Continuity, competence, and the succession of senate-confirmed agency appointees, 1989-2009.

The presidential power to appoint senior government officials has evolved from a few phrases in the second paragraph of the second section of Article II of the U.S. Constitution into an unwieldy and opaque system of rules and expectations. "The appointment power operates in a framework of studied ambiguity," Louis Fisher observes, developing through generations of "imaginative accommodations between the executive and legislative branches" (1985, 59). Demands for broad-based reform occasionally take hold, most recently through the Vacancies Reform Act of 1998, which sought to reassert the Senate's advise and consent power by placing strict (though sometimes neglected) limitations on the service of "acting" appointee (GAO 2003; Stayn 2001). Yet, as in the case of the controversy over the appointment of interim U.S. attorneys in the Department of Justice, the system's "studied ambiguity" remains unyielding. (1) Its complexity is rooted in tensions inherent in presidential administration: the drive for political control and policy competence; the dynamics of issue networks and policy domains; and the priorities and ambitions of politicians, interest groups, and administrators.

Recognizing that efforts to fix the appointee system are as inevitable as the system is, in some basic respects, unfixable, this article aims simply to provide a measure of clarity. We describe one prominent feature of agency appointee politics--continuity--based on appointee turnover, tenure, and position vacancies. (2) We develop a series of snapshots in time, illuminating the succession of the top-tier of appointees--presidentially appointed and Senate confirmed (PAS)--during the administrations of George H. W. Bush, Bill Clinton, and George W. Bush, between January 20, 1989 and January 20, 2009. Using a new data set drawn from the Office of Personnel Management, Government Accountability Office, and a variety of published sources, we analyze roughly 2,200 appointments across three presidencies, describing patterns of tenure and vacancies that form the contours of the administrative presidency. (3) We consider research linking appointee continuity and agency performance and conclude by reflecting on the nature of this relationship for research on and the practice of presidential administration.

More than three decades ago, Hugh Heclo's A Government of Strangers (1977) depicted America's transient governing elite--the presidentially appointed, Senate-confirmed administrators populating the upper echelons of federal government agencies--as the role of appointees shifted with the expanding reach of institutional presidency. Heclo observed that transience among appointees forms a stark, sometimes uneasy contrast with the continuity of senior-level career civil servants, often at the culmination of successful government careers. He saw relations between appointees and careerists shifting under the weight of thickening bureaucracies and the politicization of government administration, central themes in contemporary presidency and public administration scholarship (Goodsell 2006; Light 1995; Peters and Pierre 2004). Richard Nathan's (1983) administrative presidency and Terry Moe's (1985, 1990, 1993) politicized presidency have inspired a generation of scholars who have embraced political control of the bureaucracy as a means to fulfill Hamiltonian visions of an energetic executive. (4) Critics of politicization, by contrast, stress the value of administrative competence, contending that politicized agencies put the state's capacity for expertise and, ultimately, performance at risk (Lane 1996; Williams 1990).

Most contemporary thinking about politicization posits a trade-off between political control and administrative competence (Bawn 1995; Bertelli and Feldman 2007; Epstein and O'Halloran 1999; Miller 2005). This control-competence trade-off is analytically useful, framing a critical tension in bureaucratic structure (Waldo 1948; Weber 1958). Nevertheless, this dichotomy obscures the social foundations on which administrative institutions are built. By way of illustration, David Lewis (2008) defines politicization as the penetration of appointed positions, measured by the ratio of appointees to total agency staff. Lewis's research represents the most sophisticated empirical research on politicization in U.S. federal agencies to date. Nevertheless, this operationalization fixes on the formal features of politicization, neglecting its normative dimensions, which are represented, for example, by the erosion of professional autonomy in appointments to expert advisory committees. (5)

Consider the qualities defining appointee competence. Integrity, commitment, intellectual ability, managerial skills, personal charisma, adaptive capacities--to a greater or lesser extent, all are attributes one might associate with quality or competence. While it may be easy to agree about what incompetence looks like, competence often rests in the eye of the beholder. (6) As a consequence, the character of appointee competence is more difficult to discern than its absence. For whatever combination of reasons, a consensus now holds that the George W. Bush administration deployed a number of appointees who lacked the competence to fulfill their mandates. (7) An effort to explain the causes and consequences of Bush-era politicization by focusing on its formal artifacts risks omitting some of its most vital implications. More importantly, apart from the administration's most controversial appointments, broad segments of the Bush administration, like all presidential administrations, maintained what Heclo terms "mutual performance" between appointees and careerists. (8) Rather than a trade-off between competence and control, he finds learning, the cultivation of professional networks, and a "conditional cooperation" between appointees and bureaucrats. Time plays a critical part in the cultivation of mutual performance. (9) Heclo writes,

   Time on the job brings an opportunity not only to learn about the
   substance of programs and how to operate in the Washington networks
   but also to use what has been learned. Policy priorities are not
   immune to change, but they are most likely to respond to leaders
   who are around long enough to build support, to institutionalize
   changes in the bureaucracy, and to string together the narrow
   margins available at any one time to a strand of policy
   development. (1977, 237)

Duration is a vital component of agency leadership, as a source of both competence and the credibility of leadership commitments. Continuity among appointees promotes the accumulation of knowledge, improved judgment and accountability, and the cultivation of good-faith relations between appointees and careerists. At the same time, it is striking how little the scholarship on appointees offers when it comes to the systematic empirical observation of appointee continuity. The following section describes a new data set, which we hope will contribute to growing scholarly interest in measuring the causes and consequences of politicization and the role of appointees in the interplay between presidents and government agencies.

Appointee Tenure, Turnover, and Vacancies, 1989-2009

How long do presidential appointees serve? A common refrain holds that these "birds of passage" serve only two years, too brief a period to learn the particulars of their important positions. However, two years is a figure grounded in folk wisdom rather than systematic observation. In reality, the great diversity of appointee positions scattered across agencies resists such broad generalizations. The salience of the "two years" refrain in fact suggests just how opaque the appointment process remains, even among students of the U.S. government. For all its importance, the role of appointee continuity in agency administration has not generated a robust empirical scholarship. Until recently, with a few noteworthy exceptions (GAO 1994; Marcum et al. 2001), most analyses of political appointees in U.S. federal agencies relied on surveys (Aberbach and Rockman 2000; Ban and Ingraham 1990; Brauer 1987; Joyce 1990; Mackenzie 2001; Mackenzie and Light 1987; Mackenzie and Shogan; Pfiffner 1987). In recent years, intriguing research has examined the politics of agency appointments (Bertelli and Grose 2007; Lewis 2007; McCarty and Razaghian 1997; Nixon 2004; Peters and Pierre 2004) and the factors contributing to length of service, in particular (Wood and Marchbanks 2007).

In this section, we describe patterns of appointee succession through tenure, turnover, and position vacancies during the George H. W. Bush, Bill Clinton, and George W. Bush administrations. (10) We examine one aspect--continuity--of appointments to full-time, civilian PAS positions in all departments, single-headed independent agencies, and Executive Office of the President organizations, a total of a little more than 2,200 appointments. (11) We employ a new data set collected through formal requests to the Office of Personnel Management, Government Accountability Office, and published sources, including the Senate nominations database available through Thomas.gov, Congressional Research Service reports, contemporary news coverage accessed by LexisNexis, and other online resources.

