Iconoclast: A Neuroscientist Reveals How to Think Differently Page 11
Jacob Bernoulli, a Swiss mathematician, proved this mathematically in 1713, and although the proof itself is complex, the idea is simple. Bernoulli’s proof has come to be known as the law of large numbers, because the more measurements you make of something, the more accurate the average of these measurements becomes. But we rarely have the opportunity to make as many measurements as the law would require. We take our best shot given the available information. Since we lack the possibility of do-overs, the next best thing is to see what other people do. Because an individual’s opinion is more likely to miss the mark than a group of people rendering independent opinions, the strategy of following the crowd can be very efficient.15 So if you want to know how many jelly beans are in a jar, ask a bunch of people what they think, and average their answers.
The law of large numbers is mathematically rock solid, and the only thing that is really surprising about it is that it took so long to be discovered. But just because we have known about it for only three hundred years doesn’t mean that its effects weren’t felt long before. Perception is a statistical judgment by the brain. Given the multiple interpretations of visual stimuli, the brain chooses the most likely interpretation. The interpretation may be guided by past experience and how the individual categorizes people and objects, but the law of large numbers comes into play as well. When other individuals render opinions, the brain readily incorporates these opinions and changes its interpretation of visual information. It is far too inefficient for an individual brain to make repeated guesses about what it is seeing, and when offered the opinion of other people as potentially independent observers (whether true or not), the brain will readily assimilate this information into its own interpretation and perception.
Viewed from the perspective of evolution, the law of large numbers will give any animal that uses it a distinct survival advantage. Consider a creature whose life depends on finding food and water. One strategy would be to forage for food in the hopes of stumbling upon something good to eat. If successful, this could pay off handsomely because the animal could horde it and gain an advantage over its competitors. This would be a risky strategy with a relatively low likelihood of success but a high payoff if it worked. The law of large numbers, however, says that such an individual strategy would most likely fail. Another animal, that was perhaps a little more strategic in its thinking, would observe what other animals did before deciding its own course of action. Because a group is statistically superior to an individual, an animal that discovered this strategy would always do better than the loner. In addition to vastly increasing the likelihood of success, the second strategy is much lower in risk and does not cost the animal much energy to observe what other animals do. The power of the group is so much greater than the individual, evolution favored animals that used it, and “groupthink” became the dominant strategy for all animals that could observe each other’s behavior.
Although statistically superior to individualistic thinking, the law of large numbers is also the bane of the iconoclast. Millions of years of evolution have resulted in a human brain that has the law of large numbers hardwired into it. When Asch observed individuals capitulating to the group, they were acting in accordance with the law and simply making the statistically sound judgment that the group was more likely correct than they themselves were. When my laboratory repeated the experiment with brain imaging, we saw the law of large numbers operating at multiple levels. The first level was at the perceptual stage, but we observed the fear system kicking in, almost like a fail-safe when the individual went against the group. These are powerful biological mechanisms that make it extremely difficult to think like an iconoclast. Our brains are evolved to make judgments as quickly and efficiently as possible, and when other people’s opinions are present, the brain will incorporate them, whether we want it to or not.
Mitigating the Effects of Fear on Perception
All the strategies outlined in the previous chapter pertaining to the amygdala apply here as well. Cognitive reappraisal, for example, works to effectively look at a situation that induces fear from a different vantage point. In addition, there are other strategies that work specifically in situations where the fear of isolation has the potential for changing perception. Nobody likes to look stupid, but the pain of being the odd person out often seems worse.
Fortunately, there is a straightforward workaround for the brain’s hardwired propensity to follow the herd. A minority of one is the most extreme form of iconoclasm, because it means that the individual stands entirely alone against the crowd. One possibility is to isolate oneself so that one doesn’t have to face others’ opinions. The tactic of avoidance, however, merely postpones the inevitable confrontation of the individual against the group. Another solution, in the spirit of Feynman, is to develop a tough skin and simply not care what others think. Although this can work sometimes, it runs the risk of coming off as being aloof or antisocial. It is best suited to dominant personalities.
For most iconoclasts, change begins on a small scale. When Asch repeated his experiments, he quickly discovered that only a unanimous group was effective in getting subjects to conform. One dissenter was typically enough to break the herd effect. From the iconoclast’s perspective, this means that the most effective strategy for dealing with a group is to recruit one like-minded individual. Although two people may not be sufficient to sway the group’s opinion, having one ally is all that is needed to maintain one’s own judgment. Groups are, indeed, superior to individuals, but only when they are diverse and individuals act as individuals. Statistically, most people in a group will lie along a spectrum of opinions, but because of the social pressure to belong, these opinions contract to the social norm. The availability of a minority position breaks the stranglehold of conformity, and groups that allow for minority opinions are statistically more likely to make better decisions than groups that require unanimity.
