New in town: demographics, immigration, and the price of real estate

We link cross-sectional variation in both realized and expected state-level house price appreciation to cross-sectional variation in demographic changes. In particular, we extract two components of expected population growth: 1) a natural component due to predictable demographic changes related to fertility and mortality rates and 2) a non-natural component due to immigration. Our analysis shows that only the second component forecasts cross-sectional variation in state-level house price appreciation. We find that the sensitivity of both realized and expected returns to these demographic changes is stronger for states with greater population density, consistent with population growth actually causing the price appreciation rather than merely being correlated with some other phenomenon. We also document that building permits anticipate a portion of future population growth and house price appreciation. However, lagged measures of building activity do not subsume the ability of our expected immigration proxy to forecast price appreciation. Our findings are consistent with fundamentals driving an economically important portion of cross-sectional variation in state-level housing returns. However, markets appear to significantly underreact to the component of fundamentals that is arguably more di¢ cult for market participants to anticipate.