The Risk Return Tradeoff in the Long-Run: 1836-2003

Abstract

The risk-return tradeoff is fundamental to finance. However, while many asset pricing models imply a positive relationship between the risk premium on the market portfolio and the variance of its return, previous studies find the empirical relationship is weak at best. In sharp contrast, this study, demonstrates that the weak empirical relationship is an artifact of the small sample nature of the available data, as an extremely large number of time-series observations is required to precisely estimate this relationship. To maximize the available time-series, I employ the nearly two century history of US equity market returns from Schwert (1990), exploring the empirical risk-return tradeoff for a variety of specifications that allow for asymmetric volatility, regime-switching, and additional factors associated with intertemporal (ICAPM) hedging demands. Similar to studies that use the more recent US equity price history, conditional market volatility in the historical data is persistent and displays strong asymmetric relationships to return innovations. Further, the conditional correlation between stock and bond markets is closely related to periods of documented financial crises. Finally, in contrast to evidence based upon the recent US experience, the estimated relationship between risk and return is positive and statistically significant across every specification considered.

JEL Classification: G12, G15