Abstract
Presidents have the power to nominate the top ranks of the executive branch and they use it toward their own goals. Presidents need a mixture of experienced and loyal officers to promote bureaucratic competence and responsiveness, and advance their policies within the executive branch, but scholars’ focus has often been senatorial obstructionism rather than when presidents prefer loyal nominees and experienced nominees. This study uses a sample of 573 subcabinet nominations between 1961 and 2006, and relates presidents’ nominee choices to their policy priorities and departmental jurisdictions. It finds that presidents nominate significantly more loyalists and fewer experienced officers to higher priority departments, indicating the importance of responsiveness, as well as their suspicion of experienced officers. Presidents select more experienced nominees for lower priority departments to assure competence where they are least concerned with responsiveness. Furthermore, it finds that increased Senate ideological opposition has no significant effect on presidents’ selections.
Introduction
Being President is a daunting task and presidents rely upon many others to accomplish their goals. Following his election, Kennedy complained, “For the last four years I spent so much time getting to know people who could help me get elected President that I didn’t have time to know people who could help me, after I was elected, to be a good president” (O’Donnell & Powers, 1972, p. 270). Presidents are elected to manage the executive branch and enact their agenda, and the power to appoint executive officers provides them a means to build a team of skilled and loyal officers to assist them (King & Riddlesperger, 1996; Mackenzie, 2001). However, the Constitution also gives the Senate the final word on nominations, drawing scholarly attention away from the President the publicly visible Senate confirmations (DeRouen, Peake, & Ward, 2005; McCarty & Razaghian, 1999, 2000). Nonetheless, presidents’ decisions and policy goals are central to understanding the nomination and confirmation process of determining who holds the top political ranks of the executive branch.
Presidents use their nomination power to build effective administrations that promote their policies within the executive branch. As part of their administrative presidency strategy, they politicize the executive branch by increasing the number of political appointees and penetrating the bureaucracy with likeminded officers loyal to the presidents’ policy preferences (Aberbach, 2009; Lewis, 2009; Moe, 1985; Nathan, 1983). 1 In building their teams, presidents must prioritize their decisions about nominees depending upon their policy priorities and the jurisdictions of the departments (Light, 1982, 1999). They must decide in which departments they will emphasize experienced competence, loyal responsiveness, or both. While recent scholarship has examined the interaction between presidents, Congress, and agency ideological disposition (Lewis, 2008), there has been little discussion of presidents’ calculations about the types of officers they nominate. They know that their nominees will face confirmation hearings where opposition senators have incentives to obstruct nominations, but presidents set the process in motion by deciding the types of officers in the subcabinet.
Presidents’ calculations about the value of competence, the value of responsiveness, their own priorities, the departments’ jurisdictions, and the Senate environment determine the combination of officers who inhabit the top ranks of the bureaucracy. This study addresses this issue with analysis of the nominees chosen by presidents for subcabinet offices between 1961 and 2006. I analyze nominees’ resumes, the experience and loyalty of nominees, presidential priorities, and the Senate environment. I propose an account that places presidential strategies at its center and fills a gap in the current literature. While the Senate may exert some influence, it has been overemphasized and I find that presidents’ priorities and their efforts to control the bureaucracy determine the experience and loyalty of subcabinet officers.
Theoretical Framework
Presidents have incentives to nominate some experienced bureaucrats and some political loyalists to subcabinet positions. Heclo (1977) argued that presidents are responsible for the effective management of the executive branch and thus select experienced nominees who are better able to promote bureaucratic competence. Presidents must remain attentive to the danger of lowering agency competence when politicizing agencies (Lewis, 2008), and experienced appointees bolster agency competence because knowledge of how departments operate links subcabinet officers to careerists, and experienced officers know how to succeed within the bureaucracy (Edwards, 2001; Light, 1987). In addition to experienced officers, presidents also need responsive, aggressive officers advocating their preferences. Moe (1985) argued that presidents seek “responsive competence” rather than mere “neutral competence” to control executive power. Presidents seek to populate agencies with partisan, ideological, and personal loyalists whom they expect to be responsive and aggressive in their promotion of presidential preferences (Lewis, 2009; Moe, 1993). Presidents value bureaucratic effectiveness not for its own sake, but rather only to the extent that it advances their policies.
Personal and partisan loyalties among executive officers are crucial assets for presidents. Modern presidents, frustrated in their legislation attempts, have resorted to administrative discretion in the implementation of existing laws rather than through new legislation to achieve their goals (Cooper, 1986; Howell, 2003; Nathan, 1983), and in doing so rely upon their appointees to act as their agents (Edwards, 2001, pp. 83-84 and 101). Historically, presidents have sought to fill departments with personal, partisan, and ideological allies (Weko, 1995). Reagan aide Lyn Nofziger asserted that “As far as I’m concerned, anyone who supported Reagan is competent” (Lewis, 2008, p. 27). Aberbach and Rockman (2009) argued that aggressive politicization of the executive branch is logically connected to the “unitary executive” theory, and in selecting appointees, the Bush (43) administration sought to “implant” its “D.N.A. throughout government” (Rudalevige, 2008, p. 149). The need for bureaucratic control drives presidents to place loyal operatives within bureaucracy (Warshaw, 1997).
