Brownie, you’re doing a heck of a job.” For many, President George W. Bush’s praise for then-FEMA Director Michael Brown in the wake of Hurricane Katrina–and the often sluggish and sometimes calamitous governmental response to the disaster–captures perfectly the problems of an Administration that had lost the sure footing of its first term. But it also points to the potential problems that presidential appointees can pose, both politically and practically. It’s not often that the former head of the International Arabian Horse Association is asked to lead a major government agency, and when they are, the results–as they did after Katrina–speak for themselves.

Picking appointees is a daunting challenge for any new president. John F. Kennedy reportedly observed, “For the last four years I spent so much time getting to know people who could help me get elected president that I didn’t have time to get to know people who could help me, after I was elected, to be a good president.” The task has only grown more difficult since. According to one estimate, the number of political appointees rose from fewer than 2,000 when JFK was inaugurated to as many as 3,435 in 1980, the high-water mark. This is a stark contrast to the 100 to 200 political appointments made by leaders in Britain, France, and Germany. But viewed from another perspective, the number seems impossibly small: Many of these appointees are responsible for guiding the work of 2.5 million civilian federal employees and over 1.4 million uniformed military personnel. Adding further complexity, less than half of all federal government employees are under the traditional civil service system, confronting Administrations with varying rules and personnel systems to master.

But if “personnel is policy,” then the number of political appointees doesn’t matter; who they are does. A president’s political appointments (a minority of whom are subject to confirmation by the U.S. Senate) are significant channels for chief executives who seek both to guide agencies in desired ways and to reinforce citizen connections to the bureaucracy. Indeed, this has been a key strategy of conservatives who aspire to reduce the size and role of government; as revealed this past summer, the Bush Justice Department even screened summer interns for ideological purity.

Beyond the politics, personnel selection and management requires a sure sense of goals, priorities, and strategies. Presidents must be sensitive to the wide variety among programs and agencies, as well as to the skills and values of the pool of potential appointees. Even then, what political appointees can accomplish is likely to be constrained by the other branches of government; by the structures, processes, and agency cultures they confront; by unanticipated problems and events; and, not infrequently, by presidential inattention. Nevertheless, to understand how the federal government functions, it is critical to have a deep understanding of the appointment system.

And yet, as David Lewis ruefully comments in his new book, The Politics of Presidential Appointments: Political Control and Bureaucratic Performance, the 3,000 or so presidential appointees to executive branch agencies are of little interest outside the Beltway, at least in the absence of highly publicized missteps. Particularly as the 2008 presidential election approaches, it is worth thinking about the sorts of people and the policy priorities and administrative strategies that Senators John McCain and Barack Obama would bring with them to run the massive and diverse executive branch. While the number of political appointees may be small relative to the size of the federal bureaucracy, who is chosen to run it is critically important and may be a greater guide to what the candidates may do as president than any policy paper or speech.

This does not mean that we should be skeptical of any overtly political appointment of a longtime ally or friend. Political appointees are not necessarily a bad thing: For every Brownie, there is a Bobby Kennedy. The challenge is how to balance competence with political responsiveness, and find able public servants who have the trust of the President.

The presidential appointment process, even with its emphasis on rewarding allies and friends, has many advantages. The selection and deployment of political appointees is an important presidential tool, wielded both to control agencies and programs (“policy politicization”) and to reward campaign supporters (“patronage politicization”). “Politics” may be a dirty word to many Americans and politicization a derisive label, but politicization is also a valuable way for elected officials to guide programs and agencies in directions consistent with their objectives and ideals. It also offers additional avenues and incentives for many to participate in public service.

Nonetheless, the appointments process is a key target for political attacks. Candidates, particularly those running against incumbents, like to highlight the numbers of frequently ill-prepared appointees and career officials who are variously viewed as out of control, bound by red tape, or incompetent.

