Can we make investors smarter using a nudge? Maybe, but we can’t prove it using the most common experimental disposition effect environment.

Abstract

Investors have been shown to behave in a way that reduces their earnings by being over hesitant to sell stocks that have decreased in price and over eager to sell stocks that have increased in price, exhibiting what is known as a disposition effect. This persists even in environments that make exhibiting a disposition effect always reduce expected returns. Our study uses the most common experimental disposition effect environment to test the use of a novel nudge to reduce participants disposition effects and finds that the nudge does reduce participants’ disposition effects. However, several of our findings challenge the external and internal validity of the environment, and it is possible that the nudge only works for a subset of the population that understands the environment better. Despite the environment making diversification suboptimal, those who understand diversification (and therefore might perform better in real-world markets) perform worse in this environment due to diversifying more, indicating that participants bring their external beliefs about real world markets into the environment. We show that the optimal disposition effect in the environment is substantially negative, which critiques past studies that have used a rational benchmark of zero. We also find significantly negative disposition effects across the board for our sample, which is unique, potentially due to the inclusion of comprehension questions before trading that assisted participants to understand the environment better.