Optimal tax information with privacy concerns

Abstract Having more information about taxpayers facilitates tax enforcement, but generates utility losses due to privacy concerns. We posit that individuals put negative value on two distinct aspects of privacy: revealing their level of affluence, and revealing details of their income conditional on affluence. We characterize the enforcement-privacy tradeoff for optimal income tax policy by introducing the granularity of the tax base as a key policy variable. The first-order conditions reveal the sufficient statistics needed to describe the optimal tax rate structure and the extent of granularity. We discuss how future theoretical and empirical research could improve the characterization of optimal information.