Figure 1 summarizes tenure data during the Bush and Clinton administrations across several categories of PAS appointees. The overall median is 2.5 years, not far from conventional estimates of appointee tenure, but with several intriguing variations. A quarter of the appointees served more than 3.6 years--almost a full presidential term--while another quarter of appointees served less than 18 months. (12) Senior appointees tend to serve longer terms. The mostly cabinet-level appointees in the Level I pay category served a median of 3.3 years. Level II appointees, by contrast, generally serve much shorter terms, a median of 2.3 years, again with a quarter serving less than 18 months. A similar pattern holds for lower executive pay levels: Approximately one-quarter of appointees served less than a year and a half, and another quarter served three and a half years or more. (13)

[FIGURE 1 OMITTED]

Given profound changes in the scope and organization of American government, these figures are strikingly consistent with earlier studies. Figure 2 compares the results of the current study with two earlier studies: MacMahon and Millett's (1939) study of PAS appointee tenure from 1798 to 1938, and Stanley, Mann, and Doig's (1967) review of appointees between 1933 and 1965. Looking at a period that extends essentially from the early years of the republic to the years just prior to World War II, MacMahon and Millett find that 37% of appointees served less than two years, while a quarter served more than four. Between 1933 and 1965, appointee tenure dropped discernibly. Stanley, Mann, and Doig find half of appointees served less than two years, while only 15 % served more than four. Notably, the decades from FDR to LBJ produced more total appointments than did the prior 140 years covered by the earlier study. Expansion in the ranks of second- and third-tier appointees--undersecretaries and assistant secretaries--arguably contributed to the decline in tenure, as these lower-level appointees generally serve shorter terms. (14) Stanley, Mann, and Doig find that top-level PAS appointees served a median of roughly three and a quarter years, while second- and third-level appointees served median terms of 1.7 and 2.2 years, respectively.

The period between 1989 and 2009 produces several interesting parallels. During this period, 38% of appointees served less than two years, and 22% of appointees served more than four. Throughout, top appointees tend to serve longer. Positions classified as agency heads in Figure 1, including subcabinet bureau and department administrators, likewise typically serve longer terms, with a median of 3.2 years. Indeed, the longest-serving quarter of agency heads served more than a full presidential term, placing this group among the most durable of PAS appointees. Among the plausible interpretations of this pattern, tenure may represent a source of authority for agency leaders and the administrations they serve, enhancing the credibility of strategic commitments by extending the leadership horizon. Presidents seeking to enhance the authority of their appointees may find it advantageous to encourage agency heads to serve longer terms. Moreover, these are among the most powerful and prestigious appointments in the U.S. government. Whereas more junior positions offer several avenues for advancement, agency heads may find they have no more desirable opportunities in government service.

Among the other categories examined in Figure 1, inspectors general (IGs) serve exceptionally long periods, with a median of more than four years. A quarter of all IGs served more than seven years, and many IGs served a decade or longer. (15) Though, in general, IG positions are characterized by a high degree of continuity, some agency IGs exhibit unusually high rates of turnover. During the Clinton administration alone, for example, the Treasury Department saw three confirmed and a host of acting IGs. In general, though, IGs are durable, in part because of formal rules designed to insulate these positions from political pressure. (16)

The fixed-term category in Figure 1 consists of appointees to positions carrying terms that are fixed by statute. These include positions such as the administrator of the Rural Electrification Administration, which has a fixed term of 10 years; the director of the U.S. Mint, which has a fixed term of five years; and the director of the Office of Personnel Management, which carries a fixed-term appointment of four years. Fixed terms are usually created to insulate appointees from political pressures, and they clearly extend appointee tenure. In addition, they seem to extend appointee tenures. A little more than half of the 93 fixed-term appointees examined here served a full term or longer. With an overall category median of almost three and a half years, appointees to these positions served significantly longer than those in other PAS positions. (17)

Agency chief financial officers (CFOs), legal counsel, and positions dedicated to congressional or legislative affairs fall on the opposite end of the continuity spectrum. As many as one-quarter of appointees confirmed to these positions serve less than 16 months. In these occupational categories, a strong case can be made that appointee tenure is driven in part by external job markets (Chang, Lewis, and McCarty 2001; Wood and March-banks 2007). Accounting, law, and lobbying are marketable skills, and government service is potentially valuable in the private sector. Congressional relations appointees, for example, often move in and out of government as congressional staffers, lobbyists, and corporate legislative affairs officials.

Figure 1 also highlights variations in tenure across agencies. Short-tenure agencies, including Energy, Labor, Treasury, and the Executive Office of the President, produce medians of around two years. This is a stark contrast to the subset of long-tenure agencies, including Health and Human Services and the Central Intelligence Agency, in which appointees served a median of roughly three years, and a quarter of agency appointees served more than four. The types and number of positions in each agency presumably influence these figures, as do agency-level organizational and political factors.

Figures 3 and 4 depict a portrait of appointee succession in two major agencies during the two decades between 1989 and 2009. The Department of Health and Human Services (HHS) is among the long-tenure agencies, while the Department of Commerce is among the agencies with shorter lengths of service. Both agencies have seen significant growth in the ranks of political appointees over time. (18) HHS is the legacy of two reorganizations, the 1953 creation of the Department of Health Education and Welfare and the 1979 reorganization of that department into two new agencies, the Department of Education and HHS. Paul Light (1998, 100-102) finds that the creation of two new cabinet agencies doubled the number of undersecretaries and assistant secretaries, without a corresponding increase their policy mandates. The Department of Commerce is an older department, established in 1921, but it, too, has undergone extensive thickening in recent decades, driven in part by pressure from interest groups and Congress (Light 1998). A closer look at patterns of appointee tenure in Commerce and HHS highlights within- and cross-agency variations in administrative continuity.

The contrast between the agencies is evident in their respective cabinet secretaries. Between 1989 and 2009, HHS was headed by two well-respected health policy specialists and administrators, both of whom served almost a full presidential administration, and by two former governors. (19) During the same period, by contrast, the Department of Commerce was headed by seven confirmed and one recess-appointed cabinet secretaries, several with direct personal and partisan ties to the president. (20)

[FIGURE 3 OMITTED]

Consistent with government-wide patterns, congressional relations appointees serve comparatively short terms in both agencies. Four of the six secretaries for congressional relations serving in Commerce during this period served less than two years. Five accepted lobbying positions in private firms, and one took a similar position at the Corporation for Public Broadcasting upon leaving government service. The longest serving of the congressional affairs appointees in the Commerce Department, who served for virtually all of the George W. Bush administration, from May 2001 to January 2009, also became a lobbyist upon her departure from government service. Of the four assistant secretaries for legislative affairs in HHS during this period, six likewise became lobbyists upon leaving government.

[FIGURE 4 OMITTED]

Of the six Commerce CFOs serving between 1989 and 2009, five served less than two years, taking a variety of public and private sector positions on departure. Departing again from the pattern of rapid appointee succession, one Commerce CFO served for almost seven years of the George W. Bush administration. Three of the five HHS CFOs served less than two years, though one Clinton administration CFO served more than six.

These variations suggest a broader pattern in appointee succession seems to emerge from across these presidencies. In general, the Clinton HHS appointees served longer terms than did George W. Bush's HHS appointees. Clinton HHS appointees served an average of 3.9 years, compared with an average of 2.7 years among George W. Bush HHS appointees. By comparison, George W. Bush administration Commerce appointees exhibited greater stability than did the Clinton-era Commerce appointees. Commerce appointees in the Clinton administration served 2.5 years on average, compared with an average of 2.8 years of service in the George W. Bush administration. Without inferring too much from the broad patterns depicted here, it is worth noting that the two administrations exhibit greater appointee continuity in the agency with which the president's ideology is more closely aligned (Clinton and Lewis 2008).

