Time-varying impacts of expectations on housing markets across hot and cold phases

Abstract

This study investigates time-varying roles of people's expectations in driving the US housing price and quantity dynamics using a Threshold Vector Autoregressive model. The expectation measure, a good-time-to-buy (GTTB) index, works as the threshold indicator to classify pessimism and optimism phases, and represents the model-based measure of uncertainty. There is strong evidence for regime switches in responses to shocks across the two phases. The results show that good and bad shocks play similar roles in housing markets. Tiny responses of GTTB to both large good and bad shocks in the two regimes suggest “too bad to be believed” and “too good to be believed” patterns. The estimation is biased as volatility shocks are neglected in the housing boom and bust.