While many thinkers over the years had written about market efficiency, Fama’s was the most coherent and concise statement of the idea that the market is unbeatable. The fundamental idea behind EMH is that all relevant new information about a stock is instantly priced into the stock, making it “efficient.” Fama envisioned a large, well-developed market with many players constantly on the hunt for the latest news about companies. The process of injecting new information—a lousy earnings report, the departure of a CEO, a big new contract—is like tossing a juicy piece of fresh meat into a tank of piranhas. Before you know it, the meat has been devoured.
Since all current information is built into the stock’s price and future information is essentially unknowable, it is impossible to predict whether a stock will rise or fall. The future, therefore, is random, a Brownian motion coin flip, a drunkard’s walk through the Parisian night.
The groundwork for the efficient-market hypothesis had begun in the 1950s with the work of Markowitz and Sharpe, who eventually won the Nobel Prize for economics (together with Merton Miller) in 1990 for their work.
Another key player was Louis Bachelier, the obscure French mathematician who argued that bond prices move according to a random walk.
In 1954, MIT economist Paul Samuelson—another future Nobel laureate—received a postcard from Leonard “Jimmie” Savage, a statistician at Chicago. Savage had been searching through stacks at a library and stumbled across the work of Bachelier, which had largely been forgotten in the half century since it had been written. Savage wanted to know if Samuelson had ever heard of the obscure Frenchman. He said he had, though he’d never read his thesis. Samuelson promptly hunted it down and became enthralled with its arguments.
Since the future course of the market is essentially a 50–50 random coin flip, Bachelier had written, “The mathematical expectation of the speculator is zero.” Samuelson had already started thinking about financial markets. His interest had been piqued by a controversial speech given in 1952 by Maurice Kendall, a statistician at the London School of Economics. Kendall had analyzed a variety of market data, including stock market indexes, wheat prices, and cotton prices, looking for some kind of pattern that would show whether price movements were predictable. Kendall found no such patterns and concluded that the data series looked “like a wandering one, almost as if once a week the Demon of Chance drew a random number from a symmetrical population.” Kendall said this appeared to be “a kind of economic Brownian motion.”
Samuelson realized this was a bombshell. He made the leap embedded in Bachelier’s original paper: investors are wasting their time. Mathematically, there is no way to beat the market. The Thorps of the world should put away their computers and formulas and take up a more productive profession, such as dentistry or plumbing. “It is not easy to get rich in Las Vegas, at Churchill Downs, or at the local Merrill Lynch office,” he wrote.
At the time, Samuelson was becoming an éminence grise of the economic community. If he thought the market followed a random walk, that meant everyone had to get on board or have a damn good reason not to. Most agreed, including one of Samuelson’s star students, Robert Merton, one of the co-creators of the Black-Scholes option-pricing formula. Another acolyte was Burton Malkiel, who went on to write A Random Walk Down Wall Street.
It was Fama, however, who connected all of the dots and put the efficient-market hypothesis on the map as a central feature of modern portfolio theory.
The idea that the market is an efficient, randomly churning price-processing machine has many odd consequences. Fama postulates a vast, swarming world of investors constantly searching for inefficiencies—those hungry piranhas circling in wait of fresh meat. Without the hungry piranhas gobbling up juicy fleeting inefficiencies, the market would never become efficient. Would the piranhas exist without the fresh meat? No fresh meat, no piranhas. No piranhas, no market efficiency. It’s a paradox that continues to baffle EMH acolytes.
Another offshoot of market efficiency is that, if true, it effectively makes it impossible to argue that a market is mispriced—ever. When the Nasdaq was hovering above 5,000 in early 2000, it was impossible to argue at the time that it was in a bubble, according to EMH. The housing market in 2005, when prices for many homes in the United States had doubled or tripled in a matter of a few years? No bubble.
Despite such mind-bending conundrums, the EMH became the dominant paradigm in academia as Fama spread the gospel. It was a frontal assault on the money management industry, which was built on the idea that certain people with the right methods and tools can beat the market.
