Changes In Cyclical Behaviour Of Interest Rates

Originally published in 1994 and the recipient of the Stonier Library Award, this volume evaluates an alternative approach – the sequential filter- to managing the uncertainty inherent in the future course of the interest rate cycle. The specific hypothesis is that the sequential filter can produce valuable signals of cyclical peaks and troughs in interest rates. The analysis focusses on US interest rates from April 1953 to December 1988.

This paper investigates the behavior of the term structure of interest rates over the business cycle. In contrast to prior studies that measure the business cycle by the simple growth in aggregate economic activity, we consider an alternative measure: the deviation of aggregate economic activity from its potentially stochastic trend. We show that incorporating both an independent trend and cyclical component in consumption improves the efficiency in estimating consumption-based asset pricing models. We also find that the term spread is more informative about future changes in stochastically detrended real gross domestic product (GDP) than future growth rates in real GDP.