Buffett

by Roger Lowenstein

  But the theorists replaced the chartist voodoo with a voodoo of their own. They defused the idea that prices foretold the future, but ascribed to those same prices an unerring appraisal of the present. Prices, that is, were never wrong. They incorporated, with as much perfection as humans could manage, all there was to know of a company’s long-term prospects. Studying those prospects was therefore pointless. Thus, the theorists’ attack spread from the chartists to “fundamental analysts,” such as Buffett, who combed through corporate reports looking for undervalued stocks. Quoting Fama:

  If the random walk theory is valid and if security exchanges are “efficient” markets, then stock prices at any point in time will represent good estimates of intrinsic or fundamental values. Thus, additional fundamental analysis is of value only when the analyst has new information … or has new insights concerning the effects of generally available information.…10

  Taken at face, the qualifier offered a gaping loophole. Obviously, the analyst with neither new information nor new insights did not have an edge. But the rogue phrase “new information” was not to be taken at face. The implication was that investors with superior records must have had inside dope, or at least dope that was not widely known. Thus, the theory was intact: not even a perfect market could be expected to digest information before it was public. Indeed, the Economist, which reported each twist and turn of the theorists as received wisdom, asserted that stockpicking smarts were “either rare or nonexistent,” although, the magazine sneered, “it helps to have the kind of inside smartness that Ivan Boesky had.”11 Samuelson was explicit:

  In the same way, experience has persuaded me that there are a few Warren Buffetts out there with high rent-earning ability because they are good at figuring out which fundamentals are fundamental and which new data are worth paying high costs to get. Such super-stars don’t come cheap: by the time you spot them their fee has been bid sky high!12

  As Samuelson knew, Buffett did not charge a fee. In any event, Buffett got his “data” chiefly from annual reports, which were available to anyone. By ignoring this, Samuelson could attribute Buffett’s success to information-gathering, and thus avoid the issue of whether Buffett might have a talent for analyzing the “data.” It was as if he were merely a good librarian.

  But Samuelson, a Nobel laureate in economics, knew he was more than that. At some point after his Senate testimony, Samuelson bought a big stake in Berkshire Hathaway,13 a just-in-case hedging reminiscent of Voltaire’s deathbed acceptance of the Church. Samuelson declined to comment, but he was hardly alone. Armen A. Alchian, a noted economist at the University of California at Los Angeles, also invested in Berkshire. Yet Alchian, in a letter to a fellow economist, maintained:

  I remain convinced that no investment funds, no matter for what kind of new activity or area, can claim demonstrable superior talent. Luck, Yes. But Superior Talent, No.… No, it’s a world in which all reliably predictable events are priced right, with only surprise left. And surprise is “random” by definition.14

  But Alchian devoted considerable energy to explaining Buffett’s results. Ultimately, he decided that the secret lay in a wrinkle in Nebraska insurance law that, according to Alchian, permitted Nebraska insurers to take a greater role in monitoring their investments than out-of-state insurers.

  I attribute his success entirely to that fortunate, happenstance position, and not to any superior (relative to his competitors) skills.

  It is difficult not to marvel at Alchian’s leap of faith. Samuelson at least allowed that “Warren may be as near to a genius at investing as I have observed.”15 Yet that was also a sort of damnation. Genius was not method—and Samuelson disclaimed his method. “Warren gave a talk and said, ‘Any fool could see that the Washington Post was under priced,’ ” Samuelson noted. “I’m not a fool. I don’t find it credible.” Why, then, had Buffett bought the Post? Samuelson replied, “That’s the difference between genius and talent.” As a genius—a sort of freak—Buffett could be dismissed. The Efficient Market Theory would still hold true for mortals. Thus, Stanford’s William F. Sharpe wrote Buffett off as a “three-sigma event”—a statistical aberration so out of line as to require no further attention.

  From the foundation of the Efficient Market Theory, scholars erected the elaborate construct of modern finance. Finance is the inverse of investing; it describes the capital-raising function from the corporate point of view. It is a useful discipline, but inexact. But now, like investing, finance was seen as a quantifiable social science, far more precise than the actual world it purported to explain. One theorist depicted a formula for a stock’s return, R, with M standing for the market, a and b standing for constants that may vary from stock to stock, and u signifying a random error:16

  R= a + bM + u

  The economist hastened to add that this was only a model “in its simplest form.” There was no shortage of models as impenetrable as the Dead Sea Scrolls or a verse from the Koran.

  This “science” was grounded in the only evidence that scholars considered relevant: the data of (supposedly perfect) stock prices. It ignored all of the myriad and changing factors—such as a company’s strategy, products, market strength, and management—that are central to valuing a business in the real world. Such variables are subjective and imprecise; but they are, of course, the stuff that investor-analysts such as Buffett reckon with every day.