The quants viewed EMH as a key weapon in their arsenal: The probabilities of various movements of an efficient market could be understood through the math spawned by Brownian motion. The most likely moves were those found near the middle of the bell curve, which could be used to make forecasts about the probable future volatility of the market over the course of a month, a year, or a decade. In the financial planning community, so-called Monte Carlo simulations, which can forecast everyday investors’ portfolio growth over time, used the idea that the market moves according to a random walk. Thus, an annual gain or loss of 5 percent a year is far more likely, since it falls near the center of the bell curve. A gain or loss of 50 percent, such as the stock market crash seen in the credit crisis of 2008 (or the 23 percent single-day plunge seen on Black Monday, for that matter) was so unlikely as to be a virtual impossibility—in the models, at least. Today, nearly all large financial services firms, such as Fidelity Investments and T. Rowe Price, offer Monte Carlo simulations to investors. Thus, the insights of Bachelier more than a century earlier and prodded on by Fama had reached into the very nuts and bolts of how Americans prepare for retirement. It had also blinded them to the chance that the market could make extreme moves. Such ugly phenomena simply didn’t fit within the elegant models spawned by the quants.
EMH was in many ways a double-edged sword. On one hand, it argued that the market was impossible to beat. Most quants, however—especially those who migrated from academia to Wall Street—believed the market is only partly efficient. Fischer Black, co-creator of the Black-Scholes option-pricing formula, once said the market is more efficient on the banks of the Charles River than the banks of the Hudson—conveniently, after he’d joined forces with Goldman Sachs.
By this view, the market was like a coin with a small flaw that makes it slightly more likely to come up heads than tails (or tails than heads). Out of a hundred flips, it was likely to come up heads fifty-two times, rather than fifty. The key to success was discovering those hidden flaws, as many as possible. The law of large numbers that Thorp had used to beat the dealer and then earn a fortune on Wall Street dictated that such flaws, exploited in hundreds if not thousands of securities, could yield vast riches.
Implicitly, EMH also showed that there is a mechanism in the market making prices efficient: Fama’s piranhas. The goal was to become a piranha, gobbling up the fleeting inefficiencies, the hidden discrepancies, as quickly as possible. The quants with the best models and fastest computers win the game.
Crucially, EMH gave the quants a touchstone for what the market should look like if it were perfectly efficient, constantly gravitating toward equilibrium. In other words, it gave them a reflection of the Truth, the holy grail of quantitative finance, explaining how the market worked and how to measure it. Every time prices in the market deviated from the Truth, computerized quant piranhas would detect the error, swoop in, and restore order—collecting a healthy profit along the way. Their high-powered computers would comb through global markets like Truth-seeking radar, searching for opportunities. The quants’ models could discover when prices deviated from equilibrium. Of course, they weren’t always right. But if they were right often enough, a fortune could be made.
This was one of the major lessons Cliff Asness learned studying at the University of Chicago. But there was more.
Fama, a bulldog with research, hadn’t rested on his
efficient-market laurels over the years. He continued to churn out libraries of papers, leveraging the power of computers and a stream of bright young students eager to learn from the guru of efficient markets. In 1992, soon after Asness arrived on the scene, Fama and French published their most important breakthrough yet, a paper that stands as arguably the most important academic finance research of the last two decades. And the ambition behind it was immense: to overturn the bedrock theory of finance itself, the capital asset pricing model, otherwise known as CAPM.
Before Fama and French, CAPM was the closest approximation to the Truth in quantitative finance. According to the grandfather of CAPM, William Sharpe, the most important element in determining a stock’s potential future return is its beta, a measure of how volatile the stock is compared with the rest of the market. And according to CAPM, the riskier the stock, the higher the potential reward. The upshot: long-term investments in risky stocks tended to pay off more than investments in the ho-hum blue chips.
