Do Housing Bubbles Affect Consumption?

Standard optimzing models of consumption postulate that consumption is a function of wealth and implicitly assume that wealth is comprised of assets whose market price coincides with the fundamental price, defined as the expected present value of future dividends. We use a simple theoretical framework to show that when the market price deviates from the fundamental price, consumption behavior will depend on whether households use the market or the fundamental price to evaluate wealth. If households respond to the market price, then in an otherwise standard PIH/LCH model, the deviations from the fundamental price– what we refer to as bubbles– will affect consumption. We then empirically investigate whether housing price bubbles affect consumption expenditures using both aggregate data on durables and non-durables and the corresponding retail sales data for four regions of the U.S. The data suggest that the market price of single family homes has deviated from its fundamental price by as much as 20%-30%. In tests like those of Hall (1978), we find that housing price bubbles have predictive power for future consumption. At the aggregate level and in some regions (especially the Northeast), the impulse response function of both durables and non-durables suggests that bubble innovations can have a significant effect on consumption. The housing bubble is also found to account for over 20 percent of variations in expenditure on durables in every region considered. ∗Department of Economics, Boston College †Department of Economics, Carleton University. The authors would like to thank the SSHRC (Canada) for financial support. The first author also acknowledge grants from FCAR (Québec). Special thanks for Karen Dynan for providing us with the data. Pierre Chausse, Yong Wang, and Paramatee Vimolsimi provided research assistance. Any errors are our own. Correspondence: Department of Economics, Boston College, Chestnut Hill, MA 02167. Email: [email protected], [email protected]