As they weigh their options, there is evidence that presidents treat potential nominees’ experience as a threat to responsiveness. While subcabinet officers may be appointed to encourage responsiveness, the bureaucracy socializes its inhabitants to its needs and norms (Edwards & Wayne, 2006). Links between the White House and subcabinet officers are tenuous, and officers tend to develop their own perspective, creating the threat that they will pursue their own or their department’s interests at the president’s expense (Edwards, 2001, p. 97). Presidents express concern that once in office their appointees will “go off and marry the natives” (Cronin, 1980). Nixon complained: “It is inevitable when an individual has any government position, that after a certain length of time he becomes an advocate of the status quo; rather than running the bureaucracy, the bureaucracy runs him” (Nixon, 1972, p. 1150). A subcabinet officer who has previously worked in the bureaucracy will have developed loyalties to its norms, its status quo and the bureaucrats themselves.
The complex nature of a subcabinet officer’s experience within the executive branch suggests a trade-off for presidents seeking to promote their policies through nominations. By virtue of their positions, political appointees face cross pressures between the president and their departments they lead (Pfiffner, 2001, p. 113). Pressures within the departments compete for loyalty and threaten responsiveness. On the one hand, subcabinet officers must manage the bureaucracy to implement the president’s policies but, on the other hand, experienced officers may have developed narrow views that conflict with presidential preferences, or may have competing loyalties to careerists, relevant congressional committees, or clientele interest groups (Edwards, 2001). Presidents must weigh the costs and benefits of selecting loyalists and experienced officers, even as they consider how to account for the Senate’s environment.
Presidential Priorities and Departmental Jurisdictions
The question remains when presidents emphasize policy responsiveness or bureaucratic competence in their nominees. Presidents have policy priorities relative to different cabinet departments (Light, 1982, 1999) and some departments have jurisdictions that are simply more important to them than others. Light (1999) argues that presidents are constrained by internal and external resources and, yet, they choose to concentrate upon “issues that match their personal and political goals” (pp. 63-71). Lewis (2008) contends that presidents seek to increase the number of appointed positions in agencies where the ideological disposition differs most from their own. Gerhardt (2000) argues that the more important a position is to a president’s long-range goals, the higher value he places upon nominating his preferred person and the less likely he will compromise with senators. This suggests that politicization should exhibit patterns based upon differences between cabinet departments and the importance of their policy jurisdictions to the president’s agenda (Lewis, 2009), and presidents would be most aggressive in nominating loyalists to the departments closely associated with their priorities if they are to advance the policies most central to their agendas (Edwards, 2001).
Anticipating the Senate Environment
While the presidents’ decisions are central to understanding the confirmation process, much of the research has examined the publicly visible Senate confirmation process (McCarty & Razaghian, 1999; Ragsdale, 1996). The conventional wisdom has been that the Senate rejects nominees only in rare cases because it defers to presidents in choosing their subordinates (Cohen, 1988, p. 10; King & Riddlesperger, 1991, 1996, p. 281). Other scholars have argued that, even when confirming most nominees, senators can influence executive agencies by using hearings to make their concerns known or identify negative information about nominees as rationale for others to oppose confirmation (Gerhardt, 2000; Krutz, Fleisher, & Bond, 1998), as part of an overall strategy of competition with presidents (Mackenzie, 1981). Positive theorists argue that nominations are games of anticipation, and anticipation of senatorial obstructionism can force presidents to propose less controversial persons than their preferred nominees (Cameron, 2000, pp. 55-56). McCarty and Razaghian (1999) showed that the senators engage in dilatory tactics, especially “holds” (Binder & Smith, 1997), because they prefer the status quo over policies with the nominee in place. Senators’ preferences can be reflected in the outcome even though they confirm almost all nominees (Hill & Hammond, 1993). These works reveal aspects of the Senate environment that presidents must anticipate.
As the first mover in the game of anticipation, presidents’ actions matter and the description of process must account for their goals. Nixon (2004) studied appointees who had also served in Congress and concluded that ideological tilts in the Senate affect the ideology of nominees, but Nixon’s sample primarily consisted of cabinet secretaries rather than subcabinet officers and is thus limited in generalizability. Bertelli and Feldman (2006) argued that presidents choose nominees to offset the influence of interest groups on policies. Lewis (2008) identified patronage and policy control as two main presidential goals in politicizing agencies, and concluded that they politicize agencies when the ideological disposition of the agency differs from the president. While Lewis (2008) examined politicization through increasing the numbers of appointed positions, his work suggests that similar strategic considerations could affect the selection of appointees based upon partisanship and political attitudes (p. 208).
