Fear, Greed, and Cognitive Illusions
yrou don’t need to have been a temporarily besotted investor to realize that psychology plays an important and sometimes crucial role in the market, but it helps. By late summer 2000, WCOM had declined to $30 per share, inciting me to buy more. As “inciting” may suggest, my purchases were not completely rational. By this I don’t mean that there wasn’t a rational basis for investing in WCOM stock. If you didn’t look too closely at the problems of overcapacity and the longdistance phone companies’ declining revenue streams, you could find reasons to keep buying. It’s just that my reasons owed less to an assessment of trends in telecommunications or an analysis of company fundamentals than to an unsuspected gambling instinct and a need to be right. I suffered from “confirmation bias” and searched for the good news, angles, and analyses about the stock while avoiding the less sanguine indications.
Averaging Down or Catching a Falling Knife?
After an increasingly intense, albeit one-sided courtship of the stock (the girl never even sent me a dividend), I married it. As
13
14 J o h n Allen Paulos
its share price fell, I continued to see only opportunities for gains. Surely, I told myself, the stock had reached its bottom and it was now time to average down by buying the considerably cheaper shares. Of course, for every facile invitation I extended myself to “average down,” I ignored an equally facile warning about not attempting to “catch a falling knife.” The stale, but prudent adage about not putting too many of one’s eggs in the same basket never seemed to push itself very forcefully into my consciousness.
I was also swayed by Salomon Smith Barney’s Jack Grub- man (possessor, incidentally, of a master’s degree in mathematics from Columbia) and other analysts, who ritualistically sprinkled their “strong buys” over the object of my affections. In fact, most brokerage houses in early 2000 rated WCOM a “strong buy,” and those that didn’t had it as a “buy.” It required no great perspicacity to notice that at the time, almost no stock ever received a “sell,” much less a “strong sell,” and that even “holds” were sparingly bestowed. Maybe, I thought, only environmental companies that manufactured solar- powered flashlights qualified for these latter ratings. Accustomed to grade inflation and to movie, book, and restaurant review inflation, I wasn’t taken in by the uniformly positive ratings. Still, just as you can be moved by a television commercial whose saccharine dialogue you are simultaneously ridiculing, part of me gave credence to all those “strong buys.”
I kept telling myself that I’d incurred only paper losses and had lost nothing real unless I sold. The stock would come back, and if I didn’t sell, I couldn’t lose. Did I really believe this? Of course not, but I acted as if I did, and “averaging down” continued to seem like an irresistible opportunity. I believed in the company, but greed and fear were already doing their usual two-step in my head and, in the process, stepping all over my critical faculties.
A M athematician Plays the Stock Market 15
Emotional Overreactions and Homo Economicus
Investors can become (to borrow a phrase Alan Greenspan and Robert Shiller made famous) irrationally exuberant, or, changing the arithmetical sign, irrationally despairing. Some of the biggest daily point gains and declines in Nasdaq’s history occurred in a single month in early 2000, and the pattern has continued unabated in 2001 and 2002, the biggest point gain since 1987 occurring on July 24, 2002. (The increase in volatility, although substantial, is a little exaggerated since our perception of gains and losses have been distorted by the rise in the indices. A 2 percent drop in the Dow when the market is at 9,000 is 180 points, whereas not too long ago when it was at 3,000, the same percentage drop was only 60 points.) The volatility has come about as the economy has hovered near a recession, as accounting abuses have come to light, as CEO malfeasance has mounted, as the bubble has fizzled, and as people have continued to trade on their own, influenced no doubt by capricious lists of the fifty most beautiful (er ..., undervalued) stocks.
As with beautiful people and, for that matter, distinguished universities, emotions and psychology are imponderable factors in the market’s jumpy variability. Just as beauty and academic quality don’t change as rapidly as ad hoc lists and magazine rankings do, so, it seems, the fundamentals of companies don’t change as quickly as our mercurial reactions to news about them do.
It may be useful to imagine the market as a fine race car whose exquisitely sensitive steering wheel makes it impossible to drive in a straight line. Tiny bumps in our path cause us to swerve wildly, and we zigzag from fear to greed and back again, from unreasonable gloom to irrational exuberance and back.
16 J o h n Allen Paulos
Our overreactions are abetted by the all-crisis-all-the-time business media, which brings to mind a different analogy: the reigning theory in cosmology. The inflationary universe hypothesis holds—very, very roughly—that shortly after the Big Bang the primordial universe inflated so fast that all of our visible universe derives from a tiny part of it; we can’t see the rest. The metaphor is strained (in fact I just developed carpal tunnel syndrome typing it), but it seems reminiscent of what happens when the business media (as well as the media in general) focus unrelentingly on some titillating but relatively inconsequential bit of news. Coverage of the item expands so fast as to distort the rest of the global village and render it invisible.
Our responses to business news are only one of the ways in which we fail to be completely rational. More generally, we simply don’t always behave in ways that maximize our economic well-being. “Homo economicus” is not an ideal toward which many people strive. My late father, for example, was distinctly uneconimicus. I remember him sitting and chuckling on the steps outside our house one autumn night long ago. I asked what was funny and he told me that he had been watching the news and had heard Bob Buhl, a pitcher for the then Milwaukee Braves, answer a TV reporter’s question about his off-season plans. “Buhl said he was going to help his father up in Saginaw, Michigan, during the winter.” My father laughed again and continued. “And when the reporter asked Buhl what his father did up in Saginaw, Buhl said, ‘Nothing at all. He does nothing at all.’ ”
My father liked this kind of story and his crooked grin lingered on his face. This memory was jogged recently when I was straightening out my office and found a cartoon he had sent me years later. It showed a bum sitting happily on a park bench as a line of serious businessmen traipsed by him. The bum calls out “Who’s winning?” Although my father was a
salesman, he always seemed less intent on making a sale than on schmoozing with his customers, telling jokes, writing poetry (not all of it doggerel), and taking innumerable coffee breaks.
Everyone can tell such stories, and you would be hard- pressed to find a novel, even one with a business setting, where the characters are all actively pursuing their economic self-interest. Less anecdotal evidence of the explanatory limits of the homo economicus ideal is provided by so-called “ultimatum games.” These generally involve two players, one of whom is given a certain amount of money, say $100, by an experimenter, and the other of whom is given a kind of veto. The first player may offer any non-zero fraction of the $100 to the second player, who can either accept or reject it. If he accepts it, he is given whatever amount the first player has offered, and the first player keeps the balance. If he rejects it, the experimenter takes the money back.
Viewing this in rational game-theoretic terms, one would argue that it’s in the interest of the second player to accept whatever is offered since any amount, no matter how small, is better than nothing. One would also suspect that the first player, knowing this, would make only tiny offers to the second player. Both suppositions are false. The offers range up to 50 percent of the money involved, and, if deemed too small and therefore humiliating, they are sometimes rejected. Notions of fairness and equality, as well as anger and revenge, seem to play a role.
Behavioral Finance
People’s reactions to ultimatum games may be counterproductive, but they are at least clear-eyed. A number of psychologists in recent years have pointed out the countless ways in
A Mathematician Plays the Stock Market 17
18 J o h n Allen Paulos
which we’re all subject to other sorts of counterproductive behavior that spring from cognitive blind spots that are analogues, perhaps, of optical illusions. These psychological illusions and foibles often make us act irrationally in a variety of disparate endeavors, not the least of which is investing.
Amos Tversky and Daniel Kahneman are the founders of this relatively new field of study, many of whose early results are reported upon in the classic book Judgment Under Uncertainty, edited by them and Paul Slovic. (Kahneman was awarded the 2002 Nobel Prize in economics, and Tversky almost certainly would have shared it had he not died.) Others who have contributed to the field include Thomas Gilovich, Robin Dawes, J. L. Knetschin, and Baruch Fischhoff. Economist Richard Thaler (mentioned in the first chapter) is one of the leaders in applying these emerging insights to economics and finance, and his book The Winner’s Curse, as well as Gilovich’s How We Know What Isn’t So, are very useful com- pendiums of recent results.
What makes these results particularly intriguing is the way they illuminate the tactics used, whether consciously or not, by people in everyday life. For example, a favorite ploy of activists of all ideological stripes is to set the terms of a debate by throwing out numbers, which need have little relation to reality to be influential. If you are appalled at some condition, you might want to announce that more than 50,000 deaths each year are attributable to it. By the time people catch up and realize that the number is a couple of orders of magnitude smaller, your cause will be established.
Unfounded financial hype and unrealistic “price targets” have the same effect. Often, it seems, an analyst cites a “price target” for a stock in order to influence investors by putting a number into their heads. (Since the targets are so often indistinguishable from wishes, shouldn’t they always be infinite?)
