The Causality and Simultaneity between Price and Trading Intensity

This paper empirically examines the relationship between trading intensity and price in the US existing home markets. In particular, I have tested two recent theories, the search model by Wheaton (1990) and the "downpayment" model by Stein (1995). The search model predicts that trading volume should positively affect real home price, while the "downpayment" model predicts that, with liquidity constraints, the change of home price should positively impact trading intensity. In this paper, I have specified two relationships between trading intensity and price--a contemporaneous one and a lagged one. Using regional quarterly pooled time-series data from 1975 to 1994, the Granger causality test, Hausman test, and two-stage least square regressions are performed. The results show that there is no lagged causality from trading intensity to real price, yet trading intensity contemporaneously affects real price within a one quarter time period. Regarding the relationship between nominal price and trading intensity, the results show that while lagged nominal price affects current trading intensity, current nominal price alone does not affect trading intensity. The simultaneity between trading intensity and price is not found. In addition to the interaction between trading volume and price, the regression results in this paper indicate that mortgage rate is strongly and negatively correlated to trading intensity, but not correlated to real price. Employment shows strong effect on real price, but not on trading intensity. The percentage change of employment, in contrast, impacts trading intensity, but not real price. Thesis Supervisor: William C. Wheaton Title: Professor of Economics and Urban Studies