Unemployment and the US housing market during the Great Recession [Paper] [Slides]
This paper evaluates the role of unemployment scarring and fear thereof for the recent U.S. housing bust. I study a quantitative lifecycle model of the housing market, which features an income process that is consistent with the large and long-lasting impact of unemployment on future earnings documented in recent empirical work. The model features exogenous moving shocks consistent with survey evidence which shows that many households move for reasons unrelated to their financial situation. These shocks reduce the selection into moving, thereby amplifying the quantitative importance of unemployment shocks and tighter credit conditions in the recent bust. The reason is that movers are more sensitive to labor market and credit conditions because they are younger, have lower wealth, and less secure jobs. Housing policies such as mortgage subsidies help stabilize prices and reduce foreclosures, even if only a small fraction of homeowners receive them.
Uncertainty aversion and heterogeneous beliefs in linear models [Paper] [Slides] (with Cosmin Ilut and Martin Schneider)
This paper proposes a simple perturbation approach for dynamic models with agents who differ in their perception of exogenous
shocks. The method characterizes linear dynamics around a steady state when that steady state differs from any individual agent's
long run expectation. It applies when agents agree to disagree, as well as when they differ in aversion to Knightian uncertainty
and hence behave as if they hold different worst case beliefs. It thus provides a simple way to study effects of uncertainty on behavior
in linear models. Our leading example looks at precautionary savings, asset premia and gains from insurance in a borrower-lender model with
agents who differ in uncertainty aversion.
Work in progress
Asset prices in a labor search model with confidence shocks [Slides]
I construct a labor search model with agents who are averse to ambiguity (Knightian uncertainty). Changes in ambiguity about future profits affect firms' hiring decisions and help the model explain
cyclical movements of output and unemployment. At the same time, ambiguity helps account for the mean, volatility and predictability of excess returns on equity.
Equity premium with endogenous unemployment risk [Slides] (with Zhengyang Jiang)
Idiosyncratic, uninsurable labor income risk primarily in the form of unemployment risk plays a large role in macro models with asset pricing. We study a heterogeneous-agents labor search model with portfolio
choice that links labor and asset markets. Unemployment risk gives rise to high equity premium mostly because the job finding rate is correlated with productivity. Additionally, there is a feedback from asset
pricing to job creation: high equity premium lowers incentives to post vacancies, which in turn lowers the job finding rate and gives rise to an even higher premium.
Protection for sale with unemployment [Draft]
To match the observed patterns of trade policy over the business cycle, I study a menu-auction model of trade policy in which financial frictions give rise to unemployment. Firms face idiosyncratic liquidity
cost shocks that force some of them to bankruptcy. Tariffs raise profits providing liquidity to overcome the shocks. Both equilibrium and optimal tariffs are hump-shaped in the size of the shocks: if shocks
are small, tariffs rise (as compared to the no shock case) to save employment; if shocks are large, tariffs fall to sweep away inefficient firms. This pattern also holds in a two-country extension of the model.