Continuing our collaboration with the APSA Political Economy newsletter, today we present the third of four obituaries of prominent political economists who recently passed away, with University of Wisconsin political scientist David L. Weimer writing on William A. Niskanen, Jr.
My first encounter with William A. Niskanen, Jr. was in 1973 at my graduate school orientation at the University of California, Berkeley. Although I cannot remember a word spoken that day by any of the other faculty members, I remember what Niskanen said verbatim: “My name is William Niskanen. I hold a B.A. from Harvard College and a Ph.D. in economics from the University of Chicago. Chicago won.” Indeed it did.
Niskanen’s doctoral study in economics was followed by work as a policy analyst at the RAND Corporation, the Department of Defense, the Institute for Defense Analyses, and the Office of Management and Budget. He joined the faculty of the Graduate School of Public Policy at the University of California in 1972 and became the Chief Economist for Ford Motor Company in 1975 and briefly joined the UCLA faculty in 1980. He served on President Reagan’s Council of Economic Advisors until leaving to become chairman of the board of the Cato Institute.
His training in economics and exposure to the deficiencies of bureaucracy during his early career as a policy analyst prepared and prompted Niskanen to write one of the classics of public choice, Bureaucracy and Representative Government. He first set out his thoughts on the topic in “The Peculiar Economics of Bureaucracy,” which was published in the American Economic Review in 1968, and later modified them in important ways in “Bureaucrats and Politicians,” which appeared in the Journal of Law and Economics in 1975.
The standard approach of welfare economics at the time was to assume that government and its officers selflessly intervened to correct market failures. In Bureaucracy and Representative Government, Niskanen joined the emerging public choice movement in assuming that public officials acted upon self-interest. Throughout the book, the head bureaucrat seeks to maximize the budget allocated to the bureau by the “collective organization,” or budgetary sponsor, as a lump-sum payment for the output it produces. He began by assuming passive budgetary sponsors, but later allowed “officers” of the collective organization to act according to self-interest. Although the assumption of the budget-maximizing bureaucrat is today its primary intellectual trace in the political economy literature, the book is much richer in its treatment of bureau behavior, factor suppliers, and oversight by the budgetary sponsor. Analytically, it employs comparative statics to assess the relative efficiency of bureaus, competitive supply, monopoly (discriminating and non-discriminating), and nonprofits (discriminating and non-discriminating). It thus provides a systematic comparison of institutional arrangements, an approach that is now one of the strengths of modern political economy.
The initial and central result of the book concerned allocative inefficiency when a budget-maximizing bureaucrat with a monopoly position in supply of a good chooses output levels in response to the valuation function of a passive budgetary sponsor such that the allocated budget is at least as large as the cost of producing the valued output. Rather than limiting supply (and the corresponding budget) to the point where marginal cost equals price (as would a competitive supplier) or marginal revenue (as would a monopolist selling in a market), the bureaucrat increases the budget by pushing supply either to the point where the budget allocation just equals the total cost of supply or, in the absence of a budget constraint, to the point where the marginal valuation by the budgetary sponsor falls to zero. In the former case, all resources are used to produce output, so there is no waste in the vernacular sense, but there is allocative inefficiency because output is too high from the social perspective. In the latter case, the budget may exceed the total cost of producing the output so that there is both allocative inefficiency and what Niskanen calls “fat.”
After deciding to leave academia to become the chief economist for Ford Motor Company, Niskanen agreed to lead a weekly evening seminar for his public choice students in return for dinners. It became clear in these seminars that he was troubled because, in the most common case of a binding budget constraint, his model led only to allocative inefficiency—these bureaus produced too much but did so with technical efficiency. Drawing on a critique by Jean-Luc Migué and Gérard Bélanger (1974), Niskanen (1975) reformulated the objective function of the bureaucrat to depend on the present value of income and non-monetary perquisites. Each of these, in turn, depends on the level of output of the bureau and the difference between the budget and the minimum cost of supplying the output, or what he called the “discretionary budget” and others have called “organizational slack.” Although this objective function is general and would apply to any executive, the key insight is that whereas a private manager typically gets a share of the discretionary budget, which in this case is just profit, civil service rules do not allow the bureaucrat to claim any of the discretionary budget as income. Therefore, the bureaucrat takes it as perquisites, including easier management through not putting full effort into making sure that labor and other resources are fully employed. Indeed, the bureaucrat has an incentive to hide the discretionary budget from the budgetary sponsor so as not to reveal information about the true costs of supplying output that might undercut the funding request of the bureau in the next round of budgeting.