A Mathematician Plays th e Stock Market 19
The reason for the success of this hyperbole is that most of us suffer from a common psychological failing. We credit and easily become attached to any number we hear. This tendency is called the “anchoring effect” and it’s been demonstrated to hold in a wide variety of situations.
If an experimenter asks people to estimate the population of Ukraine, the size of Avogadro’s number, the date of an historical event, the distance to Saturn, or the earnings of XYZ Corporation two years from now, their guesses are likely to be fairly close to whatever figure the experimenter first suggests as a possibility. For example, if he prefaces his request for an estimate of the population of Ukraine with the question—“Is it more or less than 200 million people?”—the subjects’ estimates will vary and generally be a bit less than this figure, but still average, say, 175 million people. If he prefaces his request for an estimate with the question—“Is the population of Ukraine more or less than 5 million people?”—the subjects’ estimates will vary and this time be a bit more than this figure, but still average, say, 10 million people. The subjects usually move in the right direction from whatever number is presented to them, but nevertheless remain anchored to it.
You might think this is a reasonable strategy for people to follow. They might realize they don’t know much about Ukraine, chemistry, history, or astronomy, and they probably believe the experimenter is knowledgeable, so they stick close to the number presented. The astonishing strength of the tendency comes through, however, when the experimenter obtains his preliminary number by some chance means, say by spinning a dial that has numbers around its periphery—300 million, 200 million, 50 million, 5 million, and so on. Say he spins the dial in front of the subjects, points out where it has stopped, and then asks them if the population of Ukraine is more or less than the number at which the dial has stopped.
20 J o h n Allen Paulos
The subjects’ guesses are still anchored to this number even though, one presumes, they don’t think the dial knows anything about Ukraine!
Financial numbers are also vulnerable to this sort of manipulation, including price targets and other uncertain future figures like anticipated earnings. The more distant the future the numbers describe, the more it’s possible to postulate a huge figure that is justified, say, by a rosy scenario about the exponentially growing need for bandwidth or online airline tickets or pet products. People will discount these estimates, but usually not nearly enough. Some of the excesses of the dot-coms are probably attributable to this effect. On the sell side too, people can paint a dire picture of ballooning debt or shrinking markets or competing technology. Once again, the numbers presented, this time horrific, need not have much to do with reality to have an effect.
Earnings and targets are not the only anchors. People often remember and are anchored to the fifty-two-week high (or low) at which the stock had been selling and continue to base their deliberations on this anchor. I unfortunately did this with WCOM. Having first bought the stock when it was in the forties, I implicitly assumed it would eventually right itself and return there. Later, when I bought more of it in the thirties, twenties, and teens, I made the same assumption.
Another, more extreme form of anchoring (although there are other factors involved) is revealed by investors’ focus on whether the earnings that companies announce quarterly meet the estimates analysts have established for them. When companies’ earnings fall short by a penny or two per share, investors sometimes react as if this were tantamount to nearbankruptcy. They seem to be not merely anchored to earnings estimates but fetishistically obsessed with them.
Not surprisingly, studies have shown that companies’ earnings are much more likely to come in a penny or two above
A Mathematician Plays th e Stock Market 21
the analysts’ average estimate than a penny or two below it. If earnings were figured without regard to analysts’ expectations, they’d come in below the average estimate as often as above it. The reason for the asymmetry is probably that companies sometimes “back in” to their earnings. Instead of determining revenues and expenses and subtracting the latter from the former to obtain earnings (or more complicated variants of this), companies begin with the earnings they need and adjust revenues and expenses to achieve them.
Psychological Foibles, A List
The anchoring effect is not the only way in which our faculties are clouded. The “availability error” is the inclination to view any story, whether political, personal, or financial, through the lens of a superficially similar story that is psychologically available. Thus every recent American military involvement is inevitably described somewhere as “another Vietnam.” Political scandals are immediately compared to the Lewinsky saga or Watergate, misunderstandings between spouses reactivate old wounds, normal accounting questions bring the Enron-Andersen-WorldCom fiasco to mind, and any new high-tech firm has to contend with memories of the dot-com bubble. As with anchoring, the availability error can be intentionally exploited.
The anchoring effect and availability error are exacerbated by other tendencies. “Confirmation bias” refers to the way we check a hypothesis by observing instances that confirm it and ignoring those that don’t. We notice more readily and even diligently search for whatever might confirm our beliefs, and we don’t notice as readily and certainly don’t look hard for what disconfirms them. Such selective thinking reinforces the anchoring effect: We naturally begin to look for reasons
22 J o h n Allen Paulos
that the arbitrary number presented to us is accurate. If we succumb completely to the confirmation bias, we step over the sometimes fine line separating flawed rationality and hopeless closed-mindedness.
Confirmation bias is not irrelevant to stock-picking. We tend to gravitate toward those people whose take on a stock is similar to our own and to search more vigorously for positive information on the stock. When I visited WorldCom chatrooms, I more often clicked on postings written by people characterizing themselves as “strong buys” than I did on those written by “strong sells.” I also paid more attention to WorldCom’s relatively small deals with web-hosting companies than to the larger structural problems in the telecommunications industry.
The “status quo bias” (these various biases are generally not independent of each other) also applies to investing. If subjects are told, for example, that they’ve inherited a good deal of money and then asked which of four investment options (an aggressive stock portfolio, a more balanced collection of equities, a municipal bond fund, or U.S. Treasuries) they would prefer to invest it in, the percentages choosing each are fairly evenly distributed.
Surprisingly, however, if the subjects are told that they’ve inherited the money but it is already in the form of municipal bonds, almost half choose to keep it in bonds. It’s the same with the other three investment options: Almost half elect to keep the money where it is. This inertia is part of the reason so many people sat by while not only their inheritances but their other investments dwindled away. The “endowment effect,” another kindred bias, is an inclination to endow one’s holdings with more value than they have simply because one holds them. “It’s my stock and I love it.”
Related studies suggest that passively endured losses induce less regret than losses that follow active involvement. Some-
A Mathematician Plays th e Stock Market 23
one who sticks with an old investment that then declines by 25 percent is less upset than someone who switches into the same investment before it declines by 25 percent. The same fear of regret underlies people’s reluctance to trade lottery tickets with friends. They imagine how they’ll feel if their original ticket wins.
Minimizing possible regret often plays too large a role in investors’ decisionmaking. A variety of studies by Tversky, Kahneman, and others have shown that most people tend to assume less risk to obtain gains than they do to avoid losses. This isn’t implausible: Other research suggests that people feel considerably more pain after incurring a financial loss than they do pleasure after achieving an equivalent gain. In the extreme case, desperate fears about losing a lot of money induce people to take enormous risks with their money.
Consider a rather schematic outline of many of the situations studied. Imagine that a benefactor gives $10,000 to everyone in a group and then offers each of them the following choice. He promises to a) give them an additional $5,000 or else b) give them an additional $10,000 or $0, depending on the outcome of a coin flip. Most people choose to receive the additional $5,000. Contrast this with the choice people in a different group make when confronted with a benefactor who gives them each $20,000 and then offers the following choice to each of them. He will a) take from them $5,000 or else b) will take from them $10,000 or $0, depending on the flip of a coin. In this case, in an attempt to avoid any loss, most people choose to flip the coin. The punchline, as it often is, is that the choices offered to the two groups are the same: a sure $15,000 or a coin flip to determine whether they’ll receive $10,000 or $20,000.
Alas, I too took more risks to avoid losses than I did to obtain gains. In early October 2000, WCOM had fallen below $20, forcing the CEO, Bernie Ebbers, to sell 3 million shares
24 J o h n Allen Paulos
to pay off some of his investment debts. The WorldCom chatrooms went into one of their typical frenzies and the price dropped further. My reaction, painful to recall, was, “At these prices I can finally get out of the hole.” I bought more shares even though I knew better. There was apparently a loose connection between my brain and my fingers, which kept clicking the buy button on my Schwab online account in an effort to avoid the losses that loomed.
Outside of business, loss aversion plays a role as well. It’s something of a truism that the attempt to cover up a scandal often leads to a much worse scandal. Although most people know this, attempts to cover up are still common, presumably because, here too, people are much more willing to take risks to avoid losses than they are to obtain gains.
Another chink in our cognitive apparatus is Richard Thaler’s notion of “mental accounts,” mentioned in the last chapter. “The Legend of the Man in the Green Bathrobe” illustrates this notion compellingly. It is a rather long shaggy dog story, but the gist is that a newlywed on his honeymoon in Las Vegas wakes up in bed and sees a $5 chip left on the dresser. Unable to sleep, he goes down to the casino (in his green bathrobe, of course), bets on a particular number on the roulette wheel, and wins. The 35 to 1 odds result in a payout of $175, which the newlywed promptly bets on the next spin. He wins again and now has more than $6,000. He bets everything on his number a couple more times, continuing until his winnings are in the millions and the casino refuses to accept such a large bet. The man goes to a bigger casino, wins yet again, and now commands hundreds of millions of dollars. He hesitates and then decides to bets it all one more time. This time he loses. In a daze, he stumbles back up to his hotel room where his wife yawns and asks how he did. “Not too bad. I lost $5.”
