Time-Varying Properties of th Term Structure of Equity Returns
Abstract :
Since the late 1980s and early 1990s, scholars have examined the term structure of equity returns, utilizing estimates based on diverse modeling techniques and rich analytical frameworks. Recent research has primarily focused on identifying the true slope of the curve, using dividend-claim returns or yield-to-maturity measures to reconcile habit formation and long-run risk models. However, no study has yet empirically addressed the implications of the change in expected equity returns by investment horizon over time and how it compares with actual future returns. Furthermore, whether upward-sloping curves are observable only during recessions or whether changes in the slope of the curve are directly associated with changes in an observable economic variable remains unclear. This dissertation comprises two chapters that examine the time-varying properties of the term structure of equity returns, assessing their predictability over time and their relationship with the business cycle. This study based its analysis on estimates of the holding period equity return by investment horizon of the aggregate market at monthly frequencies using a first-order vector autoregressive VAR(1) process model from January 1950 to June 2023. The evidence presented in this study demonstrates that the mean-reverting properties of equity returns lead to a time-varying term structure and predict actual future equity returns across different investment horizons. I find that the term structure of expected equity returns changes frequently and forecasts actual equity returns by investment horizon up to ten years. Moreover, the results presented in this study provide conclusive evidence that upward-sloping curves are not exclusive to recessions, but extreme upward-sloping curves predict a high probability of a recession. The empirical evidence from this study confirms a direct inverse relationship between changes in the slope of the curve and investment growth rates, at least one year in advance.
Nikolaos Tessaromatis, PhD (Chair)
Enrique Schroth, PhD (Adviser)
Emmanuel Jurczenko, PhD (Member),
Mirco Rubin, PhD (Member)