Budget makers as agents: A preliminar investigation of discretionary behavior under state-contingent rewards

Abstract

Two classes of hypotheses purport to explain government spending. The conventional explanation is that spending outcomes are somehow determined by the supply of and the demand for publicly provided goods and services. Under the conventional explanation, competition either for office or through some kind of Tiebout mechanism constrains elected officials to serve the tastes and preferences of the electorate perhaps those of the median voter. Presumably, budget makers are merely loyal puchasing agents, perfectly responsive to the interests of their political masters.1 The alternative explanation is largely due to John Crecine and his students. 2 It derives from the behavioral approach of Herbert Simon, Richard Cyert, James March, and others and relies on a distinctive methodology computer simulation, based on models developed from carefully constructed decisionmaking protocols. But its most important distinguishing attribute is the explanatory role it assigns to the budget maker. As Crecine (1969: 20) explains: