Choosing between the Median-Voter and Niskanen Models : An Empirical Approach

概要

論文の詳細を見る
There are two classes of models that seek to explain the level of pubic expenditure : models that depend on the responsiveness to voter preferences (such as the median-voter model) and models that are driven by the objectives of public officials (such as the Niskanen model of bureaucratic behavior). Studies that try to explain empirically levels of public expenditure typically adopt one of these two types of models. This paper develops an expanded model in which both of these influences are present. Instead of an equilibrium that is either a median-voter outcome or a Niskanen outcome, the model has a continuum of potential outcomes in which the median-voter and Niskanen results are the polar cases. Budgetary equilibrium in the model is the result of a two-person cooperative game between politicians aiming at vote-maximization and public officials with the objective of budget maximization. The outcome is an asymmetric Nash bargaining solution between these two classes of "players." The paper sets forth a method by which this model can be estimated from actual budgetary data. This produces (among other things) an estimate of a parameter that describes the relative bargaining power or influence of the two classes of agents. It thus indicates which of the two objectives-budget maximization or vote maximization-is the more influential in determining budgetary outcomes.
関西学院大学の論文
1997-12-00