By LARRY SWEDROE
The first chapter in my book Investment Mistakes Even Smart People Make and How to Avoid Them is on overconfidence, focusing on the all-too-human tendency towards being overconfident of our skills—what could be called the “ego trap.”
Ask yourself the following questions:
Am I better than average in getting along with people?
Am I a better-than-average driver?
If you are the “average” person, you answered yes to both questions—researchers have found that about 80-90 percent of respondents answer in the affirmative to those types of questions. Obviously, 80-90 percent of the population cannot be better than average in getting along with others, nor can they be better-than-average drivers.
Psychologist Philip Tetlock spent years collecting more than 25,000 forecasts from people whose job it is to anticipate how the future will unfold — distinguished political scientists, economists, State Department consultants and television talking heads. The experts made predictions about such things as the success of the apartheid movement, secession efforts of French Canadians, and the progress of Gorbachev’s glasnost program. They did so by predicting which of three states was the most likely: The status quo would prevail, the existing trend would intensify, or the existing trend would reverse. They also indicated how confident they were.
Tetlock found that not only were there no real expert forecasters, but beyond a stark minimum, subject matter expertise translates more into overconfidence than forecasting accuracy — and the ability to spin elaborate tapestries of reasons for expecting “favourite” outcomes. He found that even when these experts were 80 percent sure they were right, they were right just 60 percent of the time; and when they were 100 percent certain they were right, they were right just 80 percent of the time. To make matters worse, Tetlock found that experts fall prey to hindsight bias, claiming after an event that they had predicted it before the fact. That helps keep alive the self-delusion that they really are experts.
Unfortunately, while only half the people can be better than average, most people believe they are above average. Overconfidence in our abilities may in some ways be a very healthy attribute. It makes us feel good about ourselves, creating a positive framework to get through life’s experiences. Unfortunately, being overconfident of our investment skills can lead to investment mistakes, such as taking more risk than we can handle (when the risk shows up, panic selling results), failing to diversify, and trading too much.
Research findings
Kent Daniel and David Hirshleifer, authors of the paper Overconfident Investors, Predictable Returns, and Excessive Trading, discussed the role overconfidence plays in investor returns. They cited a wealth of literature that demonstrates:
• People tend to be overoptimistic about their life prospects, and this optimism directly affects their financial decisions.
• Overconfidence has been documented among experts and professionals, including corporate financial officers as well as professional traders and investment bankers.
• Overconfidence includes overplacement (overestimation of one’s rank in a population on some positive dimension) and overprecision (overestimation of the accuracy of one’s beliefs). An example is the overestimation of one’s ability to predict stock market returns.
• A cognitive process that helps support overconfident beliefs is self-attribution bias—people credit their own talents and abilities for past successes while blaming their failures on bad luck. Self-attribution bias allows overconfidence to persist. When investors “get it right,” they upgrade their confidence in their beliefs; when they “get it wrong,” they fail to downgrade it.
• Individual investors trade individual stocks actively, and on average lose money by doing so. The more actively investors trade (due to overconfidence), the more they typically lose.
• The stocks individual investors buy tend to subsequently underperform, and the stocks they sell tend to subsequently outperform.
• Actively managed funds that charge high fees without delivering correspondingly high performance provides evidence that most individual investors in active funds are overconfident about their ability to select high-performing managers.
• Men are more overconfident than women in decision domains traditionally perceived as masculine, such as financial matters. Overconfidence leads to more trading. One study found that, consistent with higher confidence on the part of men, the average turnover for accounts opened by men is about 1.5 times higher than for accounts opened by women; as a result, men pay almost 1 percent more per year in higher transaction costs, and their net-of-fee returns are far lower.
• Individual investors tend to trade more after they experience high stock returns.
• Overconfidence is likely to be especially important when security markets are less liquid and when short selling is difficult or costly (i.e., limits to arbitrage are at work). When short selling becomes constrained, pessimists find it harder to trade on their views than optimists, resulting in overpricing. Thus, when overconfidence is combined with constraints on short sales, we expect the security to become overpriced.
Brad Barber, Xing Huang, Jeremy Ko and Terrance Odean, authors of the 2019 study Leveraging Overconfidence, hypothesised that “overconfident investors with a budget constraint use leverage more, trade more, and perform worse than well-calibrated investors.”
To confirm their hypothesis, they analyzed the behavior and performance of retail investors who use margin. They found that investors who trade on margin have greater overconfidence than both investors with margin accounts but no margin experience and investors with cash accounts. For example, investors with experience trading on margin were at the 65th percentile in their self-assessed financial knowledge but the 37th percentile on quizzed financial knowledge. They also found that margin account investors, especially margin experience investors, trade more, and more speculatively, than cash investors. Unfortunately, they found that they also have worse security selection ability than cash investors. The greater degree of overconfidence of margin investors not only led them to trade more often, but their stock-picking skills were even worse than the bad stock-picking skills of cash investors. The security selection skills of margin investors were so bad that their mean returns after buys were negative and returns after sells were positive. It was also interesting to note that margin investors tend to have higher incomes and wealth—their higher income levels and greater wealth may have contributed to their overconfidence. Their findings led the authors to conclude: “In sum, our evidence indicates that overconfidence—not better information—is a primary motivation for retail investors to trade, to their detriment, on margin. More generally, our analysis suggests overconfidence and leverage can be a dangerous mix.”
