The Political Public Service

Let me move first to one of my favorite topics: political appointees. Although there are several components of the political public service, those of the executive branch—the political appointees—and those of the Congress are the most important here. The complex nature of government basically mandates that these appointees meet dimensions of substantive accountability—that they be expert in the programs they will theoretically lead and analyze and will also clearly link to the substantive policy agendas of the president and the Congress. In this latter respect, they also link to representational accountability, representing various ideas and policy positions as defined by elections.

The appointive political service is, far and away, the smallest of the public service components—roughly 1% of total federal employment by most counts.¹

Those with which we are most concerned are the top political executives—the Senate confirmees—and the various levels of deputies and assistants created by Schedule C, Senior Executive Service, and other appointments. This group has grown substantially in the 60-plus years since the Brownlow Commission declared that the president should be the “center of energy… in the federal government” and political appointees his critical arm (President's Committee 1937, 114). The largest “spurt” has occurred in Schedule C appointees—the policy confidentials— and the largest growth in that group occurred relatively recently, in the years of the Carter and Reagan administrations (Ingraham 1995). Pfiffner (1996), in fact, reported an 86% increase in total numbers of Schedule C's in the years from 1979 to 1992. The numbers have not declined since then, but have continued to grow—more slowly, but into many new positions that did not exist 20 years ago and that have also multiplied rapidly.

The total number of policy critical political appointees is now about 3,000, but varies from point to point in a presidential administration as positions are filled and vacated. The second Volcker Commission estimated that President George W. Bush had nearly 3,400 positions to fill and that it took over eight months from Inauguration to initial appointment for the vast majority. This time has nearly doubled since the mid 1980s (National Commission 2003, 19). Thus, there is a “down-time” in the executive branch's political leadership that has grown steadily since 1990. The scope of the “vacancy at the top” problem, during which career civil servants assume “acting” responsibility, but no real policy authority, should not be underestimated. When Sean O'Keefe of NASA assumed office nearly two years into the Bush presidency, over 70% of the top leadership and management positions in the agency were filled by “acting” personnel (Partnership 2003). NASA may be a bit extreme, but it is not an exception in this regard.

Political appointees as a group have also been characterized as “short-timers”; that is, they serve relatively short terms in one position (Ban and Ingraham 1990). The decade of the 1980s was marked by decreasing time in position for political appointees at all ranks. Tenure was shortest, however, for the fastest growing group of appointees—the Schedule Cs. The pattern stabilized somewhat in the eight-year Clinton term, but average tenure in position continues to be less than two years.

The relatively constant state of flux is problematic in organizational terms because it creates a consistent churn in the organization. But in leadership and accountability terms, it presents a significant dilemma. Political appointees who are short-term in tenure and perspective cannot adequately—perhaps even accurately—reflect a substantive policy agenda. Nor, whatever the expertise they bring into government, can they achieve the organizational expertise necessary to achieve real policy direction and change.

This is further demonstrated by examining what politicals do in the time they do spend in their positions. Are they, as Brownlow suggested, the active executive arm of the president? Yes and no. Top political executives do serve as the chief executives of the agency and their top deputies generally have substantial policy and management responsibilities. Surveys of political appointees indicate, however, that an increasing percentage of their time is spent not in managing programs and policies, but in managing contacts with congressional staff and with reporters. This also makes them notably different from career officials who, at least officially, spend more time in management and analytical tasks. But, as Maranto notes, politicians expect to work with other politicians (and not with the career bureaucracy), because as one observed, they “speak the same language” (Maranto 2001). The significant point to be taken here is that an increased number of political appointees do exactly what we would expect appointees to do: political work. From an effective government perspective, there is some evidence that increased numbers of political appointees do not improve the performance of federal agencies in any way, but have the opposite effect (Lewis and Gilmour 2003; National Academy 2004). This clearly does not meet any accountability standard.

Congressional staff provides another dimension of political accountability. This accountability mechanism is one of oversight and entry into public organizations through casework, although it sometimes has been described as micro-management (Kettl 1988). The numbers here are substantially larger. In 2002, the formal accountability arm of the Congress—the Government Accountability Office—had about 3,200 employees. The Congressional Budget Office had slightly over 200. Congress had a committee staff of 2,000 people, and a personal staff of over 11,000. The extent to which the growth of this group is related to the earlier observation that top political appointees spend greater amounts of time in congressional relations is clear, but the point is emphasized in a recent statement by Department of Homeland Security Secretary Tom Ridge, who observed that his agency reported to 88 different committees and subcommittees.

