Conference Agenda

This paper revisits the problem of optimal public good (or public bad) provision under optimal non-linear income taxes, where the main methodological novelty is that we allow for heterogeneity in preferences and exposure to the public good, by using the perturbation approach. This generalization is shown to have crucial implications for the optimal provision rule, and in particular with respect to distributional concerns. The optimal provision rule is shown to deviate from the Samuelson rule in relation to the differences between cross-section and individual income elasticities of the marginal willingness to pay for the public good. This implies that, contrary to the conventional view, it is often optimal to take distributional concerns into account also in cost-benefit analysis, and thus not to delegate such concerns solely to the tax and transfer system.

We study theories of intergenerational justice that disentangle ethical attitudes toward discounting from ethical attitudes toward intergenerational inequalities. We call such criteria modular: each one is uniquely identified by a time-discounting function—capturing attitudes toward time—and a law-invariant aggregator—capturing attitudes toward inequality. We provide an axiomatic characterization of modularity and show its implications. Our framework nests the workhorse criteria used in applied work and yields large, tractable families of alternative criteria obtained by freely combining discounting and inequality attitudes.

Despite their desirable properties, carbon taxes have received low public support. Leading explanations attribute the rejection of carbon taxes to preferences and ideology. We show that voters make simple mistakes when reasoning about the consequences of such policies. We focus on leading versions of carbon taxes that are designed to draw wide public support by issuing dividend payments from collected revenues, resulting in a policy that is progressive and produces more winners than losers. Using experiments, we show that subjects have specific blind spots in reasoning about policies with taxes and dividends, even when described in simple terms and in incentivized, abstract settings that remove scope for preferences and ideologies. Based on these findings, we introduce a new policy that is isomorphic to a leading carbon tax proposal, but engineered to be robust to these cognitive frictions. In a U.S. survey, we find that our new policy (which we call "carbon penalty and reward") garners 30 percentage points more public support than the original proposal (known as "carbon fee and dividend"). Using independent measures, we show that this increase in support is concentrated in people who have difficulty reasoning about the policy structure.

We use a theoretical framework, a regression discontinuity design and a discrete choice model to evaluate the efficiency and distributional consequences of an environmental subsidy under imperfect competition. Our empirical setting is the largest housing retrofit subsidy in France (Ma Prime Rénov’), amounting to €3bn of annual public spending. The scheme splits eligible households into income groups, with subsidy level declining with income. Leveraging discontinuities in the heat pump subsidy schedule at income thresholds, we find that heat pump prices differ across income groups by 1.3 times the subsidy differential. Our theoretical framework identifies two mechanisms for this result, with opposing effects on relative prices: first, means-testing enables price discrimination across income categories. This adds to the progressivity of the scheme, provided demand elasticity is higher among low-income households. Secondly, the effect on prices of the subsidy differential itself can be attenuated by incomplete pass-through within each sub-market. Estimating a structural model of demand for heat pumps, we find that a relatively high redistributive effect of price discrimination and complete pass-through of the subsidy combine into higher progressivity of the means-tested scheme when discrimination is allowed, at low efficiency cost: the share of low income households in take-up increases by 23%, while the number of heat pump by €m of public spending decreases by 2%. Accounting for this mechanism in policy design is however non trivial, as the sign and magnitude of the effects depend on several demand and supply primitives: conduct parameter, elasticity and curvature of the demand curves in each income group.