57 Pages Posted: 15 May 2007
Date Written: May 2007
The key stylized facts of the housing market are positive serial correlation of price changes at one year frequencies and mean reversion over longer periods, strong persistence in construction, and highly volatile prices and construction levels within markets over time. We calibrate a dynamic model of housing in the spatial equilibrium tradition of Rosen and Roback to see whether such a model can generate these facts. With reasonable parameter values, this model readily explains the mean reversion of prices over five year periods, but cannot explain the observed positive serial correlation at higher frequencies. The model predicts the positive serial correlation of new construction that we see in the data and the volatility of both prices and quantities in the typical market, and it can account for substantial variation on construction intensity across markets. However, the model cannot explain the most volatile markets in terms of low frequency price changes. More research is needed to determine whether measurement error-related data smoothing or market inefficiency can best account for the persistence of high frequency price changes. With respect to the extremely high house price change volatility in certain coastal markets, more research is needed to ascertain whether shocks to interest rates or better measurement of local income variability can match this moment of data without appealing to some type of animal spirits.
Suggested Citation: Suggested Citation