It’s Bureaucracy’s twenty-fifth birthday. To celebrate, let’s state some basic facts that correspond with James Q. Wilson’s thinking. Americans want a lot from their government. We want more than we’ve wanted before. It doesn’t ultimately matter where these desires come from (rising standards of living? the inner logic of democracy? interest groups? politicians?). What matters is that they exist and influence government behavior. Wilson reminded his readers, in Bureaucracy and elsewhere, that we get the government we ask for—and of course he was right. Americans expect much from their government, government power changes to meet these demands, and the part of government that makes rules and administers programs—the bureaucracy—has become the dominant part of our political system. We have an “administrative state of sprawling departments and agencies,” in the words of Jonathan Turley, and it “has a larger practical impact on the lives of citizens than all the other branches combined.”
Since we’re living in a bureaucratic age, this seems to be a fine time to take down Bureaucracy from the shelf and think about its lessons. In particular, how does it answer one of our era’s more pressing questions: what’s wrong with administrative government? Conventional wisdom has held, at least since the 1980’s, that there is something bad about rule by bureaucrats (Tocqueville made the point a few years before). In recent years, conservative and libertarian scholars have added urgency to the claim, warning that bureaucratic government is increasingly arbitrary, biased, unresponsive, and incompatible with our constitutional system. Is there a bureaucracy problem identified in Bureaucracy? How does it help us with the current situation?
Much of Bureaucracy, interestingly enough, is about what the problem of bureaucracy isn’t. As Jeremy Rabkin points out in his insightful essay, the book is a corrective—of Max Weber’s theory of bureaucracy, of overgeneralizing economists, of simplistic and unflattering stereotypes about bureaucrats’ behavior. On Weber: modern bureaucrats are not disinterested rule enforcers who bring efficiency and predictability to the political system. There is no sharp (or even meaningful) distinction between administration and policy making, and, bureaucrats—like politicians–do lots of things that are arbitrary and inefficient. On economists, or at least a certain school of economists: bureaucrats are not motivated, by and large, by self-maximization. They are motivated by convoluted things like camaraderie, professional respect, agency norms, and the desire to stay out of trouble. If there is a general impulse among bureaucrats (as Rabkin points out) it is more toward autonomy than self or agency aggregation.
And as for those unflattering stereotypes: bureaucrats do not grossly overspend or fail miserably at their objectives. There never was a $435 hammer. It may not be fun to wait on line at the DMV but you’ll eventually get what you came for. As Wilson says at the conclusion of his book, we live in a county where it is “possible to get drinkable water instantly, put through a telephone call in seconds, deliver a letter in a day, and obtain a passport in a week.” The fact that Wilson closes his book with these, and other, positive observations makes it clear that he takes issue with other people’s versions of the bureaucracy problem, and has little patience for simple characterizations.
What’s the problem?
So, correctives aside, is there a bureaucracy problem? According to Bureaucracy, there are many: they include accountability problems, fairness issues, and general unresponsiveness. But mostly there is a tendency in public agencies toward stasis, an inefficiency caused by outside constraints and agency’s internal cultures. The central bureaucratic problem, in Bureaucracy, is the inability of agencies to perform critical tasks. There are many reasons agencies fail in this way, but they can be explained by three generalizations. First, there is a frequent disconnect between what front line workers do and agency objectives (a central issue in Wilson’s Varieties of Police Behavior). Second, agencies are often given vague, overly ambitious, or conflicting goals making the identification of critical tasks unlikely. Third, and probably most significantly, bureaucrats have to contend with a thicket of procedural demands, most of which reflect other people’s ideas about efficacy and fairness. Public bureaucrats can never reach goals as effectively as their private counterparts because they have many more rules and restrictions. For these, and other reasons, Wilson doesn’t spend much time bashing individual bureaucrats. In the few instances where critiques of agency heads occur, they are circumspect and gentle. His point is that they work in a system where success is quite surprising.
To be clear: Wilson argues, in Bureaucracy, that bureaucratic success is possible. His book contains examples of agencies performing critical tasks well, owing to luck, skill, or the tenacity of a particular director. But these are the exceptions. The rule is agency performance of middling quality with tendencies toward failure. This central bureaucracy problem of inefficiency has serious ramifications. Speedy mail delivery aside, there are responsibilities agencies have that are not being met sufficiently. Public services cost more than services provided by the private sector. It is awful, and very dangerous, to be an inmate at a badly run prison. Children don’t learn in poorly run schools. Agency incapacity can lose a war (see the discussion of the French in 1940) and prevent a city government from providing services to citizens (see the New York City’s Parks Department failure to build an ice skating rink under the Koch administration).
