Threshold Effects in the Relationships of REITs and Other Financial Securities in Developed Countries

Abstract

We use a Panel Smooth Transition Regression model (PSTR) to investigate the nonlinear dynamic relationship between financial variables and REITs 1 of Japan and U.S with 3-month interest rate change as threshold variable. We discuss the relationship between explained variable of REITs return and explanatory variables (10 year bond interest rate, real estate return and stock return) within two regimes. Empirical results show that the transition function is an exponential type with region one, and two regimes. In regime 1, the relationships between REITs return and two explanatory variables (10-year bond interest rate change and real estate return) are significantly positive. The relationship between REITs return and stock return is significantly negative. In regime 2, the relationships between REITs return and two explanatory variables (real estate return and stock return) are significantly positive. REITs act as hedgy against stock market downturn when the magnitude of 3-month interest rate reduces greater than 0.9886%. In the low interest rate change regime, the REITs behave more like fix income and real estate than risky stock. In the high interest rate change regime, REITs behave more like stock and real estate than fix income. The study contributes in two ways. The first is to use a Panel Smooth Transition Regression model (PSTR) to investigate the nonlinear dynamic relationship between financial variables (real estate index returns, stock index returns, and ten year bonds) and REITs of Japan and U.S with 3-month interest rate change as threshold variable. The second is to investigate at what level of 3-month interest change will induce REITs markets serve as stock market hedgy under economic shock.