The Bureaucracy Problem



James Q. Wilson, the pre-eminent American political scientist of the past half-century, passed away on March 1st at the age of 80. The breadth and depth of Wilson's scholarship were extraordinary: He refined and enlarged our understanding of urban politics, crime and law enforcement, bureaucracy, regulation, drug policy, the role of intellectuals in public life, equality and elitism, the state of the family, moral reasoning, human nature, and far more. He approached every subject with an open mind, with grace and elegance, with sober realism about the limits of knowledge and power, and with a profound understanding of how character shapes every human action and every social institution.

Many of Wilson's greatest intellectual contributions were offered in the form of essays on pressing public questions. National Affairs was honored to publish one of his last essays, "The Future of Blame," in our Winter 2010 issue. Many of his most important essays — more than 40 in all — appeared in the pages of The Public Interest; they, along with the complete archive of that journal, may be found on our web site.

Perhaps the best way to honor Wilson's memory is to bring his work to a new generation of readers. In that spirit, we reprint here one of his earliest contributions to The Public Interest. Originally published in the Winter 1967 issue, this essay was Wilson's first extended public meditation on the question of bureaucracy, which was to become one of his foremost interests. It offers a sense of Wilson's clarity of thought and writing, of his ability to reach directly to the core of complex problems, of his level-headed practicality, and of his humor and good sense. It also offers powerful evidence of the timelessness of Wilson's themes: Written 45 years ago, this essay provides some essential correctives to an argument that is very much alive in our time. We reprint it as a small token of gratitude and appreciation for a great scholar and a great American.


The federal bureaucracy, whose growth and problems were once only the concern of the right, has now become a major concern of the left, the center, and almost all points in between. Conservatives once feared that a powerful bureaucracy would work a social revolution. The left now fears that this same bureaucracy is working a conservative reaction. And the center fears that the bureaucracy isn't working at all.

Increasing federal power has always been seen by conservatives in terms of increasing bureaucratic power. If greater federal power merely meant, say, greater uniformity in government relations — standardized trucking regulations, for example, or uniform professional licensing practices — a substantial segment of American businessmen would probably be pleased. But growing federal power means increased discretion vested in appointive officials whose behavior can neither be anticipated nor controlled. The behavior of state and local bureaucrats, by contrast, can often be anticipated because it can be controlled by businessmen and others.

Knowing this, liberals have always resolved most questions in favor of enhancing federal power. The "hacks" running local administrative agencies were too often, in liberal eyes, the agents of local political and economic forces — businessmen, party bosses, organized professions, and the like. A federal bureaucrat, because he was responsible to a national power center and to a single president elected by a nationwide constituency, could not so easily be bought off by local vested interests; in addition, he would take his policy guidance from a President elected by a process that gave heavy weight to the votes of urban, labor, and minority groups. The New Deal bureaucrats, especially those appointed to the new, "emergency" agencies, were expected by liberals to be free to chart a radically new program and to be competent to direct its implementation.

It was an understandable illusion. It frequently appears in history in the hopes of otherwise intelligent and far-sighted men. Henry II thought his clerks and scribes would help him subdue England's feudal barons; how was he to know that in time they would become the agents of parliamentary authority directed at stripping the king of his prerogatives? And how were Parliament and its cabinet ministers, in turn, to know that eventually these permanent undersecretaries would become an almost self-governing class whose day-to-day behavior would become virtually immune to scrutiny or control? Marxists thought that Soviet bureaucrats would work for the people, despite the fact that Max Weber had pointed out why one could be almost certain they would work mostly for themselves. It is ironic that among today's members of the "New Left," the "Leninist problem" — i.e., the problem of over-organization and of self-perpetuating administrative power — should become a major preoccupation.

This apparent agreement among polemicists of the right and left that there is a bureaucracy problem accounts, one suspects, for the fact that non-bureaucratic solutions to contemporary problems seem to command support from both groups. The negative income tax as a strategy for dealing with poverty is endorsed by economists of such different persuasions as Milton Friedman and James Tobin, and has received favorable consideration among members of both the Goldwater brain trust and the Students for [a] Democratic Society. Though the interests of the two groups are somewhat divergent, one common element is a desire to scuttle the social workers and the public-welfare bureaucracy, who are usually portrayed as prying busybodies with pursed lips and steel-rimmed glasses ordering midnight bedchecks in public housing projects. (Police officers who complain that television makes them look like fools in the eyes of their children will know just what the social workers are going through.)

*     *     *

Now that everybody seems to agree that we ought to do something about the problem of bureaucracy, one might suppose that something would get done. Perhaps a grand re-organization, accompanied by lots of "systems analysis," "citizen participation," "creative federalism," and "interdepartmental coordination." Merely to state this prospect is to deny it.

