Investors are in denial
12 Feb 2013
Savers and investors need to be more realistic about this new, low-return world
Historic lows for interest rates and bond yields mean low prospective returns on all assets, according to London Business School authors of the 2013 Credit Suisse Global Investment Returns Yearbook, Elroy Dimson, Paul Marsh and Mike Staunton.
Professor Paul Marsh said: “Although we have been living with low rates for several years, many investors still seem in denial. Return projections by asset managers, retail savings product providers, pension funds, and even governments are often still too high. Such optimism is dangerous, because it misleads, and it also masks the need for remedial action.”
The Yearbook comprises feature articles, together with profiles of 22 national and three regional markets. It covers the five main asset classes, with a dataset that spans 113 years from 1900 to the present day. The 2013 Yearbook includes three new countries, Austria, China, and Russia.
The inclusion of Russia and China, where investors lost everything in 1917 and 1949 respectively, allows the authors to explore the impact of survivorship bias, and to estimate the long-run, historical, worldwide equity risk premium on a survivorship-free basis.
The low-return world
The first feature article in the 2013 Yearbook examines the new investment landscape that we face following the financial crisis. It shows that:
• Extrapolating the high bond returns of the last 30 years would be fantasy
• Future bond returns are likely to be low, as indicated by current yields
• Future returns on cash are likely to be even lower, at least for a generation
• When real interest rates are low, real equity returns also tend to be low
• The annualised future equity risk premium is also likely to be lower than over the second half of the 20th century
• Many investors are still expecting, and even banking on, unrealistically high returns
The second article in the Yearbook addresses the claim that equity and bond returns revert to the mean. If this were true, it would reduce risk, and provide market timing signals to help boost investors’ returns. The authors’ research reveals that:
• Higher returns have tended to follow periods when prices were low relative to fundamentals, and vice versa, when viewed over history
• But this relationship is weak, “noisy”, and fallible, giving many false signals
• Some markets take a long time to revert and some never do
• More fundamentally, when judgments about whether prices are “low” or “high” are based not on the entire history of returns, but solely on what investors would have known at the time, mean reversion loses its power
Professor Elroy Dimson explains: “Much of the evidence for mean-reversion is based on optical illusions based on hindsight. Investors – who sadly do not have the benefit of hindsight – will find that market timing, using mean reversion rules, is more likely to hurt performance than enhance it.”
The 2013 Sourcebook, which accompanies the Yearbook, provides detailed, long run investment statistics for 22 countries and three regions, including the world index. It examines risk over the long run and the historical extremes of investment performance. It documents the global long-term and shorter-term rewards for equity and bond investing.
Notwithstanding their recent rally, equities have mostly disappointed since the beginning of 2000. However, over the long run, the Sourcebook shows that they have beaten inflation, bonds and cash in every country with a continuous 113-year history.
The Sourcebook also reports:
• a size premium (the amount by which smaller companies outperform larger ones) that has been positive over long intervals and many countries
• a value premium (the amount by which value stocks outperform growth stocks) that has been larger than the size premium
• an income premium (the amount by which high-yielding stocks outperform low-yielding stocks) that has been larger than the value premium
• a momentum premium (the amount by which past winners outperform past losers) that is largest of all