Bureaucracy and the ?publicness? of local public goods

Abstract

In two seminal papers in the early 1970s, Borcherding and Deacon (1972) and Bergstrom and Goodman (1973) (hereafter referred to as BDBG) attempted for the first time to estimate the exhaustibility (or rivalness) characteristics of local public goods. Their method was based upon the examination of the effects of population change on the expenditure patterns of cities. The use of this method, however, has almost uniformly resulted in the counterintuitive conclusion that local public goods are actually pure private goods with respect to their crowding behavior. Gonzalez and Mehay (1985), for example, note that in addition to the original two articles, at least six subsequent studies have come to this conclusion. Borcherding, Bush, and Spann (1977) have argued that bureaucracy is to blame for these results. Specifically, they suggest that, in order to increase their budgets, bureaucrats may use their power to alter the composition of local government spending in favor of goods and services toward the private good end of the local public good spectrum. In addition, BBS argue that if bureaucratic power is correlated with population across cities, the BDBG methodology will be biased in the direction of pure private goods. Under such a correlation, part of the extra spending in larger cities which is erroneously attributed to greater population by the BDBG approach is actually due to greater bureaucratic power in those larger cities. In a more recent paper, Gonzalez and Mehay (1985) have suggested that the presence of bureaucracy completely invalidates the BDBG approach, necessitating an alternative and less precise methodology to estimate the publicness of local public goods.In this paper I use a new, more exact model of bureaucratic influence to argue that bureaucracy cannot be responsible for the anomalous results found in the empirical literature. The paper demonstrates that: (1) if bureaucracies can alter the composition of spending in a community, they will push for local public goods which are close to the pure public good, rather than pure private good, extreme; and (2) if bureaucratic power is correlated with population, the estimated results will be biased toward the pure public good extreme. The purpose of the paper, then, is to divert study on the question of these empirical results away from bureaucracy into more fruitful lines of research.Section 1 details the model of bureaucracy used to examine these questions. Sections 2 and 3 contain proofs of the propositions listed above. Section 4 dis-

A politico-economic approach to intergovernmental lump-sum grants

“…This result was initially stated in the seminal median voter papers of Borcherding and Deacon (1972) and Bergstrom and Goodman (1973) and endorsed by other authors who developed bureaucratic approaches: Gonzalez and Mehay (1985) and Wyckoff (1988). Here, population is a rough indicator of "needs".…”

Governmental behavior in representative democracy: a synthesis of the theoretical literature

“…Third, at the local level, larger cities tend to be central cities rather than suburbs. Consequently, as the population grows, it should be more difficult for the median voter to move to a similar community (Wyckoff 1988a). Hence, the explanatory power of the median voter model should be better in small democracies and in lower level governments (see also Turnbull andMitias 1999 andJosselin et al 2008).…”

“…They focused on the alternative hypothesis that the bureau is interested in organizational slack (also referred to as discretionary rent or fiscal residuum), defined as the difference between total revenue and the total cost of production. Wyckoff (1988a) formalizes Migué and Bélanger's (1974) theory (see also Gonzalez and Mehay 1985). The government is assumed to maximize the difference between the total revenue and the cost of producing the public good:…”

Cost Efficiency in the Provision of Educational Services: An Application of Data Envelopment Analysis

“…Thus, bureaucrats have an incentive to provide the level of outcomes that yields the highest budget rather than the level that minimizes cost. Niskanen (1975) extended his original model to allow maximization of budgetary slack, i.e., the portion of the budget in excess of the cost minimizing level (see also Migut and Btlanger, 1974;Wyckoff, 1988). If either of these models of bureaucratic supply is correct, then estimated public sector cost equations relying on cost minimization will confound cost with budgetary slack, leading to potential biases in the estimated cost function.…”