There is not one bureaucracy problem, there are several, and the solution to each is in some degree incompatible with the solution to every other. First, there is the problem of accountability or control — getting the bureaucracy to serve agreed-on national goals. Second is the problem of equity — getting bureaucrats to treat like cases alike and on the basis of clear rules, known in advance. Third is the problem of efficiency — maximizing output for a given expenditure, or minimizing expenditures for a given output. Fourth is the problem of responsiveness — inducing bureaucrats to meet, with alacrity and compassion, those cases which can never be brought under a single national rule and which, by common human standards of justice or benevolence, seem to require that an exception be made or a rule stretched. Fifth is the problem of fiscal integrity — properly spending and accounting for public money.

Each of these problems mobilizes a somewhat different segment of the public. The problem of power is the unending preoccupation of the president and his staff, especially during the first years of an administration. Equity concerns the lawyers and the courts, though increasingly the Supreme Court seems to act as if it thinks its job is to help set national goals as a kind of auxiliary White House. Efficiency has traditionally been the concern of businessmen who thought, mistakenly, that an efficient government was one that didn't spend very much money. (Of late, efficiency has come to have a broader and more accurate meaning as an optimal relationship between objectives and resources. Robert McNamara has shown that an "efficient" Department of Defense costs a lot more money than an "inefficient" one; his disciples are now carrying the message to all parts of a skeptical federal establishment.) Responsiveness has been the concern of individual citizens and of their political representatives, usually out of wholly proper motives, but sometimes out of corrupt ones. Congress, especially, has tried to retain some power over the bureaucracy by intervening on behalf of tens of thousands of immigrants, widows, businessmen, and mothers-of-soldiers, hoping that the collective effect of many individual interventions would be a bureaucracy that, on large matters as well as small, would do Congress's will. (Since Congress only occasionally has a clear will, this strategy only works occasionally.) Finally, fiscal integrity — especially its absence — is the concern of the political "outs" who want to get in and thus it becomes the concern of "ins" who want to keep them out.

Obviously the more a bureaucracy is responsive to its clients — whether those clients are organized by radicals into Mothers for Adequate Welfare or represented by congressmen anxious to please constituents — the less it can be accountable to presidential directives. Similarly, the more equity, the less responsiveness. And a preoccupation with fiscal integrity can make the kind of program budgeting required by enthusiasts of efficiency difficult, if not impossible.

Indeed, of all the groups interested in bureaucracy, those concerned with fiscal integrity usually play the winning hand. To be efficient, one must have clearly stated goals, but goals are often hard to state at all, much less clearly. To be responsive, one must be willing to run risks, and the career civil service is not ordinarily attractive to people with a taste for risk. Equity is an abstraction, of concern for the most part only to people who haven't been given any. Accountability is "politics," and the bureaucracy itself is the first to resist that (unless, of course, it is the kind of politics that produces pay raises and greater job security). But an absence of fiscal integrity is welfare chiseling, sweetheart deals, windfall profits, conflict of interest, malfeasance in high places — in short, corruption. Everybody recognizes that when he sees it, and none but a few misguided academics have anything good to say about it. As a result, fiscal scandal typically becomes the standard by which a bureaucracy is judged (the FBI is good because it hasn't had any, the Internal Revenue Service is bad because it has) and thus the all-consuming fear of responsible executives.

*     *     *

If it is hard to make up one's mind about how one wants the bureaucracy to behave, one might be forgiven if one threw up one's hands and let nature take its course. Though it may come to that in the end, it is possible — and important — to begin with a resolution to face the issue squarely and try to think through the choices. Facing the issue means admitting what, in our zeal for new programs, we usually ignore: There are inherent limits to what can be accomplished by large hierarchical organizations.

The opposite view is more often in vogue. If enough people don't like something, it becomes a problem; if the intellectuals agree with them, it becomes a crisis; any crisis must be solved; if it must be solved, then it can be solved — and creating a new organization is the way to do it. If the organization fails to solve the problem (and when the problem is a fundamental one, it will almost surely fail), then the reason is "politics," or "mismanagement," or "incompetent people," or "meddling," or "socialism," or "inertia."

Some problems cannot be solved and some government functions cannot, in principle, be done well. Notwithstanding, the effort must often be made. The rule of reason should be to try to do as few undoable things as possible. It is regrettable, for example, that any country must have a foreign office, since none can have a good one. The reason is simple: it is literally impossible to have a "policy" with respect to all relevant matters concerning all foreign countries, much less a consistent and reasonable policy. And the difficulty increases with the square of the number of countries, and probably with the cube of the speed of communications. The problem long ago became insoluble and any sensible Secretary of State will cease trying to solve it. He will divide his time instead between ad hoc responses to the crisis of the moment and appearances on Meet the Press.

The answer is not, it must be emphasized, one of simply finding good people, though it is at least that. Most professors don't think much of the State Department, but it is by no means clear that a department made up only of professors would be any better, and some reason to believe that it would be worse. One reason is that bringing in "good outsiders," especially good outsiders from universities, means bringing in men with little experience in dealing with the substantive problem but many large ideas about how to approach problems "in general." General ideas, no matter how soundly based in history or social science, rarely tell one what to do tomorrow about the visit from the foreign trade mission from Ruritania or the questions from the congressional appropriations subcommittee.

