Money Minded: How to Psychoanalyze the Stock Market
What makes investors do the wrong thing, all together, pretty much all the time?
There’s just no nice way to say it: You’re stupid with your money. You may fancy yourself a shrewd investor, but if you have normal human instincts—if you stand up and cheer at sporting events, if you follow the crowd toward the exit at the theater—then you have the instincts that make investors alternate between delirious greed and inconsolable fear. Like most of your peers, you are wired to buy high and sell low, and that’s why Richard Peterson is about to become one very rich psychiatrist.
When the markets reopened after September 11, 2001, insurance stocks plummeted as investors sold them in a panic. Even Berkshire Hathaway, Warren Buffett’s icon of caution and profitability, suffered a 9 percent drop over 10 days. (It owns a major insurance company.) A shrewd, heartlessly rational investor watching the sell-off would have calculated the long-term value of Berkshire Hathaway and realized that he was getting a great price on a great stock. Buying on September 19 at $61,400 per share, he would today be holding a stock worth more than $149,000. But at the time, the market was in the grip of emotion, not reason. Most investors decided to run for it.
Since the 1980s, psychologists, neurologists and psychiatrists have tried to apply their understanding of this kind of human behavior to the financial markets. These “neuroinvestors” are trained to diagnose phobias, compulsions—and, they hope, financial irrationality. They want to understand why, whether the market is headed into the ground or up to the heavens, everyone does the wrong thing, all together, pretty much all the time. And they aim to help the clever few profit from the synchronized panic of the investing public.
Peterson, a rapid-speaking 35-year-old Texan, may be the first person to have gone to medical school with the express purpose of learning to diagnose financial markets. “I didn’t put it in my application,” he says, “but I was looking for qualitative and emotional explanations for the way the market moves.” Now, having helped develop software that tracks the language with which reporters, analysts and bloggers talk about stocks in order to identify missed opportunities, Peterson, along with a professor of accounting and a Web entrepreneur he can’t name, will in March launch his own hedge fund, MarketPsy Capital, based in California. The fund will make money by using psychiatry and technology to understand and exploit human idiosyncrasy.
Tell Me About Your Past
At first glance, neuroinvesting looks like a bunch of Ph.D.s selling empty psychobabble to stockbrokers. But it’s the result of three centuries of work. From classical scholars like Adam Smith, who in 1759 described the ways in which humans influence one another’s behavior, to neoclassical 20th-century economists such as John Maynard Keynes, who worked to link individual choice to larger economic trends, philosophers and economists have long sought to connect the vagaries of free will with the patterns of moneymaking.
Only after the 1960s did the study of psychology become varied enough to have practical applications outside of mental health, and then, in 1979, a pair of psychologists, Amos Tversky and Daniel Kahneman, published the founding paper of neuroinvesting. Entitled “Prospect Theory: An Analysis of Decision under Risk,” it articulated the process by which people wring their hands over risky decisions—that is, financial ones. (Later, Kahneman would win the Nobel Prize for his work.)
Tversky and Kahneman posited that people tend to use irrational standards for evaluating uncertainty, and that those standards don’t fall in line with true probabilities. A classic example of such irrationality is weighing a stock’s current price against the price you paid (“I can’t sell at $42. I bought at $50!”). It’s illogical—one should calculate whether a stock is worth its current price, rather than making decisions based on hope and regret—yet every trader has done it.
Their theories remain central to most work in the field. When people have to make decisions without enough information, the way they think changes fundamentally. With sufficient information, we’re clear-headed. But investment decisions are made under pronounced uncertainty. As the amount of available information declines—the very moment investors should be analytical—people instead go with the vibes they pick up from everyone else. In short, when information is sparse, we stop using our highly tuned, very fancy human brain and go back to our good old reliable animal brain. And animals are terrible investors.
By the mid-1990s, investment houses had begun to put these kinds of ideas to work on the market. Firms like Fuller & Thaler Asset Management, founded by economics and behavioral-science professor Richard H. Thaler, one of the most visible academics in the field, were hailed for applying psychology to investor behavior, creating investment models based on psychological assessments of biases and other intangible factors. One might assume that given their particular expertise, the neuroinvestors were ready to take advantage when market participants, almost all at once, stopped reasoning and started blindly doing what everybody else was doing. They should have been prepared, in other words, for the Internet bubble. But during the dot-com bust, prominent behavioral-finance funds, including one that Thaler advised, generally either stayed out of the fray or suffered the common fate.
