Eduardo Dávila (Harvard University): Optimal Financial Transaction Taxes:
This paper characterizes the optimal linear financial transaction tax in an equilibrium model of competitive financial markets. When belief disagreement induces excess trading on assets in fixed supply, two main results arise. First, the optimal tax is positive: although a (small) transaction tax discourages all trades equally, the reduction in fundamental trading creates a second-order welfare loss, while the reduction in non-fundamental trading creates a first-order gain. Second, the cross- sectional covariance between investors’ beliefs and investors’ equilibrium portfolio tax sensitivities becomes the relevant sufficient statistic for the optimal tax, which does not depend on the actual payoff distribution. I find additional results. First, in dynamic environments, controlling for the level of static disagreement, the optimal tax is lower when investors alternate between being buyers and sellers over time. Second, when financial markets determine production in a Walrasian sense, as in a q-theory environment, a marginal tax increase creates an additional first-order distortion (positive or negative). Third, when financial markets determine production by diffusing information, a marginal tax increase creates an additional first-order loss, due to a learning externality.