The Conclusion to The Equity Risk Premium Series

equity risk premium analystFrom a review of the Literature on the Equity Risk Premium it becomes apparent that there is little consensus on the historical value, what it may be today or what it is likely to be. Today there can even be some doubt that it is the  “difference between the returns on risky stocks and the return on safe bonds”, as it was described by Dimson et al (2002 p.193). Regarding the value, Mehra and Prescott’s (1985) assertion of 6.18% is prevalent but disputed, however there is agreement on the volatility and the importance of the time frame over which it is measured. Dimson et al (2002) posed key questions about the fundamentals: the method of calculation and what it may be in the future. These authors, as well as Welch (2000), Damodaran (2011) all emphasise the importance of the Equity Risk Premium (ERP) in finance, especially with regard to its use in the CAPM. This emphasises the importance of arriving at an accurate figure.

Give this importance, Grinold, Kroner and Siegel (2011 p.53) point out that there is no agreement on the “ERPs value within an order of magnitude, or even…………whether it is negative or positive”. Taylor (www.globalfinacialdata) perhaps encapsulates this by stating that the ERP changes over time, thereby making it difficult to categorically state a historical Equity Risk Premium.  Into this argument, we also add the debate on what constitutes a risk free asset and the method of calculation, arithmetic or geometric. As we have read, it is important to use the geometric mean for the past and the arithmetic mean for the future.

Having established how difficult it is to agree on the Equity Risk Premium, what options do we have if we wish to forecast it? Dimson et al (2002) used a method of calculating the historical variance between the arithmetic and geometric means, and then applying the variance factor to a current risk proxy, which is then added to the historical geometric mean to provide prospective value. They found this variance to be 0.5 times the arithmetic mean standard deviation. From our data, we did not find this but have to accept that our historical returns may not be lognormally distributed, a key element for Dimson et al (2002).  We also considered Damodaran (2011) who proposed that the Dividend Yield may be a proxy for the ERP and that discounting future growth on an index, given certain assumptions that we discussed, may provide a prospective Equity Risk Premium. These examples served to highlight the importance of dividend reinvestment to the ERP, as we saw in a previous article.

We also looked at the influence of Behavioural Economics, with two particular facets being noticeable: narrow framing and myopic loss aversion. The former casts doubt on the fundamental principal of portfolio theory, that each investment decision is viewed as part of a balanced portfolio and not on its own merit. The second, myopic loss aversion suggests that short evaluation periods, not ideal when considering something as volatile as the Equity Risk Premium, means that investors are moving from equities to bonds, and we saw an example of this with Allianz. Having noted the importance of time periods for the ERP, we can see how there is the possibility for good returns to be missed.

Sign up for our FREE newsletter
and receive our best trading ideas and research



In the second part of the paper we looked at specific UK data from 1976 to 2012 and explained the indices used. The FTSE All Share, for equities, and the FTSE Actuaries UK Gilts Index Series for gilts. From these we derived an ERP over this period of 3.47%, well below Mehra and Prescott’s (1985) 6.18% and more in line with Dimson et al’s (2002) UK specific 2.4%. When we looked at annualised ten year returns, as opposed to the annual returns, the ERP reduced to 2.58%. This is closer to Dimson et al’s (2002) calculation but we have to consider the shorter timeframe we are looking at.

Looking at the prospective Equity Risk Premium, I adopted Dimson et al’s (2002) approach to calculate a future value based on a current (at the time) risk proxy. In practice, this was not very effective as it did not come within three standard errors of the actual, but once again we have to consider the short time period. The range for these predictions, over different time periods was 3.9% to 14.5%. Interestingly it was over the thirty year period that we calculated 3.9%. Using the techniques employed by Damodaran (2011), the Gordon Growth Model and also discounting the growth of an index, we arrived at prospective ERPs of 2.9% to 3.2%. We also derived some confidence that the Dividend Yield will be a reasonable proxy for the Equity Risk Premium. Combining these estimations, 2.9% to 3.9%, they are similar to Dimson et al (2011) and Gregory (2007) who propose a long-run ERP of 3% to 3.5%. We also read that there can be speculative events, driven by the madness of crowds which can significantly impact any ERP valuation; therefore all calculations will need to be accompanied by a degree of luck if they are going to be accurate.

To conclude, I have determined that the Equity Risk Premium is generally lower than previous values of circa 6%, especially with reference to the UK, than previously assumed. This does come with the major codicil that I am looking at a limited time period but I am slightly reassured that it is worthwhile as it includes the Great Moderation and also significant upheaval. My research also concludes that many of the uncertainties about the ERP remain unresolved and will be a source of discussion and dispute for many years to come.

I invite you to read the entire series on Equity Risk Premium.  Thanks for reading.

References

  • Damodaran, A. (2011) Equity Risk Premiums (ERP): Determinants, Estimation and Implications – The 2011 Edition   Stern School of Business
  • Dimson, E., Marsh, P., and Staunton, M. (2002) Triumph of the Optimists: 101 Years of Global Investment Returns   Princeton University Press
  • Gregory, A. How Low is the UK Equity Risk Premium? XFi Centre for Finance and Investment, University of Exeter (2007)
  • Grinold, R.C., Kroner, K.F. and Siegel, L.B. A Supply Model of the Equity Premium    Rethinking the Equity Risk Premium (Edited by P. Brett Hammond, Jr., Martin L. Leibowitz, and Laurence B. Siegel) Research Foundation of CFA Institute (2011)
  • Mehra, R. and Edward C.P. 1985, The Equity Premium: A Puzzle, Journal of Monetary Economics, v15, p145–61.
  • Welch, I. Views of Financial Economists on the Equity Premium and on Professional Controversies Journal of Business   Vol 73 No 4  2000
  • www.globalfinancialdata   Taylor, B    The Equity Risk Premium | Accessed 4th August 2012

Twitter:  @MillarAllan   @seeitmarket

Any opinions expressed herein are solely those of the author and do not in any way represent the views or opinions of any other person or entity.