Join our community of smart investors
OPINION

High risk, low return

High risk, low return
May 26, 2016
High risk, low return

Alan Moreira and Tyler Muir at Yale University point out that investment strategies that reduce equity risk when expected volatility is high would have beaten the US market for years. Investors can therefore better make money by cutting risk than by increasing it. This corroborates research by Andrew Lo at MIT. He's shown that reducing equity exposure at times of high volatility would have beaten a buy-and-hold investment in the S&P 500 over the long term.

The same is true for the UK. To establish this, I compared two simple strategies. One is simply a portfolio split 50:50 between the All-Share index and gilts, rebalanced monthly. The other is a strategy that has a 50:50 split when the VIX index - a measure of expected volatility on US and hence global stocks - is at its post-1990 average of 19.7 per cent, but which has higher equity exposure at lower levels of volatility and lower equity exposure at higher. This portfolio uses the Merton equation at the end of each month, holding risk aversion and expected equity returns constant and so varying equity weights only as expected volatility changes.

This managed volatility portfolio would have hugely outperformed the passive 50:50 allocation. £100 invested in it in December 1989 would have grown to £1,880 by now, compared with just £787 invested in the 50:50 portfolio.

Granted, the managed volatility portfolio has had slightly higher volatility. But this was 'good' volatility: more upside in good times. The worst 12-month period for the managed volatility portfolio saw a loss of 9.4 per cent, while the worst 12-monthly loss on the 50:50 portfolio was 15.8 per cent.

You should find all this surprising. If high risk means high returns, our managed volatility portfolio should have underperformed the 50:50 one, because it reduced equity exposure when risk was high.

How, then, can the opposite be the case? Why does it pay to avoid volatility?

It's because high risk - at least as measured by volatility - doesn't lead to high returns. "There is generally no empirical risk-reward relation," concluded Eric Falkenstein of Pine River Capital Management in an important paper written in 2009. If anything, the opposite is the case. Since 1990 (when VIX data began) the correlation between the VIX index at the end of one month and All-Share returns the following month has been slightly negative, at minus 0.27. A high VIX has been slightly more likely than not to lead to low returns, not high. This is the exact opposite of the 'speculate to accumulate' theory.

There's a reason for this. Increased volatility can be a sign not of higher future returns but instead of the onset of a bear market, in which any trade-off between risk and return deteriorates.

For example, the VIX was consistently above average in 2000-01. This correctly warned us to cut equity exposure just before the deflating of the tech bubble. It was low in the mid-2000s, thus predicting a bull market in equities. And it rose in 2007-08, telling us to sell equities just before the banking crisis.

Of course, no indicator is perfect. The VIX was high in the spring of 2009, causing our managed volatility portfolio to miss out on some big profits in the early stages of the recovery in stock markets. On average, though, high implied volatility has warned of poor returns and thus been a useful sell signal.

In this sense, the conventional wisdom is plain wrong. It's low risk that has led to good returns on average, not high risk.

Sadly, however, there's a problem here. It's not practical for the retail investor to implement this managed volatility strategy because the dealing costs involved in trading every month would be high.

In theory, however, fund managers who enjoy lower trading costs and who can use derivatives to manage risk should be able to construct such funds and sell them to investors cheaply - or at least, they would do so if the financial industry were not so abominably bad at beneficial innovation.

Nevertheless, there is a point here. It's that, on average, investors should avoid high volatility. Right now, the VIX is below its long-term average so this is an encouraging sign. But this will change. And when volatility does rise, it would be unwise to regard it as a buying opportunity.