Discussions about presidential bureaucratic control have been dominated by the debate between bureaucratic competence and responsiveness, but neither of these descriptions captures the full picture. Even as they recognize the need for some experienced officers, they act with concern for responsiveness rather than mere competence, and nominate more loyalists to the priority departments. Moe (1985, 1993) accurately emphasized the importance of responsiveness, but even “responsive competence” may overstate the concern for competence. Presidents seek competence, but they are also suspicious that experience threatens responsiveness. In nominations to higher priority departments, they may nominate people with less experience because they are suspicious of too much experience. However, in nominations to lower priority departments, a president may emphasize nominees’ experience, with more concern for competence and less for responsiveness, because they seek only to inefficiency or disaster rather than to alter the status quo.
Hypotheses
The previous discussion suggests a set of testable hypotheses relating presidential decisions about nominees’ loyalty and experience, presidential departmental policy priorities, and anticipation of Senate reaction. Presidents place a premium upon politicizing high-priority departments with loyalists, and thus a president will choose people whose demonstrated loyalty to him indicate that they will be responsive in higher priority departments
Hypothesis 1: A president making nominations to higher presidential priority departments will nominate potential subcabinet officers with greater loyalty to his policy agenda.
Nominees’ strong partisanship and personal connections to the president should be positively associated with nominations made to higher priority departments.
While a subcabinet officer’s responsive loyalty indicates that she would aggressively promote the president’s priorities, scholars differ on their expectations about an officer’s experience in promoting presidential policies. On the one hand, a president must nominate people whose experience ensures efficient administration (Edwards, 2001; Light, 1987). On the other hand, a president who views experience as a threat to responsiveness would nominate people with less experience to high-priority departments because the desire for responsiveness trumps the desire for competence (Moe, 1985; Pfiffner, 2001).
Hypothesis 2: A president making nominations to higher presidential priority departments will nominate potential subcabinet officers with less experience.
In departments where bureaucratic responsiveness is most important, presidents may view experience and ties to the bureaucratic status quo as a threat. This argument is not that presidents do not value some experience in high-priority departments, but that there is a significantly decreased probability of nominating experienced officers to them. Looked at another way, this hypothesis suggests that in lower priority departments where presidents do not seek to alter the status quo, presidents nominate people with more experience. They value experience in agencies where their most important goal may simply be to avoid problems or negative publicity.
Scholars have argued that presidents politicize agencies in anticipation of the Senate’s reaction (Cameron, 2000; DeRouen et al., 2005; McCarty & Razaghian, 2000; Nixon, 2004; Volden, 2002). Presidents might anticipate stronger opposition by moderating their choices for the subcabinet to people who are less loyal to their agenda (Cameron, 2000; Nixon, 2004).
Hypothesis 3: A president with stronger Senate opposition will nominate potential subcabinet officers with less loyalty to his agenda.
At the same time, Fiorina (1992) argued that divided government enhances the quality of executive appointees, suggesting that presidents anticipate strong opposition nominating people with more experience to immunize nominations from senatorial charges of inexperience.
Hypothesis 4: A president with stronger Senate opposition will nominate potential subcabinet officers with more experience.
If anticipation of ideological or partisan opposition in the Senate’s confirmation affects presidents’ choices, it should be reflected in their choices of subcabinet nominees.
Methodology
The unit of analysis for this study is individual nominations. The nominations are a sample of 573 nominations made between 1961 and 2006, drawn at random from nominations to the cabinet departments to subcabinet positions requiring Senate confirmation. 2 These offices include the positions within the executive branch that are important enough to be summarized in the Congressional Quarterly Almanac, including undersecretaries, deputy secretaries, associate attorneys general, inspectors general, administrators, assistant secretaries, deputy assistant secretaries, deputy undersecretaries, and assistant attorneys general, but not including cabinet secretaries, ambassadors, nominees to the Executive Office of the President or regulatory boards. For each year, the number of nominations was selected according to that year’s proportion of nominations to all qualifying offices from all years, and the sample of nominees represents approximately one in every 5.9, or 16.9%. 3 Nominations were included independent of whether they were confirmed because the relevant dependent variable is the president’s choice of nominee, and not whether the nominee was confirmed.
Dependent Variables
The dependent variables of interest to this study measure the loyalty and experience of the nominees, both centrally important to examining presidential promotion of responsiveness and competence. I use personal and employment experience data on the nominees to score each nominee on the Loyalty and Experience variables. Each is an additive index of the number of affirmative answers to relevant questions about their employment history and political experience at the time of their nomination. In each case, public information from the nominee’s resume at the time of nomination, as recorded in either Who’s Who in America, Who’s Who in American Politics, or in the appendix of the nominee’s committee hearing, and known to both the president and senators, was used to calculate scores.