It’s not only in casinos and the stock market that we categorize money in odd ways and treat it differently depending
A Mathematician Plays th e Stock Market 25
on what mental account we place it in. People who lose a $100 ticket on the way to a concert, for example, are less likely to buy a new one than are people who lose $100 in cash on their way to buy the ticket. Even though the amounts are the same in the two scenarios, people in the former one tend to think $200 is too large an expenditure from their entertainment account and so don’t buy a new ticket, while people in the latter tend to assign $100 to their entertainment account and $100 to their “unfortunate loss” account and buy the ticket.
In my less critical moments (although not only then) I mentally amalgamate the royalties from this book, whose writing was prompted in part by my investing misadventure, with my WCOM losses. Like corporate accounting, personal accounting can be plastic and convoluted, perhaps even more so since, unlike corporations, we are privately held.
These and other cognitive illusions persist for several reasons. One is that they lead to heuristic rules of thumb that can save time and energy. It’s often easier to go on automatic pilot and respond to events in a way that requires little new thinking, not just in scenarios involving eccentric philanthropists and sadistic experimenters. Another reason for the illusions’ persistence is that they have, to an extent, become hardwired over the eons. Noticing a rustle in the bush, our primitive ancestors were better off racing away than they were plugging into Bayes’ theorem on conditional probability to determine if a threat was really likely.
Sometimes these heuristic rules lead us astray, again not just in business and investing but in everyday life. Early in the fall 2002 Washington, D.C., sniper case, for example, the police arrested a man who owned a white van, a number of rifles, and a manual for snipers. It was thought at the time that there was one sniper and that he owned all these items, so for the purpose of this illustration let’s assume that this turned out to
26 J o h n Allen Paulos
be true. Given this and other reasonable assumptions, which is higher—a) the probability that an innocent man would own all these items, or b) the probability that a man who owned all these items would be innocent? You may wish to pause before reading on.
Most people find questions like this difficult, but the second probability would be vastly higher. To see this, let me make up some plausible numbers. There are about 4 million innocent people in the suburban Washington area and, we’re assuming, one guilty one. Let’s further estimate that ten people (including the guilty one) own all three of the items mentioned above. The first probability—that an innocent man owns all these items—would be 9/4,000,000 or 1 in 400,000. The second probability—that a man owning all three of these items is innocent—would be 9/10. Whatever the actual numbers, these probabilities usually differ substantially. Confusing them is dangerous (to defendants).
Self-Fulfilling Beliefs and Data Mining
Taken to extremes, these cognitive illusions may give rise to closed systems of thought that are immune, at least for a while, to revision and refutation. (Austrian writer and satirist Karl Kraus once remarked, “Psychoanalysis is that mental illness for which it regards itself as therapy.”) This is especially true for the market, since investors’ beliefs about stocks or a method of picking them can become a self-fulfilling prophecy. The market sometimes acts like a strange beast with a will, if not a mind, of its own. Studying it is not like studying science and mathematics, whose postulates and laws are (in quite different senses) independent of us. If enough people suddenly wake up believing in a stock, it will, for that reason alone, go up in price and justify their beliefs.
A Mathematician Plays th e Stock Market 27
A contrived but interesting illustration of a self-fulfilling belief involves a tiny investment club with only two investors and ten possible stocks to choose from each week. Let’s assume that each week chance smiles at random on one of the ten stocks the investment club is considering and it rises precipitously, while the week’s other nine stocks oscillate within a fairly narrow band.
George, who believes (correctly in this case) that the movements of stock prices are largely random, selects one of the ten stocks by rolling a die (say an icosehedron—a twenty-sided solid—with two sides for each number). Martha, let’s assume, fervently believes in some wacky theory, Q analysis. Her choices are therefore dictated by a weekly Q analysis newsletter that selects one stock of the ten as most likely to break out. Although George and Martha are equally likely to pick the lucky stock each week, the newsletter-selected stock will result in big investor gains more frequently than will any other stock.
The reason is simple but easy to miss. Two conditions must be met for a stock to result in big gains for an investor: It must be smiled upon by chance that week and it must be chosen by one of the two investors. Since Martha always picks the newsletter-selected stock, the second condition in her case is always met, so whenever chance happens to favor it, it results in big gains for her. This is not the case with the other stocks. Nine-tenths of the time, chance will smile on one of the stocks that is not newsletter-selected, but chances are George will not have picked that particular one, and so it will seldom result in big gains for him. One must be careful in interpreting this, however. George and Martha have equal chances of pulling down big gains (10 percent), and each stock of the ten has an equal chance of being smiled upon by chance (10 percent), but the newsletter-selected stock will achieve big gains much more often than the randomly selected ones.
28 J o h n Allen Paulos
Reiterated more numerically, the claim is that 10 percent of the time the newsletter-selected stock will achieve big gains for Martha, whereas each of the ten stocks has only a 1 percent chance of both achieving big gains and being chosen by George. Note again that two things must occur for the newsletter-selected stock to achieve big gains: Martha must choose it, which happens with probability 1, and it must be the stock that chance selects, which happens with probability 1/10th. Since one multiplies probabilities to determine the likelihood that several independent events occur, the probability of both these events occurring is 1 x 1/10, or 10 percent. Likewise, two things must occur for any particular stock to achieve big gains via George: George must choose it, which occurs with probability 1/10th, and it must be the stock that chance selects, which happens with probability 1/10th. The product of these two probabilities is 1/100th or 1 percent.
Nothing in this thought experiment depends on there being only two investors. If there were one hundred investors, fifty of whom slavishly followed the advice of the newsletter and fifty of whom chose stocks at random, then the newsletter- selected stocks would achieve big gains for their investors eleven times as frequently as any particular stock did for its investors. When the newsletter-selected stock is chosen by chance and happens to achieve big gains, there are fifty-five winners, the fifty believers in the newsletter and five who picked the same stock at random. When any of the other nine stocks happens to achieve big gains, there are, on average, only five winners.
In this way a trading strategy, if looked at in a small population of investors and stocks, can give the strong illusion that it is effective when only chance is at work.
“Data mining,” the scouring of databases of investments, stock prices, and economic data for evidence of the effectiveness of this or that strategy, is another example of how an
inquiry of limited scope can generate deceptive results. The problem is that if you look hard enough, you will always find some seemingly effective rule that resulted in large gains over a certain time span or within a certain sector. (In fact, inspired by the British economist Frank Ramsey, mathematicians over the last half century have proved a variety of theorems on the inevitability of some kind of order in large sets.) The promulgators of such rules are not unlike the believers in bible codes. There, too, people searched for coded messages that seemed to be meaningful, not realizing that it’s nearly impossible for there not to be some such “messages.” (This is trivially so if you search in a book that has a chapter 11, conveniently foretelling many companies’ bankruptcies.)
People commonly pore over price and trade data attempting to discover investment schemes that have worked in the past. In a reductio ad absurdum of such unfocused fishing for associations, David Leinweber in the mid-90s exhaustively searched the economic data on a United Nations CD-ROM and found that the best predictor of the value of the S8cP 500 stock index was—a drum roll here—butter production in Bangladesh. Needless to say, butter production in Bangladesh has probably not remained the best predictor of the S&P 500. Whatever rules and regularities are discovered within a sample must be applied to new data if they’re to be accorded any limited credibility. You can always arbitrarily define a class of stocks that in retrospect does extraordinarily well, but will it continue to do so?
I’m reminded of a well-known paradox devised (for a different purpose) by the philosopher Nelson Goodman. He selected an arbitrary future date, say January 1, 2020, and defined an object to be “grue” if it is green and the time is before January 1, 2020, or if it is blue and the time is after January 1, 2020. Something is “bleen,” on the other hand, if it is blue and the time is before that date or if it is green and the
A Mathematician Plays th e Stock Market 29
30 J o h n Allen Paulos
time is after that date. Now consider the color of emeralds. All emeralds examined up to now (2002) have been green. We therefore feel confident that all emeralds are green. But all emeralds so far examined are also grue. It seems that we should be just as confident that all emeralds are grue (and hence blue beginning in 2020). Are we?
