Publications
We develop a quantitative theory of business cycles with coordination failures. Because of demand complementarities and increasing returns, firms seek to coordinate production and multiple equilibria arise. We use a global game approach to discipline equilibrium selection and show that the unique dynamic equilibrium exhibits multiple steady states. Coordination on high production may fail after a large transitory shock, pushing the economy in a quasi-permanent recession. Our calibrated model rationalizes various features of the 2007-2009 recession and its recovery. Government spending, while generally harmful, can increase welfare when the economy is transitioning between steady states. Other policy instruments are preferable to fix miscoordination.
This paper explores whether rational herding can generate endogenous aggregate fluctuations. We embed a tractable model of rational herding into a business cycle framework. In the model, technological innovations arrive with unknown qualities, and agents have dispersed information about how productive the technology really is. Rational investors decide whether to invest based on their private information and the investment behavior of others. Herd-driven boom-bust cycles arise endogenously in this environment when the technology is unproductive but investors' initial information is unusually optimistic. Their overoptimism leads to high investment rates, which investors mistakenly attribute to good fundamentals, leading to a self-reinforcing pattern of higher optimism and higher investment until the economy reaches a peak, followed by a crash when agents ultimately realize their mistake. We calibrate the model to the U.S. economy and show that it can explain boom-and-bust cycles in line with episodes like the dot-com bubble of the 1990s.
We study optimal dynamic lockdowns against Covid-19 within a commuting network. Our framework integrates canonical spatial epidemiology and trade models, and is applied to cities with varying initial viral spread: Seoul, Daegu and NYC-Metro. Spatial lockdowns achieve substantially smaller income losses than uniform lockdowns. In NYM and Daegu—with large initial shocks—the optimal lockdown restricts inflows to central districts before gradual relaxation, while in Seoul it imposes low temporal but large spatial variation. Actual commuting reductions were too weak in central locations in Daegu and NYM, and too strong across Seoul.
Supplementary material: replication package, Vox column.
"Optimal Transport Networks in Spatial Equilibrium" Paper Slides Matlab Toolbox (joint with Pablo Fajgelbaum), July 2020, Econometrica, Vol.88, No.4, pages 1411-1452
We study optimal transport networks in spatial equilibrium. We develop a framework consisting of a neoclassical trade model with labor mobility in which locations are arranged on a graph. Goods must be shipped through linked locations, and transport costs depend on congestion and on the infrastructure in each link, giving rise to an optimal transport problem in general equilibrium. The optimal transport network is the solution to a social planner's problem of building infrastructure in each link. We provide conditions such that this problem is globally convex, guaranteeing its numerical tractability. We also study cases with increasing returns to transport technologies in which global convexity fails. We apply the framework to assess optimal investments and inefficiencies in observed road networks in European countries.
Supplementary material: Online Appendix, Supplementary Material, Replication files, Application to US
Julia Implementation of the toolbox by Sebastian Krantz
We develop a theory of endogenous uncertainty and business cycles in which short-lived shocks can generate long-lasting recessions. In the model, higher uncertainty about fundamentals discourages investment. Since agents learn from the actions of others, information flows slowly in times of low activity and uncertainty remains high, further discouraging investment. The economy displays uncertainty traps: self-reinforcing episodes of high uncertainty and low activity. While the economy recovers quickly after small shocks, large temporary shocks may have long-lasting effects on the level of activity. The economy is subject to an information externality but uncertainty traps may remain in the efficient allocation. Embedding the mechanism in a standard business cycle framework, we find that endogenous uncertainty increases the persistence of large recessions and improves the performance of the model in accounting for the Great Recession.
Supplementary material: Online Appendix
"Uncertainty and Unemployment" Paper, November 2017, Econometrica, Vol.85, No.6, pages 1675-1721 (previously circulated as "Uncertainty, Productivity and Unemployment in the Great Recession")
This paper studies the impact of time-varying idiosyncratic risk at the establishment level on aggregate unemployment fluctuations and on the labor market over the period 1972-2009. I build a tractable search-and-matching model of the labor market with firm dynamics and heterogeneity in productivity and sizes, in which I introduce time-varying idiosyncratic volatility. The model features directed search and allows for endogenous separations, entry and exit of establishments, and job-to-job transitions. I show, first, that the model can replicate salient features of the behavior of firms at the microeconomic level. Second, I find that the introduction of time-varying idiosyncratic volatility improves the fit of search-and-matching models for a range of business cycle moments. In a series of counterfactual experiments, I then show that time-varying idiosyncratic risk is important to account for the magnitude of fluctuations in aggregate unemployment for past US recessions, including in particular the recessions of 1990-1991 and 2001. Though the model can account for about 40% of the total increase in unemployment, uncertainty alone does not seem sufficient to explain the magnitude and persistence of unemployment during the 2007-2009 recession.