Among the appointee positions organized around issue expertise, the demands of specialization and limited external job markets presumably favor longer tenures. Where issue networks are robust and agency professionals retain credible claims to expert autonomy, we expect greater continuity among appointees. (21) The three administrators of the National Oceanic and Atmospheric Administration during this period, for example, each served almost a full presidential administration.

Where salience or controversy is high and administrators do not have recourse to strong claims of the value of professional autonomy, however, appointee continuity is clearly affected. For example, the director of the U.S. Census Bureau, also in the Commerce Department, is a long-standing expert appointment that in recent years has become the focus of considerable political controversy, particularly surrounding the proposed used of statistical sampling to enhance the accuracy of census data. The terms of all five Census Bureau directors during this period were shaped, in part, by that controversy. The Bush administration appointed Barbara Bryant during recess almost a year after the administration took office, delayed by conflict over an earlier proposed nominee (Barringer 1990). The second Clinton administration Census Bureau director, Martha Farnsworth Riche, resigned unexpectedly in 1998, a critical time for the pending 2000 census, remaining vacant until the second term of the George W. Bush administration. (22)

Likewise, in HHS, the surgeon general is an expert position that became the focus of public controversy and conflict between presidents and Congress. Surgeon General Jocelyn Elders, for example, was forced to resign in 1994 because of controversy over comments she made about drug legalization and contraception. After her departure, conflict between President Clinton and the Republican Senate majority left the position vacant for almost four years.

The commissioner of the Social Security Administration (SSA) offers a useful contrast. For several decades during the twentieth century, the Social Security commissioner offered a noteworthy example of appointee continuity and influence (Derthick 1979). Yet, in recent decades, resource constraints and a fragmenting political consensus placed new pressures on the agency. In Agency Under Stress, Martha Derthick (1990, 122-23) observes that appointee turnover and vacancies during the 1970s and 1980s, a period when the programs administered by the SSA suffered increasingly fragmented political support, contributed to the erosion of the agency's capacity. (23) Responding to this perceived link between leadership continuity and organizational capacity, Congress in 1995 legislated a five-year fixed term for the position of commissioner when it made the SSA an independent agency, removing it from HHS (Koitz 1989, 1994). In other words, Congress sought a structural solution to the problem of appointee continuity. The logic of fixed-term appointments holds that they insulate positions from politics and enhance leadership credibility by extending the appointee's leadership horizon. (24)

Though the link between appointee continuity and agency performance is implicit in the creation of fixed-term positions, this empirical relationship has received little systematic treatment. The following section considers existing scholarship on the hypothesized relationship appointee continuity and agency performance.

Appointee Continuity and Agency Performance

Much of the concern over the continuity of agency appointees centers on the perceived consequences for agency performance. Such concerns garner public attention when issues of national security are at stake. In 1960, concern that shortening appointee terms would erode the capacity of the national security establishment prompted passage of a Senate resolution stating that "individuals appointed to administrative and policymaking posts should be willing to serve for a period long enough to permit them to contribute in their assigned fields." (25) In recent years, lengthy vacancies that have disrupted relations between agency careerists and elected politicians have also garnered attention. (26) The 9/11 Commission report noted that "the new [George W. Bush] administration did not have its deputy cabinet officers in place until the spring of 2001, and the critical subcabinet officials were not confirmed until the summer or later. The new administration--like others before it--did not have its team on the job until at least six months after taking office" (2004, 422). (27) The 2001 Brookings Institution Presidential Appointee Initiative highlighted this problem by tracking the time it took the new administration and Congress to make the first round of cabinet and subcabinet appointments. The 2008-9 presidential transition benefited from efforts in the aftermath of 9/11 to speed the placement of new administration appointees, but concerns about appointee continuity again took center stage, this time focusing particularly on the Obama administration's ability to fill key Treasury Department positions in the face of economic crisis.

Even still, evidence linking appointee continuity and agency performance remains largely anecdotal. David Lewis's empirical work is a noteworthy exception (see Gilmour and Lewis 2006; Lewis 2007). Using the Program Assessment Rating Tool (PART) of the Office of Management and Budget (OMB), Gilmour and Lewis (2006) find that programs administered by Senate-confirmed appointees scored systematically lower on portions of the PART related to management and "results" than did programs administered by career Senior Executive Service members. Lewis (2007) finds that agency experience and longer tenures contribute to differences in PART performance between agencies headed by appointees and careerists. Many agency observers distrust PART scores as indicators of organizational performance, however. PART ratings are designed to evaluate programs rather than agencies, and, despite significant efforts on the part of some in the OMB to counter this impression, PART is regarded as one component of the George W. Bush administration's larger ideological agenda (Dull 2006; Radin 2006). As the political contentiousness surrounding the use of PART scores illustrates, defining effective administration is a daunting proposition. Alexander Pope's refrain, "For forms of government let fools contest; What'er is best administered is best," is a popular quotation for dinner speeches because it is so ambiguous.

Looking beyond the U.S. federal government, Boyne et al. (2008) employs externally validated annual performance measures mandated for all British local governments to find that executive succession positively influences performance where prior performance is low but negatively influences performance when prior performance is high (Boyne et al. 2008). The authors also find that individual and organizational contingencies, such as changing political leadership and whether the executive was elevated from within, significantly moderate the effects of executive turnover.

The robust literature examining the relationship between tenure and performance in private organizations also offers useful evidence. Classic studies of professional baseball (Grusky 1963) and college basketball (Eitzen and Yetman 1972) that find a negative correlation between frequent coaching turnover and team performance led to a broad research literature highlighting a curvilinear relationship between executive tenure and organizational performance (Kesner and Sabora 1994). At short tenures, task-based learning occurs at a steep rate, and thus the length of executive tenure positively influences performance. In the case of a basketball coach, the chances for success increase steadily, but over time, the benefits of tenure plateau and begin to drop (Hambrick and Fukutomi 1991). (28) Long tenured executives may lose motivation, become overconfident or rigid in their habits, or remain wedded to outmoded strategies and practices. In general, however, executive team tenure is positively associated with strategic persistence and organizational conformity (Finkelstein and Hambrick 1990; Hambrick, Geletkanycz, and Frederickson 1993). Therefore, tenure may tend to promote stability and trust while, at the same time, increasing the risk that an organization will fail to adapt to changing conditions. (29) Executives may get "stale in the saddle," becoming rigid in their thought processes and unresponsive to the demands of the organization's strategic environment (Katz 1982; Miller 1991).

