Abstract:
Using state-level data for the U.S. housing market over the period of 1975:Q1-2012:Q2, we show that the consumption-wealth ratios derived from aggregate wealth (cay) and disaggregate (i.e. financial and housing) wealth (cday) are strong predictors of real housing returns (and their volatility). Additionally, we find that, barring the extreme ends of their respective conditional distributions, such effect is stronger for housing return volatility than housing returns. All in all, our findings show that state-level regressions can recover a large degree of heterogeneity that country-level exercises typically ignore. Such heterogeneity is prominent not only in terms of consumption smoothing behavior, but also with regard to housing return predictability.