A natural objection is that these color words grue and bleen are very odd, being defined in terms of the year 2020. But were there aliens who speak the grue-bleen language, they could make the same charge against us. “Green,” they might argue, is an arbitrary color word, being defined as grue before 2020 and bleen afterward. “Blue” is just as odd, being bleen before 2020 and grue from then on. Philosophers have not convincingly shown what exactly is wrong with the terms grue and bleen, but they demonstrate that even the abrupt failure of a regularity to hold can be accommodated by the introduction of new weasel words and ad hoc qualifications.
In their headlong efforts to discover associations, data miners are sometimes fooled by “survivorship bias.” In market usage this is the tendency for mutual funds that go out of business to be dropped from the average of all mutual funds. The average return of the surviving funds is higher than it would be if all funds were included. Some badly performing funds become defunct, while others are merged with betterperforming cousins. In either case, this practice skews past returns upward and induces greater investor optimism about future returns. (Survivorship bias also applies to stocks, which come and go over time, only the surviving ones making the statistics on performance. WCOM, for example, was unceremoniously replaced on the S&P 500 after its steep decline in early 2002.)
The situation is rather like that of schools that allow students to drop courses they’re failing. The grade point averages of schools with such a policy are, on average, higher
A Mathematician Plays th e Stock M arket 31
than those of schools that do not allow such withdrawals. But these inflated GPAs are no longer a reliable guide to students’ performance.
Finally, taking the meaning of the term literally, survivorship bias makes us all a bit more optimistic about facing crises. We tend to see only those people who survived similar crises. Those who haven’t are gone and therefore much less visible.
Rumors and Online Chatrooms
Online chatrooms are natural laboratories for the observation of illusions and distortions, although their psychology is more often brutally basic than subtly specious. While spellbound by WorldCom, I would spend many demoralizing, annoying, and engaging hours compulsively scouring the various WorldCom discussions at Yahoo! and RagingBull. Only a brief visit to these sites is needed to see that a more accurate description of them would be rantrooms.
Once someone dons a screen name, he (the masculine pronoun, I suspect, is almost always appropriate) usually dispenses with grammar, spelling, and most conventional standards of polite discourse. Other people become morons, idiots, and worse. A poster’s references to the stock, if he’s shorting it (selling shares he doesn’t have in the hope that he can buy them back when the price goes down), put a burden on one’s ability to decode scatological allusions and acronyms. Any expression of pain at one’s losses is met with unrelenting scorn and sarcasm; ostensibly genuine musings about suicide are no exception. A suicide threat in April 2002, lamenting the loss of house, family, and job because of WCOM, drew this response: “You sad sack loser. Die. You might want to write a note too in case the authorities and your wife don’t read the Yahoo! chatrooms.”
32 J o h n Allen Paulos
People who characterize themselves as sellers are generally (but not always) more vituperative than those claiming to be buyers. Some of the regulars appear genuinely interested in discussing the stock rationally, imparting information, and exchanging speculation. A few seem to know a lot, many are devotees of various outlandish conspiracy theories, including the usual anti-Semitic sewage, and even more are just plain clueless, asking, for example, why they “always put that slash between the P and the E in P/E, and is P price or profit.” There were also many discussions that had nothing directly to do with the stock. One that I remember fondly was about someone who called a computer help desk because his computer didn’t work. It turned out that he had plugged the computer and all his peripheral devices into his surge protector, which he then plugged into itself. The connection with whatever company was being discussed I’ve forgotten.
Taking advice from such an absurdly skewed sample of posters is silly, of course, but the real-time appeal of the sites is akin to overhearing gossip about a person you’re interested in. It’s likely to be false, spun, or overstated, but it still holds a certain fascination. Another analogy is to listening to police radio and getting a feel for the raw life and death on the streets.
Chatroom denizens form little groups that spend a lot of time excoriating, but not otherwise responding to, opposing groups. They endorse each other’s truisms and denounce those of the others. When WorldCom purchased a small company or had a reversal in its Brazilian operations, this was considered big news. It was not nearly as significant, however, as an analyst changing his recommendation from a strong buy to a buy or vice versa. If you filter out the postings drenched in anger and billingsgate, you find most of the biases mentioned above demonstrated on a regular basis. The posters are averse to risk, anchored to some artificial number,
A M athematician Plays th e Stock Market 33
addicted to circular thought, impressed by data mining, or all of the above.
Most boards I visited had a higher percentage of rational posters than did WorldCom’s. I remember visiting the Enron board and reading rumors of the bogus deals and misleading accounting practices that eventually came to light. Unfortunately, since there are always rumors of every conceivable and contradictory sort (sometimes posted by the same individual), one cannot conclude anything from their existence except that they’re likely to contribute to feelings of hope, fear, anger, and anxiety.
Pump and Dump, Short and Distort
The rumors are often associated with market scams that exploit people’s normal psychological reactions. Many of these reactions are chronicled in Edwin Lefevre’s 1923 classic novel, Reminiscences of a Stock Operator, but the standard “pump and dump” is an illegal practice that has gained new life on the Internet. Small groups of individuals buy a stock and tout it in a misleading hyperbolic way (that is, pump it). Then when its price rises in response to this concerted campaign, they sell it at a profit (dump it). The practice works best in bull markets when people are most susceptible to greed. It is also most effective when used on thinly traded stocks where a few buyers can have a pronounced effect.
Even a single individual with a fast Internet connection and a lot of different screen names can mount a pump and dump operation. Just buy a small stock from an online broker, then visit the chatroom where it’s discussed. Post some artful innuendoes or make some outright phony claims and then back yourself up with one of your pseudonyms. You can even maintain a “conversation” among your various screen names,
34 J o h n Allen Paulos
each salivating over the prospects of the stock. Then just wait for it to move up and sell it quickly when it does.
A fifteen-year-old high school student in New Jersey was arrested for successfully pumping and dumping after school. It’s hard to gauge how widespread the practice is since the perpetrators generally make themselves invisible. I don’t think it’s rare, especially since there are gradations in the practice, ranging from organized crime telephone banks to conventional brokers inveigling gullible investors.
In fact, the latter probably constitute a vastly bigger threat. Being a stock analyst used to be a thoroughly respectable profession, and for most practitioners no doubt it still is. Unfortunately, however, there seem to be more than a few whose fervent desire to obtain the investment banking fees associated with underwriting, mergers, and the other quite lucrative practices induces them to shade their analyses—and “shade” may be a kind verb—so as not to offend the companies they’re both analyzing and courting. In early 2002, there were well-publicized stories of analysts at Merrill Lynch exchanging private emails deriding a stock that they were publicly touting. Six other brokerage houses were accused of similar wrongdoing.
Even more telling were records from Salomon Smith Barney subpoenaed by Congress indicating that executives at companies generating large investment fees often personally received huge dollops of companies’ initial public offerings. Not open to ordinary investors, these hot, well-promoted offerings quickly rose in value and their quick sale generated immediate profits. Bernie Ebbers was reported to have received, between 1996 and 2000, almost a million shares of IPOs worth more than $11 million. The $1.4 billion settlement between several big brokerage houses and the government announced in December 2002 left little doubt that the practice was not confined to Ebbers and Salomon.
A Mathematician Plays the Stock Market 35
In retrospect it now seems that some analysts’ ratings weren’t much more credible than the ubiquitous email invitations from people purporting to be Nigerian government officials in need of a little seed money. The usual claim is that the money will enable them and their gullible respondents to share in enormous, but frozen foreign accounts.
The bear market analogue to pumping and dumping is shorting and distorting. Instead of buying, touting, and selling on the jump in price, shorters and distorters sell, lambast, and buy on the decline in price.
They first short-sell the stock in question. As mentioned, that is the practice of selling shares one doesn’t own in the hope that the price of the shares will decline when it comes time to pay the broker for the borrowed shares. (Short-selling is perfectly legal and also serves a useful purpose in maintaining markets and limiting risk.) After short-selling the stock, the scamsters lambast it in a misleading hyperbolical way (that is, distort its prospects). They spread false rumors of writedowns, unsecured debts, technology problems, employee morale, legal proceedings. When the stock’s price declines in response to this concerted campaign, they buy the shares at the lower price and keep the difference.
Like its bull-market counterpart, shorting and distorting works best on thinly traded stocks. It is most effective in a bear market when people are most susceptible to fear and anxiety. Online practitioners, like pumpers and dumpers, use a variety of screen names, this time to create the illusion that something catastrophic is about to befall the company in question. They also tend to be nastier toward investors who disagree with them than are pumpers and dumpers, who must maintain a sunny, confident air. Again there are gradations in the practice and it sometimes seems indistinguishable from some fairly conventional practices of brokerage houses and hedge funds.
36 J o h n Allen Paulos
Even large stocks like WCOM (with 3 billion outstanding shares) can be affected by such shorters and distorters although they must be better placed than the dermatologically challenged isolates who usually carry on the practice. I don’t doubt that there was much shorting of WCOM during its long descent, although given what’s come to light about the company’s accounting, “short and report” is a more faithful description of what occurred.
