Aaron Wildavsky

The Self-Evaluating Organization

Why don’t organizations evaluate their own activities? Why don’t they seem to manifest rudimentary self-awareness? How long can people work in organizations without discovering their objectives or determining how well they have been carried out? I started out thinking it was bad for organizations not to evaluate, and I ended up wondering why they ever do it. Evaluation and organization, it turns out, are somewhat contradictory. Failing to understand that incompatibility, we are tempted to believe in absurdities, much in the manner of mindless bureaucrats who never wonder whether they are doing useful work. If instead we asked more intelligent questions, we would neither look so foolish nor be so surprised.

Policy analysis is what information systems are not

The task of analysis is to create problems, preferences tempered by possibilities, which are worth solving. A difficulty is not necessarily a problem; that depends on what I can do about it, including whether it is worth my while to try. My inability to go to Mars, a famous gap between aspirations and actuality, is not a problem but a longing to overcome my limitations. My inability to explain the influence of the tides on the rise and fall of the stock market is not a problem unless I have a hypothesis suggesting how I might influence factors by which the two events might be linked. Only by suggesting solutions, such as programs linking governmental resources with social objectives, can we understand what might be done. Policy analysis involves creating problems that are solvable by specific organizations in a particular arena of action. A problem in policy analysis, then, cannot exist apart from a proposed solution, and its solution is part of an organization, a structure of incentives without which there can be no will to act.

Policy as Its Own Cause

Why do we feel that public policy problems never seem to be solved? As knowledge and skill grow in society, why do efforts to control public policies lag behind their ability to surprise us? Why don’t organizations that promote public policies seem to learn from experience? If they do try, why do their actions lead to ever larger numbers of unanticipated consequences? One answer, I will argue, lies in the growing autonomy of the policy environment. Because policy is evermore its own cause, programs depend less on the external evironment than on events inside the sectors from which they come. The rich inner life of public agencies helps explain why there appears to be so much change for its own sake. If bureaucracies are the principal opponents of change, as is often alleged, however, how can they also be its chief sponsors? How, if major sectors of public policy can control their internal response to external events, does the world outside specifically affect organizational behavior in government? If external forces matter, why do organizational responses often appear to have so little relation to what actually goes on out there in society? Why, in a word, do supposed solutions turn into perplexing problems? Because the Law of Large Solutions in Public Policy —when the solution dwarfs the problem as a source of worry—is inexorable.

A Bias Toward Federalism

To govern this nation do we want an operative federal structure? If so, where do we want the balance between national and state power to be drawn and on which issues? Under a national regime, states and localities carry out national instructions; the problem is how to improve their obedience. In a federal regime, states and localities are disobedient. The operational meaning of federalism is found in the degree to which the constituent units disagree about what should be done, who should do it, and how it should be done. In a word, federalism is about conflict. It is also about cooperation, that is, the terms and conditions under which conflict is limited. A federal regime, therefore, cannot be coordinated from the center any more than it can be controlled or coerced. Coordination, as we have seen, does not necessarily imply a coordinator. Under an operative federalism, coordination occurs by interaction among many governments, not by intellectual cogitation by a single one. Federalism means mutuality, not hirearchy, multiple rather than single causation, a sharing instead of a monopoly of power. One can determine if the federal beast is alive only by whether it kicks—and then whom it kicks and who kicks it back. The rationality, responsiveness, and responsibility of a regime, its overall decency and effectiveness, should not depend on its appearance (it may appear untidy) but on results.

Analysis as Craft

A lot of “stake claiming” goes on in defining policy analysis. The landscape of our knowledge is surveyed and boundaries that delimit the domain of each discipline are drawn: “this belongs to political science, that belongs to economics.” Over past centuries, the great empires of theology, geometry, and natural history have broken up, spawning a multitude of disciplinary fiefdoms. New alliances, formed on marginal lands, claim independence: econometrics, social psychology, political economy. Subdisciplinary groups coalesce, border disputes flare, while intrepid basic researchers of each discipline fan out in search of virgin territory on which to plant their flags. Explorers bearing the ensign of policy analysis seem bewildered by this scramble for territory. They expropriate lands claimed by political scientists decades ago and more recently by planners and public administrators. They skirt the edges of economics, law, organizational theory, and operations research. Some seek refuge in these disciplines. Others wait for a Moses to lead them out of the wilderness to the promised land of professionalism. Still others, being more nationalistic, want to carve out a “policy analytic” domain. But where? Establishing a discipline in the interstices of disciplines already distinct is risky; the new map is likely to reveal an impossibly gerrymandered state composed of marginal lands already contested by others.

Citizens as Analysts

Citizenship has been studied from almost every standpoint except that of participation in public policy. The influence of citizens on the making and changing of policy, citizens’ power (or the lack thereof) over public officials who make and administer policy, and the ability of the general public to hold these elites accountable, even citizens’ ability to create and dissolve government itself, have been subject to scrutiny. But participation as part of policy (with the exception of Michael Lipsky’s “Street Corner Bureaucrats”) has been neglected.

Opportunity Costs and Merit Wants

If opportunity costs are about cabbages, merit wants are about philosopher kings. These rival doctrines may speak in the language of economics but the guiding hand behind them is that of politics. The great questions are there: Who should rule? How should they rule? Who will govern the governors? How might subjective preferences be converted into collective rationality? In short, the question of whether decisions should be made by social interaction, with many minds contributing, or by intellectual cogitation, with an elite acting, in effect, as a single intelligence, will reappear here in different doctrinal form. The better to get to know them, we shall approach opportunity costs analytically, by decomposition, and historically, by evolution, for from these seemingly simple garden-variety notions of cost and merit, we shall peel off layers of ideology. If our economic onion causes more than a few tears, we can say only that it’s better to be sad theoretically than actually sorry for absentmindedly eliminating economics from public policy.

Coordination Without a Coordinator

Should the United States government spend for war or for peace? Crudely expressed, that was a major dividing line in the debates over public policy in the late fifties through the sixties and seventies. The course followed by enlightened liberal opinion was clear: much more for welfare programs designed to help deserving groups, such as the poor, the aged, and minorities; much less for military preparedness or, worse still, for actually fighting wars. Less clear but still pronounced was a preference for a strategy specifying income over service for transferring money rather than goods, so that people could exercise more choice. No specific decision declared that government would do these things and no announcement came that this reallocation of resources would be governmental policy from a chosen day onward. We have no Domestic Welfare Day like an Independence Day. That these trends were not summed up and announced as deliberate governmental policy one day may account for their continued lack of recognition. People are still waiting for this revolution in public policy to come when it has already been. That we may not like it as much as we thought (viz. “Strategic Retreat on Objectives”) doesn’t mean it didn’t happen.

Distribution of Urban Services

Are policy analysts “hired guns”? Do they sell themselves to the highest bidder, and recommend whatever their clients desire? By the very nature of the craft, this task actually is not easy to do. Analysts, when asked to help solve a problem, are likely to reformulate it: this problem cannot be solved within our limitations, but here is one like it that we can do something about. By altering the means, the ends are altered, whether that is acknowledged or not.