Research

Abstract: This paper characterizes the effects of ambiguity aversion under dispersed information. The equilibrium outcome is observationally equivalent to a Bayesian forecast of the fundamental with increased sensitivity to signals and a pessimistic bias. This equivalence result takes a simple form that accommodates dynamic information and strategic interactions. Applying the result, we show that ambiguity aversion helps rationalize the joint empirical pattern between the bias and persistence of inflation forecasts conditional on household income. In a policy game à la Barro and Gordon (1983) with ambiguity-averse agents, the policy rule features higher average inflation and increased responsiveness to fundamentals.

Abstract: We show that households' private information on future income can be identified from the correlation between consumption growth and future income growth conditional on current income growth. Employing PSID data, we find that this conditional correlation is positive and significant. We use this evidence to structurally estimate a standard incomplete markets model and discover that US households possess enough advance information to reduce their income forecast errors by 15%. This significantly affects the measurement of consumption insurance. With advance information, 25% more income shocks pass through to consumption on average, and more than twice as much for the 5% asset poorest.

Abstract:  We study optimal fiscal policy to address climate change and inequality. We theoretically characterize optimal carbon and income taxes and quantify them for the US economy with a climate model calibrated to DICE. In contrast to the representative-agent setting, we find that (i) tax distortions have a negligible effect on the optimal carbon tax; (ii) inequality only slightly reduces it; (iii) the revenue from carbon taxes is optimally split halfway between reducing tax distortions and increasing transfers. Unlike the double-dividend policy, optimal carbon taxation has progressive welfare effects and low-income households benefit even in the short run.

Abstract: Individuals with higher income levels are less likely to declare bankruptcy and face higher unsecured credit limits. I present a theory of unsecured consumer credit in which: (i) the bankruptcy structure has direct counterparts in the actual US bankruptcy law; (ii) agents face a quantitatively relevant income process; and (iii) credit limits are increasing with income levels. In a simple example, where credit limits can be solved analytically, I show that if agents face the same standard income process, the default punishment is more severe to those who receive a negative shock. As a result, these agents declare bankruptcy less often which leads to a counterfactual prediction with respect to the correlation between credit limits and income levels. I propose a solution: if agents face heterogeneous income processes with different growth rates which they learn about during their lives, the agents more likely to default are the ones who are negatively surprised about what they learn. I, then, calibrate a quantitative life-cycle model to match properties of individual income processes and the US bankruptcy system. I find that the model matches untargeted correlations between income and credit limits on unsecured consumer credit only when the income processes are allowed to be heterogeneous with respect to their growth rates.

Abstract: With dispersed information, how much can agents learn from past endogenous aggregates such as prices or output? In a rational-expectations equilibrium, if general equilibrium effects are strong enough, aggregates no longer perfectly reveal underlying fundamentals. In this confounding regime, the effects of informational frictions are persistent over time, and the aggregate outcome displays an initial under-reaction followed by a delayed over-reaction relative to its perfect-information counterpart. In a standard New Keynesian model, we show that endogenous information aggregation helps bring the model predictions on aggregate forecasts closer to the data.

Abstract: We study optimal fiscal policy in a standard incomplete-markets model with uninsurable idiosyncratic income risk, where a Ramsey planner chooses time-varying paths of proportional capital and labor income taxes, lump-sum transfers (or taxes), and government debt. We find that: (1) short-run capital income taxes are effective in providing redistribution since the tax base is relatively unequal and inelastic; (2) an increasing pattern of labor income taxes over time mitigates intertemporal distortions from capital income taxes; (3) the optimal policy increases overall transfers, calibrated initially to the US welfare system, by roughly 50 percent; (4) two-thirds of the welfare gains come from redistribution and the remaining third come mostly from insurance; and (5) redistribution also leads to a more efficient allocation of labor via wealth effects on labor supply—lower productivity households can afford to work relatively less.

Abstract: We show that the equilibrium policy rule in beauty-contest models is equivalent to that of a single agent's forecast of the economic fundamental. This forecast is conditional on a modified information process, which simply discounts the precision of idiosyncratic shocks by the degree of strategic complementarity. The result holds for any linear Gaussian signal process (static or persistent, stationary or non-stationary, exogenous or endogenous), and also extends to network games. Theoretically, this result provides a sharp characterization of the equilibrium and its properties under dynamic information. Practically, it provides a straightforward method to solve models with complicated information structures.

*Previously circulated as "Infinite Higher Order Beliefs and First Order Beliefs: An Equivalence Result."