Of course, it’s not that Thaler and his peers failed to spot dot-com irrationality. They knew it was all nuts. But it’s one thing to know that insanity has taken hold, and another to predict when the stampede will subside so you can profit off the return to normalcy. The neuroinvestors, like everyone else, weren’t able to predict when their clients should walk away from the table.
The problem was that their models were too vague. “There’s been more progress on the individual cognitive psychology over the last 20 years than on mass psychology,” Thaler told the New York Times in 2001, at the end of a down year. “If we had a model that would have been able to tell us when the Internet bubble would break, believe me, we would have used it . . . But nobody has the tools, at least as far as I know, that allow you to forecast these changes in these moods.” Peterson thinks he may have exactly those tools.
How Does It Make You Feel?
His father was a professor of finance, and Peterson grew up talking about the stock market at dinner. As an undergraduate, he studied electrical engineering at the University of Texas, and in his senior year he built a piece of statistical software for predicting stock performance. “I found that mechanical systems were most correct when I was most reluctant to believe them,” he remembers. “When it said ‘Sell,’ my reaction was always ‘What? This thing is going to the moon!'”
Looking for a deeper understanding of investor behavior, he applied to medical school in 1995, and in his off-time at the University of Texas he busied himself with further investment experiments. “I started trading on my intuition, and my performance dropped dramatically,” he says. “On paper I was 70 percent accurate, but trading my own money”—which is more nerve-wracking—”I was 30 percent accurate. I figured I needed to understand my emotional state.” He began to watch conversations about investing on early Internet message boards. “I looked at the language that was making people buy stocks.”
Peterson embarked on his psychiatric training at a time when behavioral finance had fallen out of favor on Wall Street, and he began reevaluating the model. As he sees it, the rise of high-speed computing and new forms of analysis make it possible to pursue a more sophisticated form of behavioral finance. “Younger academics are being taught experimental techniques and statistical tools that create good ways to run real experiments,” he says.
A wave of research like Peterson’s—made possible in large part by the widespread digitization of data and the ready availability of powerful computers—is on display in new publications like the quarterly journal of the Institute of Behavioral Finance, founded in 1998. Its contributors are drawn from all over the worlds of finance and academia, and they’re all trying to pin down a very nebulous subject.
Much of the research relates to risk aversion and the ways it gets investors into trouble. A 2001 paper by Robert Prechter [see preceding page] uses neurological theories about the difference between the neocortex (which is thought to control reason) and the basal ganglia and limbic system (instinct and the emotions) to conclude that our instinct to pursue acceptance and alliance—”herding”—while helpful for survival in the wild, is counterproductive in investing. On the market, so few people know what they’re talking about, Prechter argues, that following the loudest opinion just sets off a mindless stampede. Just ask Warren Buffett, who was trampled after September 11th even though events didn’t damage his company’s fundamental value.
The tendency to throw good money after bad is another repeated theme. In a 2004 paper, “Holding on to the Losers: Finnish Evidence,” Mirjam Lehenkari and Jukka Perttunen, researchers at the University of Oulu in Finland, looked at the transactions of all individual investors on the national stock market. Not only do investors hang on to stocks long after they should cut their losses, but the research showed that the tendency continues even in a bear market, where there’s no relief in sight. As the ship turns over, the crew grips the railing more tightly, confident their luck will change, rather than bailing out.
Same Time Next Week
Although these findings help predict the behavior of individuals, they don’t answer a money manager’s questions: Where the heck is the market going, when do I get in, and when do I get out? That’s what Peterson aims to address.
“In traditional behavioral finance, you find your advantages in overreaction and underreaction,” Peterson explains. “You’re looking for deviation from an equilibrium point based on the price at that time. But those guys are just looking at history—markets change dynamically.” Instead, Peterson realized, he needed a way of tracking the mood on Wall Street as it moved.
The software Peterson developed with his unnamed partner (“he’s an established Web 2.0 guy,” he hints) offers a way of anticipating shifts in the market. The software churns through the daily maelstrom of blog posts and earnings reports through which the public volunteers its expectations. It scans the language they use—millions of words each day—and weights each relevant word based on its historical relationship to swings in the market. The software then makes predictions about likely reversals of fortune.
Several scholars—notably Paul Tetlock, an assistant professor of finance at the University of Texas, and Feng Li, an assistant professor of accounting at the University of Michigan—have demonstrated that tracking certain positive or negative terms (“risky,” for example) in corporate reports and newspaper columns, and then tailoring stock selections to the frequency of these terms’ appearance, results in a portfolio that significantly outperforms the market. Peterson hopes to reproduce results like these in the real world, searching for a wider range of signals of market sentiment in a broader variety of sources.