Variations across settings and conditions help to fill out the broader pattern. The quality of organizational leadership and performance prior to executive succession, for example, exerts a strong influence on post-succession performance. Shen and Cannella (2002) find that the relationship between departing executive tenure and post-succession firm performance takes the form of an inverted U-shaped curve. The departure of both short-tenured and long-tenured CEOs negatively influences firm performance, whereas performance increases in the middle of the distribution. In recently acquired firms, where short tenures may be linked to adaptability in a dynamic and uncertain context, executive team tenure--and presumably the accumulated knowledge and trust that are associated with tenure--nonetheless is a strong positive predictor of post-acquisition performance (Bergh 2001; Krishnan, Miller, and Judge 1997). Characteristics of the executives themselves are both shaped by and, in turn, help to shape the relationship between tenure and performance. (30)

Scholarship on the impact of position vacancies is much more limited, in part because of the distinctive nature of appointees in the U.S. government. The emphasis in the 9/11 Commission report and elsewhere on vacancies during the transition between administrations, on "hitting the ground running," speaks to important challenges faced by new administrations. Less attention has been dedicated to the vacancies resulting from midterm appointee departures, which also result in substantial gaps in agency leadership, particularly in the context of conflict between presidents and Congress. In contrast to the vacancies that are typical of a first-year presidential administration, midterm vacancies are scattered and varied. Though particular midterm appointee vacancies occasionally become a focus of public debate (often tied to claims of administrative incompetence), midterm vacancies are typically a function of politics at the level of agencies, committees, and issue networks, where there is little support for systemic reform. In his case study of politicization and performance in the Federal Emergency Management Agency (FEMA), Lewis (2008) observes the impact of ongoing, rotating appointee vacancies on agency leadership and morale. The expansion in the number of appointee positions as a consequence of the Department of Homeland Security reorganization resulted in vacancies of one-third of FEMA's appointee positions by the fall of 2004 (Lewis 2008, Table 6.1).

[FIGURE 5 OMITTED]

To illustrate how patterns of appointee vacancies shape agency leadership, Figure 5 graphs the percentage of Commerce and HHS PAS positions occupied between 1989 and 2009. (31) Several waves of position turnover and vacancies are discernable. The first, during the early months of the first Bush administration, is defined in part by the retention of a number of Reagan appointees, most of whom ultimately turned over in the first year of the new administration. The number of confirmed Bush appointees climbs slowly, reaching 90% in both agencies during the middle of 1991. Commerce yields a particularly pronounced increase, from less than half to essentially fully staffed in a year, about a year stable at that level, to an equally sharp decline that is likely related, among other factors, to the departure of Secretary of Commerce Robert Mosbacher to join the president's reelection campaign. Appointee holdovers in HHS yield comparatively higher rates early on, growing more slowly but remaining at nearly 80% until September 1992.

The first year of the Clinton administration creates a second discernable wave. Both agencies undergo a swift and near complete turnover among PAS appointees. Again, the two agencies exhibit different trajectories. By December 1993, HHS climbs to 100% of positions confirmed, while Commerce barely exceeds half. Both agencies are fully staffed by the middle of 1994, declining again by late 1995. During the first year of Clinton's second term, a smaller wave of appointee departures and vacancies is discernable. This second-term reshuffling of agency leadership results in a shallower, but longer period of appointee vacancies. Rates of confirmed appointees climb in both agencies, again more steeply in HHS, and remain stable from 1998 through the end of the administration. During this period, HHS is nearly fully staffed, declining slightly in the waning months; long vacancies in Commerce yield rates that are discernibly lower. The George W. Bush administration continues the pattern: substantial turnover and vacancies in the first months of the administration, followed by a period of near full staffing, and small decreases in appointee occupancy in the second administration. Again, the Clinton HHS exhibits greater continuity than does the George W. Bush administration HHS, while the George W. Bush administration Commerce Department exhibits less turnover and fewer vacancies than does the Commerce Department under Clinton.

The patterns of appointee turnover, tenure, and position vacancies depicted here point to potentially fruitful research examining the linkages between appointee continuity and organizational performance. While they present empirical challenges, we believe these questions offer an important direction for research on presidential administrative leadership. Continuity arguably forms a vital part of administrative competence and capacity, and research on executive succession across settings suggests that the turnover of administrators plays an important role in organizational performance. Though our effort here is limited, the following section concludes with a handful of thoughts on the implications of our analysis for research on and the practice of presidential leadership.

Conclusion

Three decades ago, Hugh Heclo observed the enduring tensions of appointee-careerist relations set against two developments in the institutions governing the presidential appointment power. The dual forces of politicization and bureaucratization continue to reshape the contours of the institutional presidency. This article has described one broad set of consequences of these developments based on indicators of appointee continuity--tenure, turnover, and position vacancies. The patterns of appointee continuity described here bring to light a potentially rich source of data through which to further develop the study of the institutional presidency.

Continuity is presumably a function of a variety of factors: bargaining among presidential, congressional, interest group, and agency actors for policy influence; the construction of agency competence and the social institutions structuring issue networks; and decisions by appointees as agents responding to material and professional incentives (Golden 2000; Wood and Marchbanks 2007). Looking at particular positions brings to light the variety of factors contributing to appointee continuity. A high degree of policy conflict in an agency may lead appointees to leave their posts more quickly, or high-conflict agencies may attract ideologues who are intent on serving out their full term in order to advance a cause. Agencies with highly technical positions might attract appointees who serve longer because their jobs require a great deal of expertise gained over time. Some agencies might have particularly harmonious or particularly dysfunctional cultures that either attract or repel appointees (Weeks 2004). Discerning a relationship--contingent on these factors--between appointee continuity and agency performance presents significant unresolved empirical challenges. However, the research of Lewis (2007, 2008), Boyne et al. (2008), and others suggests the substantial opportunity to shed new light on the role of the appointments power in agency politics and performance.

Among the empirical challenges to research in this area, appointee vacancies are often obscured by the absence of systematic measurement. Positions and titles frequently shift, complicating the measurement of vacancies over time (GAO 1994). Moreover, despite mandated agency reporting under the Vacancies Reform Act of 1998, reliable and useful information on appointee vacancies is not systematically available from government sources. (32) Like tenure, the impact of vacancies on organizational performance no doubt varies. From the perspective of some actors, vacancies are not always undesirable. Agency staff, congressional committees, and interest groups may prefer the stability and experience of an acting official drawn from among long-serving agency administrators. The acting official may have an established reputation for good-faith relations, perhaps even a sense of partnership. The absence of a presidential appointee may enhance the influence of other parties conducting agency business day to day. However, substantial anecdotal evidence suggests that vacancies--in general--tend to erode the capacity of agencies to plan and adapt.

A broad consensus holds that the rules governing appointee succession impose costs on potential appointees that increase vacancies and shorten service. Over the past 20 years, a number of commissions and task forces have issued reports criticizing the length of the appointment process, the level of ethical and political scrutiny of potential appointees, and the complexity and quantity of the paperwork that appointees are required to submit (U.S. Senate 2001). Academic studies have associated these costs with longer vacancies and difficulty in attracting well-qualified candidates to political executive positions (Presidential Appointee Initiative 2001). These reports have called on Congress and the president to streamline the appointment process and reduce the number of PAS positions. (33)

Debate will undoubtedly continue, and observers will justifiably maintain a healthy skepticism about the prospects for comprehensive reform. One matter that does not require anything approaching comprehensive reform, but would enhance the transparency of the appointment process, is improving the reliability and usefulness of the information mandated under the Vacancies Reform Act of 1998. The simple truth behind this failure of government transparency is that none of the actors with a direct stake in appointee politics--committees, agencies, the White House, interest groups--sees an obvious benefit from greater transparency. Recent innovations in the analysis and transfer of information, enabled particularly by Internet communications, illustrate how improving the clarity of information might enhance transparency and political deliberation. (34) While no reform will erase the institutional politics complicating appointee succession, at least we might agree that careful description and analysis holds the potential to promote more enlightened public debate regarding what remains a "quiet" but important problem in American democratic governance. (35)

AUTHORS' NOTE: We wish to thank Henry Hogue for his significant contribution to this project; the Government Accountability Office, the Office of Personnel Management, and Nolan McCarty for generously making their data available; and Tara Bryan, Michael Keeney, and Alejandra Palma for their able research assistance.

