The efficiency of the supply of non-market output; the surplus maximizing bureaucrat under no versus perfect price discrimination

More than twenty years ago (1971), William A. Niskanen completed Bureaucracy and Representative Government. The central hypothesis in the book was that bureaucrats act to maximize their budgets, because it is in their personal interest to do so, and that the strategies they adopt contribute substantially to the overall growth of the state. Since 1971 the most instructive and far-reaching criticism of Niskanen's book has been that it fails to integrate the legislature into its formal framework. Niskanen (1991) acknowledged the validity of this criticism and modified his ideas, but did not present them in a formal model. He recognized that the legislator is not passive and that he will play an active role in the bargaining process with the bureaucrat. He also modified his thesis about the behavior of the bureaucrat.The bureaucrat, according to Niskanen (1991), would act to maximize the bureau's "discretionary budget" (surplus). Niskanen, however, admitted that he had "$\...$not worked out the full implications of this assumption for the budget-output outcomes of the bargaining process with a surplus-maximizing bureaucrat and, to (his) knowledge, neither (had) any other scholar.$\...$.This issue, however, is not yet resolved and should be the highest priority for subsequent research" (Niskanen, 1991, p. 23). The main goal of the dissertation is to work out the budget-output outcomes of the bargaining process between the legislator and the bureaucrat for the surplus-maximizing bureau. The legislator-bureaucrat relationship will be modeled as a bilateral monopoly, employing already well-developed methods in microeconomics and game theory. In the model it will be assumed that the legislator and the bureaucrat collude to maximize the bureau's surplus. This surplus will then be spent in ways that serve the interest of the bureaucrat and the political authorities. In summary the main results suggest that the surplus-maximizing legislator/bureaucrat, as described by Niskanen 1991, will generally seek a budget that is too large, the output will never be oversupplied, and will only be supplied in the "correct" quantity, as compared with a competitive solution and Pareto optimality, if the surplus-maximizing legislator/bureaucrat can engage in both perfect output price and factor cost discrimination. Furthermore, their production methods will, as long as the legislator/bureaucrat maximize joint surplus, generally be efficient and the difference between the minimum cost of providing the bureaucratic services and the budget may rather represent a welfare transfer from the legislator's constituents to the surplus-maximizing agents than a uniform inefficiency predicted by Niskanen (1991).