and transaction costs are reasonable-there is no evidence of consistendy superior or inferior retums to market participants.

Two lines of statistical research have developed purporting to show the efficiency of markets. The first is a group of studies that indicate money managers do not outperform markets, one of the two facets of the semi-strong form of EMH. The second is that markets quickly adjust to new information because of the rapid (and presumably correct) interpretation of events by investors. This tenet assumes diat thousands of analysts, money managers, and other sophisticated investors search out and analyze all available information, constandy keeping prices in line with value."

First lets examine the EMH premise that investors cannot consistently make money by using public information to find superior investments. Since it is difficult to assess how investors analyze information to determine undervalued stocks, tests of this premise focus on whether groups of investors have eamed superior retums. Because the information is readily available, the group that most frequendy serves as guinea pigs is the managers of mutual funds. The research, some of which we have examined, showed that mutual funds do not outperform the major averages, whether risk-adjusted or not (although the risk-adjusted studies are now certainly open to question), which supports the efficient market hypothesis. But wait.

An even more demanding challenge lay direcdy ahead. For the hypothesis to be invalid, not all investors, but only a reasonable number, are required to beat the market over time. The statistics of the original mutual fund researchers in the sixties and early seventies failed to tum up such above-average performance by any investors, the essential evidence required to make the EMH case.

On closer examination, the efficient market victory vanished. Studies we reviewed in chapter 14 demonstrated that the standard risk-adjustment tools the researchers used were too imprecise to detect even major fund outperformance of the averages. One showed, for example, that using Jensens technique (one of the important mutual fund investigators) only one manager of the 115 measured demonstrated superior performance at a 95% confidence level (the lowest statistical level normally acceptable)."

Even to be flagged on the screen, die manager had to outperform the market by 5.83% annually for 14 years. When we remember a top manager might beat the market by IM or 2% a year over this length of time, the retums required by Jensen to pick up managers outperforming the averages were impossibly high. Only a manager in the league of a Buf-fett or Templeton might make the grade. One fund outperformed the

Those Dreadful Anomalies

market by 2.2% a year for 20 years, but according to Jensens calculations, this superb performance was not statistically significant.

In another study using standard risk adjustment techniques, the researchers showed it was not possible at a 95% confidence level to say a portfolio up more than 90% over ten years was better managed than another portfolio down 3%. It was also noted that "given a reasonable level of annual outperformance and variability (volatility), it takes about 70 years of quarterly data to achieve statistical significance at the 95% confidence level."

One researcher, in an understatement, noted fliat the problem lay in weak statistical tools. Corroborating these findings, Lawrence Summers, later Deputy Secretary of the Treasury in the Clinton Administration, estimated that it would take 50,000 years worth of data to disprove the theory to tiie satisfaction of the stalwarts. Indeed the EMH tools were so weak that it proved impossible to delineate even outstanding performance, which by sheer coincidence was the one thing tiiat would invalidate the hypothesis.

In spite of this and other evidence, the conclusions of Jensens mutual fund study are still used to support the main premise of efficient markets. "There is very littie evidence," wrote Jensen at the time, "that any individual fund was able to do significantiy better tiian that which we expected from mere random chance."

Even though risk measurements used by academic researchers were shown by Fama and others to be valueless, mutual fund performance has not been recalculated by EMH defenders to exclude their now refuted definitions of risk. We have just seen how tiiese measurements resulted in outstanding performance not being detected. The records of most managers who consistentiy outperformed tiie market were wiped out by statistical gobbledygook.

The ghosts of beta and other academic risk measurements still walk the night, defending EMH and weeding out any above-average performance not permitted by the theory. This is not the only instance of such tactics being employed by the true believers, as we shall see.

Revenants and errors notwithstanding, superior performance could not be completely eliminated by the believers. Well look at some of the ghost-busters next.

cording to die liypotliesis, none could do so with regularity. The theorists, as we saw in the last section, focussed primarily on mutual funds, and, although the risic adjustment techniques are questionable, the evidence did show that funds as a group did little better or worse than the market over time.

However, this was not the final hurdle because, as noted, not all funds have to do better or worse consistently. In fact EMH makes a stronger statement: that no group of investors or any investment strategy can do better than the market over time. Thats where die trouble starts.

There are mediods, supported by a large body of evidence, that consistently do better than the market, as do some mutual funds and money managers. The $63 billion Magellan Fund, for example, with over a million shareholders, and under three separate money managers, has outperformed the market for well over a decade. So did John Templeton and John Neff, the latter running billions of dollars for the Windsor Fund for over two decades. How are these outstanding records possible using only public information? Is it sheer chance, as EMH adherents are forced to claim? If it is, we must look at how many other institutional investors have outperformed, using statistics that can actually detect superior performance, not filter it out inadvertently as Jensens methods did. Whether large numbers of professional investors outperform or underperform the market is not settled, as the efficient marketeers would like to think; its wide open. The emphatic statement by these academics that the research available corroborates its central tenet, that professionals perform in hne with the market, is impossible to accept without further evidence.

However, while the jury may still be out on the last point, it has come in with a unanimous decision on another. The verdict is solidly against EMH. The tenet that managers do not outperform or underperform a market benchmark has a corollary: there is no method or system that can consistendy provide higher retums over time. This may be the Waterloo of the hypothesis.

As we saw in chapters 6 to 9, a considerable body of literature demonstrates that contrarian strategies have produced significantly better retums than the market over many decades. The explanation for diis explicitly contradicts die central tenet of EMH-that people behave with almost omniscient rationality in markets. As students of scientific method would anticipate, findings that threatened the existing beliefs were not met with gariands and victory parades by the new conquistadors.