In the cases of both variables, it is difficult to assign values to measures of abstract concepts such as loyalty or relevant experience. 4 Even presidents and their staffs after extended interviews cannot be certain how loyal appointees will prove to be, and senators use interviews and hearings to determine whether nominees will put presidential interests ahead of senators’ concerns. Loyalty is measured as the number of “yes” answers to the following two questions:
1. Did the nominee have personal ties to the president as indicated by past employment with presidential staff or campaign?
2. Did the nominee have strong partisanship as indicated by holding a political party office or job?
Question 1 captures whether the nominee can be expected to have loyalty based upon a personal relationship to the president (Bertelli & Juenke, 2004; Cohen, 1988; Edwards, 2001). 5 Question 2 captures whether the nominee has strong partisan or ideological commitment to the president’s agenda. Affirmative answers to these questions measure a nominee’s personal, partisan, or ideological loyalty to the president’s agenda. Unless otherwise indicated, I use the term “loyalist” to indicate a nominee who answers “yes” to either Question 1 or 2.
Relevant experience for the variety of positions within the executive branch is difficult to quantify, but these measures are based upon past scholarly works and readings of confirmation hearings. 6 In order for a subcabinet officer to be successful, there is no substitute for experience within the executive branch or the specific department and to be savvy in the ways of Washington, DC (Edwards, 2001; King & Riddlesperger, 1996; Light & Thomas, 2000). Experience is measured as the number of “yes” answers to the following two questions:
3. Did the nominee have past executive branch employment?
4. Did the nominee have past employment with the relevant department?
Question 3 captures whether the nominee has any federal agency experience. Question 4 captures whether the nominee has experience with the specific department, indicating knowledge of that department’s unique workings and familiarity with issues related to it. In effect, a score of 1 indicates a nominee with some federal bureaucracy experience and a score of 2 indicates experience in the department to which she is nominated. Unless otherwise indicated, I use the term “experienced” to indicate a nominee who answers “yes” to either Question 3 or 4. 7
Independent Variables
Presidential priorities affect the selection of nominees, making the importance of the department’s policy jurisdiction to the president’s agenda the most important independent variable in this study. This study measures presidential priorities relative to different departments by scoring each department for each term as a proportion between 0 and 1 based upon policy statements in the State of the Union Addresses and Budget Messages of the President. 8 Light (1982, 1999) argued that presidents reveal their policy agendas in their public statements, specifically the State of the Union Addresses. The Budget Message of the President communicates the president’s top priorities in a public document to the legislature and is frequently more thorough than the State of the Union Address. The more frequently a president mentions a department, its officers or the policy jurisdiction of a department, the higher priority score is assigned to nominations to the department.
This study’s measure combines two proportions derived from content analysis of these documents to assign scores that measure the extent to which a department’s policy jurisdiction is featured in these public statements of presidents. The first measure is derived from Baumgartner and Jones’s 9 work on State of the Union Addresses that coded policy categories for each sentence in the addresses. The second measure is derived specifically for this study by coding each paragraph of each term’s first State of the Union Addresses and Budget Message of the President. Each paragraph was coded according to whether it had policy-relevant statements and whether those statements were either implicit or explicit references to cabinet departments, 10 and then each department was scored according to the average proportion of implicit and explicit statements divided by the total number of policy-relevant statements related to that department. 11
While both measures are blunt indicators of presidential priorities relative to cabinet departments, combining two similar measures increases the reliability of the measure (Crano & Brewer, 1986). On its face, the Presidential Priority Department variable is a combined measure that captures the importance of the policy jurisdiction to a president’s efforts to advance his priorities within the bureaucracy. For example, in Johnson’s second term during his advocacy for his “Great Society,” the Department of Health, Education and Welfare had a Presidential Priority score of 0.40, compared to the highest nonsocial service department, the Treasury Department, with a score of 0.12. During Carter’s efforts to address the “Energy Crisis,” the Department of Energy had a Presidential Priority score of 0.22 compared to 0.04 for the Justice Department. During Reagan’s second term, the Treasury Department had a Presidential Priority score of 0.32 compared to 0.14 for the Commerce Department and 0.04 for the Department of Agriculture. The Presidential Priority measure quantifies the importance that each department’s subcabinet offices have relative to the president’s policy agenda.