References

9/11 Commission. 2004. Final Report of the National Commission on Terrorist Attacks Upon the United States. New York: W.W. Norton.

Aberbach, Joel D., and Bert A. Rockman. 2000. In the Web of Politics: Three Decades of the U.S. Federal Executive. Washington, DC: Brookings Institution Press.

Ban, Carolyn, and Patricia W. Ingraham. 1990. "Short-Timers: Political Appointee Mobility and Its Impact on Political-Career Relations in the Reagan Administration." Administration & Society 22 (May): 106-24.

Barringer, Felicity. 1990. "Census Chief Believes in Numbers and Optimism." New York Times, April 19.

Bawn, Kathleen. 1995. "Political Control versus Expertise: Congressional Choices about Administrative Procedures." American Political Science Review 89 (March): 62-73.

Bergh, Donald D. 2001. "Executive Retention and Acquisition Outcomes: A Test of Opposing Views on the Influence of Organizational Tenure." Journal of Management 27 (October): 603-22.

Bertelli, Anthony, and Sven Feldman. 2007. "Strategic Appointments." Journal of Public Administration Research and Theory 17 (January): 19-38.

Bertelli, Anthony, and Christian Grose. 2007. "Agreeable Administrators? The Public Positions of Cabinet Secretaries and Presidents." Presidential Studies Quarterly 37 (June): 228-47.

Bolton, John R. 2008. "Our Politicized Intelligence Services." Wall Street Journal, February 5.

Boyne, George A., Oliver James, Peter John, and Nicolai Petrovsky. 2008. "Does Chief Executive Succession Affect Public Service Performance?" Paper prepared for the 13th Annual Research Conference of the Association for Public Policy Analysis and Management, Los Angeles, November 6-8.

Brauer, Charles. 1987. "Tenure, Turnover and Postgovernment Employment Trends of Presidential Appointees." In The In-and-Outers: Presidential Appointees and Transient Government in Washington, ed. G. Calvin Mackenzie. Baltimore: Johns Hopkins University Press, 174-94.

Chang, Kelly, David Lewis, and Nolan McCarty. 2001. "The Tenure of Political Appointees." Paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago, April 19-22.

Clinton, Joshua D., and David A. Lewis. 2008. "Expert Opinion, Agency Characteristics, and Agency Preferences." Political Analysis 16 (Winter): 3-20.

Derthick, Martha. 1979. Policymaking for Social Security. Washington, DC: Brookings Institution.

--. 1990. Agency under Stress: The Social Security Administration in American Government. Washington, DC: Brookings Institution.

Dull, Matthew. 2006. "Why PART? The Institutional Politics of Presidential Budget Reform." Journal of Public Administration Research and Theory 16 (2): 187-215.

Dull, Matthew. 2009. "Results-Model Reform Leadership: Questions of Credible Commitment." Journal of Public Administration Research and Theory 19 (April): 255-84.

Edwards, George C., III. 2001. "Why Not the Best? The Loyalty-Competence Tradeoff in Presidential Appointments." In Innocent until Nominated: The Breakdown of the Presidential Appointments Process, ed. G. Calvin Mackenzie. Washington, DC: Brookings Institution Press, 311-36.

Eitzen, D. Stanley, and Norman R. Yetman. 1972. "Managerial Change, Longevity, and Organizational Effectiveness." Administrative Science Quarterly 17: 110-16.

Epstein, David, and Sharyn O'Halloran. 1999. Delegating Powers: A Transaction Cost Politics Approach to Policy Making under Separate Powers. New York: Cambridge University Press.

Finkelstein, Sidney, and Donald C. Hambrick. 1990. "Top Management Team Tenure and Organizational Outcomes: The Moderating Role of Managerial Discretion." Administrative Science Quarterly 35 (September): 484-503.

Fisher, Louis. 1985. Constitutional Conflicts Between Congress and the President. Princeton, NJ: Princeton University Press.

Freidson, Eliot. 2001. Professionalism: The Third Logic. Chicago: University of Chicago Press.

Friedman, Stewart D., and Paul Olk. 1995. "Four Ways to Choose a CEO: Crown Heir, Horse Race, Coup d'etat and Comprehensive Search." Human Resource Management Journal 34 (Spring): 141-64.

Friedman, Stewart D., and H. Singh. 1989. "CEO Succession and Stockholder Reaction: The Influence of Organizational Context and Event Context." Academy of Management Journal 32: 718-44.

Gilmour, John B., and David E. Lewis. 2006. "Does Performance Budgeting Work? An Examination of the Office of Management and Budget's PART Scores." Public Administration Review 66 (September/October): 742-52.

Golden, Marissa Martino. 2000. What Motivates Bureaucrats? Politics and Administration During the Reagan Years. New York: Columbia University Press.

Goodsell, Charles T. 2006. "Relations Between Political Appointees and Career Officials: Principal-Agent or Moral Equals?" Presidential Studies Quarterly 36 (June): 323-26.

Grusky, Oscar. 1963. "Managerial Succession and Organizational Effectiveness." American Journal of Sociology 69: 21-31.

Hambrick, Donald C., and G. D. Fukutomi. 1991. "The Seasons of a CEO's Tenure." Academy of Management Review 16 (October): 719-42.

Hambrick, Donald, C., Marfa A. Geletkanycz, and J. W. Frederickson. 1993. "Top Executive Commitment to the Status Quo: Some Tests of Its Determinants." Strategic Management Journal 15: 401-18.

Harris, Shane. 2007. "Homeland Security: The Coming Storm." National Journal, June 2.

Heclo, Hugh. 1977. A Government of Strangers: Executive Politics in Washington. Washington, DC: Brookings Institution.

Hogue, Henry B. 2003. Presidential Appointments to Full-Time Positions in Independent and Other Agencies During the 107th Congress. CRS Report no. RL31435. Washington, DC: Congressional Research Service.

--. 2008. Temporarily Filling Presidentially Appointed, Senate-Confirmed Positions. CRS Report no. RS21412. Washington, DC: Congressional Research Service.

Hood, Christopher, Oliver James, B. Guy Peters, and Colin Scott. 2004. Controlling Modern Government: Variety, Commonality, and Change. Cheltenham, UK: Edward Elgar.

Institute of Medicine. 2007. The Future of Drug Safety: Promoting and Protecting the Health of the Public. Washington, DC: National Academies Press.

Joyce, Philip G. 1990. "An Analysis of the Factors Affecting the Employment Tenure of Federal Political Executives." Administration & Society 22 (May): 127-45.

Katz, Ralph. 1982. "The Effects of Group Longevity on Project Communication and Performance." Administrative Science Quarterly 27 (March): 81-104.

Kesner, Idalene E, and Terrence C. Sebora. 1994. "Executive Succession: Past, Present and Future." Journal of Management 20 (Summer): 327-72.

Khademian, Anne M. 2002. Working with Culture: How the Job Gets Done in Public Programs. Washington, DC: CQ Press.

King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton, NJ: Princeton University Press.

Koitz, David. 1989. What Would It Mean to Make the Social Security Administration an Independent Agency? Washington, DC: Congressional Research Service. May 12:89-309 EPW.

--. 1994. Social Security: Making It an Independent Agency. Washington, DC: Congressional Research Service. August 16: IB-86-12C.

