Estimating the Social Marginal Cost of Public Funds : A Micro-data Approach

This study proposes a method to estimate the social marginal cost of public funds (SMCF) that utilizes household data. First, we recast the existing SMCF formula to highlight the role of individual marginal costs of public funds (IMCF) and their respective distributional weights. Second, we elaborate on aspects of the distributional weights. In particular, we derive the weights from a utility function with constant-elasticity substitution and a social welfare function with constant-inequality aversion, and compare them to the conventional distributional weights that are frequently used in distributional cost-benefit analysis. Third, using a Japanese micro-data set, we estimate the wage elasticity of labor supply and the corresponding IMCFs on a household basis. This then not only enables us to examine how the elasticities and the IMCFs are distributed but also allows us to explore how they are related to income level and to see how valid the assumptions made in the previous studies are. Lastly, as examples, we use the SMCF estimates to examine the social desirability of the progressive tax system in Japan as compared to an optimized flat tax