Welfare Analysis Meets Causal Inference

We describe a frame work for empirical welfare analysis that uses the causal estimates of a policy’s impact on net government spending. This framework provides guidance for which causal effects are (and are not) needed for empirical welfare analysis of public policies. The key ingredient is the construction of each policy’s marginal value of public funds (MVPF). The MVPF is the ratio of beneficiaries’ willingness to pay for the policy to the net cost to the government. We discuss how the MVPF relates to “traditional” welfare analysis tools such as the marginal excess burden and marginal cost of public funds. We show how the MVPF can be used in practice by applying it to several canonical empirical applications from public finance, labor, development, trade, and industrial organization.

Should Benefit-Cost Analysis Include a Correction for the Marginal Excess Burden of Taxation?

This paper provides an overview of theoretical, empirical, and practical arguments in favor of or against a correction for the marginal excess burden of taxation (MEB). Benefit-cost analysis (BCA) should be used to compare the costs and benefits of a policy measure and its major alternatives, and whenever relevant, also to compare different ways of financing this. The best pragmatic approach is then to assume first that a policy measure is financed out of general tax revenues and then that the MEB of these taxes is broadly counterbalanced by the benefits of redistribution of these taxes. The latter assumption is consistent with the preferences for equality in a country’s current tax system. It is a simple and politically neutral assumption, and it implies that the marginal cost of public funds is equal to 1 and that no correction is needed in BCAs for the MEB. This shortcut assumption does not imply that the tax system is optimal or that BCAs should be distributionally weighted. Choosing an alternative source of financing, i.e., other than general tax revenues, should be regarded as a separate policy measure that should be analyzed separately in a BCA, including its distortionary and distributive effects.