Supplementary material: Online Appendix
Working papers
We study how political preferences shaped California's High-Speed Rail (CHSR), a large transportation project approved by referendum in 2008. Across census tracts, support for the project responded significantly to the projected economic gains at the time of voting, as measured by a quantitative model of high-speed rail matched to CHSR plans. Instrumenting with random station placements on feasible routes, we estimate that 0.1%-0.2% projected economic gains swayed 1% of votes at the median tract. Given this elasticity, a revealed-preference approach comparing the CHSR with counterfactual designs identifies strong policymakers' preferences for political support. A politically-blind planner would have placed the stations nearer to California’s dense metro areas, doubling the projected economic gains.
"Time-to-build, Echoes and Delays in Production Networks" Paper (joint with Mathieu Taschereau-Dumouchel). February 2025.
We study how time-to-build and delivery lags affect the propagation of sectoral and aggregate shocks in an economy with input-output sectoral linkages. Time-to-build significantly contributes to the persistence of shocks and its effects are highly heterogeneous across sectors. We study delay shocks and show that bottleneck sectors for dynamic propagation can be identified by the product of their supplier and buyer centralities. Shocks propagate asynchronously through the network and generate echoe effects and endogenous fluctuations. We show that endogenous fluctuations are due to the presence of loops in the network (ie.e., direct and indirect cycles) and characterize the Fourier spectrum of the economy as a function of the network's dominant cycle lengths and weights. Dynamic sectoral comovements are complex and can be decomposed into the network's dominant paths.
"Zipf's Law and Fractal City Networks" Draft coming soon
This paper revisits the celebrated Zipf's law for city-size distributions from the lens of optimal transport networks. The theory first establishes that a fractal city network with population mobility can naturally lead to a city-size distribution that follows a power law with a coefficient equal to minus the network's fractal dimension. We then provide conditions under which a) fractal city networks are optimal and b) the fractal dimension is close to 1. The theory is empirically tested using variation across US MSA's and highly disaggregated population and commuting flow data.
"Aggregate Demand and the Dynamics of Unemployment" Paper (joint with Mathieu Taschereau-Dumouchel), May 2019
We introduce an aggregate demand externality into the Mortensen-Pissarides model of equilibrium unemployment. Because firms care about the demand for their products, an increase in unemployment lowers the incentives to post vacancies which further increases unemployment. This positive feedback creates a coordination problem among firms and leads to multiple equilibria. We show, however, that the multiplicity disappears when enough heterogeneity is introduced in the model. In this case, the unique equilibrium still exhibits interesting dynamic properties. In particular, the importance of the aggregate demand channel grows with the size and duration of shocks, and multiple stationary points in the dynamics of unemployment can exist. We calibrate the model to the U.S. economy and show that the mechanism generates additional volatility and persistence in labor market variables, in line with the data. In particular, the model can generate deep, long-lasting unemployment crises.
"Learning to Coordinate" Slides (joint with Mathieu Taschereau-Dumouchel), June 2014
We study a class of dynamic global games where agents learn from both exogenous and endogenous sources of information. Because endogenous information sources such as quantities and prices aggregate private information in a non-linear fashion, the amount of information provided in each period varies with the outcome of the coordination game. We show that a particular type of information cascades arises in this context. The more similar are the actions taken by agents, the less informative endogenous signals become. As a result, the economy may display bubble-like behavior with exuberant periods of economic activity followed by brutal crashes; as well as slow recoveries from crises, as agents take time to learn the new fundamentals.
"Optimal Redistributive Policy in a Labor Market with Search and Private Information" Paper (joint with Mathieu Taschereau-Dumouchel), October 2012
We study the design of optimal policies in a frictional model of the labor market with private information about skills. Heterogeneous, risk-averse agents look for a job in a labor market characterized by an aggregate matching technology. Firms post vacancies but cannot observe each agent’s productivity. This paper emphasizes the importance of general equilibrium effects in policy design and focus on the non-observability of workers’ underlying skills as the main information friction. Our mechanism design approach shows that the constrained optimal allocation can be implemented by policy instruments such as a non-linear tax on wages, a non-constant unemployment insurance and firm subsidies. We calibrate our model to the US economy and characterize the welfare gains from the optimal policy and its effects on output, employment and the wage distribution. Our findings suggest that the optimal policy under a utilitarian government features a negative income tax, a more generous unemployment insurance for low-skilled workers and higher marginal tax rates, which results in a higher participation in the labor market and a lower unemployment rate. This paper also shows that a government with a higher taste for redistribution would favor policies with more European characteristics: heavier taxation and more generous unemployment insurance, which result in a lower output and slightly higher unemployment rate.