Fama and French cranked up their Chicago supercomputers and ran a series of tests on an extensive database of stock market returns to determine how much impact the all-important beta actually had on stock returns. Their conclusion: none.
Such a finding was nothing short of lobbing a blazing Molotov cocktail into the most sacred tent of modern portfolio theory. Decades of research were flat-out wrong, the two professors alleged. Perhaps even more surprising were Fama and French’s findings about the market forces that did, in fact, drive stock returns. They found two factors that determined how well a stock performed during their sample period for 1963 to 1990: value and size.
There are a number of ways to gauge a company’s size. It’s generally measured by how much the Street values a company through its share price, a metric known as market capitalization (the price of a company’s shares times the number of shares). IBM is big: it has a market cap of about $150 billion. Krispy Kreme Doughnuts is small, with a market cap of about $150 million. Other factors, such as how many employees a company has and how profitable it is, also matter.
Value is generally determined by comparing a company’s share price to its book value, a measure of a firm’s net worth (assets, such as the buildings and/or machines it owns, minus liabilities, or debts). Price-to-book is the favored metric of old-school investors such as Warren Buffett. The quants, however, use it in ways the Buffetts of the world never dreamed of (and would never have wanted to), plugging decades of data from the CRSP database into computers, pumping it through complex algorithms, and combing through the results like gold miners sifting for gleaming nuggets—flawed coins with hidden discrepancies.
Fama and French unearthed one of the biggest, shiniest nuggets of all. The family tree of “value” has two primary offspring: growth stocks and value stocks. Growth stocks are relatively pricey, indicating that investors love the company and have driven the shares higher. Value stocks have a low price-to-book value, indicating that they are relatively unloved on Wall Street. Value stocks, in other words, appear cheap.
Fama and French’s prime discovery was that value stocks performed better than growth stocks over almost any time horizon going back to 1963. If you put money in value stocks, you made slightly more than you would have if you invested in growth stocks.
Intuitively, the idea makes a certain amount of sense. Imagine a neighborhood that enjoys two kinds of pizza—pepperoni and mushroom. For a time both pizzas are equally popular. But suddenly mushroom pizzas fall out of favor. More and more people are ordering pepperoni. The pizza man, noticing the change, boosts the price for his pepperoni pizzas and, hoping to encourage more people to buy his unloved mushroom pies, lowers the price. The price disparity eventually grows so wide that more people gravitate toward mushroom, leaving pepperoni behind. Eventually, mushroom pizzas start to gain in price, and pepperonis decline—just as Fama and French predicted.
Of course, it’s not always so simple. Sometimes the quality of the mushrooms are on the decline and the neighborhood has a good reason for disliking them, or the flavor of the pepperoni has suddenly improved. But the analysis showed that, according to the law of large numbers, over time value stocks (unloved mushrooms) tend to perform better than growth stocks (pricey pepperoni).
Fama and French also found that small stocks tended to fare better than large stocks. The notion is similar to the value and growth disparity, because a small stock is intuitively unloved—that’s why it’s small. Large stocks, meanwhile, often suffer from too much love, like a celebrity with too many hit movies on the market, and are due for a fall.
In other words, according to Fama and French, the forces pushing stocks up and down over time weren’t volatility or beta—they were value and size. For students such as Asness, the message was clear: money could be made by focusing purely on these factors. Buy cheap mushroom pizzas (small ones) and short jumbo pepperonis.
For the cloistered quant community, it was like Martin Luther nailing his Ninety-five Theses to the door of the Castle Church in Wittenberg, overturning centuries of tradition and belief. The Truth as they knew it—the holy CAPM—wasn’t the Truth at all. If Fama and French were right, there was a New Truth. Value and size were all that mattered.
Defenders of the Old Truth rallied to the cause. Fischer Black, by then a partner at Goldman Sachs, leveled the most damning blast, writing, “Fama and French … misinterpret their own data,” a true smackdown in quantdom. Sharpe argued that the period Fama and French observed favored the value factor, since value stocks performed extremely well in the 1980s after the market pummeling in the previous decade of oil crises and stagflation.
