Federalism--by which I mean more than federal-state relationships, and in particular mean all relationships between relatively centralized governments and their relatively decentralized subdivisions--allows different jurisdictions to compete for residents, both firms and individuals. Those who embrace federalism point to the benefits it confers by allowing jurisdictions to vary the bundle of goods and services they offer when the contents of that bundle generate only intrajurisdictional effects. The underlying assumption is that federalism induces subnational jurisdictions to attract residents by offering a preferred public good at a particular tax price. (1) The resulting competition, in theory, has many benefits. It maximizes preference satisfaction, as individuals migrate to jurisdictions that offer the public goods that they desire. It allows efficient delivery of local public goods as competitive markets for residents both drive down monopoly tax prices that public entities might otherwise be able to demand and provide public officials signals of desired services. It reduces the size of government because the resulting homogeneity (1) reduces the costs of monitoring officials who are charged with providing public goods consistent with residents' preferences and (2) represses logrolling among groups with diverse interests that might otherwise divert public resources to private use. And it diminishes corruption within government because that same homogeneity causes expenditures for illicit activity--which almost by definition deviates from majoritarian preferences--to be more salient.
A corollary of these characteristics of federalism, and perhaps the primary objective of federalism itself, is that interjurisdictional competition simultaneously acts as a bulwark against strong centralized governments. (2) Allowing potential residents to migrate to jurisdictions that offer desired goods and services and that are motivated to limit expenditures on rent-seeking activities obviously serves liberty interests that otherwise could be quashed by centralized governments that dictated expenditures over a broader jurisdictional scope. (3)
Apart from pure liberty interests, federalism with these characteristics solves the puzzle that arises once one recognizes that any government strong enough to protect the contract and property rights essential to the functioning of well-ordered markets also can interfere in the operation of those markets. Federalism, in Weingast's useful terminology, becomes "market-preserving" where decentralized governments retain substantial autonomy over the economy within their sphere. (4) Meanwhile, the national government ensures the flow of goods across jurisdictions through centralized doctrines such as a negative commerce clause. (5)
The conditions under which we would be fully able to achieve these objectives are highly stylized, as has been well-rehearsed since Charles Tiebout first articulated them. (6) Perfect sorting of the type ideal federalism requires entails the presence of individuals who possess substantial mobility and minimal attachments to their residences for reasons other than the provision of public goods. (7) Potential residents must have substantial choice among a wide array of jurisdictions whose proffered bundle of goods and services is relatively well recognized and imposes no external effects on other jurisdictions. (8) Thus, if we posit a population of omniscient automatons who live on dividend income and have no interpersonal relationships, the theory could work pretty well.
But even with imperfect satisfaction of the conditions, competition among relatively centralized governments in a federal system promises certain benefits. There is at least some empirical evidence that lower taxes and superior services in some jurisdictions produce efficiency-generating responses in the tax rates and services of neighboring jurisdictions and that the presence of multiple jurisdictions in a metropolitan area constrains the size of government. …