Another reason is that good people are in very short supply, even assuming we knew how to recognize them. Some things literally cannot be done — or cannot be done well — because there is no one available to do them who knows how. The supply of able, experienced executives is not increasing nearly as fast as the number of problems being addressed by public policy. All the fellowships, internships, and "mid-career training programs" in the world aren't likely to increase that supply very much, simply because the essential qualities for an executive — judgment about men and events, a facility for making good guesses, a sensitivity to political realities, and an ability to motivate others — are things which, if they can be taught at all, cannot be taught systematically or to more than a handful of apprentices at one time.

This constraint deserves emphasis, for it is rarely recognized as a constraint at all. Anyone who opposed a bold new program on the grounds that there was nobody around able to run it would be accused of being a pettifogger at best and a reactionary do-nothing at worse. Everywhere except in government, it seems, the scarcity of talent is accepted as a fact of life. Nobody (or almost nobody) thinks seriously of setting up a great new university overnight, because anybody familiar with the university business knows that, for almost any professorship one would want to fill, there are rarely more than five (if that) really top-flight people in the country, and they are all quite happy — and certainly well paid — right where they are. Lots of new business ideas don't become profit-making realities because good business executives are both hard to find and expensive to hire. The government — at least publicly — seems to act as if the supply of able political executives were infinitely elastic, though people setting up new agencies will often admit privately that they are so frustrated and appalled by the shortage of talent that the only wonder is why disaster is so long in coming. Much would be gained if this constraint were mentioned to Congress before the bill is passed and the hopes aroused, instead of being mentioned afterward as an excuse for failure or as a reason why higher pay scales for public servants are an urgent necessity. "Talent is Scarcer Than Money" should be the motto of the Budget Bureau.

*     *     *

If administrative feasibility is such a critical issue, what can be done about it? Not a great deal. If the bureaucracy problem is a major reason why so many programs are in trouble, it is also a reason why the problem itself cannot be "solved." But it can be mitigated — though not usually through the kinds of expedients we are fond of trying: Hoover Commissions, management studies, expensive consultants, coordinating committees, "czars," and the like. The only point at which very much leverage can be gained on the problem is when we decide what it is we are trying to accomplish. When we define our goals, we are implicitly deciding how much, or how little, of a bureaucracy problem we are going to have. A program with clear objectives, clearly stated, is a program with a fighting chance of coping with each of the many aspects of the bureaucracy problem. Controlling an agency is easier when you know what you want. Equity is more likely to be assured when overall objectives can be stated, at least in part, in general rules to which people in and out of the agency are asked to conform. Efficiency is made possible when you know what you are buying with your money. Responsiveness is never easy or wholly desirable; if every person were treated in accordance with his special needs, there would be no program at all. (The only system that meets the responsiveness problem squarely is the free market.) But at least with clear objectives we would know what we are giving up in those cases when responsiveness seems necessary, and thus we would be able to decide how much we are willing to tolerate. And fiscal integrity is just as easy to insure in a system with clear objectives as in one with fuzzy ones; in the former case, moreover, we are less likely to judge success simply in terms of avoiding scandal. We might even be willing to accept a little looseness if we knew what we were getting for it.

The rejoinder to this argument is that there are many government functions which, by their nature, can never have clear objectives. I hope I have made it obvious by now that I am aware of that. We can't stop dealing with foreign nations just because we don't know what we want; after all, they may know what they want, and we had better find out. My argument is advanced, not as a panacea  —  there is no way to avoid the problem of administration — but as a guide to choice in those cases where choice is open to us, and as a criterion by which to evaluate proposals for coping with the bureaucracy problem.

Dealing with poverty — at least in part — by giving people money seems like an obvious strategy. Governments are very good at taking money from one person and giving it to another; the goals are not particularly difficult to state; measures are available to evaluate how well we are doing in achieving a predetermined income distribution. There may be many things wrong with this approach, but administrative difficulty is not one of them. And yet, paradoxically, it is the last approach we will probably try. We will try everything else first — case work, counseling, remedial education, community action, federally financed mass protests to end "alienation," etc. And whatever else might be said in their favor, the likelihood of smooth administration and ample talent can hardly be included.

Both the White House and the Congress seem eager to do something about the bureaucracy problem. All too often, however, the problem is described in terms of "digesting" the "glut" of new federal programs — as if solving administrative difficulties had something in common with treating heartburn. Perhaps those seriously concerned with this issue will put themselves on notice that they ought not to begin with the pain and reach for some administrative bicarbonate of soda; they ought instead to begin with what was swallowed and ask whether an emetic is necessary. Coping with the bureaucracy problem is inseparable from rethinking the objectives of the programs in question. Administrative reshuffling, budgetary cuts (or budgetary increases), and congressional investigation of lower-level boondoggling will not suffice and are likely, unless there are some happy accidents, to make matters worse. Thinking clearly about goals is a tough assignment for a political system that has been held together in great part by compromise, ambiguity, and contradiction. And if a choice must be made, any reasonable person would, I think, prefer the system to the clarity. But now that we have decided to intervene in such a wide range of human affairs, perhaps we ought to reassess that particular tradeoff.