Krishnan, Hema A., Alex Miller, and William Q. Judge. 1997. "Diversification and Top Management Team Complementarity: Is Performance Improved by Merging Similar or Dissimilar Teams?" Strategic Management Journal 18 (May): 361-74.

Lane, Larry. 1996. "Public Administration and the Problem of the Presidency." In Refounding Democratic Public Administration, eds. Gary Wamsley and James E Wolf. Thousand Oaks, CA: Sage, 225-59.

Lewis, David A. 2007. "Testing Pendleton's Premise: Do Political Appointees Make Worse Bureaucrats?" Journal of Politics 69 (November): 1073-88.

--. 2008. The Politics of Presidential Appointments: Political Control and Bureaucratic Performance. Princeton, NJ: Princeton University Press.

Light, Paul C. 1995. Thickening Government: Federal Hierarchy and the Diffusion of Accountability. Washington, DC: Brookings Institution.

--. 1998. The Tides of Reform: Making Government Work, 1945-1995. New Haven, CT: Yale University Press.

Mackenzie, G. Calvin, ed. 2001. Innocent until Nominated: The Breakdown of the Presidential Appointments Process. Washington, DC: Brookings Institution Press.

Mackenzie, G. Calvin, and Paul C. Light. 1987. Presidential Appointees, 1964-1984. Ann Arbor, MI: Inter-University Consortium for Political and Social Research.

Mackenzie, G. Calvin, and Robert Shogan, eds. 1996. Obstacle Course: The Report of the Twentieth Century Fund Task Force on the Presidential Appointment Process. New York: Twentieth Century Fund Press.

MacMahon Arthur W., and John D. Millett. 1939. Federal Administrators: A Biographical Approach to the Problem of Departmental Management. New York: Columbia University Press.

Maranto, Robert. 2005. Beyond a Government of Strangers: How Career Executives and Political Appointees Can Turn Conflict to Cooperation. Lanham, MD: Lexington Books.

Marcum, Cheryl Y., Lauren R. Sager Weinstein, Susan D. Hosek, and Harry J. Thie. 2001. Department of Defense Political Appointments: Positions and Process. Santa Monica, CA: RAND, National Defense Research Institute. http://www.rand.org/publications/MR/MR1253/[accessed May 14, 2009].

McCarty, Nolan, and Rose Razaghian. 1997. "Advice and Consent: Senate Response to Executive Branch Nominations, 1885-1996." American Journal of Political Science 43: 1122-43.

Miller, Danny. 1991. "Stale in the Saddle: CEO Tenure and the Match between Organization and Environment." Management Science 37 (January): 34-52.

Miller, Gary J. 2000. "Above Politics: Credible Commitment and Efficiency in the Design of Public Agencies." Journal of Public Administration Research and Theory 10 (April): 289-328.

--. 2005. "The Political Evolution of Principal-Agent Models." Annual Review of Political Science 8: 203-25.

Moe, Terry M. 1985. "The Politicized Presidency." In The New Direction in American Politics, eds. John E. Chubb and Paul E. Peterson. Washington, DC: Brookings Institution Press, 235-71.

--. 1990. "Political Institutions: The Neglected Side of the Story." Journal of Law, Economics and Organization 6:213-53.

--. 1993. "Presidents, Institutions and Theory." In Researching the Presidency: Vital Questions, New Approaches, eds. George C. Edwards III, John H. Kessel, and Bert A. Rockman. Pittsburgh: University of Pittsburgh Press, 337-85.

Nathan, Richard P. 1983. The Administrative Presidency. Hoboken, NJ: Wiley.

National Commission on the Public Service. 1989. Leadership for America: Rebuilding the Public Service. Washington, DC: National Commission on the Public Service.

Nixon, David C. 2004. "Separation of Powers and Appointee Ideology." Journal of Law, Economics and Organization 20: 438-57.

North, Douglass C. 1994. "Institutions and Credible Commitment." Working paper, Washington University in St. Louis.

Peters, B. Guy, and Jon Pierre, eds. 2004. Politicization of the Civil Service in Comparative Perspective: The Quest for Control. London: Routledge.

Pfiffner, James P. 1987. "Political Appointees and Career Executives: The Democracy-Bureaucracy Nexus." Public Administration Review 47 (January/February): 57-65.

Presidential Appointee Initiative. 2001. To Form a Government: A Bipartisan Plan to Improve the Presidential Appointments Process. Washington, DC: Brookings Institution.

Radin, Beryl A. 2006. Challenging the Performance Movement: Accountability, Complexity and Democratic Values. Washington, DC: Georgetown University Press.

Roberts, Patrick. 2006. "FEMA and the Prospects for Reputation-Based Autonomy." Studies in American Political Development 20: 57-87.

Shen, Wei, and Albert A. Cannella, Jr. 2002. "Political Dynamics within Top Management and Their Impacts on CEO Dismissal Followed by Inside Succession." Academy of Management Journal 45: 1195 -208.

Stanley, David T., Dean E. Mann, and James W. Doig. 1967. Men Who Govern. Washington, DC: Brookings Institution.

Stayn, Joshua L. 2001. "Vacant Reform: Why the Vacancies Reform Act of 1998 Is Unconstitutional." Duke Law Journal 50: 1511-29.

U.S. Congress. House Committee on Homeland Security. 2007. "Critical Leadership Vacancies Impede the U.S. Department of Homeland Security." http://homeland.house.gov/SiteDocuments/ 20070709112923-81091.pdf [accessed May 14, 2009].

U.S. Congress. Senate. Committee on Governmental Affairs. 2001. Presidential Appointment Process: Reports of Commissions That Studied the Staffing of Presidential Administrations: A Summary of Their Conclusions and Recommendations for Reform. 107th Cong., 1st sess.

U.S. General Accounting Office (GAO). 1994. Political Appointees: Turnover Rates in Executive Schedule Positions Requiring Senate Confirmation. GAO/GGD-94-115FS. Washington, DC: Government Printing Office.

U.S. Government Accountability Office (GAO). 2003. Federal Vacancies Reform Act: Key Elements for Agency Procedures for Complying with the Act. GAO-03-806. Washington, DC: Government Printing Office.

Vobejda, Barbara. 1998. "Embattled Census Director Quits; Riche Leaves in Midst of Sampling Dispute." Washington Post, January 13.

Waldo, Dwight. 1948. The Administrative State: A Study of the Political Theory of American Public Administration. New Brunswick, NJ: Transaction, 2006.

Weber, Max. 1958. From Max Weber: Essays in Sociology. Eds. and trans. H. H. Gerth and C. Wright Mills. New York: Oxford University Press.

Weeks, John. 2004. Unpopular Culture: The Ritual of Complaint in a British Bank. Chicago: University of Chicago Press.

Williams, Walter. 1990. Mismanaging America: The Rise of the Anti-Analytic Presidency. Lawrence: University Press of Kansas.

Williamson, Elizabeth. 2007. "OMB Offers an Easy Way to Follow the Money." Washington Post, December 13.

Wood, B. Dan, and Miner P. Marchbanks III. 2007. "What Determines How Long Political Appointees Serve?" Journal of Public Administration Research and Theory 18 (July): 375-96.

Zegart, Amy B. 2007. Spying Blind: The CIA, the FBI, and the Origins of 9/11. Princeton, NJ: Princeton University Press.