Presidents must anticipate senatorial assertion of their confirmation powers to delay or even defeat nominations. Even a single senator can delay confirmations to pressure presidents to alter their choices (McCarty & Razaghian, 2000). The senator at the filibuster pivot is the key voter in confirmations (Johnson & Roberts, 2004), because he represents the vote crucial to stopping a filibuster against a nominee (Binder & Smith, 1997; Krehbiel, 1998). Mackenzie (2001) argues that filibusters are the key to senatorial confirmation opposition (pp. 31-32). This study operationalizes Senate Opposition as the absolute difference between the president’s first dimension DW-Nominate score and the filibuster pivot’s DW-Nominate score. The greater the distance between the two, the greater would be the ideological opposition to nominations.
Control Variables
Other aspects of a subcabinet nomination can affect strategic calculations. Previous research has argued that presidents and senators treat nominations to foreign policy departments different from domestic departments, and I statistically control for the difference. The reason for this control is twofold. First, there is evidence that Congress has been more deferential on foreign policy and given generally more leeway, and there has been greater bipartisan agreement about nominations to the defense and state departments, as well as fewer partisans and more experienced nominees to those departments (Cronin, 1980; Ragsdale, 1996). Bertelli and Feldmann (2006) argued that the nature of the interest group environment and oversight related to these agencies may alter presidential choices (p. 33). Second, my presidential priority variable is based upon public statements in which foreign policy discussion accounts for a large proportion of the text and would swamp differences between other policy areas. To account for the different strategic considerations and the disproportions in references, I employ two dichotomous variables, Defense and State, equal to 1 if the nomination is to the relevant foreign policy department, and 0 if not.
Lewis (2008) examined the effects of relationship between presidents, Congress, and the ideological predisposition of agencies in politicization of the bureaucracy. He argued that presidents politicize by trying to change the number of appointees with the goals of patronage or policy control. To enhance political control, presidents seek to increase the number of appointees in agencies that they perceive as ideologically different from themselves. Thus, conservative presidents tend to politicize liberal departments and liberal presidents tend to politicize conservative departments. While Lewis includes many other offices and examines the number of offices rather than the types of nominees, the ideological predisposition of the agency is still an important consideration. To account for those effects on subcabinet nominee choices, I use Lewis’s categories based upon extensive expert surveys to code a dichotomous, Department Disposition, variable equal to 1 when the president is a conservative making a nomination to a liberal department or a liberal making a nomination to a conservative department.
Bureaucratic control literature emphasizes the importance of presidents “hitting the ground running” in appointments (Edwards, 2001; Mackenzie, 2001; McCarty & Razaghian, 2000), and to account for the importance of the first 3 months, I use a dichotomous Honeymoon variable equal to 1 if the nomination was made during the first 3 months of the administration and 0 if not. In a broader context, King and Riddlesperger (1996) found that the Senate expedites popular presidents’ nominations and others have found that presidents can use popularity to counter Senate delays (Johnson & Roberts, 2004; Krutz et al., 1998). I account for the public support by the most recent previous Gallup presidential approval rating prior to the nomination as a whole number between 1 and 100 with the Presidential Approval variable. Finally, presidents use appointments to build electoral support by choosing nominees from key electoral groups (Gerhardt, 2000; Pfiffner, 2001). Because electoral calculations may affect choices, I control for first-term presidents with a dichotomous Face Election variable equal to 1 during the first term and 0 when not.
This study employs two additional statistical controls: the rank of the position and the availability of nominees. McCarty and Razaghian (1999, 2000) found that the senators are less tolerant of fellow senators’ dilatory tactics on more salient, higher ranking offices, resulting in shorter confirmation times for these nominations. Consistent with their ranking system, I control for the effect of nomination to higher office with a dichotomous Rank variable. Nominations to lower ranked offices, such as assistant secretaries, deputy assistant secretaries, deputy undersecretaries, and assistant attorneys general are scored as 0 for Rank, and nominations to higher offices such as undersecretaries, deputy secretaries, associate attorney generals, and inspectors general are scored as 1. Furthermore, presidents with more prospective nominees to choose from should have more leeway than presidents with fewer, and the vast majority of nominees whose political affiliation can be determined from their resumes are from the president’s party. 12 Thus, a president whose party has controlled the executive branch more in recent years has more potential choices, and thus more leeway (Lewis, 2008, p. 124), and it should affect the number of loyalist and experienced nominees. This study operationalizes the availability of nominees with an Availability variable equal to the number of years in the previous 12 that a president’s party has held the executive branch. 13
Finally, administrative presidency scholars have noted that Nixon’s administration marks a turning point in presidential strategies. Nixon initiated a strategy of advancing his policies through administrative procedures, reorganization, centralization, and politicizing agencies (Nathan, 1983), and later presidents have also employed such administrative strategies to further their policy goals (Aberbach & Rockman, 2009; Howell, 2003; Moe, 1989). The subcabinet nominee sample begins with the Kennedy administration in 1961 as the first administration to begin his presidency with control of his own personnel office (Mackenzie, 1990), and I account for Nixon’s development of the administrative presidency with a dichotomous variable, Post-1968, equal to 1 for Nixon and subsequent presidents and 0 for Kennedy and Johnson.
