Accounting for private information

The realization of uncertainty faced by workers over the lifecycle can be either public or private information. This has important consequences on the degree of risk sharing and ultimately on the degree of lifetime inequality. Recent literature in public finance has focused on the implications of private information shocks on efficient allocations. In this paper we quantify the importance of private information in accounting for inequality. In particular we determine how much private information is needed to account for the distribution of income and consumption in the US. To answer this question we explore the quantitative implications of a class of economies with informational frictions. We consider two dynamic models: a standard Mirrlees economy with exogenous skill shocks and a taste shock economy with endogenous labor supply. In both economies the shocks have a component that is privately observable and another that is publicly observable. We solve the informational constrained problem for these environments and calibrate the degree of private information necessary to match the distributions of consumption, income and hours in the US. In both these models private information shocks introduce two important implications on lifetime consumption and income. They introduce an endogenous fanning out of continuation utilities. This translates in consumption and income inequality rising over time without any increase in the variance of the shocks. The presence of private information also reduces the consumption response to income innovations with respect to the full information case. In models with incomplete markets this response is large when compared with the data. Preliminary results indicate that in both models we need a large fraction of the shocks being public information to be consistent with features observed in the data.