Unfortunately, after Enron, WorldCom, Tyco, and the others, even an easily generated whiff of malfeasance can cause investors to sell first and ask questions later. As a result, many worthy companies are unfairly tarred and their investors unnecessarily burned.
Averaging Down or Catching a Falling Knife?
After an increasingly intense, albeit one-sided courtship of the stock (the girl never even sent me a dividend), I married it. As
13
14 J o h n Allen Paulos
its share price fell, I continued to see only opportunities for gains. Surely, I told myself, the stock had reached its bottom and it was now time to average down by buying the considerably cheaper shares. Of course, for every facile invitation I extended myself to “average down,” I ignored an equally facile warning about not attempting to “catch a falling knife.” The stale, but prudent adage about not putting too many of one’s eggs in the same basket never seemed to push itself very forcefully into my consciousness.
I was also swayed by Salomon Smith Barney’s Jack Grub- man (possessor, incidentally, of a master’s degree in mathematics from Columbia) and other analysts, who ritualistically sprinkled their “strong buys” over the object of my affections. In fact, most brokerage houses in early 2000 rated WCOM a “strong buy,” and those that didn’t had it as a “buy.” It required no great perspicacity to notice that at the time, almost no stock ever received a “sell,” much less a “strong sell,” and that even “holds” were sparingly bestowed. Maybe, I thought, only environmental companies that manufactured solar- powered flashlights qualified for these latter ratings. Accustomed to grade inflation and to movie, book, and restaurant review inflation, I wasn’t taken in by the uniformly positive ratings. Still, just as you can be moved by a television commercial whose saccharine dialogue you are simultaneously ridiculing, part of me gave credence to all those “strong buys.”
I kept telling myself that I’d incurred only paper losses and had lost nothing real unless I sold. The stock would come back, and if I didn’t sell, I couldn’t lose. Did I really believe this? Of course not, but I acted as if I did, and “averaging down” continued to seem like an irresistible opportunity. I believed in the company, but greed and fear were already doing their usual two-step in my head and, in the process, stepping all over my critical faculties.
A M athematician Plays the Stock Market 15
Emotional Overreactions and Homo Economicus
Investors can become (to borrow a phrase Alan Greenspan and Robert Shiller made famous) irrationally exuberant, or, changing the arithmetical sign, irrationally despairing. Some of the biggest daily point gains and declines in Nasdaq’s history occurred in a single month in early 2000, and the pattern has continued unabated in 2001 and 2002, the biggest point gain since 1987 occurring on July 24, 2002. (The increase in volatility, although substantial, is a little exaggerated since our perception of gains and losses have been distorted by the rise in the indices. A 2 percent drop in the Dow when the market is at 9,000 is 180 points, whereas not too long ago when it was at 3,000, the same percentage drop was only 60 points.) The volatility has come about as the economy has hovered near a recession, as accounting abuses have come to light, as CEO malfeasance has mounted, as the bubble has fizzled, and as people have continued to trade on their own, influenced no doubt by capricious lists of the fifty most beautiful (er ..., undervalued) stocks.
As with beautiful people and, for that matter, distinguished universities, emotions and psychology are imponderable factors in the market’s jumpy variability. Just as beauty and academic quality don’t change as rapidly as ad hoc lists and magazine rankings do, so, it seems, the fundamentals of companies don’t change as quickly as our mercurial reactions to news about them do.
It may be useful to imagine the market as a fine race car whose exquisitely sensitive steering wheel makes it impossible to drive in a straight line. Tiny bumps in our path cause us to swerve wildly, and we zigzag from fear to greed and back again, from unreasonable gloom to irrational exuberance and back.
16 J o h n Allen Paulos
Our overreactions are abetted by the all-crisis-all-the-time business media, which brings to mind a different analogy: the reigning theory in cosmology. The inflationary universe hypothesis holds—very, very roughly—that shortly after the Big Bang the primordial universe inflated so fast that all of our visible universe derives from a tiny part of it; we can’t see the rest. The metaphor is strained (in fact I just developed carpal tunnel syndrome typing it), but it seems reminiscent of what happens when the business media (as well as the media in general) focus unrelentingly on some titillating but relatively inconsequential bit of news. Coverage of the item expands so fast as to distort the rest of the global village and render it invisible.
Our responses to business news are only one of the ways in which we fail to be completely rational. More generally, we simply don’t always behave in ways that maximize our economic well-being. “Homo economicus” is not an ideal toward which many people strive. My late father, for example, was distinctly uneconimicus. I remember him sitting and chuckling on the steps outside our house one autumn night long ago. I asked what was funny and he told me that he had been watching the news and had heard Bob Buhl, a pitcher for the then Milwaukee Braves, answer a TV reporter’s question about his off-season plans. “Buhl said he was going to help his father up in Saginaw, Michigan, during the winter.” My father laughed again and continued. “And when the reporter asked Buhl what his father did up in Saginaw, Buhl said, ‘Nothing at all. He does nothing at all.’ ”
My father liked this kind of story and his crooked grin lingered on his face. This memory was jogged recently when I was straightening out my office and found a cartoon he had sent me years later. It showed a bum sitting happily on a park bench as a line of serious businessmen traipsed by him. The bum calls out “Who’s winning?” Although my father was a
salesman, he always seemed less intent on making a sale than on schmoozing with his customers, telling jokes, writing poetry (not all of it doggerel), and taking innumerable coffee breaks.
Everyone can tell such stories, and you would be hard- pressed to find a novel, even one with a business setting, where the characters are all actively pursuing their economic self-interest. Less anecdotal evidence of the explanatory limits of the homo economicus ideal is provided by so-called “ultimatum games.” These generally involve two players, one of whom is given a certain amount of money, say $100, by an experimenter, and the other of whom is given a kind of veto. The first player may offer any non-zero fraction of the $100 to the second player, who can either accept or reject it. If he accepts it, he is given whatever amount the first player has offered, and the first player keeps the balance. If he rejects it, the experimenter takes the money back.
Viewing this in rational game-theoretic terms, one would argue that it’s in the interest of the second player to accept whatever is offered since any amount, no matter how small, is better than nothing. One would also suspect that the first player, knowing this, would make only tiny offers to the second player. Both suppositions are false. The offers range up to 50 percent of the money involved, and, if deemed too small and therefore humiliating, they are sometimes rejected. Notions of fairness and equality, as well as anger and revenge, seem to play a role.
Behavioral Finance
People’s reactions to ultimatum games may be counterproductive, but they are at least clear-eyed. A number of psychologists in recent years have pointed out the countless ways in
A Mathematician Plays the Stock Market 17
18 J o h n Allen Paulos
which we’re all subject to other sorts of counterproductive behavior that spring from cognitive blind spots that are analogues, perhaps, of optical illusions. These psychological illusions and foibles often make us act irrationally in a variety of disparate endeavors, not the least of which is investing.
Amos Tversky and Daniel Kahneman are the founders of this relatively new field of study, many of whose early results are reported upon in the classic book Judgment Under Uncertainty, edited by them and Paul Slovic. (Kahneman was awarded the 2002 Nobel Prize in economics, and Tversky almost certainly would have shared it had he not died.) Others who have contributed to the field include Thomas Gilovich, Robin Dawes, J. L. Knetschin, and Baruch Fischhoff. Economist Richard Thaler (mentioned in the first chapter) is one of the leaders in applying these emerging insights to economics and finance, and his book The Winner’s Curse, as well as Gilovich’s How We Know What Isn’t So, are very useful com- pendiums of recent results.
What makes these results particularly intriguing is the way they illuminate the tactics used, whether consciously or not, by people in everyday life. For example, a favorite ploy of activists of all ideological stripes is to set the terms of a debate by throwing out numbers, which need have little relation to reality to be influential. If you are appalled at some condition, you might want to announce that more than 50,000 deaths each year are attributable to it. By the time people catch up and realize that the number is a couple of orders of magnitude smaller, your cause will be established.
Unfounded financial hype and unrealistic “price targets” have the same effect. Often, it seems, an analyst cites a “price target” for a stock in order to influence investors by putting a number into their heads. (Since the targets are so often indistinguishable from wishes, shouldn’t they always be infinite?)
A Mathematician Plays th e Stock Market 19
The reason for the success of this hyperbole is that most of us suffer from a common psychological failing. We credit and easily become attached to any number we hear. This tendency is called the “anchoring effect” and it’s been demonstrated to hold in a wide variety of situations.
