The Marginal Value of Public Funds in a Federation

Our objective is to establish a new framework for quantifying the welfare effects of fiscal policies in an open economy, with an emphasis on state and local governments in a federal system. To do this, we develop a model of fiscal policy with benefit-spillovers, firm/household mobility, congestion effects, and capitalization resulting from changing taxes and expenditures among interacting local jurisdictions. We then propose using the marginal value of public funds (MVPF) as a way of capturing the welfare effects of policies that induce interjurisdictional mobility and spillovers. Because a federal planner internalizes interjurisdictional externalities, the MVPF for a federal and local planner can diverge substantially, and we derive theoretical conditions for when the local MVPF will be the larger or the smaller. We provide guidance on the additional empirical components of the marginal value of public funds necessary to understand the welfare effects of policies in a federal system. Although the formulas for the MVPF appear more complex, we derive the necessary assumptions to use house price capitalization as a sufficient statistic to estimate both the federal and local MVPF. We illustrate the key differences of social and local MVPFs using four empirical applications: decentralized wealth taxation, bidding-for-firm subsidy competition, higher education and K-12 educational spending. The examples quantify why local governments may not implement policies preferred by a federal planner, and instead adopt ones that have lower social value