Policy with Dispersed Information

This paper studies optimal taxation in a class of economies in which agents have dispersed private information regarding aggregate shocks (commonly-relevant fundamentals such as aggregate productivity and demand conditions). The dispersion of information opens the door to inefficiencies that need not have obtained otherwise and that may manifest themselves in excessive non-fundamental volatility (overreaction to common noise), excessive cross-sectional dispersion (overreaction to idiosyncratic noise), or suboptimal social learning (low quality of information contained in financial prices, macroeconomic data, and other indicators of economic activity). In either case, a novel role for policy emerges. We thus seek to identify policies that permit the government to manipulate how agents utilize their various sources of information, without requiring the government to monitor these sources of information. Our key result is that this can be achieved by appropriately designing the contingency of marginal taxes on aggregate activity. This contingency is not essential when agents have private information regarding only idiosyncratic shocks, but becomes essential once agents have private information regarding aggregate shocks. It permits the government to control the reaction of the economy to noise, as well as to improve the quality of information in prices and macro data.