Jeremy Rabkin agrees with Wilson on the central bureaucracy problem. His essay reminds that bureaucrats muck things up, and have mucked up a lot lately. The Obamacare rollout was a disaster, Benghazi ambassadors were left unsafe, and the secret service can’t adequately protect the White House or the President. The NSA is careless with highly sensitive information, the Veteran’s Administration falsifies records, and our Intelligence Services underestimated the capacity of ISIS. Rabkin explains these screw-ups with Wilsonian arguments. Agencies are burdened with complicated demands. They receive conflicting directives. The bureaucratic system is riddled with bad incentives, encouraging people to downplay problems instead of recognizing and addressing them. Bureaucracy is helpful in the contemporary context, according to Rabkin, because it shows that government agencies are fundamentally ineffective–especially when they take on complex and ambitious objectives.
How do we fix it?
And what about solutions? Here, Wilson and Rabkin diverge. For Wilson, there are no easy solutions to the problems of bureaucracy, just as there are no straightforward problems. He tells us and tells us again that it’s impossible to “fix” a system whose defects reflect ingrained preferences, and that any attempt at ambitious reform (through central planning, through more resources, through new organizational charts) is doomed to certain failure. In a section Rabkin mentions called “a few modest suggestions that would make a small difference,” Wilson lists ways to improve things on the ground and on the margins: give agency leaders more leeway. Encourage clear missions. Privatize as many aspects of an agency’s work as possible. Judge by results. Keep the size and responsibility of agencies small and focused. Carve out a little space for policy experimentation.
Rabkin’s approach is considerably more direct. He’d like to see the President fire more agency officials. Bureaucrats should be held responsible for screw-ups which, for Rabkin, means more Congressional scrutiny and punishment to follow. When incompetent bureaucrats stay on the job, he says, the effect is corrosive. Agency morale is lowered, whistleblowers stop coming forward, politicians and the public lose trust in agencies, and agencies become even less capable of handling tasks and solving problems. This makes for an interesting contrast with Wilson who stresses the importance of getting culture and incentives right and downplays individual culpability. (When Wilson talks about security lapses at embassies, for example, he explains it by referring to State Department priorities—not particular agents’ mistakes.) Indeed, using Wilson’s logic, Rabkin’s approach would exacerbate the bureaucracy problem, since a system of more scrutiny and punishment would probably increase cautiousness among bureaucrats and other kinds of CYA behavior. One also wonders how much effect a new commitment to firings would have in a federal bureaucracy where less than 1% of positions are filled by political appointees and most officials are shielded by civil service protections.
Another kind of problem
At the end of the day, though, this difference between Wilson and Rabkin is secondary, and their commonalities more important than differences. Both think modern bureaucracies are inherently inefficient (though Wilson warns against overstating the problem). Both define the bureaucracy problem as essentially under performance: the failure of bureaucrats to meet critical tasks and objectives. And both describe a political system where under performance is not penalized, or penalized very rarely. All good enough—but unsatisfying. This picture of hamstrung, incompetent agents doesn’t capture the current reality. Feckless bureaucrats exist, to be sure, but there is lots of evidence that bureaucrats can be quite energetic. In fact, this may be more of a problem than agency incompetence
In a 2010 book by Eric Posner and Adrian Vermule the authors argue that, in crisis, legislative power flows to the executive, and from the executive to particular agencies. The 9/11 attacks of 2001 and the financial crisis of 2008 resulted in broad delegations of power to the Department of Homeland Security, the NSA, the CIA, the Treasury, and the Federal Reserve, and, as history shows, these agencies proved quite effective using it. (You might not like your Treasury Secretary purchasing billions of dollars worth of mortgage-related securities, or federal agents monitoring your cell phone calls, but they are examples of how much agencies can do in an unrestricted environment.) But the phenomenon of broad delegation of power to agencies and ambitious agency action is, of course, not restricted to times of crisis. A quick review of agency behavior over the last few years produces many examples of energetic bureaucratic action: the EPA has strict new rules about carbon emission. The FCC is pushing for net neutrality. The IRS confiscates funds of business owners on the mere suspicion of crime. Even Obamacare, which Rabkin uses as an example of bureaucratic failure, can be used as an example of prowess: the Department of Health and Human Services, after all, has established a health insurance marketplace, insurers are participating in it, and Medicaid has been expanded in twenty-seven states.
