Working Papers

Endogenous Uncertainty and Credit Crunches (with Ludwig Straub), accepted at Review of Economic Studies.

We propose a theory of endogenous uncertainty caused by financial distress. Because uncertainty reinforces financial distress, temporary shocks can persistently cut off firms from external funding.


Robust Predictions for DSGE Models with Incomplete Information (with Ryan Chahrour), AEJ: Macroeconomics, 15(1), January 2023, pp. 173–208.

We establish bounds on the contribution of information shocks to aggregate fluctuations. The bounds are robust across all information sets agents may have. When agents care about micro and macro fundamentals, information shocks account for up to 51% of the U.S. business cycle. When agents only care about the latter, this figure drops to 3%.

Mismatch Cycles (with Isaac Baley and Ana Figueiredo), Journal of Political Economy, 130(11), November 2022, pp. 2943–2984.

We build a business cycle model with directed job search, multidimensional sorting, and learning about skills. During recessions, highly mismatched jobs are destroyed but also created. The predictions are consistent with evidence from the NLSY79 and O*NET.

A Quantitative Theory of Political Transitions (with Lukas Buchheim), Review of Economic Studies, 87(4), July 2020, pp. 1726–1756.

We develop a quantitative theory of repeated political transitions caused by revolts and reforms. The model generates a process of political transitions that looks remarkably close to the data.

Dynamic Oligopoly Pricing: Evidence from the Airline Industry (with Caspar Siegert), International Journal of Industrial Organization, 71, July 2020.

We explore how pricing dynamics in the European airline industry vary with competition and customer heterogeneity. The documented patterns are consistent with intertemporal price discrimination.

Optimal Delegated Search with Adverse Selection and Moral Hazard, Theoretical Economics, 11(1), January 2016, pp. 253–278.

I study a model of delegated search subject to two frictions: Only the agent knows the ex-ante distribution of search revenues, and search is non-contractible. Second-best search is too short and is implemented by a menu of simple bonus contracts.