Niskanen recognized that the full assessment of bureaucratic inefficiency required attention to the interests of the budgetary sponsor. When the sponsor could be viewed as a unitary actor, say holding the preferences of the median member, the outcome depended on bargaining within a bilateral monopoly. He was also sympathetic to more nuanced treatments of the interests of the budgetary sponsor and the nature of oversight, such as that provided by Gary Miller and Terry Moe (1983).
The empirical record of the usefulness of Niskanen’s model of bureaucratic behavior is mixed—see André Blais and Stéphane Dion (1991) for summaries of the first two decades of empirical tests. Clearly, the empirical assessment of the implications of the budget-maximizing bureaucrat is complicated by the need to make assumptions about the nature of oversight by the budgetary sponsor. The assessment becomes even more difficult when one follows Niskanen (1975) in allowing bureaucratic preferences to be over income and perquisites rather than just the budget. Indeed, Niskanen (1991) recognized that some bureaucrats may be zealots for the output of the bureau, and described himself as a zealot for efficiency when he worked in the bureaucracy.
Bureaucracy and Representative Government was an amazing contribution in light of the tools available at the time. Niskanen relied almost exclusively on neoclassical mircoeconomic analysis of a complicated relationship that he nonetheless saw as both strategic and inherently involving imperfect information, features that later were more naturally addressed with game theoretic tools by Jeffrey Banks (1989) and Banks and Barry Weingast (1992). As a student and practitioner of operations research, he was familiar with the original forms of game theory but once confided to me that he thought it had become a “dry hole” in terms of social science innovation. (Interestingly for someone so influential in the public choice movement, he also expressed misgivings to me about social choice theory, which he thought was “Byzantine.”) Nonetheless, he went a long way toward anticipating the key elements of agency theory, which has now become a common framing for bureaucracy.
I take the liberty of offering two additional personal observations about Niskanen. First, the sometimes harsh and cold logic of his public persona belied the warmth, gentility, and generousness I observed in his interaction with students. He was always open to discussing issues with students, even when they held very different views from his.
Second, he followed his beliefs even when they were costly to him personally. In the course he taught us on policy analysis, students asked him about how best to avoid ethical conflicts between doing good analysis and accommodating the preferences of clients. His response was, “Keep your bags packed.” A few years later, we saw him live by these words when he left his position as chief economist at Ford rather than prepare analyses to support import quotas on automobiles, which he viewed as bad public policy.
Banks, Jeffrey S. (1989) Agency Budgets, Cost Information, and Auditing. American Journal of Political Science 33(3), 670–699.
Banks, Jeffrey S. and Barry R. Weingast (1992) The Political Control of Bureaucracies under Asymmetric Information. American Journal of Political Science 36(2), 509–524.
Blais André and Stéphane Dion, eds. (1991) The Budget Maximizing Bureaucrat: Appraisals and Evidence. (Pittsburgh: University of Pittsburgh Press).
Migué, Jean-Luc and Gérard Bélanger (1974) Toward a General Theory of Managerial Discretion. Public Choice 17(Spring), 27–51.
Miller, Gary J. and Terry M. Moe (1983) Bureaucrats, Legislators, and the Size of Government. American Political Science Review 77(2), 297–322.
Niskanen, William A., Jr. (1991) A Reflection on Bureaucracy and Representative Government. In Blais André Blaise and Stéphane Dion, eds., The Budget Maximizing Bureaucrat: Appraisals and Evidence. (Pittsburgh: University of Pittsburgh Press), 13–31.
_(1975) Bureaucrats and Politicians. Journal of Law and Economics 18(3), 617–643.
_ (1971) Bureaucracy and Representative Government (Chicago: Aldine Atherton)
_ (1968) The Peculiar Economics of Bureaucracy. American Economic Review 58(2), 293–305.