In one of the rare analyses of accountability perceptions and values among congressional staff, Barbara Romzek (2000) found that accountability mechanisms were perceived almost solely in terms of individual member's interests. The Framers, in their wildest dreams, could not have imagined a secretary of one of the largest federal agencies trying to satisfy the information and political demands of 88 committees and their many members—to say nothing of balancing those demands against those of the president and his other appointees. Ironically, by splintering Congressional oversight so profoundly, true accountability suffers.

The Career Public Service

Then we move to the career service. This part of the service is an odd byproduct of American constitutionalism: a public service that as Waldo noted is “proscribed from politics” but operates “in its interstices” (Waldo 1948). In 2004, the total size of this workforce was 2,700,000, or just under two million if the Postal Service is excluded. Nearly 700,000 of these employees are in the Department of Defense (Office of Personnel Management 2004).

Both in its composition and in the nature of its work, this group has changed over the past two decades. The change in composition is related, at least in part, to representational accountability: since the early 1960s, the public service has been viewed as an employer more favorable than the private sector to women and minorities. This “target of opportunity” provides one clear way for public organizations to reflect the composition of the society at large, to represent their interests, and to respond to their concerns (Rosenbloom and Krislov 1981). The link is somewhat troubling theoretically, however. The career service is, after all, “thrice removed” from electoral processes (Mosher 1968). In reality, the link is also far from perfect. In 2000, for example, women constituted about 22% of the membership of the top levels of federal agencies (an increase of 12% in a decade), but a much larger percentage on lower rungs of organizational ladders and in society at large. Persons of color and other minorities increased from 7% to about 13% at the top levels of federal agencies in the same period (Government Accountability Office 2001), but again, numbers are higher at lower levels of the organization and in the broader public. Federal organizations, broadly writ, are fairly representative, but their leadership is not.

Another area in which diversity is an issue is the overall age distribution of the federal career service. Several factors, including the reinvention and downsizing cuts of the 1990s, associated buy-outs, hiring freezes, and the long-term career nature of the federal service, have contributed to a steadily aging federal workforce. In 1991, the modal age for federal supervisors and managers was 44 years; in 2003 it was 54. In that same year, over 72% of federal managers were 45 and above (National Academy 2004).

Not surprisingly, this makes potential retirements a major issue for the federal service. Some agencies, such as NASA, project a retirement rate as high as 60% in the next five years in management and leadership ranks. Retirement rates are projected to be near or above 50% in such critical agencies as the Department of Homeland Security. Of course, all organizational populations age. But for the career federal service, the problem is especially troublesome for two reasons. In the 1990s, as government was being downsized and experienced managers and leaders left, a hiring freeze precluded refreshing the ranks. New hires are now coming in, but probably not fast enough. In this critical time for effective government, there is an experience gap in the middle of many federal agencies and a potential experience drain at the top (National Academy 2004).

This has occurred as the nature of the work changed significantly. Technology and the demand for scientific expertise catapulted the federal service into new terrain and new skill demands. Some, such as the Internal Revenue Service, struggled very publicly, but it was clearly not alone in facing new challenges. Consider that in 1950 more than half of classified civil service employees government wide were in lower level clerical jobs. By the year 2000, the comparable number was only 15%. Ten percent of the service in 2000 was in scientific and engineering positions. Other professional and technical positions—and the grades they occupy in the federal classification scheme—exploded (Office of Personnel Management 2002). The federal service lagged, however, in the ability to recruit, in the ability to meet or compete with private sector pay, and in the ability to retain expertise once it had been acquired. The Patent and Trademark Office, fundamental to both public and private sector ability to sustain innovation and change, found that its ability to retain new patent examiners lasted roughly for the full education and training period. After that, the expertise—now more valuable for its understanding of government procedures—moved to the private sector. This was widely true across government for experience and expertise ranging from Air Force pilots to NASA engineers.

