John Authers, Financial Times
“If you can keep your head when all about are losing theirs, it’s just possible that you haven’t grasped the situation.” That was how the American satirist Jean Kerr updated Rudyard Kipling’s poem, and new financial research suggests she was on to something.
It also adds to the growing supply of data suggesting we should re-examine the traditional financial view that investment is about trading off risk and return, with greater risk ultimately rewarded with greater return.
According to new research by Alan Moreira and Tyler Muir, two academics at Yale University’s School of Management, the correct response to an increase in volatility — and with it, risk — is to exit the market. The time to re-enter is only when the volatility has already started to subside.
This means not buying until the market has bottomed and started to recover, and also implies selling when the market has already started to fall. But over time, it still beats the returns from simply buying the market and holding it. And it also rather worryingly contradicts the widespread belief, based on much research, that higher volatility creates a great opportunity for brave contrarians to buy at the bottom.
Their research involved a strategy which on average is 100 per cent exposed to the stock market. It then adjusts the amount it holds in the market at the end of each month, according to the volatility experienced in that month. As volatility increases compared to its average, it leaves stocks and goes into cash. In periods of very low volatility it can borrow to invest. The same exercise was performed for the different factors which have beaten the market over time — value stocks, smaller companies and stocks with positive momentum — and it was also applied to the “carry trade” strategy in foreign exchange, and to a range of non-US stock markets.
In all cases, using timing to exit the market as volatility increases turned out to raise long-term returns compared to a straight “buy and hold” strategy. Further, the advantage widened when the academics looked at risk-adjusted returns. Using the Sharpe ratio — the standard measure for judging returns compared to the amount the returns vary — volatility-timing improved returns in all the US factors the academics examined, and in 16 of the 20 OECD countries.
During the most famous US market breaks, the volatility-timing portfolio led the market at all times, avoiding much of the savage drawdowns seen after the 1929 Great Crash, and during the 2008 Great Recession.
In 2008, the year of Lehman, the volatility-timing portfolio was barely ever down. Even though it failed to buy at the bottom in March 2009, and went against the wisdom of many pundits, led by Warren Buffett, that this was a time to buy, it maintained its advantage. However, by 2014, after years of strong returns on the market, the buy-and-hold portfolio had almost caught up. Full details are available in the paper, which is available free online.
Do these findings mean the complete reversal of our notion of the risk-return trade-off? Not necessarily, but they do show that we need to pay more attention to its dynamics over time. Markets tend to grind upwards steadily over long periods of time, and correct downwards in short and more violent bursts. The premium for taking risk exists, but that premium is not necessarily available when the volatility is happening.
Why does the volatility-timing anomaly exist? As the effect has been documented long into the past, it is not just a story of modern-day investment institutions and the perverse pressures on them. It is something more fundamental, probably rooted in human nature.
One explanation is that we are slow to recognise risk when it appears, so that the market generally recovers after the first flare-up of volatility. That gives the chance to get out in time. At the other end, the trauma of recent loss allows fear to persist.
A practical implication is that this approach should offer a clear-cut way to make money. If this research gains acceptance, expect “smart beta” funds that use volatility to time exposure to markets. And these ideas should be very useful for the new wave of so-called “robo-advisers”, which allocate clients’ assets following a few simple rules. Expect wealth managers to program their robots to lose their heads when all about them are losing theirs.
Read full article in Financial Times