“We’ve been doing simulations since 2006,” Peterson says. He began by finding roughly two dozen stocks about which there was enough available discussion, and from sources—analysts, reporters, bloggers—that had track records to monitor. The number of potential stocks was also, Peterson knew, limited by liquidity: There had to be enough active trading in them to offer a nimble way of getting in and getting out.
Then, from the list of potential stocks, he chose two that the software had identified as inspiring high expectations in investors, and two that inspired low expectations. Using an algorithm that predicts changes in value, he placed simulated bets against and for them, respectively—known as going short and going long. In effect, he was swimming against what his software revealed to be the tide of public opinion.
Peterson won’t reveal which stocks he picked, but he offers a few others that the simulation analyzed. “Right when it started running, Yahoo was the most positive of all stocks” in the way it was written about in the press, “and I remember there was a very rosy article about them, one of many,” he says. And then “they reported earnings, investors were extremely disappointed, and the stock dropped 20 percent. The model predicted it—basically, that Yahoo couldn’t meet the crowd’s expectations. Cisco, meanwhile, was something people were very pessimistic about—it was all over the media—and then its earnings met expectations and the stock jumped 15 percent. We identify opportunities where people are thinking the wrong way.”
In the end, Peterson’s long-short strategy simulation came up with a 34 percent return between July 2006 and November 2007. (In that period, the S&P 500 earned a 13 percent return.) But four stocks aren’t enough to make a successful portfolio: If one of them crashes for impossible-to-anticipate reasons, money will be lost no matter what the other three stocks are doing. And a year isn’t long enough to assure investors that his approach works. Eventually, as the program becomes more sophisticated and has a larger historical database to refer to, the list of stocks and potential plays will grow. But the proof, as Peterson repeatedly emphasizes, will be in the results. “The initial investors are happy with the current strategy,” he writes in an e-mail. “The later investors will be convinced by performance.”
When it comes to the real magic that makes MarketPsy Capital different—the way its software times the trade—Peterson won’t get into details. It’s fairly simple to detect irrationality in the market as a whole, and even to detect it in a particular stock, as Thaler pointed out in 2001, but before now it hasn’t been possible to reliably predict when that irrationality will stop. No one ever really knows when it’s time to cash in their chips.
Peterson only hints at how his software solves the problem: “There are systematic ways that people mis-expect, and they can be detected in language.” He admits to certain technical barriers—it’s been difficult to find sources for the quantity of information his software requires, and it’s still building a reliable historical database for each term it encounters—but he insists the bugs are coming out and that the software will soon be capable of even more complicated predictions. “More advanced artificial-intelligence strategies look good in statistical testing, but they aren’t ready to be implemented into trading until this summer.”
The fund has enough seed money from investors to put his approach to the test this spring. If it produces a good track record—which is harder in the real world, where real people are coping with anxieties about losing real money, than in simulations—more investors will surely come. But will they stay when the next dot-com-style hysteria whirls through the market?
And if neuroinvesting takes hold on a large scale, and investors resist the ingrained urge to follow each other around like sheep, going against the crowd may no longer be as profitable as it seems to be today. But don’t bet on it. Herding, as is true of most of our irrational habits, has been with us since we walked upright. It’s unlikely we’ll wise up anytime soon.
Robert Armstrong is a financial analyst in New York City. Jacob Ward is deputy editor of Popular Science.
Hee-yah, Investors! Get Along There!
Don’t Go With Your Gut
Robert Prechter, president of Elliott Wave International, a financial-market forecasting firm, wrote in 2001 that the herding impulse that evolved in humans is a crippling investment handicap. Yet, he points out, it’s universal. Individual investors [see chart above] “commit more money to the market as it rises and less as it falls, behavior opposite from that which would generate profits.” And professionals can’t help, because (as shown by the cash held at their institutions as related to the level of the S&P 500 Composite index) they herd in the market’s direction along with the public.
How the Software Reads Warning Signs
Scanning the Language
The software developed by Richard Peterson and his partners scans media including blogs, earnings reports and analyst research every minute to arrive at hourly forecasts about where certain stocks are headed. Using a library of keywords—”exuberance,” “bankruptcy,” “crisis of confidence”—and calculations based on the historical relationship between those words and sources and movement on the market, it develops a “risk rating” for the stock in question.
The rating ranges from –5 (stocks likely to go up) to +5 (likely to go down), with the timing of the predicted change depending on how long they’ve been in that risk category and what words are being used to describe them. Seventy percent of the stocks Peterson covers fall between –2 and +2. It’s the stocks that fall outside this middle range that offer the greatest opportunity to make money as they move.