The disposition effect messes with you
The disposition effect famously describes the tendency of investors to sell winners too early and hold on to losers for too long. This is obviously bad for performance because investments exhibit momentum and winners tend to keep going up after the investor sold them while losers tend to keep going down. From a pure performance perspective, investors should do the opposite and if you live in a country where there are capital gains taxes, you definitely should do the opposite since tax loss harvesting can improve your after-tax returns substantially.
Apparently, the disposition effect not only reduces your performance, it also messes with your head. Think about it. If you sell a winner and lock in the gains, it makes you feel good about your investment decisions and you can start boring all your friends with your stories of investment prowess. Heck, we live in the 21st century, you probably will start a Youtube channel and become an investment influencer based on your stellar track record.
Never mind that the rest of your portfolio performs poorly, your friends and followers won’t know that because they can’t see it. And apparently you won’t either, because you measure your performance and your skill based on the returns of investments you sold, not the whole portfolio.
Katrin Gödker, Terrence Odean, and Paul Smeets set up a couple of lab experiments where 301 investors had to invest in several stocks with predetermined probabilities of positive or negative returns. Then a random selection of participants was forced to sell some stocks in every round. In the condition called ‘selling gains’ participants were forced to sell stocks at a profit just like the disposition effect would predict, while in the ‘selling losses’ condition, participants were forced to sell stocks at a loss.
In the final round of the experiment, the participants were then asked to rate if they thought their investment returns were in the top half of all participants. Of course, we expect roughly 50% of participants to rate themselves above average and if you look at the self-assessments of women in the chart below, you can see that his is roughly correct. The average of the two bars is about 50%. For men, meanwhile, more than 50% thought they were above average. But what did we expect…
The bars below are corrected for the actual performance of the portfolio so any deviation from the 50% level shows the degree of overconfidence and underconfidence induced by the disposition effect. And it is clear that men and women who were forced to sell stocks at a profit were much more likely to think their performance was above average even if it was not. Which provides us once again with an important lesson about investing.
No, not that one.
It’s that we should hire women as investment managers. On average they are less overconfident than men.
Overconfidence triggered by the disposition effect
Source: Gödker et al. (2023)