Two features distinguish residential real estate from financial assets: households’ consumption demand for a dwelling and the indivisibility of properties. Because of credit market imperfections, the indivisibility implies that the purchase of a home requires a significant amount of liquid wealth up front. Furthermore, the only homeownership option for households with limited resources is to buy a starter home as a first step on the property ladder. It is only by continuing to accumulate wealth that households can eventually trade up. The performance of their highly leveraged investment in the starter home is key to their ability to accumulate wealth and hence affects the timing of their move up the property ladder. This paper builds upon these features of housing consumption to address two fundamental issues concerning housing market dynamics: the large predictable housing price swings and the co-movement of prices and transactions. We develop a life-cycle model where households are heterogeneous with respect to income and preferences, and mortgage lending is restricted by a down-payment requirement. The first contribution of the paper is to point out the critical role of the income of young households to housing market fluctuations. Since the income of young households is more volatile than per capita income, the variable typically used in the empirical housing literature, this goes some way towards explaining “excessive” price volatility. The second contribution of the paper is the characterization of a mechanism by which credit constraints not only amplify income shocks, but also affect the timing of households’ moves in a way that explains the co-movement of prices and transactions. The price overshooting prediction of our model is in agreement with empirical findings based on US city-level data. Our modeling framework is also able to account for the dramatic boom-bust episodes which seem to follow credit market liberalizations. Modeling agent’s life cycle allows us to rationalize the concomitant cohort specific changes in mortgage demand and owner-occupancy rates. In particular, the model sheds light on the causes of the U.K. housing boom-bust cycle which followed the credit market liberalization of the early Eighties. The paper first provides empirical evidence in support of the building blocks of the model and discusses alternative approaches proposed in the literature. We then explain the price and transactions dynamics in the simplest possible version of our model before extending it in various directions. Finally, we present empirical evidence in support of the predictions of the model.
Publication Date
Financial Markets Group Discussion Papers DP 375