If an experimenter asks people to estimate the population of Ukraine, the size of Avogadro’s number, the date of an historical event, the distance to Saturn, or the earnings of XYZ Corporation two years from now, their guesses are likely to be fairly close to whatever figure the experimenter first suggests as a possibility. For example, if he prefaces his request for an estimate of the population of Ukraine with the question—“Is it more or less than 200 million people?”—the subjects’ estimates will vary and generally be a bit less than this figure, but still average, say, 175 million people. If he prefaces his request for an estimate with the question—“Is the population of Ukraine more or less than 5 million people?”—the subjects’ estimates will vary and this time be a bit more than this figure, but still average, say, 10 million people. The subjects usually move in the right direction from whatever number is presented to them, but nevertheless remain anchored to it.
You might think this is a reasonable strategy for people to follow. They might realize they don’t know much about Ukraine, chemistry, history, or astronomy, and they probably believe the experimenter is knowledgeable, so they stick close to the number presented. The astonishing strength of the tendency comes through, however, when the experimenter obtains his preliminary number by some chance means, say by spinning a dial that has numbers around its periphery—300 million, 200 million, 50 million, 5 million, and so on. Say he spins the dial in front of the subjects, points out where it has stopped, and then asks them if the population of Ukraine is more or less than the number at which the dial has stopped.
20 J o h n Allen Paulos
The subjects’ guesses are still anchored to this number even though, one presumes, they don’t think the dial knows anything about Ukraine!
Financial numbers are also vulnerable to this sort of manipulation, including price targets and other uncertain future figures like anticipated earnings. The more distant the future the numbers describe, the more it’s possible to postulate a huge figure that is justified, say, by a rosy scenario about the exponentially growing need for bandwidth or online airline tickets or pet products. People will discount these estimates, but usually not nearly enough. Some of the excesses of the dot-coms are probably attributable to this effect. On the sell side too, people can paint a dire picture of ballooning debt or shrinking markets or competing technology. Once again, the numbers presented, this time horrific, need not have much to do with reality to have an effect.
Earnings and targets are not the only anchors. People often remember and are anchored to the fifty-two-week high (or low) at which the stock had been selling and continue to base their deliberations on this anchor. I unfortunately did this with WCOM. Having first bought the stock when it was in the forties, I implicitly assumed it would eventually right itself and return there. Later, when I bought more of it in the thirties, twenties, and teens, I made the same assumption.
Another, more extreme form of anchoring (although there are other factors involved) is revealed by investors’ focus on whether the earnings that companies announce quarterly meet the estimates analysts have established for them. When companies’ earnings fall short by a penny or two per share, investors sometimes react as if this were tantamount to nearbankruptcy. They seem to be not merely anchored to earnings estimates but fetishistically obsessed with them.
Not surprisingly, studies have shown that companies’ earnings are much more likely to come in a penny or two above
A Mathematician Plays th e Stock Market 21
the analysts’ average estimate than a penny or two below it. If earnings were figured without regard to analysts’ expectations, they’d come in below the average estimate as often as above it. The reason for the asymmetry is probably that companies sometimes “back in” to their earnings. Instead of determining revenues and expenses and subtracting the latter from the former to obtain earnings (or more complicated variants of this), companies begin with the earnings they need and adjust revenues and expenses to achieve them.
Psychological Foibles, A List
The anchoring effect is not the only way in which our faculties are clouded. The “availability error” is the inclination to view any story, whether political, personal, or financial, through the lens of a superficially similar story that is psychologically available. Thus every recent American military involvement is inevitably described somewhere as “another Vietnam.” Political scandals are immediately compared to the Lewinsky saga or Watergate, misunderstandings between spouses reactivate old wounds, normal accounting questions bring the Enron-Andersen-WorldCom fiasco to mind, and any new high-tech firm has to contend with memories of the dot-com bubble. As with anchoring, the availability error can be intentionally exploited.
The anchoring effect and availability error are exacerbated by other tendencies. “Confirmation bias” refers to the way we check a hypothesis by observing instances that confirm it and ignoring those that don’t. We notice more readily and even diligently search for whatever might confirm our beliefs, and we don’t notice as readily and certainly don’t look hard for what disconfirms them. Such selective thinking reinforces the anchoring effect: We naturally begin to look for reasons
22 J o h n Allen Paulos
that the arbitrary number presented to us is accurate. If we succumb completely to the confirmation bias, we step over the sometimes fine line separating flawed rationality and hopeless closed-mindedness.
Confirmation bias is not irrelevant to stock-picking. We tend to gravitate toward those people whose take on a stock is similar to our own and to search more vigorously for positive information on the stock. When I visited WorldCom chatrooms, I more often clicked on postings written by people characterizing themselves as “strong buys” than I did on those written by “strong sells.” I also paid more attention to WorldCom’s relatively small deals with web-hosting companies than to the larger structural problems in the telecommunications industry.
The “status quo bias” (these various biases are generally not independent of each other) also applies to investing. If subjects are told, for example, that they’ve inherited a good deal of money and then asked which of four investment options (an aggressive stock portfolio, a more balanced collection of equities, a municipal bond fund, or U.S. Treasuries) they would prefer to invest it in, the percentages choosing each are fairly evenly distributed.
Surprisingly, however, if the subjects are told that they’ve inherited the money but it is already in the form of municipal bonds, almost half choose to keep it in bonds. It’s the same with the other three investment options: Almost half elect to keep the money where it is. This inertia is part of the reason so many people sat by while not only their inheritances but their other investments dwindled away. The “endowment effect,” another kindred bias, is an inclination to endow one’s holdings with more value than they have simply because one holds them. “It’s my stock and I love it.”
Related studies suggest that passively endured losses induce less regret than losses that follow active involvement. Some-
A Mathematician Plays th e Stock Market 23
one who sticks with an old investment that then declines by 25 percent is less upset than someone who switches into the same investment before it declines by 25 percent. The same fear of regret underlies people’s reluctance to trade lottery tickets with friends. They imagine how they’ll feel if their original ticket wins.
Minimizing possible regret often plays too large a role in investors’ decisionmaking. A variety of studies by Tversky, Kahneman, and others have shown that most people tend to assume less risk to obtain gains than they do to avoid losses. This isn’t implausible: Other research suggests that people feel considerably more pain after incurring a financial loss than they do pleasure after achieving an equivalent gain. In the extreme case, desperate fears about losing a lot of money induce people to take enormous risks with their money.
Consider a rather schematic outline of many of the situations studied. Imagine that a benefactor gives $10,000 to everyone in a group and then offers each of them the following choice. He promises to a) give them an additional $5,000 or else b) give them an additional $10,000 or $0, depending on the outcome of a coin flip. Most people choose to receive the additional $5,000. Contrast this with the choice people in a different group make when confronted with a benefactor who gives them each $20,000 and then offers the following choice to each of them. He will a) take from them $5,000 or else b) will take from them $10,000 or $0, depending on the flip of a coin. In this case, in an attempt to avoid any loss, most people choose to flip the coin. The punchline, as it often is, is that the choices offered to the two groups are the same: a sure $15,000 or a coin flip to determine whether they’ll receive $10,000 or $20,000.
Alas, I too took more risks to avoid losses than I did to obtain gains. In early October 2000, WCOM had fallen below $20, forcing the CEO, Bernie Ebbers, to sell 3 million shares
24 J o h n Allen Paulos
to pay off some of his investment debts. The WorldCom chatrooms went into one of their typical frenzies and the price dropped further. My reaction, painful to recall, was, “At these prices I can finally get out of the hole.” I bought more shares even though I knew better. There was apparently a loose connection between my brain and my fingers, which kept clicking the buy button on my Schwab online account in an effort to avoid the losses that loomed.
Outside of business, loss aversion plays a role as well. It’s something of a truism that the attempt to cover up a scandal often leads to a much worse scandal. Although most people know this, attempts to cover up are still common, presumably because, here too, people are much more willing to take risks to avoid losses than they are to obtain gains.
Another chink in our cognitive apparatus is Richard Thaler’s notion of “mental accounts,” mentioned in the last chapter. “The Legend of the Man in the Green Bathrobe” illustrates this notion compellingly. It is a rather long shaggy dog story, but the gist is that a newlywed on his honeymoon in Las Vegas wakes up in bed and sees a $5 chip left on the dresser. Unable to sleep, he goes down to the casino (in his green bathrobe, of course), bets on a particular number on the roulette wheel, and wins. The 35 to 1 odds result in a payout of $175, which the newlywed promptly bets on the next spin. He wins again and now has more than $6,000. He bets everything on his number a couple more times, continuing until his winnings are in the millions and the casino refuses to accept such a large bet. The man goes to a bigger casino, wins yet again, and now commands hundreds of millions of dollars. He hesitates and then decides to bets it all one more time. This time he loses. In a daze, he stumbles back up to his hotel room where his wife yawns and asks how he did. “Not too bad. I lost $5.”
