Optimal Non-linear Taxation and the Design of Education Policy

optimal_tax.pdf431 KB


This paper studies the design of education policy in an optimal non-linear tax model with asymmetric information. It shows that both heterogeneity in ability and risky human capital investment (or the combination of the two) can provide a theoretical justification for government intervention in education. The sign of the optimal policy is exclusively determined by the Hicksian coefficient of complementarity. Specifically, when education increases (decreases) exposure to risk, or equivalently, when the wage elasticity of education is increasing (decreasing) in ability, the optimal policy is to tax (subsidise) education. But when heterogeneity and risk are combined, the sign of the optimal policy is indeterminate. Numerical results suggest that the magnitude of the optimal policy will depend on the strength of the insurance and redistributive motives. Income-contingent loans or educationdependent taxes and subsidies can implement the optimum.
Last updated on 07/11/2021