At an institutional level, the implications are clear: committees should not be required to arrive at a unanimous decision. Dissension must be encouraged. Although it is standard committee practice to go around a table and vote, this often results in an Asch effect because individuals have varying degrees of confidence in their judgments. A more effective strategy is to have individuals provide a numerical rating. This works well for binary decisions, where someone might rate zero for option A and ten for option B. The distance from the midpoint of the scale reflects the strength of their opinion. It also works well for decisions in which options must be ranked. And although not typically the norm for committees, closed balloting alleviates much of the stigma of social isolation.
Managers often do not like to hear these suggestions because they imply a certain reticence on the part of their employees. Individuals are hired and promoted according to their ability to perform a job and act independently. But even a board of directors contains a wide range of personality types. The most effective way for a group to make a decision is by aggregating the opinions of independent individuals. It also follows that a group with a lot of diversity among its members is more likely to arrive at a good decision than a group that is composed of members who are alike.
On an individual level, there are several effective strategies for mitigating fear. In addition to cognitive reappraisal, extinction is a useful approach. In general it is impossible to stay fearful of something for a long period of time. The prefrontal cortex can inhibit fear through repeated exposure. This works well when the fear is well defined and can be experienced on a repetitive basis. In committee situations, for example, the individual who is afraid of looking stupid in front of a group must force himself to voice his opinion. It can be painful at times, but it is only through repetition and practice that the fear response becomes attenuated and fear no longer clouds perception.
Finally, there is the strategy advocated by Martin Luther King. In his strategy, which is closely related to cognitive reappraisal, King appealed to the rational part of the brain to make the amygdala shut up. Re
alizing that fear is the enemy of civil rights, King made fear itself his target. Safety in numbers helps. But the real change must occur within the individual’s mind. Fortunately, fear is easily recognizable. One only needs to listen to the body’s responses to know that one is scared. But once fear is recognized the individual must bring online cognitive processes to deconstruct what the fear is. Only when the fear is broken down into its component pieces can it be eliminated. The key is recognizing the fear in the first place and not to make judgments while under the influence of fear.
Think of fear like alcohol. It impairs judgment. You shouldn’t make any decisions while under its influence.
FIVE
Why the Fear of Failure
Makes People Risk Averse
Making money in the stock market is so simple a monkey
could do it. Here’s the secret: Buy low and sell high.
—David Dreman
IN THE LAST TWO CHAPTERS, we saw how fear affected both perception and decision making. Fear prevents people from taking action, and even worse, fear changes the way they see the world. Fear touches everything that we do on a daily basis, but it is perhaps no more clearly evident than in the stock market. Here, we see fear manifest in all its glory.
Fear can be boiled down to three types. First, fear of the unknown, as in “What’s the deal with this Chinese company I’ve never heard of that my broker is pushing on me?” Second, fear of failure. In finance, this masquerades as “risk,” but we all know it as the fear of losing money. And finally, the fear of looking stupid. For many, there is nothing worse than your neighbor making a killing on a stock that you sold too soon. And if you run an investment fund, there really is nothing worse than being outperformed by your competitors or (God forbid) the S&P 500.
Few arenas are more punishing to the iconoclast than the stock market. If you’re in the market, you confront head-on the stark reality that when you invest money, you take a risk. I will get to the financial approach to risk shortly, but for now, think of risk simply as the odds of failure. Risky investments have a relatively high likelihood of tanking. Unfortunately, for most people, the emphasis falls on failure, and the fear of failure prevents them from taking risk, even when it can be profitable to do so. As we shall see, there are biological reasons for this behavior that originate in the brain’s distortion of perception, especially under the influence of fear. It is so common that simply not behaving this way makes for an iconoclast.
If you’re a fund manager, and you fancy yourself an iconoclast, think twice, because the stock market is an ideal setting to weed out the true iconoclasts from the pretenders. The New York Stock Exchange (NYSE) had a collective market capitalization of $21.2 trillion at the end of 2005. In a typical month, about $1 trillion in stocks changed hands. In 2002, 85 million people (42 percent of the U.S. adult population) either were directly invested in the NYSE or owned stocks that were traded on the NYSE through mutual funds and retirement plans. And this represents just the NYSE, not to mention the NASDAQ or foreign exchanges. With so many individuals actively participating in the market, the odds of being an iconoclast are slim, to say the least. Moreover, the public availability of fund performance makes it straightforward to see who the iconoclasts really are. The successful Wall Street iconoclast is the fund manager who beats the market on a consistent basis.
