Why it pays to cut a loss

Recouping losses

Many investors never use stop losses. They tend to view them solely as a tool for short-term traders who often use leverage (borrowed money) and can be wiped out by a few big, ill-timed losses. If you’re investing for years ahead, your portfolio is diversified and you don’t use leverage, the threat from the odd big loss doesn’t seem vast.

However, even long-term investors should have a clear philosophy for deciding when to cut an investment. If you buy a stock that drops 50%, it’s possible that the market is overreacting – but it’s more likely that you made a bad decision. In that situation, you could hang on and hope for a turnaround, you could sell, or you could buy more. So which is the smartest bet? A recent book by fund manager Lee Freeman-Shor of Old Mutual provides some interesting insight.

Freeman-Shor studied how the different managers that his multi-manager “Best Ideas” fund invested in coped with losing trades. He found that investors who simply hung on performed worst, because the turnaround rarely came. Out of 1,866 investments in total, 946 lost money and 131 of those lost more than 40% before they were sold. Some of those 131 stocks recovered strongly after the manager sold, but just 21 doubled and none made it back to break even (if a stock has fallen 50%, it needs to double to get back to break even – see the chart above). So if you take a big loss, the odds are against you.

Managers who sold when their stocks fell by a fixed amount did much better. The key was to sell at a level that avoided severe losses without being caught out by short-term volatility. Set the stop loss too high and the chances of bailing out of a share that would have made money was much greater: out of 421 trades sold for a loss of up to 10%, 249 would have gone on to be profitable. Given that average annual volatility for the UK market over the past ten years has been around 20%, somewhere between 20% and 30% probably makes sense – any stock that falls by more than 30% in a year is being abnormally volatile.

Interestingly, the most successful managers were those who sometimes bought more when the shares fell, rather than always selling. However, this is risky (you can end up throwing money at a share as it enters long-term decline) and requires investors to be ruthless about ditching stocks when they no longer have conviction. Either way, it’s clear that doing nothing is rarely the best move.