In The Politics of Presidential Appointments, Lewis, a political scientist now at Vanderbilt University, delves into an assessment of the pros and cons of the process through a systematic and balanced examination of the available evidence. The result is impressive: a careful, nuanced, and thoughtful analysis of a major part of government that few citizens know much about. Lewis reminds us that the tension between competent and loyal government employees is nearly as old as the United States itself. Andrew Jackson championed the rotation system (or “spoils,” according to its critics) as a way to reinvigorate government and reward partisans (though many professionals continued in government across presidential administrations). The 1883 Pendleton Act sought to banish spoils by creating the federal merit system, but elected officials devised new ways of assuring loyalty and support. This includes the “burrowing” of appointed officials into the career service at the end of administrations; the 1940 Ramspeck Act’s guarantee of executive branch jobs to congressional staffers whose bosses lost elections; the Eisenhower Administration’s Willis Directive, which invited Republican officials to suggest names for the civil service; and an executive order that created Schedule C jobs “of a policy determining or confidential nature” for political appointees at a variety of levels.

The number of presidential appointments (those requiring Senate confirmation; appointed members of the Senior Executive Service, or SES, and its predecessors; and Schedule C appointments) varied considerably between 1960 and 2004. Contrary to what some critics might believe, they peaked both in number and as a percent of government managers in 1980. Political appointments declined under Bill Clinton, although Schedule C and political SES appointees have increased in both number and percent under George W. Bush, especially in the departments of Commerce, Education, and Labor. At the same time, the percentage of politically appointed managers differs considerably across departments and agencies, tending to be highest in places like the Small Business Administration and the departments of Education, Labor, and Housing and Urban Development, and lowest in NASA, the Social Security Administration, and the departments of the Treasury, Defense, and Veterans Affairs.

Republican and Democratic Presidents alike have engaged in politicization, though their techniques vary somewhat. Not surprisingly, the numbers and percentages of political appointees increase most when there is party turnover in the presidency. They are higher as well when the same party controls both Congress and the White House. Presidents take policy orientations seriously, engaging in policy politicization in agencies they think are pursuing goals inconsistent with their own views. Republican presidents thus focus on more “liberal” entities like HUD’s Public and Indian Housing Bureau and the EPA, while Democratic presidents emphasize more “conservative” bastions like the Department of the Interior. Conversely, patronage appointments tend to cluster in more liberal locations for Democratic presidents and more conservative ones for Republican ones.

It is of some value to have a more accurate understanding of how, when, and why politicization does (and does not) take place. One can be heartened to learn of Lewis’s finding that, for example, less politicization evidently takes place in programs and units that perform complex or technical tasks. Yet these are academic questions. What matters to citizens is whether politicization makes any difference in what government does. Do presidents and their supporters achieve the responsiveness they desire? What effect is there on the quality of government services and programs?

The answers are surprising. On the one hand, it is difficult for presidential appointees to change the direction of entire agencies. Scholars have found, for example, that even after the Reagan Administration employed budget cuts, reductions-in-forces (RIFs), and multiple appointments to diminish the power of the EPA, enforcement of most environmental laws continued, and efforts even increased in some parts of the country as states stepped in to fill the void. Others have discovered little impact of politicization efforts at the field level in Bureau of Land Management projects. Meanwhile, scholars like UCLA political scientist Amy Zegart conclude that organizational, rather than individual, deficiencies provide the stronger explanations for why the 9/11 attacks were not stopped by U.S. intelligence agencies.

Lewis takes a somewhat different tack, scrutinizing the relationships between politicization and agency performance using measurements generated by the Office of Management and Budget with the Program Assessment Rating Tool (PART). He finds that programs and agencies with relatively few political appointees (such as the National Science Foundation) are among the highest scoring, while four of the 10 programs with the lowest PART scores are in the most politicized department, Education (along with Interior, HUD, Justice, and Veterans Affairs). Even when controlling for differences in program content and management difficulties, programs managed by political appointees perform less well than those run by career officials. Career workers responding to surveys of executive branch officials distributed by the Office of Personnel Management in 2002 and 2004 mostly reinforced these findings. The units receiving the highest evaluations for the quality and commitment of their leadership tended to be the least politicized: Veterans Affairs and NASA, and the most poorly evaluated–bureaus in HUD and the Education Department–were the most politicized. Yet this doesn’t mean that all political appointees are disasters.