Subcabinet Nomination Sample
The nominations analyzed in this study are a sample of nominations to sixteen cabinet departments, including health, education, and welfare counted twice because it was split into the education and health and human services in 1979. Republican presidents made 59.3% of the nominations, reflecting the additional years that Republicans controlled the White House. Of the nominations in the sample, 76.0% were made by presidents facing reelection, 13.6% were made during the first 3 months of their administrations, and 22.1% of the nominations were to higher ranking posts in the subcabinet.
Figure 1 reports the number of nominees in the sample by administration. The number of subcabinet nominations per administration varies from 26 for Ford to 100 for Clinton, and generally reflects more nominations for presidents who served longer as well as a trend of increasing numbers of subcabinet offices over time. While there may be theoretical justifications for sampling an equal number of nominees from each administration (e.g., by increasing the probability that a Kennedy nomination would be in the sample compared to a Reagan nomination), this study sampled proportionally across administrations to equally weight each nomination rather than each president and to remain consistent with earlier studies of the nomination and confirmation processes that weight each nomination equally across administration (McCarty & Razaghian,1999; Nixon, 2004).
Subcabinet nominees in sample by presidential administration.
Figure 2 reports the number of nominations in the sample by department. The number generally varies according to age and the size of the department, with only 5 nominations to the Department of Homeland Security and up to 77 nominations to the Department of Defense. The Departments of Commerce with 66 nominations and the Treasury Department with 52 nominations led the domestic policy departments, although the total nominations combined from Departments of Health, Education, and Welfare and the two departments it divided into (education and health human services) equals treasury with 52 nominations.
Subcabinet nominees in sample by agency.
The relationship between nominee loyalty and experience is important in the examination of presidential selection of nominees who are loyalists, experienced, neither, or both. Table 1 reports the cross-tabulation of loyalist and experienced nominees. While scholars have tended to treat loyalty and experience as a trade-off between “hacks” and “bureaucrats” (Dionne, 2001), the relationship between nominee loyalty and experience is more complicated than a simple trade-off. Edwards (2001) argued that there is no necessary reason to view expertise and loyalty as opposites.
Subcabinet Nominees’ Loyalty by Experience (Whole Sample). a
Cells report the proportion of the whole nominee sample as a percentage with the raw total below in parentheses.
On the one hand, in any given nominees’ career, experience in the executive branch would take time away from the types of activities such as campaign and partisan work that indicate loyalty. On the other hand, there are not disproportionate numbers of nominations in the upper right and lower left cells of Table 1 that would indicate a trade-off, and there are a substantial number of nominees (15.7%) who are both loyalists and experts. 14 This study cannot resolve whether presidents necessarily must choose between loyalty and experience. However, there are a substantial number of nominees who are both, indicating that although presidents might often see loyalty and experience as a trade-off, they do not necessarily have to choose between the two and often do not.
Statistical Method
This study uses ordered logistical regression for its primary analysis of presidential choices of nominees. Given the nature of the dependent variables, Experience and Loyalty as discrete and ordered categories, ordered logistic estimations of the relationship are the appropriate statistical methods and are preferable to ordinary least squares regression (Borooah, 2002). Each coefficient for ordered logistic regressions is interpreted as the additive change in the log of the odds ratio produced by a one-unit change in the independent variable. This study offers two equations for the two dependent variables, Loyalty and Experience, using the same set of independent variables for each equation. 15
Findings
Presidents’ selections of subcabinet nominees reflect strategic decisions about advancing their priorities as well as nominees’ loyalty and experience, and without as much concern for senatorial opposition as previous scholarship has suggested. According to these findings, presidents making subcabinet nominations to their higher priority departments emphasize the need for responsiveness by nominating significantly more loyalists, and in practice presidents view nominees’ experience with suspicion and nominate fewer experienced nominees to their higher priority departments. This finding also means that presidents making nominations to lower priority departments de-emphasize responsiveness, and emphasize competence by selecting significantly more experienced nominees. At the same time, the strength of Senate opposition has no significant effect, indicating that anticipation of Senate opposition plays little if any role in presidential choices. Presidents do not simply nominate a mixture of loyalists and experts, as other works suggest, but rather their choices reflect calculations about how best to advance their higher priorities.
While previous scholarship on subcabinet offices found that presidents desire some combination of experience and loyalty to promote their policies, these findings reveal the strategic nature of their choices. Presidents are most aggressive in politicizing higher priority departments. The findings in Table 2 indicate that presidents emphasize responsiveness by nominating more loyal nominees to higher priority departments.