This has serious implications. The career civil service is the part of the public service upon which the most attention to accountability has been lavished. Members of the career service, moreover, are the stewards of democracy, with the responsibility—in Aberbach, Putnam, and Rockman's lovely phrase, to “take the dreams of politicians and bring them gently to earth.” They have enormous power. Accountability in this component of the service is absolutely fundamental, but so is having the experience, the skills, and the capacity to do the job. Without that, meaningful accountability is impossible. The experience gap in the middle of many public organizations is threatening; the experience drain at the top is threatening as well.

For the career service, the greatest challenge to longer term accountability, however, may be that federal agencies—already challenged internally—are rapidly assuming responsibility for ever increasing numbers of contracts for tasks that are increasingly complex. Remaining the accountability center for an ever more “hollow” government is already problematic. The trend line points to it becoming more so.

The Contractor Public Service

And so, let me move to contracting. Contracting for services for, and in the name of, government explicitly dons the cloak that lines all accountability systems—legal accountability. But it implicitly wears others—performance and profit. While the performance dimension is now being applied with some zeal to the career service as well, it is the base foundation and rationale for contractors.

Contracting for service and expertise has a long and—in many ways—quite successful history at all levels of government in the United States. Contractors provide laundry and dining facilities; they build airplanes, helicopters, roads, and missiles; they design information management systems, payroll systems, and change strategies. Universities and organizations such as the Rand Corporation conduct critical and extensive research for government agencies. Contractors build and maintain space shuttles. They have served as intelligence analysts, linguists, and as critical and flexible logistical support for troops in Bosnia, Afghanistan, and Iraq.

There have obviously been long-term concerns about the growth of the contractor workforce, however. As early as 1962, President John F. Kennedy requested that a commission—the Bell Commission—look into the implications of the expanding practice. In his request, Kennedy asked: 1) which criteria should be used in determining whether to employ a contractor; and 2) how to ensure effective monitoring and accountability? Speaking specifically of research and development activities, the commission responded, “we need to be particularly sensitive to the cumulative effects of contracting out Government work. A series of actions to contract out important activities, each wholly justified when considered on its own merits, may when taken together begin to erode the government's ability to manage” (Commission, in Nelson 1968, 203). The panel noted further that relying on contractors for critical expertise and knowledge did not ultimately retain that key knowledge for government unless the contractors remained on board for extensive periods—a clear challenge to the substantive accountability of the career service.

The prescience of these observations is evident today. Indeed, the numbers of contractors have grown so large that counting them accurately is a futile endeavor. There is considerable dispute about how this might even be done. Guttman (2002) provides a useful perspective and some evidence, noting that the Pentagon is unable to accurately account for the total numbers of contractors in even one of its units—the Army—and that estimates vary about total cost. Pentagon analyses demonstrate that using the average cost of a comparable civil servant's salary (and dividing total contractor budgets by that number) is not an accurate way to count contract employees, for example. This is so not because contract employees cost less than members of the civil service, but because they cost more—probably substantially more (Daily Briefing 2001). Remarkably, the Army itself can do no better than estimate its total contract numbers to somewhere between 100,000 and 600,000 (Guttman 2002).

Two organizations very much in the public eye demonstrate further dimensions of the issue. NASA has historically relied heavily on a contractor workforce. This is an organization with a record of both dazzling success and heartbreaking failure. Romzek and Dubnick note that prior to the Challenger tragedy in 1986, NASA's budget had been cut by half; its staff by over 40% (Romzek and Dubnick 1987). Similar observations were made about the Columbia tragedy in 2003. In that year, the General Accountability Office noted that NASA spends over $12 billion per year on contracted activities. Since 1990, however, GAO has also placed NASA is on its “high risk” list for contract management capacity. The retirement and recruiting problems the agency confronts now will only compound the problem (Government Accountability Office 2003).

Or consider the Department of Home-land Security. At its creation, this department posed one of the most complex organizational challenges ever to confront the federal government. Under the continuing threat of crisis, 22 agencies—each with its own history, culture, and mission—needed to quickly become a cohesive whole. Reporting to scores of different congressional committees makes the task exponentially more difficult. The demand for new expertise, skills, and capacity is immense. So was—and is—the use of contractors. As the organization pondered implementation of the path-breaking new management flexibilities provided by Congress, one large management consulting firm managed the entire planning and consultation process.