A closer look at FEMA (to which Lewis devotes a full chapter) adds detail to some of the broad brush strokes of this general assessment. From the start, the agency has been “hampered by design.” Not only does it have a relatively large number of political appointees, but the agency rarely has been a top priority for presidents or carried much prestige in Washington, making it difficult to attract high-quality appointees. Moreover, until recently, FEMA has had to juggle three competing tasks: civil defense, natural disaster response, and, after September 11, terrorism response. This turbulent context clearly has affected the agency’s performance. But the lack of qualified leaders and the overpoliticization of FEMA appointments have been at the core of the problem: No FEMA director had any emergency management experience between 1979 and 1992; relationships between appointees and career officials were poor, just as the recent OPM surveys would predict; and their performance were even worse. FEMA’s problematic responses to several disasters between 1989 and 1992 (Hurricanes Hugo and Andrew, the Loma Prieta Earthquake) damaged both President George H.W. Bush’s reelection chances and agency morale and reputation.

Yet the performance of Bill Clinton’s FEMA director, James Lee Witt, underscores the potential value and contribution of politicization–and complicates any one-sided assessment of the process. Witt, unlike other appointees, had a background in emergency management, and he introduced numerous reform initiatives. Most important, according to Lewis, Witt reduced the number of appointees in the organization, relying on career officials and others with emergency management experience to populate senior levels of FEMA. Under George W. Bush, the pendulum swung back. The numbers of appointees in the agency again increased. The first director, Joseph Allbaugh–who ran Bush’s 2000 campaign–had some emergency management experience, but many viewed the post as a consolation prize for a losing candidate for the position of White House chief of staff.

Even though by most accounts FEMA responded well during the aftermath of September 11, the attacks eclipsed the agency’s primary missions. Once it was absorbed into the new Department of Homeland Security, Allbaugh resigned, having lost his direct link to the White House. The structural change presaged the poor response to Katrina: The move into DHS seriously damaged morale at FEMA and led to considerable turnover. FEMA also lost much of its preparedness staff and budget to other parts of the Department. And the move slowed FEMA’s preparation of a National Response Plan, while fostering confusion about lines of authority in response to disasters.

Myriad lessons can be drawn from this rich narrative. Certainly, political appointees like Witt, with relevant substantive backgrounds, managerial savvy, political acumen, and strong policy commitment, as well as ties to the President, can reinvigorate and guide even a troubled agency. Yet that alone is insufficient when policy priorities change dramatically and organizational structures and resource flows shift. Lewis usefully cautions against imagining that either simple discovery of the “right” leader or slashing political appointees alone can guarantee dramatic improvement.

Political philosopher Sheldon Wolin has called for the nurturing of “a counterelite of democratic public servants” who would combine expertise with a commitment to democratic values. Lewis might well observe that the federal government is already full of such people, and that political appointees add a vital “leavening” and a shot of democratic guidance to the bureaucracy. Working with career public servants, political appointees can interject the values and views of elected officials and challenge the inertia and entrenched cultures of some agencies and programs. The question, in any case, is how to achieve such a balance.

It is not too early to begin asking such questions. Reportedly, both the McCain and the Obama campaigns are preparing lists of possible nominees so that they could be quickly vetted after the election and then confirmed once the new President is inaugurated. At the same time, senior Defense Department officials have been asked to stay on until their replacements are in place, and career officials are moving into number two slots in the DHS so they will be in position as political appointees leave. This sort of advance preparation for taking over key parts of the executive branch is crucial.

One hopes as well that such personnel and policy planning is taking place in other areas, possibly drawing on the work of the White House Transition Project (whose board I serve on), as well as other initiatives to gather reflections and “lessons learned” from previous administrations. The candidates and their advisers should target agencies and programs that will be central to the new President’s priorities, and they should explore whether and what sorts of politicization are called for in different cases. Here it may be useful to recall what Witt demonstrated in FEMA: Sometimes less politicization can reap greater benefits by more fully engaging career officials and harnessing their experience and expertise. Then again, Witt was a friend of Bill Clinton’s from Arkansas.