  One might have imagined that a few business schools might have turned to Buffett’s reports, at least as a guide, in discussing such topics. But save for an occasional guest lecture, Buffett was a nonperson in academia. This hit him in a sensitive spot. While most investors are content to make the money, Buffett very much wanted recognition for being right. It was important to Buffett, ever the teacher, that Graham—and Buffett—be seen as a useful model. Quoting from a letter:

  In my opinion, the continuous 63-year arbitrage experience of Graham-Newman Corp., Buffett Partnership, and Berkshire illustrates just how foolish EMT is.… the three organizations traded hundreds of different securities … we did not have to dig for obscure facts … we simply acted on highly-publicized events.…17

  In what for Buffett must have been an excruciating defection, Graham himself, near the end of his life, voiced strong doubts about whether, given the growing abundance of stock research, security analysis would still pay off. Shortly before his death, in 1976, Graham told an interviewer:

  I am no longer an advocate of elaborate techniques of security analysis in order to find superior value opportunities. This was a rewarding activity, say, 40 years ago, when our textbook “Graham & Dodd” was first published; but the situation has changed.…18

  Since Graham had also attacked the “random walk,” and specifically the notion of efficient prices, only six months before,19 the best that one may surmise is that he continued to recognize inefficiency in a theoretical sense, but doubted that one could take advantage of it. Buffett never alluded to Graham’s semi-apostasy, but then Buffett always idealized Graham, and Graham was always braver as a teacher than as an investor.

  In contrast, neither Buffett nor the campus theologians entertained any ambiguity. As Buffett noted, economists treated the theory as “holy scripture.”20 Richard Brealey and Stewart Myers’s Principles of Corporate Finance, the most widely used textbook of recent times, captured the unquestioning spirit. The authors spoke of the “discovery” of efficient markets as though the theory had been a natural element awaiting its Marie Curie. Their claim for it was remarkably sweeping:

  In an efficient market you can trust prices. They impound all available information about the value of each security.21

  One wonders how a flesh-and-blood trader would react to Brealey and Myers’s idealized depiction of the stock market. Their traders were ever calm, their speculators ever dispassionate:

  In an efficient market there are no financial illusions. Investors are unromantically concerned with the firm’s cash flows and the portion of those cash flows
to which they are entitled.22

  Scholars did subject the theory to cross-examination, but in an entirely trivial sense. They studied various exceptions to the appearance of random pricing, but these exceptions—so-called anomalies—were formulaic in nature, such as a pattern of stocks showing gains in January, or on certain days of the week, or at smaller companies. The discovery of such a less-than-random pattern wide enough to leave the theory with a pinprick prompted cries of scholarly amazement and feverish demands for more such detailed, arcane studies. Yet in this vast epistemological literature there were no studies of the “anomaly” that might have left the theory with a mortal puncture—the consistently superior record of Buffett and of others such as Keynes, Graham, John Templeton, Mario Gabelli, John Neff, and Peter Lynch, to cite only the better-known names.

  The records of such investors were ignored or wished away. Tony Thomson, an investment manager at Bankers Trust, airily dismissed Buffett as a “red herring.” Buffett’s record signified nothing; absent seventy-five years of quarterly data, Thomson argued, one couldn’t “establish” whether he had done it with brains or luck. “Thus, the jury is out on Mr. Buffet [sic].”23 Burton G. Malkiel, a Princeton economist, popularized this notion in his best-selling A Random Walk down Wall Street:

  But while I believe in the possibility of superior professional investment performance, I must emphasize that the evidence we have thus far does not support the view that such competence exists.…24

  Malkiel saw no evidence of “competence” beyond that of coin-flippers able to disguise their luck as talent. “God Almighty,” Malkiel proclaimed, “does not know the proper price-earnings multiple for a common stock.”25

  This fetching comment introduced a straw man. Graham-and-Dodders did not claim to know the proper price for a stock. Theirs was a rough science, at best. What they said was that on occasion a stock was so out of line that one could leap in without any claim to precision. Such instances might be rare. Graham-and-Dodd investors typically owned only a dozen or so stocks from among the thousands available. But those few could make one rich. Quoting Buffett:

  Observing correctly that the market was frequently efficient, [EMT adherents] went on to conclude incorrectly that it was always efficient. The difference between these propositions is night and day.26

  One striking contrast was in the rival camps’ definition of “risk.” Risk, to Buffett, was the risk of paying more than a business would prove to be worth. And the range of variables was nearly infinite. Was a company dependent on too few customers? Did the chairman drink? Since the sum (or even the number) of such risks could not be figured with precision, Buffett looked for companies—the very few companies—in which the risks seemed tolerable even allowing for error.

  The theorists recognized no such nuances; risk, in their view, was measurable. Since stock prices were right, they simply assumed that the foreseeable risks in a business were incorporated in its price. Every change in outlook was immediately matched by a change in price. Therefore, the best proxy for the “riskiness” of an investment was the historical riskiness of its stock. Risk, then, equaled price volatility. It was defined in precise mathematical terms, as the degree to which a stock had bounced around, relative to each bounce of the market as a whole. As if to sanctify its algebraic properties, it was christened with a Greek letter, beta. A stock with a beta of 1.0 bounced around as much as the general market; one with a beta of 1.2 was quantifiably more volatile, and one with a beta of 1.5 more volatile still.