Policy Promotion Regressions.
Cells report the regression ordered logistic coefficients (left) and SEs (parentheses).
Significant at the p < .05 level.
Significant at the p < .01 level.
The significant positive coefficient for the Presidential Priority Department variable in the responsiveness model (Table 2, left column) indicates that presidents nominate significantly more loyal officers to their higher priority departments (p = .027). Hypothesis 1 is confirmed. Presidents politicize cabinet departments by nominating subcabinet officers whose loyalty motivates them to aggressively promote the presidential preferences. The president’s policy goals are central to their first move in the subcabinet confirmation game.
The conditional probability estimates in Figure 3 report the extent of the effect of presidential priorities upon the probability of choosing subcabinet nominees whose backgrounds indicate their loyalty to the president. In order to aid interpretation of these results, it displays the predicted probabilities of nominations at different levels of loyalty (one or two) based upon the importance of the department to the president’s agenda (other variables set at their means).
Conditional probabilities of nominating loyalist subcabinet officers.
Across the whole range of the values of the independent variable, the probability of nominating someone in the most loyal category (loyalty = 2) more than triples, increasing from 3.1% to 10.4%, and the probability of nominating someone in the lower loyalty category (loyalty = 1) more than doubles, increasing from 16.3% to 36.4%. Overall, presidential priorities are associated with a 27.3% increase in the probability of nominating a loyalist to the subcabinet across the range of data in the sample.
Anticipation of the Senate environment is not an important determinant of nominee loyalty. Senate opposition does not significantly reduce nominee loyalty (p = .589). Hypothesis 3 is rejected. While other studies have found that Senate opposition decreases success in confirmation (King & Riddlesperger, 1991) and delays confirmations processes (McCarthy & Razaghian, 1999, 2000), presidents appear willing to risk these costs to populate the bureaucracy with subcabinet officers responsive to their preferences. Furthermore, the average length between nomination and confirmation for loyalists and nonloyalists is only 3.5 days longer (50.3 and 46.8 days, respectively) and is not statistically significant. 16 Presidents choose their nominees without significant concern for opposition in the Senate.
Presidents choose people with significantly more experience to subcabinet offices in their lower priority departments, and significantly less experience to offices in higher priority departments. The significant negative coefficient for the Presidential Priority Department variable in the competence model (Table 2, right column) indicates that they select experienced nominees in departments where they do not seek to change the status quo, but nominate people with less experience to their higher priority departments (p = .022). Hypothesis 2 is confirmed. Presidents seek competence in lower priority departments and select experienced nominees better at managing and motivating careerists. However, they seek responsiveness even at the potential cost of reduced competence with less experience in higher priority departments, indicating the importance to presidents of gaining responsive implementation of their policies.
The statistical estimates in Figure 4 report the extent of the effect of presidential priorities upon nominee experience. To aid interpretation of these results, it displays the predicted probabilities for different levels of nominee experience based upon how important the department’s jurisdiction is to the president’s agenda. Setting all other variables at their means, the probability of nominating someone in the most experienced category (experience = 2; i.e., to a department in which they have previously served) is cut by more than half, decreasing from 42.0 % to 18.7%, while the probability of nominating someone in the lower category (experience = 1; i.e., any executive branch experience) drops from 24.6% to 20.1%. 17
Conditional probabilities of nominating experienced subcabinet officers.
Across the range of the sample, presidential priorities are also associated with a 27.9% decrease in the probability of nominating someone with executive branch experience. Figure 4 illustrates the extent to which presidents select nominees with more experience in lower priority departments where they are more concerned with administrative competence and less concerned with promoting responsiveness.
This finding supports the view that presidents nominate fewer experienced subcabinet officers to their higher priority departments because they view previous experience with suspicion. Specifically, presidents demonstrate a significant decrease in the probability of choosing nominees who have served in the same department. A potential nominee’s previous experience in any department may suggest some decrease in responsiveness, but a nominee returning to a department where she has already served would be that much more likely to be influenced by that departmental status quo contrary to presidential preferences. The sharp decline of 23.3% across the range of the data indicates that presidents worry that nominees with previous experience in the same department are less responsive. Contrary to some assumptions, presidents do not view a potential nominee’s previous experience within a department as always an asset, but rather they view it with suspicion in nominations to their higher priority departments and worry about their agents “going native” where policies matter most. However, in other departments, less important to their agendas, presidents would still nominate experienced people to promote administrative competence.
