| 摘要: | Current literature about the relationship between interest rate spreads and economic activity has focused on the interest rate spread for long-term bonds such as 10-year Treasury bond. The importance of the interest rate spread for medium-term and short-term Treasury bonds as upcoming key predictors has been neglected. This study, therefore, aims to explore whether the medium-long-term (7-year) spread, the medium-short-term (5-year) spread, and the short-term spread (3-year) are as informative as the long-term (10-year) spread in predicting recessions.
The variables used in this study include real GDP, consumer price index (CPI), money supply (M2), and the aforementioned four spreads. The data spanning over the period 1961.Q1-2011.Q3 are collected at the quarterly interval from the Taiwan Economic Journal. The unit-root test is first performed on each of the variables to detect stationarity. Four regression models are then specified to examine the relationship between real GDP growth and the four spreads and the significance of this relationship overtime.
Evidence indicates that the spread lagged by at least the first five quarters exert significantly positive influences on real GDP growth for the four spreads if the lags are included as regressors individually. However, only the spreads lagged by one quarter remain significantly positive for the four spreads if all the lags are included as regressors in the model simultaneously. The medium-long-term spread, the medium-short-term spread, and the short-term spread are as informative as the long-term spread in predicting recessions. Evidence further shows that the positive relationship had turned to be weaker in Period II than In Period I for the four spreads. These findings emphasize the interest rate spread as an economic indicator for the Fed as well as investors to predict recession or future economic activity.
Current literature about the relationship between interest rate spreads and economic activity has focused on the interest rate spread for long-term bonds such as 10-year Treasury bond. The importance of the interest rate spread for medium-term and short-term Treasury bonds as upcoming key predictors has been neglected. This study, therefore, aims to explore whether the medium-long-term (7-year) spread, the medium-short-term (5-year) spread, and the short-term spread (3-year) are as informative as the long-term (10-year) spread in predicting recessions.
The variables used in this study include real GDP, consumer price index (CPI), money supply (M2), and the aforementioned four spreads. The data spanning over the period 1961.Q1-2011.Q3 are collected at the quarterly interval from the Taiwan Economic Journal. The unit-root test is first performed on each of the variables to detect stationarity. Four regression models are then specified to examine the relationship between real GDP growth and the four spreads and the significance of this relationship overtime.
Evidence indicates that the spread lagged by at least the first five quarters exert significantly positive influences on real GDP growth for the four spreads if the lags are included as regressors individually. However, only the spreads lagged by one quarter remain significantly positive for the four spreads if all the lags are included as regressors in the model simultaneously. The medium-long-term spread, the medium-short-term spread, and the short-term spread are as informative as the long-term spread in predicting recessions. Evidence further shows that the positive relationship had turned to be weaker in Period II than In Period I for the four spreads. These findings emphasize the interest rate spread as an economic indicator for the Fed as well as investors to predict recession or future economic activity. |