The Department of Justice’s recent efforts to improve policing is another instructive example. Attorney General Eric Holder has made it clear that he wants police reform to be one of the Civil Rights Division’s highest priorities. He has certainly met that objective. Lawyers in the Special Litigation Section have been extremely active, in recent years, visiting cities with problem police departments, investigating police practices, threatening to sue cities, and designing consent decrees. The result: multi-year, multifaceted and multi million dollar monitoring plans that overhaul police policies, police training, police record keeping, and police investigations. But the most profound effects of DOJ activism are independent of actual supervision. The conditions in consent decrees are becoming the new best practices in big city police departments—an indication that police leaders are interested in improvement, to be sure, but also in avoiding the hassle of federal intervention. Are bureaucrats in the Special Litigation Section bumbling incompetents? Hardly. They are changing the way cities across the county think about policing.
Is efficient agency action a good or bad thing? As Wilson liked to say, it depends. Energetic bureaucrats affecting police reform are a great if you are a civil rights advocate in Seattle; less so if you are a front line officer resentful of new use of force standards. On a deeper level, though, agency effectiveness is itself highly problematic, particularly if it’s unmoored from legislative control and direction. Agency decisions have always had profound effects on individuals, businesses, and local governments. But we’re living in a time where agency officials increasingly act on their own volition, and agency actions are increasingly insulated from scrutiny and appeal–raising all sorts of questions about how power in a democracy should be allocated and how we hold officials responsible for their actions. Broad delegation of power to agencies may be inevitable in an era where government does many complex things. But we don’t have to like the situation. Bureaucrats handling their assignments well seems more troubling than bureaucratic bumbling since more damage can be done in the pursuit of grand objectives. The essential bureaucracy problem of our time is not bureaucratic inefficiency—it is the fact that bureaucrats can do so much, so often, so effectively.
Why doesn’t Wilson take up the problem of bureaucratic power in Bureaucracy? It was a different era in administrative government, to be sure, but there were plenty of energetic agencies around to talk about in 1989. Instead of stressing the power and discretion given to bureaucrats, he stresses their limitations, and particularly the power Congress has to direct and monitor their actions. This omission is particularly odd in light of one of Wilson’s earlier essays where he discusses the serious problems that arise when political authority is transferred to administrators. Wilson saw clearly enough, in 1975, that bureaus were wielding broad discretionary powers, and those powers were being used to favor particular ends and interests.
A couple of suggestions about the omission. One has to do with inclination. Wilson was, throughout his career, less interested in big theoretical schemes than he was with concrete cases. By identifying the bureaucracy problem as a disconnect between goals and tasks—a problem of skill and conditions–it allowed him to discuss agency action from (what he called) a “bottom up” approach, that is, on a case by case basis. If, on the other hand, he had identified the bureaucracy problem as something more pernicious—anti-democratic government, client politics, the collapse of legal norms—the solutions, by necessity, would need to be more ambitious. Bureaucracy is a book about agency behavior, not the recovery of the lost Constitution.
The other reason Wilson may not have talked about the “real” bureaucracy problem in Bureaucracy is because it’s not really a book about problems. I was struck, this time around, how much of the book is about conditions, and the importance of culture (yes, culture) in shaping and directing our actions. As a final birthday act, it might be worth thinking about Bureaucracy more as a work of sociology than a work of political science (though Wilson would dislike that distinction), and how it tries to explain the tension between the human capacity for choice with the limits of our reason. This might not be as fun as planning a constitutional revolution, or the sacking of John Koskinen, but it’s a more fitting way to commemorate this important occasion.
Jeremy Rabkin has written a fine essay about the continuing relevance of James Q. Wilson’s 1989 book Bureaucracy: What Government Agencies Do and Why They Do It. I have been fortunate enough to benefit from Wilson’s analysis in my own writing on the Justice Department’s Office of the Solicitor General. His framework showed why the…
The 25th anniversary of James Q. Wilson’s Bureaucracy: What Government Agencies Do and Why They Do It marks an appropriate occasion to reflect on the contributions of this work to our understanding of bureaucratic behavior and performance, and the extensive—and, at least in some areas, growing—presence of the administrative state in the lives of American…
All the participants in this discussion seem to agree that James Q. Wilson’s book, Bureaucracy, still offers valuable insights, a quarter century after its initial publication. At the same time, we all seem to agree that Wilson’s book didn’t prepare readers for the scale of dysfunction we now see in the federal bureaucracy. We have…