This is the original text of my latest FT column which appeared today. Because of space constraints, sometimes the column has to be cut to fit the available space, so I offer here the full version as originally written. (No newspaper is going to turn away an advertisement in the current environment, so writers have to accept the demands of their medium!).
“The official guardians of the truth about historic returns from equities and bonds have delivered their verdict on current market conditions. “While investors should keep faith with stocks” say Elroy Dimson, Paul Marsh and Mike Staunton, authors of the annual London Business School/Credit Suisse Global Investment Returns Yearbook, “they should not harbour fantasies of an immediate return to either previous (and with hindsight unrealistic) market levels, or to previous high rates of return”.
“Markets” the three academics go on “are likely to take a long time to recover from the battering they have received during the credit and banking crisis”. They estimate that the odds on the FTSE 100 index regaining its previous high before 2013 as no better than 50-50, and the probability that it does not do so within 10 years as a far from insignificant 25%.
In other words, investors who rely on history as a guide to future returns should allow for disappointment. Although mean reversion is a staple feature of stockbroker economics, it is far from infallible. “Our evidence” say the LBS trio “is consistent with the view that it is hard to improve on extrapolation from the longest history that is available at the time forecast is being made”. On their analysis, this points to a forward looking equity risk premium of around 3.0%-3.5% per annum.
In the context of predicting future market returns, the academics are on solid ground in rooting their projections of market returns on long run historic probabilities, rather than on the more simplistic heuristics, typically based on much more recent data, favoured by the broking community. The truth is that at any point in time the market’s future behaviour encompasses a range of outcomes that no single point estimate (let alone any salesman’s pitch) can adequately capture.
The dispersions around any single point estimate can be extremely wide. History in due course provides a verdict on what happens, but the direct causal link between starting point and later outcome is tenuous. The best investors can hope for in practice is that market valuations become so extreme that the ex ante odds on a favourable outcome move from being evenly balanced to strongly positive.
This is what happens at tops and bottoms in the market cycle, with the caveat that there is no guarantee even then that any trend must eventually reverse. The Russian stock market, to take one extreme example, simply went to zero following the 1917 revolution. There never can be any certainty that the assumptions on which investors have based their decisions for many years will continue. The history of the equity risk premium itself (rarely observed in the first half of the twentieth century, rarely out of evidence in the second) is testimony to that.
Such uncertain dynamics make the current markets particularly interesting. No practitioner needs an academic to tell them that the 2008 market meltdown was a near unprecedented event in terms of its depth, its global nature and the extraordinary correlations of returns across most markets and asset classes. The range between the best and worst performing stock market last year, for example, was narrower than in any previous severe market crash, making nonsense of many diversification strategies.
Making forward projections in these conditions is more than usually unreliable. The rules of the game are changing, with increased government intervention, higher taxation and unprecedented State and central bank intervention in the financial system all likely to have an impact on market performance over the next few years. The risk of policy errors remains high.
While a perspective based solely on historical data would suggest that equities are now at or just below fair value, that government bonds at today’s prices are only for the brave and that all fiat currencies must be a sell against gold, only the last view is wholly persuasive. Even Jeremy Grantham of GMO, as passionate an advocate of mean reversion as you will find, acknowledges in his latest quarterly letter that while the barometer for equities has swung from stormy to fair, that may not be reflected in market performance any time soon.
The current market crisis would be a historical oddity if it ended here, given the propensity of markets to overshoot on the other side of fair value. That could make equities fall further and bonds generate decent returns even from their current negligible starting yields. After last week’s market action, which saw the Dow Jones Industrial Average flirt with an important technical floor, its November lows, some would argue that the process has already begun.
As a dyed-in-the-wool contrarian, Mr Grantham bemoans the fact that the odds on further falls in sterling, the US housing market and risk aversion (to name but three of his “sure things” a year ago) are no longer sufficiently good to justify a big bet. A market environment in which the odds are evenly balanced is an uncomfortable one for anyone whose livelihood depends on holding – and acting on – a bold and unambiguous market view. But that is where we seem to be today, as the worried tone of many of my professional friends attests. There is a case for buying equities today, but the evidence is not yet clear-cut.
Jonathan Davis 