Join our community of smart investors

Uncertain about uncertainty

Investors have mixed attitudes to uncertainty. This is not as irrational as it seems
December 5, 2019

It’s possible that one source of uncertainty might diminish next week with the result of the general election. Which raises a question: do investors really hate uncertainty as much as is widely thought?

Ever since the work of University of Chicago economist Frank Knight in 1921, economists have distinguished between risk and uncertainty. Risk is, in Donald Rumsfeld’s useful words, a “known unknown” – a probability we can quantify such as the chance of a die rolling a six. Uncertainty, however, cannot be quantified. The question: “what will the next government do?” is an uncertainty. We can’t quantify it.

For a long time, economists have thought that people hate uncertainty even more than they hate risk. This was first established by Daniel Ellsberg back in 1957. He asked economist colleagues to bet on whether a red, black or yellow ball would be drawn from an urn and found that many preferred to bet when the numbers of balls in the urn was known than when it was unknown. They preferred risk to uncertainty. Subsequent experiments have confirmed that many of us share this preference.

This is no mere laboratory finding. It explains a lot of investors’ behaviour such as:

  • There’s a strong correlation between political uncertainty (as measured by Scott Baker, Nick Bloom and Steven Davis) and the dividend yield. Political uncertainty cuts share prices.
  • Investors, wherever they are, prefer stocks based in their home country to overseas ones. They regard the latter as more unfamiliar and more uncertain than companies they see every day.
  • The University of Aberdeen’s Hans Hvide has shown that people disproportionately own shares in the industry they work in, but these stocks don’t on average outperform others. They own such shares because they feel familiar, and so less uncertain, than shares in industries they don’t know so well.
  • Investors favour fund managers with good track records, even though these aren’t always a reliable guide to future performance, as investors in Woodford Equity Income know too well. A good track record does, however, reduce perceived uncertainty. And investors like that.
  • Many mature companies that supply familiar products are on low yields, such as Diageo (DGE), Unilever (ULVR), Reckitt Benckiser (RB.) and Associated British Foods (ABF). People pay high prices for companies that are well-known and so seem to offer less uncertainty.

There’s lots of evidence, then, that people hate uncertainty and that this affects share prices. The evidence is not, however, overwhelming. Sometimes investors actively prefer uncertainty. For example:

  • Aim stocks have underperformed mainline ones since the 1990s, suggesting that investors have caused them to be overpriced. If they really hated uncertainty, however, they would have avoided such shares because they offer more uncertainty than others as they tend to be smaller, younger and less well-known.
  • Even when annuity rates were decent there was less demand for them than one would expect, given that annuities abolish uncertainty about investment returns and about whether we'll outlive our wealth (they also protect us from uncertainty about inflation if we buy an index-linked annuity).
  • Investors have flocked into cryptocurrencies even though these offer very uncertain returns – although economists at the University of Glasgow have pointed out that such demand comes disproportionately from less knowledgeable investors.

The evidence that investors hate uncertainty, therefore, seems mixed. But in fact, it’s not. There are two different things going on here.

One is that we sometimes like uncertainty when it offers the potential for big gains, even if this potential is unquantifiable. This explains why some of us like Aim stocks and cryptocurrencies. It also explains why we hate political uncertainty. It doesn’t offer such upside. The best we can hope for is tolerably adequate government.

We do, however, hate uncertainty when it creates the danger that we’ll regret our choices. We stick to familiar stocks and funds because these are the default option, and we’ll kick ourselves if we stray from that into unfamiliar territory.

These two facts explain why we sometimes see bubbles in assets that offer uncertainty. Occasionally, regret works in favour of such assets. If everybody seems to be buying, we might fear missing out and so regret not buying. When this happens, bubbles develop.

Whatever the result of Thursday’s election, however, we’ll see – at best – the removal of only one element of uncertainty. Many others will remain – not just about Brexit (a Conservative majority brings with it the danger of a no-deal Brexit in 2020), but also about next year’s US presidential election. And of course the usual economic and market uncertainty will remain. Investors who do hate uncertainty, then, are out of luck.