MATTHEW DULL and PATRICK S. ROBERTS

Virginia Tech

(1.) Henry Hogue observes that the George W. Bush administration used the Vacancies Act as one of multiple authorities in succession allowing the president to retain "acting" officials essentially without limitation. Hogue notes, "By using more than one authority, the [Bush] Administration has been able to place unconfirmed individuals in these positions for longer periods of time than would have been possible if only one authority had been used. Since 2003, U.S. Attorney vacancies have been filled temporarily through appointment by the Attorney General and also under the provisions of the Vacancies Act. As of July 23, 2007, the most recent date for which comprehensive data are available, at least 21 U.S. Attorney vacancies had been filled through sequential appointments under first one of these authorities and then the other" (2008, 5).

(2.) Continuity, as it we define it here, includes appointee turnover, measuring what is elsewhere termed "rotation" (Lewis 2008), "transience" (Heclo 1977), and "succession" (Boyne et al. 2008), and position vacancies, defined as positions unoccupied by a Senate-confirmed appointee. Heclo observes that "[t]he single most obvious characteristic of Washington's political appointees is their transience" (1977, 103). Yet, there is also the problem of vacancies particular to PAS positions, subject to action by both the president and the Senate, as well as the famously byzantine paperwork related to mandated background investigations (Mackenzie 2001). Analyzing vacancies is a thorny empirical challenge, as it requires the systematic review of positions over time. However, we contend they represent a distinct but closely related phenomenon, which also contributes to the continuity of agency administration.

(3.) Put another way, our objective is to contribute to a vital, growing research agenda through descriptive rather than causal inference (King, Keohane, and Verba 1994).

(4.) In Beyond a Government of Strangers, Robert Maranto (2005) offers a spirited defense of politicization, arguing that the expanding ranks of appointees promote bureaucratic responsiveness and accountability.

(5.) The basic distinction here is between formal institutional arrangements and institutions as "regularized behavior patterns." It is worth noting that in his influential analysis of politicization, Terry Moe employs the latter definition (1985, 236 n. 1). Lewis (2008) develops a case study examining the most celebrated recent example of the control-competence trade-off in a federal agency administration--the role of appointees to the Federal Emergency Management Agency (FEMA) and the perceived shortcomings of FEMA's response to Hurricanes Katrina and Rita. By contrast, Lewis offers Clinton appointee James Lee Witt, who served as FEMA director from 1993 to 2001, reducing by one-third the number of FEMA appointees (a move reversed following the Department of Homeland Security reorganization). As Lewis's treatment suggests, this reduction in formal politicization was embedded in a broader strategy of cultivating organizational reputation, culture, and expertise (Khademian 2002; Roberts 2006).

(6.) "Quality, like beauty, is in the eye of the beholder. Evaluations of quality depend heavily on what the evaluator wants from the bureaucracy," observes presidency scholar George Edwards (2001).

(7.) Indeed, some have argued that the Bush administration employed incompetence and incapacity to advance the president's agenda--if an agency lacks capacity, it is less likely to oppose the White House's position. John Bolton (2008), who served in several Republican administrations, wanted Congress and the White House to pressure the Central Intelligence Agency (CIA) to refrain from publicly releasing conclusions about the likelihood that other nations will develop nuclear weapons. The CIA's analytic capacity, he argued, limited the president's options in negotiating with other countries. Walter Williams (1990) argues that staffing cuts eroding agency capacity for analysis during the Reagan administration represented a similar strategy.

(8.) Political power, Heclo observes, is "one of the clearest but also one of the most limited assets for political leadership in bureaucracy" (1977, 195).

(9.) Douglass North observes that leadership duration can be a critical factor in the maintenance of credible leadership commitments governing market economies. He notes, "Time is crucial ... since uncertainty about the ruler's behavior can only be mitigated by the ruler establishing a reputation and by learning on the part of constituents" (1994, 8-9). Miller (2000) and Dull (2009) develop extensions of this logic to the politics of agency leadership. Heclo's analysis is similarly grounded in the comparative study of political economy and government institutions. Hood et al. (2004) and Peters and Pierre (2004) offer useful perspectives on current scholarship.

(10.) We examine appointments to full-time, civilian PAS positions in all departments, single-headed independent agencies, and Executive Office of the President organizations. Data were collected through formal requests to the Office of Personnel Management and Government Accountability Office and were verified against published sources, including the Senate nominations database available through Thomas.gov, Congressional Research Service reports, contemporary news coverage accessed by LexisNexis, and online resources. Several PAS positions are omitted, including U.S. attorney and U.S. marshal positions in the Department of Justice; Foreign Service and diplomatic positions in the Department of State; officer corps positions in the civilian uniformed services of the National Oceanic and Atmospheric Administration in the Department of Commerce, and of the Public Health Service in the Department of Health and Human Services; and the officer corps in the military services. Excluded from the analysis are multilateral banking organizations, legislative branch agencies, the Appalachian Regional Commission, and the Court Services and Offender Supervision Agency for the District of Columbia (Hogue 2003, 37). Also omitted from the calculation of appointee tenure and turnover are appointments confirmed prior to January 20, 1989, which were held over into President George H. W. Bush's term. Recess appointments are included.

(11.) We do not count service in an "acting" position because we are interested in whether a position is filled with someone who holds the full authority of the office. When a position is vacant, other appointees or careerists may be given some of the duties of the vacant position temporarily, but these substitutes do not exercise these duties with the full authority of the position. Several PAS positions are omitted, including U.S. attorney and U.S. marshal positions in the Department of Justice; Foreign Service and diplomatic positions in the Department of State; officer corps positions in the civilian uniformed services of the National Oceanic and Atmospheric Administration in the Department of Commerce, and of the Public Health Service in the Department of Health and Human Services; and the officer corps in the military services. Excluded from the analysis are multilateral banking organizations, legislative branch agencies, the Appalachian Regional Commission, and the Court Services and Offender Supervision Agency for the District of Columbia (Hogue 2003, 37). Also omitted from the calculation of appointee tenure and turnover are appointments confirmed prior to January 20, 1989, which were held over into President George H. W. Bush's term. Recess appointments are included.

(12.) We excluded from the analysis about 25 appointments of fewer than 60 days. These very brief appointments are often symbolic in nature, perhaps recognizing the service of a senior deputy or an administrator who has served for an extended period in an acting capacity. Secretary of State Lawrence Eagleburger, for example, was a career diplomat and senior member of President George H. W. Bush's foreign policy team. Secretary Eagleburger formally served less than six weeks as secretary of state under a recess appointment in the waning weeks of the Bush presidency.

(13.) The findings here are roughly consistent with Maranto's (2005, 116) analysis of a sample of Reagan administration appointees. Maranto correctly observes that data on position turnover overstates the disruption by not accounting for appointees transferring within the agency.

(14.) During this period, the number of cabinet secretaries remained essentially constant, while the numbers of undersecretaries and assistant secretaries more than doubled (Light 1995, Table A1-A2).

(15.) During the period examined here, for example, John Layton served as Department of Energy IG from 1986 to 1998, John Martin served as Environmental Protection Agency IG from 1983 to 1997, and Patrick McFarland served as Office of Personnel Management IG from 1990 to 2007.

(16.) Though presidents may remove IGs as they can other agency appointees, if they are removed, the president "shall communicate the reasons for any such removal to both Houses of Congress" (5 U.S.C. Appx. [section] 3[b]).

(17.) We code multiple reappointments to a position by a single appointee as a one period of service.