  The theorists’ logic now proceeded on a dizzying spiral. Investors did not like risk. Therefore, the investors who bought high-beta stocks, being ever-rational, must have done so because such stocks held the prospect of above-average returns—indeed, with returns that would exceed the norm in precise arithmetic relation to their betas. Conversely, investors in lower-beta stocks would “pay” for the past tranquillity of such issues by accepting lower returns. Indeed, since there could be no free lunch, the only way to consistently earn a higher profit was by accepting added “risk”—that is, higher betas.27 Now the totemic nature of modern finance begins to emerge. The only factor necessary to calculate the expected relative return on a stock was its beta. Nothing about the fundamentals of a company mattered; the one number, beta, computed from past stock prices, was the only relevant issue. “What is your beta?” the scholars asked. It was like a mantra. And Wall Street analysts slavishly paid heed. Virtually every brokerage in the country required its analysts to assess their stocks in terms of “betas” and of “risk-adjusted”—meaning “beta-adjusted”—returns.

  To Graham,28 and to Buffett, this was a madness. That a stock bounced around did not make it risky to a long-term holder. In fact, beta turned “risk” on its head. Consider that when Buffett invested in the Washington Post, the market was valuing the Post at about $80 million. Had the stock fallen by half before his purchase, it would have been more volatile—and hence, to an Efficient Market Theorist, “riskier.” Buffett tartly observed, “I have never been able to figure out why it’s riskier to buy something at $40 million than at $80 million.”29

  Columbia Business School brought the two sides face to face in 1984, the fiftieth anniversary of Graham and Dodd’s textbook. Buffett was asked to speak on behalf of Graham-and-Dodders, the University of Rochester’s Michael C. Jensen for the theorists. A devout believer, Jensen had written in 1978:

  I believe there is no other proposition in economics which has more solid empirical evidence supporting it than the Efficient Market Hypothesis.30

  Indeed, he had warned dissenters that the theory was “accepted as a fact of life, and a scholar who purports to model behavior in a manner which violates it faces a difficult task of justification.”31

  Since Jensen’s encyclical, the theory, and particularly the concept of beta, had come under attack. Now, at Columbia, Uris Hall was stacked with Buffett’s fellow investors. Surveying the crowd, Jensen adopted a gracious tone. He felt “like a turkey must feel at the beginning of a turkey hunt.”

  But the turkey was hard to pin down. First, Jensen made an absolutist claim: no profit could be systematically earned from analyzing public information. Then he gave some ground. Exceptions had popped up; he seemed to grant that a superior analyst might exist. Still, he ridiculed the profession in general. People consulted security analysts, like priests, owing to their “psychic demand for answers.”

  When science responds by saying there is no known answer, or worse yet, there is no answer, people are not satisfied.… In such circumstances people are perfectly willing to make up answers, or even to pay others to make them up for then.32

  Security analysts, then, were the spiritual descendants of “the medicine man, mystics, astrologers, gurus.” As opposed to these pagan stockpickers, Jensen claimed for “science” the Efficient Market Theory. Noting that the “star pupils of Graham and Dodd” were in attendance, Jensen still claimed that it was hard to tell if “any” were really superior, due to the well-known “selection-bias problem.”

  If I survey a field of untalented analysts all of whom are doing nothing but flipping coins, I expect to see some who have tossed two heads in a row and even some who have tossed ten heads in a row.

  Buffett could not have asked for a better setup. The failure of most money managers to do better than coin-flippers had been invoked at every turn. But on reflection, there was less to this than met the eye. Since pros accounted for most of the trading, the average stockpicker could not do better than average. The meaningful question was, were there enough gaps in the market’s efficiency so that some preidentifiable group could beat the market, and beat it consistently?

  Borrowing from Jensen, Buffett envisioned a “national coin-flipping contest.” Each day, everyone in the United States flipped a coin, with those who flipped tails continually dropping out. After twenty days, only 215 flippers would be left.

  Now this group will probably start getting a little puffed up about this, human natur
e being what it is. They may try to be modest, but at cocktail parties they will occasionally admit to attractive members of the opposite sex what their technique is, and what marvelous insights they bring to the field of flipping.… But then some business school professor will probably be rude enough to bring up the fact that if 225 million orangutans had engaged in a similar exercise, the results would be much the same.…33

  But what if the surviving orangutans largely came from the “same zoo,” located, conveniently, in Omaha? One would suspect that the zookeeper had something to do with it. Buffett’s proposition was that a concentration of head-flippers did hail from one zoo—from the “intellectual village” of “Graham-and-Doddsville.” Then he laid out the records of nine Graham-and-Dodd money managers—that is, all of those with whom Buffett had had a personal, and long, association. Their particular tastes varied from cigar butts (Walter Schloss) to franchise stocks (Bill Ruane). But each of the nine had beaten the market over an extended period. And each had been preidentified as a Graham-and-Dodder; that is, each had spent his career looking for discrepancies between market price and intrinsic value. None had paid attention to whether stocks did better on Monday or Thursday, or in January or August.