It’s not only in casinos and the stock market that we categorize money in odd ways and treat it differently depending
A Mathematician Plays th e Stock Market 25
on what mental account we place it in. People who lose a $100 ticket on the way to a concert, for example, are less likely to buy a new one than are people who lose $100 in cash on their way to buy the ticket. Even though the amounts are the same in the two scenarios, people in the former one tend to think $200 is too large an expenditure from their entertainment account and so don’t buy a new ticket, while people in the latter tend to assign $100 to their entertainment account and $100 to their “unfortunate loss” account and buy the ticket.
In my less critical moments (although not only then) I mentally amalgamate the royalties from this book, whose writing was prompted in part by my investing misadventure, with my WCOM losses. Like corporate accounting, personal accounting can be plastic and convoluted, perhaps even more so since, unlike corporations, we are privately held.
These and other cognitive illusions persist for several reasons. One is that they lead to heuristic rules of thumb that can save time and energy. It’s often easier to go on automatic pilot and respond to events in a way that requires little new thinking, not just in scenarios involving eccentric philanthropists and sadistic experimenters. Another reason for the illusions’ persistence is that they have, to an extent, become hardwired over the eons. Noticing a rustle in the bush, our primitive ancestors were better off racing away than they were plugging into Bayes’ theorem on conditional probability to determine if a threat was really likely.
Sometimes these heuristic rules lead us astray, again not just in business and investing but in everyday life. Early in the fall 2002 Washington, D.C., sniper case, for example, the police arrested a man who owned a white van, a number of rifles, and a manual for snipers. It was thought at the time that there was one sniper and that he owned all these items, so for the purpose of this illustration let’s assume that this turned out to
26 J o h n Allen Paulos
be true. Given this and other reasonable assumptions, which is higher—a) the probability that an innocent man would own all these items, or b) the probability that a man who owned all these items would be innocent? You may wish to pause before reading on.
Most people find questions like this difficult, but the second probability would be vastly higher. To see this, let me make up some plausible numbers. There are about 4 million innocent people in the suburban Washington area and, we’re assuming, one guilty one. Let’s further estimate that ten people (including the guilty one) own all three of the items mentioned above. The first probability—that an innocent man owns all these items—would be 9/4,000,000 or 1 in 400,000. The second probability—that a man owning all three of these items is innocent—would be 9/10. Whatever the actual numbers, these probabilities usually differ substantially. Confusing them is dangerous (to defendants).
Self-Fulfilling Beliefs and Data Mining
Taken to extremes, these cognitive illusions may give rise to closed systems of thought that are immune, at least for a while, to revision and refutation. (Austrian writer and satirist Karl Kraus once remarked, “Psychoanalysis is that mental illness for which it regards itself as therapy.”) This is especially true for the market, since investors’ beliefs about stocks or a method of picking them can become a self-fulfilling prophecy. The market sometimes acts like a strange beast with a will, if not a mind, of its own. Studying it is not like studying science and mathematics, whose postulates and laws are (in quite different senses) independent of us. If enough people suddenly wake up believing in a stock, it will, for that reason alone, go up in price and justify their beliefs.
A Mathematician Plays th e Stock Market 27
A contrived but interesting illustration of a self-fulfilling belief involves a tiny investment club with only two investors and ten possible stocks to choose from each week. Let’s assume that each week chance smiles at random on one of the ten stocks the investment club is considering and it rises precipitously, while the week’s other nine stocks oscillate within a fairly narrow band.
George, who believes (correctly in this case) that the movements of stock prices are largely random, selects one of the ten stocks by rolling a die (say an icosehedron—a twenty-sided solid—with two sides for each number). Martha, let’s assume, fervently believes in some wacky theory, Q analysis. Her choices are therefore dictated by a weekly Q analysis newsletter that selects one stock of the ten as most likely to break out. Although George and Martha are equally likely to pick the lucky stock each week, the newsletter-selected stock will result in big investor gains more frequently than will any other stock.
The reason is simple but easy to miss. Two conditions must be met for a stock to result in big gains for an investor: It must be smiled upon by chance that week and it must be chosen by one of the two investors. Since Martha always picks the newsletter-selected stock, the second condition in her case is always met, so whenever chance happens to favor it, it results in big gains for her. This is not the case with the other stocks. Nine-tenths of the time, chance will smile on one of the stocks that is not newsletter-selected, but chances are George will not have picked that particular one, and so it will seldom result in big gains for him. One must be careful in interpreting this, however. George and Martha have equal chances of pulling down big gains (10 percent), and each stock of the ten has an equal chance of being smiled upon by chance (10 percent), but the newsletter-selected stock will achieve big gains much more often than the randomly selected ones.
28 J o h n Allen Paulos
Reiterated more numerically, the claim is that 10 percent of the time the newsletter-selected stock will achieve big gains for Martha, whereas each of the ten stocks has only a 1 percent chance of both achieving big gains and being chosen by George. Note again that two things must occur for the newsletter-selected stock to achieve big gains: Martha must choose it, which happens with probability 1, and it must be the stock that chance selects, which happens with probability 1/10th. Since one multiplies probabilities to determine the likelihood that several independent events occur, the probability of both these events occurring is 1 x 1/10, or 10 percent. Likewise, two things must occur for any particular stock to achieve big gains via George: George must choose it, which occurs with probability 1/10th, and it must be the stock that chance selects, which happens with probability 1/10th. The product of these two probabilities is 1/100th or 1 percent.
Nothing in this thought experiment depends on there being only two investors. If there were one hundred investors, fifty of whom slavishly followed the advice of the newsletter and fifty of whom chose stocks at random, then the newsletter- selected stocks would achieve big gains for their investors eleven times as frequently as any particular stock did for its investors. When the newsletter-selected stock is chosen by chance and happens to achieve big gains, there are fifty-five winners, the fifty believers in the newsletter and five who picked the same stock at random. When any of the other nine stocks happens to achieve big gains, there are, on average, only five winners.
In this way a trading strategy, if looked at in a small population of investors and stocks, can give the strong illusion that it is effective when only chance is at work.
“Data mining,” the scouring of databases of investments, stock prices, and economic data for evidence of the effectiveness of this or that strategy, is another example of how an
inquiry of limited scope can generate deceptive results. The problem is that if you look hard enough, you will always find some seemingly effective rule that resulted in large gains over a certain time span or within a certain sector. (In fact, inspired by the British economist Frank Ramsey, mathematicians over the last half century have proved a variety of theorems on the inevitability of some kind of order in large sets.) The promulgators of such rules are not unlike the believers in bible codes. There, too, people searched for coded messages that seemed to be meaningful, not realizing that it’s nearly impossible for there not to be some such “messages.” (This is trivially so if you search in a book that has a chapter 11, conveniently foretelling many companies’ bankruptcies.)
People commonly pore over price and trade data attempting to discover investment schemes that have worked in the past. In a reductio ad absurdum of such unfocused fishing for associations, David Leinweber in the mid-90s exhaustively searched the economic data on a United Nations CD-ROM and found that the best predictor of the value of the S8cP 500 stock index was—a drum roll here—butter production in Bangladesh. Needless to say, butter production in Bangladesh has probably not remained the best predictor of the S&P 500. Whatever rules and regularities are discovered within a sample must be applied to new data if they’re to be accorded any limited credibility. You can always arbitrarily define a class of stocks that in retrospect does extraordinarily well, but will it continue to do so?
I’m reminded of a well-known paradox devised (for a different purpose) by the philosopher Nelson Goodman. He selected an arbitrary future date, say January 1, 2020, and defined an object to be “grue” if it is green and the time is before January 1, 2020, or if it is blue and the time is after January 1, 2020. Something is “bleen,” on the other hand, if it is blue and the time is before that date or if it is green and the
A Mathematician Plays th e Stock Market 29
30 J o h n Allen Paulos
time is after that date. Now consider the color of emeralds. All emeralds examined up to now (2002) have been green. We therefore feel confident that all emeralds are green. But all emeralds so far examined are also grue. It seems that we should be just as confident that all emeralds are grue (and hence blue beginning in 2020). Are we?
A natural objection is that these color words grue and bleen are very odd, being defined in terms of the year 2020. But were there aliens who speak the grue-bleen language, they could make the same charge against us. “Green,” they might argue, is an arbitrary color word, being defined as grue before 2020 and bleen afterward. “Blue” is just as odd, being bleen before 2020 and grue from then on. Philosophers have not convincingly shown what exactly is wrong with the terms grue and bleen, but they demonstrate that even the abrupt failure of a regularity to hold can be accommodated by the introduction of new weasel words and ad hoc qualifications.