Given that they oversee more than $46 trillion in assets globally, an important question is: Are pension plan sponsors subject to the same behavioral mistakes that individual investors have been found to make?
Pension plan managers
Plan sponsors implement investment policy by hiring outside managers and, in some cases, managing certain assets in house. Those responsible for directing large plans tend to be highly experienced investment professionals with advanced degrees, supported by pension consultants, and well informed on both the markets and state-of-the-art portfolio management procedures. Unfortunately, the empirical research evidence demonstrates that they are subject to the same behavioural errors as individual investors, including the mistake of overconfidence — mistakes that lead to poor performance. In their 2008 study, The Selection and Termination of Investment Management Firms by Plan Sponsors, Amit Goyal and Sunil Wahal examined the subject. Their database covered approximately 3,700 plan sponsors from 1994 to 2003. They found:
• Plan sponsors hire investment managers after those managers earn large positive excess returns up to three years prior to hiring.
• Post-hiring excess returns are indistinguishable from zero.
• If plan sponsors had stayed with the fired investment managers, their returns would have improved—the fired managers outperformed their replacements!
Their findings are consistent with those of other studies on pension plan performance, such as The Performance of U.S. Pension Funds, A Panel Study of U.S. Equity Pension Fund Manager Style Performance and Picking Winners? Investment Consultants’ Recommendations of Fund Managers. The empirical research demonstrates that institutional investors tend to be trend followers, selecting managers after periods of three to five years of strong active returns only to be “rewarded” with reversals of performance. They all found no evidence to suggest that either pension plan sponsors or the consultants they hire add value by their selection of actively managed funds. That raises the question: Why do they continue to engage in the unproductive behavior of hiring and firing active managers? Scott Stewart, author of Performance, Perception, and Manager Selection, published in the April 2022 issue of The Journal of Portfolio Management, sought the answer to that question.
Examining the results of three investment surveys, Stewart found that while “the empirical evidence in the finance literature clearly demonstrates that institutional investors do not add value from selecting managers … plan sponsors are highly confident in their manager selection abilities.” For example, one survey conducted at the 2014 CFA Annual Conference asked the question, “Do you think that institutional investors add value from their manager selection?” Only 11 percent of the 2,041 respondents answered no. Another survey conducted at Financial Research Associates conferences in 2015 and 2016 asked the same question of 32 pension plan sponsors and consultants, of whom only 6 percent answered no. These plan sponsors either were unaware of the empirical evidence, or they believed that their own fund manager’s selection skills (or those of the consultants they hired) were above average (a sign of overconfidence).
Stewart hypothesised that their overconfidence resulted from their failure to evaluate the performance of the managers they fired. In support of his hypothesis, his analysis of data collected from a survey of 100 U.S. corporate and public pension plan officers responsible for more than $750 billion in assets requesting information on experience, training, investment processes and performance found “a near-zero statistical relationship between surveyed pension officers’ self-confidence and their belief in the importance of evaluating performance of investment decisions.” Stewart found that less than 11 percent of plan sponsors showed little confidence in their ability to select managers. He added: “Institutional investors continue to pursue faulty manager selection because they lack rigorous feedback mechanisms and, as a result, are unaware that their processes actually lose value.” He concluded: “Plan sponsors’ confidence in their manager selection is misplaced because it is not based on effective reviews of investment processes.” Stewart recommended “that plan sponsors collect and compare data on terminated managers with replacement managers to determine the true effectiveness of their selection processes.”
Also of interest was that Stewart found that greater investment experience did not lead to superior performance. Instead, it led to greater overconfidence, with its predictably negative results. The surveys also showed that on average plan officers agree they are confident in selecting managers. Sadly, they showed that plan officers also agree (though to a lesser extent) that their performance is “very good” — despite the evidence to the contrary.
Summary
Stewart’s findings on pension plan sponsors are entirely consistent with the research demonstrating that individual investors are overconfident about their investment skills, leading to poor results. Before you conclude that this is true of others, but not you, remember that the research demonstrates it doesn’t matter what the question is — whether you are a better-than-average driver, or you are liked more by others than the average person, or you are smarter than average — the typical survey finds that 80 percent or more believe they are above average. Overconfidence is particularly dangerous to investors because it can lead to excessive trading, a failure to diversify sufficiently to minimise idiosyncratic risks for which investors are not compensated, and to select active managers that underperform. Forewarned is forearmed.
© The Evidence-Based Investor MMXXIV. All rights reserved. Unauthorised use and/ or duplication of this material without express and written permission is strictly prohibited.