A History of the Equity Risk Premium and its Estimation

Abstract

The estimation of the equity (or market) risk premium has become a cottage industry, both for academics and professionals. It is recognized as a key economic or financial parameter for a variety of interests and applications, yet opinions as to its magnitude either historical or projected vary widely. But not as widely as was once the case. There has been general acceptance, a consensus if you will, that historical equity return premia overstate what was anticipated or expected and that a large component of the historical equity return premium constitutes unanticipated capital gains. This paper explores the history of this idea and examines the role this it plays (or not) in a variety of recent and current methods of estimating the equity risk premium. Using a methodology similar to Fama and French (2002) but presaged in Copeland (1982) I describe the behavior of ex post and ex ante risk premia for the period 1872 to 2013 and estimate the equity risk premium going forward for the next 10 years (2014-2023). I also discuss various issues in the estimation of the equity risk premium, such as geometric versus arithmetic mean, top down versus bottom up forecasts of the equity risk premium, and whether to use dividend yields or P/E multiples in accounting for unanticipated capital gains. I conclude that the arithmetic equity risk premium as usually calculated is significantly overstated and that the current and expected future ERP is in the range of 3-4 percent for both geometric and arithmetic means, though likely in the upper half of this range.