Intergenerational Transfer of Human Capital and Optimal Education Policy

We study the design of education policies (subsidies and public education) when parents' investment in education is motivated by warm‐glow altruism and determines the probability that a child has a high ability. The optimal subsidy is not necessarily positive. It is determined by two conflicting terms: a Pigouvian term (warm‐glow altruists do not properly account for the impact of education on future generations) and a “paternalistic” effect (the warm‐glow term may not be fully included in social welfare). Finally, total crowding out of private expenditure (for one of the types) by public education may be desirable.