In their headlong efforts to discover associations, data miners are sometimes fooled by “survivorship bias.” In market usage this is the tendency for mutual funds that go out of business to be dropped from the average of all mutual funds. The average return of the surviving funds is higher than it would be if all funds were included. Some badly performing funds become defunct, while others are merged with betterperforming cousins. In either case, this practice skews past returns upward and induces greater investor optimism about future returns. (Survivorship bias also applies to stocks, which come and go over time, only the surviving ones making the statistics on performance. WCOM, for example, was unceremoniously replaced on the S&P 500 after its steep decline in early 2002.)
The situation is rather like that of schools that allow students to drop courses they’re failing. The grade point averages of schools with such a policy are, on average, higher
A Mathematician Plays th e Stock M arket 31
than those of schools that do not allow such withdrawals. But these inflated GPAs are no longer a reliable guide to students’ performance.
Finally, taking the meaning of the term literally, survivorship bias makes us all a bit more optimistic about facing crises. We tend to see only those people who survived similar crises. Those who haven’t are gone and therefore much less visible.
Rumors and Online Chatrooms
Online chatrooms are natural laboratories for the observation of illusions and distortions, although their psychology is more often brutally basic than subtly specious. While spellbound by WorldCom, I would spend many demoralizing, annoying, and engaging hours compulsively scouring the various WorldCom discussions at Yahoo! and RagingBull. Only a brief visit to these sites is needed to see that a more accurate description of them would be rantrooms.
Once someone dons a screen name, he (the masculine pronoun, I suspect, is almost always appropriate) usually dispenses with grammar, spelling, and most conventional standards of polite discourse. Other people become morons, idiots, and worse. A poster’s references to the stock, if he’s shorting it (selling shares he doesn’t have in the hope that he can buy them back when the price goes down), put a burden on one’s ability to decode scatological allusions and acronyms. Any expression of pain at one’s losses is met with unrelenting scorn and sarcasm; ostensibly genuine musings about suicide are no exception. A suicide threat in April 2002, lamenting the loss of house, family, and job because of WCOM, drew this response: “You sad sack loser. Die. You might want to write a note too in case the authorities and your wife don’t read the Yahoo! chatrooms.”
32 J o h n Allen Paulos
People who characterize themselves as sellers are generally (but not always) more vituperative than those claiming to be buyers. Some of the regulars appear genuinely interested in discussing the stock rationally, imparting information, and exchanging speculation. A few seem to know a lot, many are devotees of various outlandish conspiracy theories, including the usual anti-Semitic sewage, and even more are just plain clueless, asking, for example, why they “always put that slash between the P and the E in P/E, and is P price or profit.” There were also many discussions that had nothing directly to do with the stock. One that I remember fondly was about someone who called a computer help desk because his computer didn’t work. It turned out that he had plugged the computer and all his peripheral devices into his surge protector, which he then plugged into itself. The connection with whatever company was being discussed I’ve forgotten.
Taking advice from such an absurdly skewed sample of posters is silly, of course, but the real-time appeal of the sites is akin to overhearing gossip about a person you’re interested in. It’s likely to be false, spun, or overstated, but it still holds a certain fascination. Another analogy is to listening to police radio and getting a feel for the raw life and death on the streets.
Chatroom denizens form little groups that spend a lot of time excoriating, but not otherwise responding to, opposing groups. They endorse each other’s truisms and denounce those of the others. When WorldCom purchased a small company or had a reversal in its Brazilian operations, this was considered big news. It was not nearly as significant, however, as an analyst changing his recommendation from a strong buy to a buy or vice versa. If you filter out the postings drenched in anger and billingsgate, you find most of the biases mentioned above demonstrated on a regular basis. The posters are averse to risk, anchored to some artificial number,
A M athematician Plays th e Stock Market 33
addicted to circular thought, impressed by data mining, or all of the above.
Most boards I visited had a higher percentage of rational posters than did WorldCom’s. I remember visiting the Enron board and reading rumors of the bogus deals and misleading accounting practices that eventually came to light. Unfortunately, since there are always rumors of every conceivable and contradictory sort (sometimes posted by the same individual), one cannot conclude anything from their existence except that they’re likely to contribute to feelings of hope, fear, anger, and anxiety.
Pump and Dump, Short and Distort
The rumors are often associated with market scams that exploit people’s normal psychological reactions. Many of these reactions are chronicled in Edwin Lefevre’s 1923 classic novel, Reminiscences of a Stock Operator, but the standard “pump and dump” is an illegal practice that has gained new life on the Internet. Small groups of individuals buy a stock and tout it in a misleading hyperbolic way (that is, pump it). Then when its price rises in response to this concerted campaign, they sell it at a profit (dump it). The practice works best in bull markets when people are most susceptible to greed. It is also most effective when used on thinly traded stocks where a few buyers can have a pronounced effect.
Even a single individual with a fast Internet connection and a lot of different screen names can mount a pump and dump operation. Just buy a small stock from an online broker, then visit the chatroom where it’s discussed. Post some artful innuendoes or make some outright phony claims and then back yourself up with one of your pseudonyms. You can even maintain a “conversation” among your various screen names,
34 J o h n Allen Paulos
each salivating over the prospects of the stock. Then just wait for it to move up and sell it quickly when it does.
A fifteen-year-old high school student in New Jersey was arrested for successfully pumping and dumping after school. It’s hard to gauge how widespread the practice is since the perpetrators generally make themselves invisible. I don’t think it’s rare, especially since there are gradations in the practice, ranging from organized crime telephone banks to conventional brokers inveigling gullible investors.
In fact, the latter probably constitute a vastly bigger threat. Being a stock analyst used to be a thoroughly respectable profession, and for most practitioners no doubt it still is. Unfortunately, however, there seem to be more than a few whose fervent desire to obtain the investment banking fees associated with underwriting, mergers, and the other quite lucrative practices induces them to shade their analyses—and “shade” may be a kind verb—so as not to offend the companies they’re both analyzing and courting. In early 2002, there were well-publicized stories of analysts at Merrill Lynch exchanging private emails deriding a stock that they were publicly touting. Six other brokerage houses were accused of similar wrongdoing.
Even more telling were records from Salomon Smith Barney subpoenaed by Congress indicating that executives at companies generating large investment fees often personally received huge dollops of companies’ initial public offerings. Not open to ordinary investors, these hot, well-promoted offerings quickly rose in value and their quick sale generated immediate profits. Bernie Ebbers was reported to have received, between 1996 and 2000, almost a million shares of IPOs worth more than $11 million. The $1.4 billion settlement between several big brokerage houses and the government announced in December 2002 left little doubt that the practice was not confined to Ebbers and Salomon.
A Mathematician Plays the Stock Market 35
In retrospect it now seems that some analysts’ ratings weren’t much more credible than the ubiquitous email invitations from people purporting to be Nigerian government officials in need of a little seed money. The usual claim is that the money will enable them and their gullible respondents to share in enormous, but frozen foreign accounts.
The bear market analogue to pumping and dumping is shorting and distorting. Instead of buying, touting, and selling on the jump in price, shorters and distorters sell, lambast, and buy on the decline in price.
They first short-sell the stock in question. As mentioned, that is the practice of selling shares one doesn’t own in the hope that the price of the shares will decline when it comes time to pay the broker for the borrowed shares. (Short-selling is perfectly legal and also serves a useful purpose in maintaining markets and limiting risk.) After short-selling the stock, the scamsters lambast it in a misleading hyperbolical way (that is, distort its prospects). They spread false rumors of writedowns, unsecured debts, technology problems, employee morale, legal proceedings. When the stock’s price declines in response to this concerted campaign, they buy the shares at the lower price and keep the difference.
Like its bull-market counterpart, shorting and distorting works best on thinly traded stocks. It is most effective in a bear market when people are most susceptible to fear and anxiety. Online practitioners, like pumpers and dumpers, use a variety of screen names, this time to create the illusion that something catastrophic is about to befall the company in question. They also tend to be nastier toward investors who disagree with them than are pumpers and dumpers, who must maintain a sunny, confident air. Again there are gradations in the practice and it sometimes seems indistinguishable from some fairly conventional practices of brokerage houses and hedge funds.
36 J o h n Allen Paulos
Even large stocks like WCOM (with 3 billion outstanding shares) can be affected by such shorters and distorters although they must be better placed than the dermatologically challenged isolates who usually carry on the practice. I don’t doubt that there was much shorting of WCOM during its long descent, although given what’s come to light about the company’s accounting, “short and report” is a more faithful description of what occurred.
Unfortunately, after Enron, WorldCom, Tyco, and the others, even an easily generated whiff of malfeasance can cause investors to sell first and ask questions later. As a result, many worthy companies are unfairly tarred and their investors unnecessarily burned.
Comentarii
Trimiteți un comentariu