Fiscal Inequivalence, Incentive Equivalence and Pareto Efficiency in a Decentralized Urban Context

We develop a simple urban model consisting of a center and a suburb, populated by a fixed number of identical individuals who can migrate costlessly between jurisdictions. Land ownership ranges from rent sharing to an absentee landlord who owns all urban land. We discuss laissez-faire equilibrium and show that it is Pareto efficient. We then introduce local governments which are only concerned about the welfare of their residents, and which can collect head and land taxes. Whereas the former is purely local, the latter will extend to both jurisdictions (if individuals own a positive share of the entire urban land) and beyond (if there is an absentee landlord). This produces a simple kind of fiscal inequivalence—tax-exporting. Since local governments do not provide public goods, their role is limited to stopping the flow of resources leaving their jurisdiction. They then redistribute this revenue as a head subsidy to the residents of their jurisdiction. If it just happens that the head subsidy is equal between jurisdictions, the equilibrium is both Pareto efficient and a preferable to laissez-faire from the viewpoint of urban residents. If however, it varies between jurisdictions, the equilibrium is Pareto inefficient: the jurisdiction providing a higher per head subsidy becomes overpopulated. We thus arrive at a classic market failure in the literature of competing regions. Correction involves aggregate interjurisdictional transfers to equalize head subsidies which, in the literature, must be undertaken by a central government. Following Myers [J. Public Econ., 43, 107–121 (1990)], we treat this as an assumption rather than a necessity. We show that if aggregate transfers become a local instrument, the Nash equilibrium will be both Pareto efficient and preferable to laissez-faire. This happens because free mobility compels noncooperative local governments to act in a cooperative way. We then extend this result to public goods which need be neither purely public (congestion) nor purely local (spillovers). We conclude by discussing implications for the existing urban literature.