant | 64.38 (49.27) | 66.38 (48.84) | 68.54 (49.55) | 66.12 (49.08) | 65.78 (49.12) |
| N | 457 | 457 | 457 | 457 | 457 |
| Districts | 90 | 90 | 90 | 90 | 90 |
| Wald χ2 | 28.27 (0.00) | 29.16 (0.00) | 26.91 (0.00) | 27.62 (0.00) | 30.18 (0.00) |
| Area under ROC | 0.72 | 0.69 | 0.69 | 0.69 | 0.69 |
| Model 8 | Model 9 | Model 10 | Model 11 | Model 12 |
| White Male | -1.33 (.43)** | -0.74 (.40)* | -1.10 (.46)** | -1.47 (.40)** | 0.21 (.58) |
| Ideological Dist. | -3.01 (1.03)** | -2.56 (.61)** | -2.63 (.61)** | -2.64 (.61)** | -2.47 (.63)** |
| Avg. Filings | 0.085 (.044)* | 0.47 (.08)** | 0.088 (.046)* | 0.081 (.044)* | 0.12 (.04)** |
| Avg. Workhours | -0.017 (.015) | -0.02 (.02) | -0.01 (.01) | -0.017 (.016) | -0.014 (.014) |
| Top 15 Law School | -0.52 (.38) | -0.43 (.37) | -0.50 (.37) | -1.40 (.80)* | -0.49 (.42) |
| Dem. Pres. Appt. | -0.71 (.38)* | -0.67 (.38)* | -0.69 (.37)* | -0.71 (.38)* | 1.09 (.66) |
| WM * Ideo. Dist | 0.57 (1.31) | — | — | — | — |
| WM * Avg. Filings | — | -0.38 (.11)** | — | — | — |
| WM* Avg. Workhours | — | — | -0.01 (.03) | — | — |
| WM * Top 15 | — | — | — | 1.21 (.88) | — |
| WM * Dem. Pres. | — | — | — | — | -3.06 (.95)** |
| Border District | -0.65 (.84) | -1.60 (.97)* | -0.64 (.85) | -0.60 (.87) | -0.81 (.75) |
| Transition Year | -2.24 (1.02)** | -3.23 (.84)** | -2.28 (1.06)** | -2.31 (1.07)** | -2.05 (1.07)** |
| Year Leaving | -0.02 (.02) | -0.02 (.02) | -0.02 (.02) | -0.02 (.02) | -0.01 (.02) |
| Constant | 34.51 (43.23) | 37.43 (44.27) | 36.23 (42.89) | 36.71 (44.52) | 23.72 (43.98) |
| N | 457 | 457 | 457 | 457 | 457 |
| Districts | 90 | 90 | 90 | 90 | 90 |
| Wald χ2 | 85.74 (0.00) | 86.41 (0.00) | 74.63 (0.00) | 76.05 (0.00) | 81.32 (0.00) |
| Area under ROC | 0.83 | 0.84 | 0.83 | 0.83 | 0.84 |