Nevertheless, Fama and French’s New Truth began to take hold.
Aside from the theoretical bells and whistles of the paper, it had a crucial impact on the financial community: by bringing down the CAPM, Fama and French opened the floodgates for a massive wave of fresh research as finance geeks started to sift through the new sands for more gleaming golden nuggets. Cliff Asness was among the first in line.
In time, the findings had a more sinister effect. More and more quants crowded into the strategies unearthed by Fama and French and others, leading to an event the two professors could never have anticipated: one of the fastest, most brutal market meltdowns ever seen.
But that was years later.
One day in 1990, Asness stepped into Fama’s office to talk about an idea for a Ph.D. dissertation. He was nervous, wracked by guilt. Fama had given him the greatest honor any student at the University of Chicago’s economics department could hope for: he’d picked Asness to be his teaching assistant. (Ken French, Fama’s collaborator, also sang Asness’s praises. Fama and French were known to say that Asness was the smartest student they had ever seen at Chicago.) Asness felt he was double-crossing a man he’d come to worship as a hero.
The phenomenon Asness was considering as a dissertation topic flew in the face of Fama’s beloved efficient-market hypothesis. Combing through decades of data, Asness believed he had discovered a curious anomaly in a trend driving stock prices. Stocks that were falling seemed to keep falling more than they should, based on underlying fundamentals such as earnings, and stocks that were rising often seemed to keep rising more than they should. In the parlance of physics, the phenomenon was called “momentum.”
According to the efficient-market hypothesis, momentum shouldn’t exist, since it implied that there was a way to tell which stocks would keep rising and which would keep falling.
Asness knew that momentum was a direct challenge to Fama, and he expected a fight. He cleared his throat.
“My paper is going to be pro-momentum,” he said with a wince.
Fama rubbed his cheek and nodded. Several seconds passed. He looked up at Asness, his massive forehead wrinkled in concentration.
“If it’s in the data,” he said, “write the paper.”
Asness was stunned and elated. Fama’s openness to whatever the data showed was a remarkable display of intellectual hone
sty, he felt.
He started crunching the numbers from Chicago’s extensive library of market data and noticed a variety of patterns showing long-and short-term momentum in stocks. At first Asness didn’t realize he’d made a profound discovery about hidden market patterns that he could exploit to make money. He was simply thrilled that he could write his dissertation and graduate. The money would come soon enough.
In 1992, as Asness buckled down on his dissertation on momentum, he received an offer to work in the fixed-income group at Goldman Sachs. A small but growing division at Goldman, called Goldman Sachs Asset Management, was reaching out to bright young academics to build what would become one of the most formidable brain trusts on Wall Street.
Asness’s first real job at Goldman was building fixed-income models and trading mortgage-backed securities. Meanwhile, he spent nights and weekends toiling away at his dissertation and thinking hard about a choice he’d have to make: whether to stay in academia or pursue riches on Wall Street.
His decision was essentially made for him. In January 1992, he received a call from Pimco, the West Coast bond manager run by Bill Gross. A billionaire former blackjack card counter (in college he’d devoured Beat the Dealer and Beat the Market), Gross religiously applied his gambling acumen to his investment decisions on a daily basis. Pimco had gotten hold of Asness’s first published research, “OAS Models, Expected Returns, and a Steep Yield Curve,” and was interested in recruiting him. Over the course of the year, Asness had several interviews with Pimco. In 1993, the company offered him a job building quantitative models and tools. It was an ideal position, Asness thought, combining the research side of academia with the applied rigor of Wall Street.
Goldman, upon learning about the offer, offered him a similar job at GSAM. Asness took it, reasoning that Goldman was closer to home in Roslyn Heights.
“So you’re taking the worse job because you’re a mama’s boy, huh?” his Pimco recruiter quipped.
The Quants: How a New Breed of Math Whizzes Conquered Wall Street and Nearly Destroyed It Page 10
