Blogs | December 14, 2016

Trading chances to screw up

Investors Chronicle economics writer Chris Dillow spotlights 2016 research on some classic investing errors.


We know that people often go wrong in their investment decisions. New economic research in 2016 has highlighted some reasons for this.

Let’s start with a paradox. Thanks to the internet, investors now have more and potentially better information available to them than ever before. But there’s scant evidence decision-making has improved. Why is this?

Confidence levels
Nikos Askitas at Bonn's Institute for the Study of Labour says it is because “an abundance of data may be counterproductive”. Having more information can make us overconfident; this can embolden us to take bad decisions.

A study of Swedish entrepreneurs by Rasmus Toft-Kehler at Copenhagen Business School has shown this. Toft-Kehler finds that novice and experienced businesspeople alike can be quick to walk away from failing projects: the former because they become quickly disheartened, the latter because they have learned to know a bad bet when they see one.

Entrepreneurs with a middling amount of experience stuck with failing businesses and so lost money. This was because they had learned bad times are sometimes temporary, but not how to distinguish a run of misfortune from a genuinely bad project.

Know all, know nothing?
Being moderately well-informed, therefore, can be even worse than knowing nothing. The information itself can mislead us. Robert Metcalfe and colleagues at the University of Chicago established this with a neat experiment. They got foreign exchange traders to test a new trading platform.

They put an artificial asset they called a leveraged dollar fund on this platform, alongside the usual assets. This was set up to be an especially attractive trade, as it rose a lot when the US dollar rose but fell only a little when it fell.

They then split the traders into two groups. One group got second-by-second prices of the artificial dollar fund; the other got prices only every four hours. After two weeks of trading - long enough to learn that the fund was a good trade - the traders who got four-hourly price data held more of the dollar fund than those who got second-by-second data. The traders who only got data every four hours earned much higher profits as a result.

Noise as a distraction
Better-informed traders thus did worse. This is because second-by-second data has too much “noise” – data that is random and therefore does not reflect what’s really happening. This disguises an asset’s attractions. Having less data sometimes means getting less noise and so more signal – and this can encourage better decisions.

Knowledge, of course, is not enough. We also need the discipline to act on it. Other new experiments, however, show that this can be lacking. Economists at the University of Munich got people to trade an artificial asset under laboratory conditions: one virtue of such experiments is that they allow us to know for sure an asset’s true value. So we can say when it is mispriced.

They then put one group in a situation which would reduce their self-control. And they found that this group traded the asset at higher prices. This tells us that poor self-control can contribute to overpricing.

Bold – or even desperate?
But what might cause low self-control in the real world? One thing is past success.

Alex Krumer at the University of St Gallen shows that men who have succeeded get more testosterone which emboldens them to gamble more. This, he says, “can create price bubbles in financial markets, because success in a first investment leads men to increase their willingness to take additional risks.”

Another source of poor self-control can be simple desperation. Separate experiments by Carmen Wang at Harvard Business School and Tobias Rotheli at the University of Erfurt have shown that people take more risks if returns on safe assets are low or non-existent – even if the payoff to doing so is poor.

Those pundits who believe low interest rates have caused shares to become overvalued at least have experimental data on their side.

Place your best bet
What’s more, some mistakes seem remarkably persistent not only across time but across markets too. Maximilian Franke at the University of Ulm studied thousands of bets on football matches across Europe between 2006 and 2014. He found a systematic pattern: gamblers bet too much on outsiders and not enough on favourites.

This habit isn’t confined to football. It has long existed in horse racing and even in tennis. We also find it in stock markets. Investors pay too much for shares with a small chance of big returns: so-called lottery stocks are therefore overpriced and deliver low returns on average.

You might imagine that if investors make so many mistakes it would be easy to make money by exploiting this fact. You’d be wrong, as some work by Heiko Jacobs at the University of Mannheim and Sebastian Muller at the German Graduate School of Management in Heilbronn has found.

They studied many so-called stock market anomalies, such as the tendency for less risky shares to do well or for firms with lots of non-cash profits to do badly. They found that, at least outside the US, this continued even after academics had announced their discovery of these tendencies. Investors, it seems, don’t learn from others’ mistakes.

This might be because it is risky to buy apparently cheap stocks – after all, they can get even cheaper. Or it might be because it’s difficult and risky to sell overpriced shares. Whatever the reason, the markets remain inefficient.

All this has implications. If investors are systematically mistaken, the share market will do a bad job of overseeing company bosses. Research by Rene Stulz at Ohio State University and Kathleen Kahle at the University of Arizona finds this is the case.

They show there are fewer and less profitable firms listed on the US share market than 20 or 40 years ago. So the question is, then: what use are stock markets?

InvestingBiasResearch

Chris Dillow
Chris Dillow

Investors Chronicle writer and economist

Have your say

Do you know how much money you really earn and spend each month?