Do Credit Conditions Move House Prices?

Did credit drive the 2000s housing cycle? The existing literature’s findings range from credit having no effect to credit explaining most of the cycle. We show that these disparate results hinge on the extent to which landlords absorb credit-driven demand, which depends on the degree of housing market segmentation. We develop a model that nests cases between the extremes of no segmentation and perfect segmentation typically considered, estimate an elasticity that pins down the degree of segmentation, and use it to calibrate our model. We find credit standards played an important role, explaining 32 percent to 53 percent of the boom. (JEL E32, E44, G51, R21, R31)