Morningstar reports statistics on over two thousand mutual funds. Despite the required warnings about past performance, the prospectuses of most of these funds paint a rosy picture of that fund’s investment strategy. But the dirty secret of Wall Street is that, in fact, hardly any funds consistently do well. According to Standard & Poor’s, only 10.8 percent of large-capitalization (large-cap) funds maintained a top-half ranking over five consecutive twelve-month periods.1 That’s top-half—just consistently better than average. The figure was even lower for mid- and small-cap funds. If we look for the real standouts—say, consistently in the top 25 percent of funds over five consecutive years—we find a grand total of three large-cap funds (1.12 percent) that make the grade.
The low percentage of funds that consistently do better than their peers indicates that little consistency exists in the market. These dismal results suggest that the warning that past performance is no guarantee of future results might be interpreted more accurately as this: past performance has nothing to do with future performance. If this were the case, then a fund’s ranking relative to its peers would be random from one year to the next, and whether a fund was in the top half in a given year would be like flipping a coin. Taking a top-half fund and flipping a hypothetical coin four more times means that 6.25 percent of funds would be expected to have consecutively good performance—by chance alone. The fact that 8–10 percent of funds eke out this feat means that fund performance is only a little better than coin flipping.
But this sets the bar fairly low. Once transaction fees are taken into account, actively managed funds rarely beat the market as a whole. Nothing could be further from iconoclastic investing than putting money in an index fund. Index funds contain stocks in proportion to their relative weight of major market indexes such as the S&P 500, which tracks large-cap companies, or the Russell 2000, which tracks small-cap companies.
The beauty of the market lies in the tenuous balance between buyers and sellers. The very definition of a market means that for every buyer there must be a seller. So for every person buying a stock, there is someone else who wants to get rid of it. Who is right? Who is the iconoclast?
The Economics of Risk
Returning to the randomness of mutual fund performance, we might ask what a reasonable person would be willing to pay for what amounts to little more than successive coin flips. Consider the following, which is a guaranteed way to make money. You walk into a bar and offer to pay $20 to anyone who is willing to take this bet. The bet goes like this: the person who accepts the bet places $2 on the table. You will flip a coin. If it comes up heads, the other person gets the $20 and the game is over (while you keep the $2, but realizing a net loss of $18). If it comes up tails, the taker of your bet must double the money in the pot, and you flip the coin again. Heads, you take the pot; tails, the taker must double down again. The game continues until the first heads appears.
Think that $20 is too much to pay for such a bet? How much would you be willing to pay?
The issue boils down to how much this game is worth. It’s like buying a lottery ticket. The most rational, and mathematically correct, way to calculate the value of a lottery is to multiply the payoff by the odds of its occurrence. On the first round, you’ve got a $2 payoff on the table, and the odds of winning it are one-half. Two dollars times one-half equals $1, so the expected value of the first round is $1. So far, this does not seem like a good bet. If you make it to the second round, the payoff doubles to $4, but the odds of this occurring are one-half times one-half, or one-fourth. The expected value of the second round is $4 times one-fourth, or again $1. The odds of making it to the third round (three tails in a row) drop to one-eighth, and the payoff doubles again, to $8. Thus, the expected value of every round is $1.
The expected value of the entire game is simply the sum of the value of each round. Since there are potentially an infinite number of rounds (albeit increasingly unlikely but with payoffs increasing exponentially), the expected value of the game is infinity. Therefore, a rational person should be willing to pay any amount of money to play this game. But, of course, nobody does.
The fact that people are unwilling to wager anything significant on this game, despite the mathematical rigor of the determination of its value, underscores the fundamentally irrational way that humans deal with risky decisions. This game, known as the St. Petersburg paradox, was articulated by the eighteenth-century Swiss mathematician Daniel Bernoulli, and his explanation of why people are unwilling to play this game forms the basis of the modern economic approach to risk.2
Bernoulli proposed an elegant solution to the paradox. He suggested that t
he reason people are unwilling to play this game stems from the fact that they don’t value money in a linear manner. To get around this limit, Bernoulli introduced the idea of utility. The value of something, be it a new car or a $100 bill, is governed not by its price, but by the utility it yields. Utility is the subjective benefit that a person experiences. The price, according to Bernoulli, depends only on the thing itself, but the utility it confers to someone depends on the individual. This makes intuitive sense. A $100 bill confers more utility to a pauper than to a rich man. To account for this observation, Bernoulli suggested that money has diminishing marginal utility. The more you have, the less utility each additional dollar adds. An extremely wealthy person would experience very little increase in utility from getting more money. This may seem irrational, and it is, but then again, not playing the St. Petersburg lottery is itself an irrational act.
Bernoulli proposed that the utility of money follows a logarithmic curve, which has the property of flattening out the higher you go. If people make decisions about money according to the utility they get, as opposed to the actual face value, then the mathematical logic assumes a different form. Although the exponentially decreasing odds of successive tail flips are balanced by the exponential doubling of the pot, the utility of the pot does not keep pace with the decrease in odds. The utility of the game is no longer infinite, which means that every person will have some finite price they are willing to pay to play.