(18.) Our discussion of government "thickening," drawn from Light's (1998) analysis, refers to the growth in both the number of people in government and the layers of hierarchy that separate them. HHS and Commerce are both rated as among the "thickest" agencies by Light. His detailed examination of administrative thickening highlights the varied sources and consequences of the expanding ranks of appointees. His discussion of the interest group, congressional, and bureaucratic pressures that promoted thickening in the Commerce Department through the debate leading to the establishment of the undersecretary for travel and tourism and in the Department of Health and Human Services through the creation of the assistant secretary for family support illustrate the broader patterns at work (Light 1995, 142-46).

(19.) George H. W. Bush appointee Louis Sullivan was a prominent physician who helped establish the Morehouse School of Medicine. Donna Shalala, secretary of HHS from 1993 to 2001, was a prominent university administrator at the University of Wisconsin and the University of Miami before and after her service. Both George W. Bush administration secretaries were former governors Tommy Thompson, former governor of Wisconsin, and Mike Leavitt, former governor of Utah, who previously served as administrator of the Environmental Protection Agency.

(20.) Two secretaries of commerce served under George H. W. Bush. Robert Mosbacher served from January 1989 to January 1992, when he resigned to become chairman of the president's reelection campaign. He was replaced by Secretary Barbara Franklin, a prominent executive and active Republican, who served during the administration's final year. Clinton's first secretary of commerce, former Democratic National Committee chair Ron Brown, served three years before his untimely death in 1996 in an airplane crash. Facing a Republican Congress, Clinton made a recess appointment of longtime advisor and U.S. Trade Representative Mickey Kantor as the next secretary. Less than a year later, at the beginning of Clinton's second term, Kantor was replaced by Secretary Bill Daley. The son and brother of powerful Democratic Chicago mayors, Daley took office promising reduced political influence. He served from 1997 to 2000, resigning to become chairman for the presidential election campaign of Vice President Al Gore. He was replaced by former Democratic congressman Norman Mineta, who served the last six months of the Clinton administration. George W. Bush appointed two secretaries of commerce: Donald Evans, a longtime friend and chairman for the Bush-Cheney 2000 campaign, and former Kellogg CEO Carlos Gutierrez.

(21.) Freidson's (2001) logic of professionalism, rooted in credible claims to expert autonomy, offers a useful framework for thinking about this point.

(22.) Reflecting on the controversy over the use of sampling by the Census Bureau, one knowledgeable observe noted, "The difficulty in that position is being stuck between the scholarly community, advising her how to take a good census, and the political community, reacting to political forces" (Vobejda 1998).

(23.) Derthick notes, "The practical result of the presidential appointment process was not so much to supply leadership to the SSA as to deprive it of any. Assuming that an appointee is new for a year, the SSA had a new leader or no formally approved leaders for well over half the time in the twelve-year period of this analysis" (1990, 122-23).

(24.) Notably, the SSA administrator is also protected from presidential removal except for cause, and this strengthens that insulation. Likewise, the Institute of Medicine's 2007 report on reform of the Federal Drug Administration recommends a fixed six-year term for that agency's commissioner, citing similar concerns about leadership continuity and expert capacity.

(25.) Senate Resolution 338. Congressional Record, vol. 106, pt. 12, 86 Cong. 2 sess. (1960), 15705 (quoted in Stanley Mann and Doig 1967, 55).

(26.) A recent Department of Defense-commissioned RAND analysis of PAS tenure in that department from 1947 to 1999 finds increasing numbers of PAS positions, variations in length of tenure, and steady increases in vacancy rates over fifty years, from approximately 2% during the Truman administration to more than 20% during Clinton's first term (Marcum et al. 2001).

(27.) During 2007, the Department of Homeland Security (DHS) developed a transition plan to ensure that career managers are capable of leading the department until appointees from the next administration are brought up to speed. The leader of the transition planning effort, Deputy Secretary Michael Jackson, resigned in 2007 after it became clear Michael Chertoff would remain DHS secretary and that Jackson would not assume the department's top job (Harris 2007). Vacancies in top DHS leadership positions are the focus of a recent analysis by the majority staff of the U.S. House Committee on Homeland Security (2007). We wish to thank Justin S. Vaughn and Jose D. Villalobos for bringing this report to our attention.

(28.) Lewis (2007) also observes a nonlinear pattern. The marginal impact of tenure on performance is greatest at short tenures and subsequently levels off.

(29.) Amy Zegart (2007) offers a similar explanation for the intelligence community's failure to adapt to the terrorist threat.

(30.) For example, are the successors "insiders" or "outsiders"? Are they drawn from within the agency, or are they allies of a retiring executive, outsiders, or even competitors of an executive who has been forced out? (Friedman and Olk 1995; Friedman and Singh 1989). Shen and Cannella (2002) find, for example, that where successors are former competitors of the departing executive, high turnover among the organization's top leadership positively influences organizational performance, but when the new CEO is an outsider, subsequent turnover has a strong negative effect on performance.

(31.) Only positions in existence for a full presidential administration are included (Commerce, N = 24; HHS, N = 18). Figures are based on counts of confirmed appointees on the first day of each month from February 1, 1989, to January 1, 2009.

(32.) Prompted by conflict over the Clinton administration's use of "acting" officials to avoid Senate confirmation, the Vacancies Reform Act of 1998 placed limits on acting terms and mandated that all agencies report appointee vacancies to Congress and the General Accounting Office (GAO, now Government Accountability Office). The GAO developed a Web site designed to make these data public. However, agency compliance has been uneven (GAO 2003), and the GAO interface is difficult to navigate. See the GAO vacancies Web site at http://www.gao.gov/legal/vacancies.html.

(33.) Reports have recommended identifying nominees early in the presidential transition process; greater control by cabinet heads of selecting subordinates; better guidance for nominees during the nomination and confirmation process; fewer requirements for financial disclosure, conflict-of-interest disclosure, and post-employment restrictions; and streamlining and standardizing committee requirements for financial disclosure. For example, in 1996, the Twentieth Century Fund Task Force on Presidential Appointments published a report recommending two reforms. First, the report recommended reducing the number of PAS positions by a third. Second, it recommended that "[a]ppointments to most advisory commissions and routine promotions of military officers, foreign service officers, [and] public health services officers, except those at the very highest ranks ... cease to be presidential appointments and cease to require Senate confirmation" (Mackenzie and Shogan 1996, 9).

(34.) One notable example of the capacity of work in this area to effect change in government has been the construction of the OMB's USA Spending program (http://www.usaspending.gov/), mandated under the Federal Funding, Accountability, and Transparency Act using software developed by the group OMB Watch for its own Web site (http://www.fedspending.org). The role of OMB staffer and management reform advocate Robert Shea helps to illustrate the unlikely alliances that often characterize successful reform initiatives (Williamson 2007).

(35.) National Commission on the Public Service (1989).

Matthew Dull is an assistant professor in the Center for Public Administration and Policy at Virginia Tech. His research interests include public administration, executive politics, and the politics of performance.

Patrick S. Roberts is an assistant professor in the Center for Public Administration and Policy at Virginia Tech. His research interests include the history of how the federal government prepares for disaster. His work has appeared in Studies in American Political Development.

FIGURE 2. Appointee Tenure: Years of Service During Three Periods.

                        <1     <2    >4

1798-1983               15%    37%   25%

N=539
MacMahon and
Millett (1939)

1933-1965               20%    51%   15%

N=589
Stanley, Mann, and
Doig(1967)

1989-2009               13%    38%   22%

N=2,278

Note: Table made from bar graph.