Yale University's Nick Barberis and colleagues show that high risk does indeed lead to higher returns. However, it is not risk in the sense of volatility or beta that matters. Instead, they measured risk by using prospect theory. This idea, developed by Nobel laureate Daniel Kahneman and the late Amos Tversky, says that people recognise gains and losses in a different way than orthodox economics predicts. In particular, they overweight the small chances of big gains or losses.
This changes what constitutes a risky stock. From the perspective of prospect theory the least attractive assets are those which offer the small chance of a big gain but little chance of big profits; this is why retail investors tend to buy insurance policies rather than sell them. The most attractive assets, by contrast, are those offering the chance of a big upside even if this comes at the expense of larger chances of poor returns: these are lottery-type stocks.
Professor Barberis and colleagues ranked stocks according to their attractiveness from the point of view of prospect theory. They found that the 10 per cent of stocks least attractive to a prospect theory investor outperformed those most attractive by an annual average of 15 per cent in the US between 1931 and 2010. This is true even controlling for other risks such as beta and size. And, they say, it is also true in most international stock markets as well as the US.
This tells us that investors do value stocks by implicitly using prospect theory. They buy stocks which are attractive from a prospect theory point of view. This means such stocks are highly priced and so offer poorer subsequent ones than stocks which prospect theory investors avoid.
In this sense, high risk does mean high returns - as long as we measure risk using prospect theory rather than conventional yardsticks such as volatility.
This explains some things which are perplexing from the point of view of conventional measures of risk.
One is the underperformance of Aim stocks and newly-floated companies. Such stocks are usually riskier than others by conventional measures and so should outperform according to theory. From a prospect theory perspective, however, these stocks are actually attractive as they offer the small chance of immense upside; they could be the 'next Google'. This means investors overpay for such stocks, with the result that their long-run performance is poor.
Another thing it explains is the good performance of defensives. These don't offer big, quick upside: you'll not double your money quickly in Glaxo or National Grid. But they do have an obvious downside. Being liquid stocks, they can be easily dumped in bad times and so could fall sharply on days when investors panic. This makes them unattractive to prospect theory investors, which means they are under-priced for other investors.
You might object here that prospect theory is an irrational way of assessing risk. And it's certainly wrong to consider stocks in isolation rather than in terms of their contribution to your overall portfolio.
Maybe. But in the real world it seems that people do behave in a way consistent with prospect theory. And it is real behaviour that matters, not some textbook ideal. And from this perspective, there is indeed a trade-off between risk and return.