This empirical support for presidential suspicion of previous executive branch experience should be kept in perspective. The evidence does not support the case that presidents entirely avoid nominating experienced subcabinet officers. First, theoretically, presidents could choose to never nominate people with any experience, but they often do. Second, the finding that availability increases nominee experience, as indicated by the significant coefficient for the Availability variable in the competence model (Table 2, right column), indicates that presidents choose more experienced nominees when more are available (p = .003). Presidents may be cautiously suspicious, but they clearly realize the value of some experience among their subcabinet officers to even higher priority departments. While there is no set minimal standard for proportion of experienced subcabinet officers, in practice subcabinet nominees with some executive branch experience still make up 62.0% of the nominees in the sample. This presidential strategic behavior makes sense in that it portrays presidents as avoiding experienced nominees in their highest priority departments, but ensuring bureaucratic competence in departments whose jurisdictions matter little to them. In these departments, presidents have little disagreement with the status quo and hope that no bureaucratic incompetence there hurts their administration’s reputation. While they avoid appointing “too many” people with “too much” experience to higher priority departments, they find some experienced people whom they trust.
The evidence also indicates that Senate opposition strength does not have a significant effect upon nominee experience. The coefficient for Senate opposition (Table 2, right column) is not significantly different than 0 and is in the wrong, negative direction (p = .250). Hypothesis 4 is rejected. Senators publicly state that they see their role in confirmations as assuring minimal levels of competence among subcabinet officials, but greater ideological opposition to the president does not, in itself, result in him choosing nominees with more experience (King & Riddlesperger, 1996). Senators may enforce a minimum standard of competence but ideological differences with the president do not significantly alter the standard or affect presidential decisions.
In addition to identifying the ways presidential priorities affect the strategic choices of presidents, these findings reveal other strategic aspects of subcabinet nominations. The need to politicize quickly officers increases nominee loyalty. Scholars have frequently noted the importance of presidents seizing control of the bureaucracy early in their administration (Mackenzie, 2001; McCarty & Razaghian, 2000; Moe, 1985, 1989, 1993; Nathan, 1983) by placing loyal subcabinet officers quickly. The significant coefficient for the Honeymoon variable (p = .049) in the positive direction for the responsiveness regression indicates that presidents act quickly to gain control of the bureaucracy through the nomination of officers loyal to his policy agenda.
Finally, the coefficient estimates for the Post-1968 variable are consistent with previous research on the administrative presidency since Nixon. The significant positive coefficient in the responsiveness model (p = .015) and the significant negative coefficient in the competence model (p = .026) indicate that Nixon and more recent presidents have been more likely to nominate subcabinet officers who were loyalists and who were less likely to have executive branch experience. These results should be interpreted with caution because the nature of ordered logistic regression calculations means they are a comparison of the Kennedy and Johnson nominees to Nixon and more recent presidents, but the findings are consistent with a picture of changing strategies by presidents in the modern era. Nixon and his predecessors politicized the executive branch by nominating significantly more loyalists and fewer experienced officers.
Conclusion
This President-centered account argues that understanding the President is the key to understanding the nomination and confirmation processes that determine who inhabits the top political ranks of the bureaucracy. While other studies have enhanced the understanding of the role of obstructionist senators, this study illuminates the strategies of the President who sets the process in motion with his selection of his nominees, the great majority of whom become executive officers. While Senate opposition delays confirmation and decreases success in a few cases, senatorial opposition has no significant, discernable effect on presidential choices. Scholars argue that presidents choose subcabinet nominees with an eye toward the Senate, but presidents appear willing to risk delays or rare defeats to win the confirmation of responsive loyalists who will advance their preferred policies. These findings, along with high confirmation rates indicate that presidents largely succeed in their domination of the confirmation process.
Presidential decisions and goals should be central to understanding subcabinet appointments. Attention to the actual nominees that presidents choose and the strategic selection of experienced or loyal officers reveals the centrality of presidential decisions and policy goals. Whereas other research has concluded that presidents need some combination of loyalists and experts or has overemphasized the Senate’s role, these findings specify that in practice presidents primarily seek bureaucratic responsiveness by populating the departments, especially those central to their agendas, with loyalists. Presidents value loyal officers and they act strategically to politicize the executive branch with subcabinet officers whose partisan or personal loyalties ensure that they will responsively promote presidential preferences. Presidents value experienced subcabinet officers, but they are suspicious that they agents will be unresponsive or even resistant to their policy preferences once embedded in the bureaucracy. Thus, presidents nominate more experienced people to the departments where policies are not central to their policy agendas to promote efficient administration there, but they nominate significantly fewer experienced people and more loyalists to the departments whose priorities are important to their agendas. In both cases, the central determinants of who inhabits subcabinet offices are presidents and their goals rather than the Senate environment, and these findings empirically demonstrate the connection between presidential policy priorities and their strategic politicization of executive branch agencies.
Footnotes
Acknowledgements
The author thanks Lawrence Jacobs for his invaluable insight and advice on this research. I would also like to thank Timothy Johnson, Jeffrey Cohen and Karen Hult for their help and encouragement on earlier versions of this paper.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.