Agglomeration, tax competition and local public goods supply

The purpose of this paper is to develop a framework for analysing local public goods supply and tax competition between jurisdictions in a context where there are gains from geographic agglomeration and where labour is imperfectly mobile. We capture agglomeration gains in the simplest possible manner, by assuming that individuals consume a bundle of locally produced differentiated products. Because consumers value variety, and the range of products available will be larger the larger the local market, this creates agglomeration gains. These will be reinforced if there are economies of scale in the supply of goods provided by local authorities – i.e. if local authorities provide pure public goods or private goods with scale economies. The agglomeration forces are counteracted by residential preferences. We assume that individuals differ in their residential preferences; i.e. the mobility of individuals differ. In the paper we use the framework to look at a two-community equilibrium, and focus on the case in which there is a stable interior equilibrium (with settlement in both communities). In an interior equilibrium, each community will gain by attracting new residents. The communities will therefore compete for residents. The instruments available are publicly provided goods and local tax rates. The result is a systematic bias towards overprovision of publicly provided goods that the most mobile indivi-duals value more highly than the less mobile ones, and towards underprovision of goods with the opposite characteristics. Whether or not there will be a bias towards lower tax rates depends on whether the willingness to pay for the average publicly provided good increases or decreases with the mobility of individuals. What is new in this paper compared to models with pure fiscal externalities is that the results hold even if there are no economies of scale in the publicly provided goods. As most goods provided by local authorities are of that kind, we feel that our model is a more meaningful framework for analysing the competition between communities.