Using a quantitative heterogeneous agents macro-housing model and detailed microdata, this paper studies the drivers of the 2006–2011 housing bust, its spillovers to consumption and the credit market, and the ability of mortgage rate interventions to accelerate the recovery. The model features tenure choice between owning and renting, rich portfolio choice, long-term defaultable mortgages, and endogenously illiquid housing from search frictions. The equilibrium analysis and empirical evidence suggest that the deterioration in house prices and liquidity, transmitted to consumption via balance sheets that vary in composition and depth, is central to explaining the observed aggregate and cross-sectional patterns.

This paper shows that a macro model with segmented financial markets can generate sizable movements in housing prices in response to changes in credit conditions. We establish theoretically that reductions in mortgage rates always have a positive effect on prices, whereas the relaxation of loan-to-value constraints has ambiguous effects. A quantitative version of the model under perfect foresight accounts for about one-half of the observed price increase in the United States in the 2000s. When we include shocks to expectations about housing finance conditions, the model’s ability to match house values improves significantly. The framework reconciles the observed disconnect between house prices and rents since, in general equilibrium, financial shocks can decrease rents and increase prices.