Mandatory or Voluntary? Optimal Public Good Funding

I study an economy where a government and heterogeneous individuals can dedicate part of their resources to the funding of public goods. What is the optimal combination between government and voluntary contributions? When individuals make different donation choices only because they have different income, relying on voluntary contributions is Pareto-improving. This is true independently of the efficiency and redistributive properties of the public goods to be funded. When donation choices are influenced by unobserved characteristics (e.g., altruism, ideology), I provide simple policy rules for deciding whether governments or private contributions should be the marginal funding source, based on the price elasticity of giving and the redistributive preferences of the government. In particular, I show that the commonly applied unit-elasticity rule can align with Rawlsian preferences. I study how these policy rules adapt in presence of a market failure (leaky donations) and a government failure (uniform tax subsidies for different causes). Numerical simulations using French data reveal that limiting tax discrimination across causes significantly increases government funding, as direct government contributions compensate for the lack of flexibility of the tax system. These funding rules apply for both optimal and arbitrary provision of public goods.