Overconfidence in Tanks and Banks
Malcolm Gladwell’s recent article in the New Yorker argues that one of the key causes of the financial crisis was overconfidence among the men at the top, a phenomenon that ranges from the battlefields of history to the banks of the 21st century (“Cocksure: Banks, battles, and the psychology of overconfidence”, New Yorker, 27 July 2009). The important role of overconfidence even in high stakes decision-making is strongly supported by my book “Overconfidence and War: The Havoc and Glory of Positive Illusions”. For all the structural and intellectual obstacles to the effective conduct of war (or banking or any other human endeavor), humans—especially men—are excellent at creating their own obstacles with a remarkably pervasive and powerful overconfidence.
The problem with overconfidence is measuring it. After a disaster, it is easy to find misjudgments that look like overconfidence, given the outcome. However, misjudgments often precede successes too, but are forgotten by history. Gladwell’s example of Gallipoli illustrates this problem well. For all the apparent overconfidence leading to a lack of men and artillery, the British nevertheless “had a ten-to-one advantage over the Turks and could easily have seized the highlands overlooking the bay.” If they taken the initiative and won this “easy” battle, we would not be questioning the lack of men and machines at all.
Still, overconfidence can be tested in controlled scientific experiments. We ran wargames in which men (but not women) turned out to significantly overconfident about their chances of winning (“Overconfidence in Wargames”). Furthermore, those who were more overconfident were more likely to launch unprovoked attacks on their enemies. These confident and aggressive subjects also tended to score highly on personality tests of narcissism. The pattern that Galdwell describes is certainly apparent in the lab as well as in Gallipoli and Wall Street.
The big question is why we have these biases in the first place. From an evolutionary perspective, any bias that causes what are essentially decision-making “errors” leading to disasters should have been stamped out by natural selection. However, for all the havoc that overconfidence reaps, it also appears to bring glory. Not only does a little bit of overconfidence appear to improve resolve, perseverance, and effective bluffing, it is also disproportionately found in people who rise to positions of power. To reach the top (or even want it), you have to believe in yourself, believe that your ideas are better than everyone else’s, and you have to keep on believing that in the face of grinding daily criticism telling you exactly the opposite. Confidence is both a virtue and a vice, simultaneously propelling us towards great—even apparently impossible—things like landing on the moon, but equally to blame when we fall.