Optimal Structure of Local Governments
The preceding chapters-most notably Chapter2-have studied whether an exogenously fixed number of mobile individuals in a federal state is allocated efficiently across a politically predetermined number of local jurisdictions. What the previous analysis has left out of consideration is determining the optimal size of individual jurisdictions or, in other words, the optimal number of individuals living in these jurisdictions. Of course, such questions concerning the optimal governmental structure in a federal state cannot be ignored when analyzing the problem of whether fiscal decentralization secures an optimal allocation. As long as utility of individuals can be increased by a restructuring of jurisdictional boundaries, there is scope for a Pareto improvement. There are several empirical examples of such a restructuring, indicating that costs and benefits of changes in jurisdictional boundaries are on the agenda of the political debate in federal states. For instance, the number of local governments in Germany was reduced significantly during the 1970s, and Henderson (1985) indicates that most American cities have grown through annexation. There is also an ongoing debate in Germany of integrating some small Bundesländer into a larger jurisdiction in order to reduce costs in the public sector.1 It is therefore necessary to include another dimension in the characterization of an optimal allocation. We will speak of an optimal allocation in this chapter when, in addition to the efficient supply of local public goods and the efficient distribution of a certain number of individuals across jurisdictions, each jurisdiction has its optimal population size and thus there is an optimal number of jurisdictions in the federal state.
There are several important directions of research trying to determine the optimal structure of local governments in a federal state. In the original world of Tiebout (1956), there were no locationally fixed factors like land and hence____________________