Elroy Dimson, Paul Marsh, and Mike Staunton of London Business School and authors of Triumph of the Optimists (Princeton University Press, 2002) are responsible for the marvelous Credit Suisse (SIX:CSGN) Global Investment Returns Yearbook, now in its nineteenth annual edition. The yearbook was distributed to Credit Suisse clients and is not for sale to the general public, but a 40-page summary of the 251-page yearbook is available online.
The authors worked with 118 years’ worth of data. So this is no year in review; it’s 118 years in review, complete with myriad tables and graphs. Divided into five chapters, the yearbook covers long-run asset returns, risk and risk premiums, factor investing, private wealth investments, and 26 individual markets (23 countries plus world, world ex-USA, and Europe).
One of the questions the authors ask is whether equity premium is predictable. Using their long-run global database, they “adopt three approaches, each of which involves analyzing whether the equity risk premium (ERP) relative to bills in a particular year can help us to predict the annualized ERP over the subsequent five years.” They found, as one might expect, that “for strategic asset allocation, we learn relatively little (and nothing statistically significant) from recent annual performance about future equity premiums.” They conclude that “to forecast the long-run equity premium, it is hard to beat extrapolation that takes into account the longest history available when the forecast is being made.”
The authors also assess various forms of factor investing. Both over the long run and across different countries, size, value, income, momentum, and volatility have generated sizable premiums. But the authors write that “the theory of why such premiums should exist, or what types of risk they are rewarding, is admittedly weak. Furthermore, if they are generated by behavioral traits, behavior can change, especially as awareness of these factors—and their popularity—increases.”