摘要
According to search theory, transaction volume possesses the function of price discovery and reflects information more rapidly than price does. However, the findings of previous empirical studies differ considerably. In this study, a theoretical model is first established to analyze the potential information lag of transaction volume during pessimistic speculation. Data on the UK housing market are collected to conduct an empirical analysis of the responses of housing transaction volume to different market conditions. The results show that transaction volume responds to market information more quickly than does housing prices. However, under increasing market uncertainty, transaction volume lags four periods before reflecting the effect of the uncertainty. Moreover, this study performs a rolling window bootstrap Granger causality test, revealing that price leads volume during the period in which transaction volume fails to reflect an immediate rise in market uncertainty. An increase in market uncertainty reduces transaction volume. In addition, once transaction volume drops below a specific threshold, it loses its information content and price discovery function, extending the lead-lag gap with housing prices by two periods. The present study proposes a simple method for determining the informativeness of housing transaction volume.