Overconfidence, Overweighting, and Overtrading — Normal Investor Behavior? Quite Possibly

Nicholas C. Barberis

When more than 400 investors gather to hear you speak on behavioral finance, there is a an air of anticipation — and at the 67th CFA Institute Annual Conference in Seattle, Nicholas Barberis, Stephen and Camille Schramm Professor of Finance at the Yale School of Management, certainly lived up to the hype.

Setting his stall out from the start, Barberis acknowledged that this would not be a talk merely justifying the case for behavioral finance — that had already been done. Instead, his goal was to encourage the audience to consider three ideas he sees as playing a key role in investor behavior:

  • Representativeness
  • Overconfidence
  • Prospect theory


Tackling representativeness first, Barberis referred to it as the “belief in the law of small numbers,” a mindset that draws individuals to rely too heavily on data samples that the more dispassionate may consider too small or too niche to warrant significant weighting. Illustrating his point, he outlined a “hot hand” scenario of the final seconds of a crucial basketball match. Your team has a free throw, and if they score they will win the game. There are two players who can take the free throw — one has scored highly today and is in top form, and the other is your season star but his game appears slightly cooler today. Who should take the shot? Many would say give the ball to the first player. But Barberis said that choice could be indicative of a short-term view, where the most recent data is deemed to outweigh prior information. If we extrapolate the players’ results for the whole season, who gains our trust?

He applied this concept to the stock market by highlighting a common behavior pattern where investors overweight recent data to predict future scenarios — such as investors in “good times” becoming caught up in a wave of positive expectancy, convinced that rising markets will continue to rise, and self-fueling fear of desolation when markets fall — when in fact longer-term data often shows high past returns are followed by low returns and low past returns are followed by high.


Addressing overconfidence, Barberis turned his attention to overtrading — a behavior commonly seen in retail investors and seasoned professionals alike. Overtrading, he said, can be easily identified and rarely leads to enhanced results. It’s a view supported by Brad Barber and Terrance Odean’s study of 66,000 households trading through a large discount brokerage, which showed that over a five-year period, the most active investors, on average, underperformed a range of benchmarks. A study by Mark Grinblatt and Matti Keloharju also found a significant link between overconfidence and overtrading.

Perhaps the most startling application of overtrading, Barberis suggested, can be seen in excessive M&A activities, in which overconfident CEOs feel sure their personal vision and unique strategies will deliver a more prosperous future for both organizations. In fact, analysis shows mergers rarely create value for the acquirer, with the stock of the bidder frequently underperforming over the subsequent three years.

Prospect Theory

Barberis contrasted prospect theory with the idealistic rational approach to risk outlined in the expected utility framework, in which our decision-making process would be guided by:

  • considering the different possible future outcomes;
  • deciding how good or bad each outcome will make them feel;
  • weighting each outcome by its probability, p; and
  • selecting a course of action.

Prospects theory’s two core features — loss aversion and probability weighting — outline how individuals think in terms of potential gains and losses, weight probabilities in a nonlinear way, and are much more sensitive to potential losses. One simple example of this, the overweighting of tail events, can be witnessed in the comfort people derive from holding comprehensive insurance policies or valid lottery tickets.

While loss aversion is better known, Barberis urged investors to pay careful attention to probability weighting. This, he said, predicts that the skewness of an asset’s returns will be priced — even idiosyncratic skewness. Positively skewed assets will be overpriced and will earn low average returns. Negatively skewed assets will be underpriced and will earn high average returns. One application Barberis cites was the review of average returns, ranked from high to low. While some average returns are puzzlingly high (such as the equity premium on the aggregate stock market), other average returns are puzzlingly low (such as the average return on IPO stocks in the five years after issue).

According to Barberis, the concept of probability weighting provides an elegant framework for understanding such peculiarities. The aggregate market has negatively skewed returns under probability weighting and should therefore have a high average return. On the other hand, IPO stocks have positively skewed returns and should therefore have a low average return under probability weighting.

Are markets driven purely by investors’ behavior traits and thought processes? Far from it. But covering a wealth of subtopics in great detail, and responding to a multitude of audience questions with refreshing objectivity, Barberis’s presentation provided a great opportunity for attendees to dig deeper, question assumptions, and challenge the speaker directly. While many may remain skeptical of elements of behavioral finance, its place as a central factor in the analysis of financial markets and assets is undoubted.

If you liked this post, consider subscribing to the CFA Institute Annual Conference blog.

Please note that the content of this site should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute.

Photo credit: W. Scott Mitchell

This entry was posted in Behavioral Finance, News and tagged . Bookmark the permalink.