Interest Rates and Non-Fundamental Fluctuations in Home Values

Fluctuations in housing prices are relevant to wealth accumulation, labor mobility, consumption, macroeconomic volatility, and financial market stability. However, it is ex ante difficult to know when housing price movements are due to fundamentals, such as changes in the user cost of capital, versus irrational exuberance. I propose combining the canonical urban economics Alonso-Muth-Mills model and Poterba (1984, 1990) asset pricing equation to form grounded theoretical expectations about the impact of changes in the user cost of capital on home values. In this framework, I show rental prices and rental expenditures to be endogenous to interest rates, which limits the applicability of conventional price-to-rent ratios. Concretely, rental prices should decrease when interest rates go down. Using the model I can express expected changes in home values that result from changes in user costs as simple functions of the demand elasticity for space, the supply elasticity of housing, and the initial share of land relative to prices in a city. The simple formula can be used to diagnose and underwrite home valuations under the null hypothesis of a common shock in the user cost of capital and to benchmark the plausibility of local deviations from the theory based on idiosyncratic demand shocks. Empirically, I find that housing supply elasticities (as calculated by Saiz, 2009) and land shares as of 1990 predicted 50% of the variance in price growth during the boom period (2000-2005). Expected relative home price growth, based on the fundamental parameters, was largely consolidated and insensitive to the subsequent bust cycle (2005-2008). However, deviations from expected growth mean-reverted dramatically during the bust period.