As the department moved beyond the planning stage, another major contract was let—this time to manage the implementation of the flexibilities, but also to manage consolidation of information technology systems within the department. Even the scope of work is troubling here. Two very disparate tasks were bundled: combining information technology systems from 22 agencies and implementing management flexibilities. Indeed, the contractor selected is not known for its management expertise, but rather for scientific and technological skills.

A second dimension, however, is also present: the sensitive nature of DHS's work suggests hyper-careful attention to accountability. Yet, as recently as 2002 the contractor selected in this case was placed on the Project on Government Oversight's top 10 list of “repeat offenders” for misconduct and alleged misconduct in the administration of government contracts (Project on Government Oversight 2002). The same study found that in 1999, just 43 contractors received nearly half of all contract dollars awarded.

This is the troubling scenario that the Bell Commission warned against in 1962, when it declared: “Basic functions of management cannot be transferred to any contractor if we are to have proper accountability for the performance of public functions and the use of public funds” (1968, 201). Legal contracts are difficult to enforce when the scope of work is unclear, when it is difficult to create and monitor performance standards, and when those managing the contracts are overmatched in expertise and overwhelmed by sheer numbers of contractors (Choi and Heinrich 2004). In fact, recent work suggests that long-term contracts become “relational”; that is, the final scope of work is tailored to contractor capabilities, rather than the real needs of the contracting agency—the government (Romzek and Johnston 2002).

You Still Talking to Somebody?

Well, who are we trying to talk to? We know well that the public service has changed in the past quarter century, as has the environment in which it operates. In this time, the federal service has been reformed, reinvented, transformed, downsized, and hollowed out. It has lost leadership potential and critical expertise. Political appointees have increased in numbers and the kinds of positions to which they can be appointed have grown. Contracting out has mushroomed and contractors now perform tasks that profoundly cloud the line between “government” and “nongovernmental” work. At the same time, the challenges of public work, the dramatic increase in threat and uncertainty, the global consequences of actions taken in and by the U.S., and the now common and well recognized demands for collaboration and networked governance have also grown. The ‘disarticulation of the state”—George Frederickson's term (1999)—has placed yet unforeseen demands on public administration and public administrators, as well as on theories of governance. Substantive and representational accountability systems have been severely challenged by the new profit and performance accountability of the contract force.

Finding solutions to these problems will be difficult, but it is very, very important. Let me suggest some starting points for all three components.

For the political service:

1. Carefully examine where political appointees are placed and take a hard look at whether they and their positions are really needed.
2. Address the short tenure problem, including a careful look at where turnover is the highest and why. In other words, who leaves appointive posts most quickly?
3. Pay attention to expertise needs and demands. Political appointees may expect to talk to other politicals, but does that activity contribute to greater effectiveness in government? We should ask.

For the career service:

1. Address the representation dimension more aggressively and more consistently. Full representativeness at leadership levels is critical.
2. Continue to focus on capacity problems. Sustainable expertise is absolutely fundamental to effectiveness and accountability in this part of the public service.
3. Tailor performance expectations to real capacity and include the demands of managing large numbers of contractors in those expectations. We are being unfair to the career service when we allow political rhetoric to set performance and accountability standards.

For the contract service:

1. Count!!! Recognize the size and increasing significance of this group and their growing detachment from established accountability systems.
2. Create better monitoring capabilities in the career service. Pay attention to the monitoring deficit. This relates directly to career capacity.
3. Return to the warnings of the Bell Commission to examine what contractors do, how they do it, and to whom they report performance—or lack thereof. Ask the key question: no matter how central working for profit may be in the broader economy, where and how does it fit into the public service?

Again, there is no issue more central to political science and public administration than democratic accountability. In 1936, John Gaus wrote that “the responsibility of an individual civil servant to the hierarchy of his superiors—is confused” (Gaus 1936, 31). In 2004, we have not clarified that accountability; we have muddled it further.

Just as the public service is now the sum of multiple parts, so accountability systems that govern it must add up to something coherent and greater than each of the individual efforts. It is, as Nelson has observed, an irony of American history that countless past efforts to control public bureaucracy have actually resulted in diminished control. But even that irony does not accommodate current reality. Reclaiming the challenge of finding an effective accountability framework for the future—one that fits all three components of the public service—is of fundamental significance to effective government, to important research, to good teaching, and to our responsibilities as engaged citizens. It is time for us to accept the challenge.