Spatial effects and house price dynamics in the USA

While an understanding of spatial spillovers and feedbacks in housing markets could provide valuable information for location decisions, little known research has examined this issue for the US Metropolitan Statistical Areas (MSAs). Also, it is unknown whether there can be differences in the spatial effects before and after a major housing “bust”. In this paper we examine spatial effects in house price dynamics. Using panel data from 363 US MSAs for 1996 to 2013, we find that there are significant spatial diffusion patterns in the growth rates of urban house prices. Lagged price changes of neighboring areas show greater effects after the 2007-08 housing crash than over the entire sample period of 1996-2013. In general, the findings are robust to controlling for potential endogeneity, and for various spatial weights specifications (including contiguity weights and migration flows). These results underscore the importance of considering spatial spillovers in MSA-level studies of housing price growth.