A Semi-Defense of the Efficient-Market Hypothesis

It would be an awful thing for investors if the Supreme Court took away their right to sue as a class in securities-fraud cases. Oral arguments are tomorrow.
Yeah, Eugene Fama got a Nobel. So did Knut Hamsun. Photographer: Scott Olson/Getty Images

The case of Halliburton Co. vs. Erica P. John Fund Inc., scheduled for oral argument tomorrow in the U.S. Supreme Court, may produce one of the most important securities-law rulings in decades. Either the court will curtail the right of investors to sue as a class in securities-fraud cases, or it won't. I hope it's the latter, but here's the rub. The precedents securing that right have been based on the Supreme Court's endorsement of a questionable theory: the efficient-market hypothesis.

I've long viewed the theory, which holds that the prices of publicly traded securities reflect all publicly available information, with great skepticism, based on personal experience. Too many times I've seen journalists and others write articles that have sent companies' stock prices soaring or plunging, based solely on information from public records. If the theory had it right, this shouldn't happen, or at least not very often. Yet it goes on all the time.

My old friend Abraham Briloff, a Baruch College accounting professor who wrote dissections of corporate financial statements in Barron's for 45 years, tanked so many companies' stocks during his career, using nothing but information from their own disclosures, that the phenomenon came to be known as the Briloff effect.

In my first Heard on the Street column as a reporter for the Wall Street Journal, I wrote about a Dallas insurer called Unistar Financial Services in 1999. The American Stock Exchange halted trading in Unistar's stock the morning the article appeared. It never traded there again, and the shares soon were worthless. The company was a complete fraud. But every bit of information in the article came from publicly available records. Markets often are efficient, I figured, but not always.

So I'm torn about the issues presented by Halliburton vs. Erica P. John Fund. I view the ability of investors to sue as a class as a clear benefit to the capital markets. It helps keep corporate leaders honest when they communicate with the investing public. But this has been a good outcome based on a weak premise.

There has been much speculation in certain legal circles that the court may reverse its precedent and make it much harder for small investors to recover damages for securities fraud in many instances. In the suit, the John Fund is the lead plaintiff accusing the Houston-based oilfield-services provider of securities fraud.

The efficient-market theory is the basis for the fraud-on-the-market theory long recognized by U.S. courts. Traditionally, when someone sues another person for fraud, the plaintiff must prove reliance on the defendant's misstatements. Practically speaking, though, it would be almost impossible to demonstrate reliance in a securities-fraud lawsuit against a public company that has thousands of shareholders. So the fraud-on-the-market theory made for a convenient shortcut. The court ruled in 1988 in a case called Basic vs. Levinson that reliance could be presumed, on the grounds that investors relied on the integrity of the securities' market prices, making reliance on a defendant's allegedly fraudulent misstatements irrelevant for litigation purposes.

Yet the efficient-market theory has been under constant attack. For example: After University of Chicago economics professor Eugene Fama was awarded his Nobel Prize last year as the father of the efficient-market theory, Roger Lowenstein wrote in an article for Bloomberg View that "the Flat Earth Society has all but disappeared, but the efficient-market hypothesis is alive and well."

Lowenstein's sentiment is widely held. He noted Fama's own words from a version of his doctoral thesis almost 50 years ago. "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," Fama wrote. Lowenstein went on to say, "This idea is contradicted by the view that human behavior is often irrational (or imperfect) and that, therefore, the market not infrequently gets it wrong. It is also contradicted by the many investors who have exploited mispricings to beat the market and by the many examples of investor folly or bubbles."

Lowenstein's points were on the mark. Even Fama has given some ground on this subject. A week before he was awarded his Nobel, Fama was interviewed at a conference in Chicago by John Nersesian of Nuveen Investments, where he acknowledged that markets aren't "perfectly efficient." That is a perfectly reasonable position.

Just as we know that markets aren't flawlessly efficient, we also know that if a company's stock collapses immediately after a huge accounting fraud is revealed, it makes sense that the integrity of the market price had been corrupted in that instance. To deny shareholders the right to sue as a class would needlessly punish the wrong people.

The efficient-market hypothesis is flawed, but it's not all wrong. Markets tend to be extremely efficient over the long run. (To be sure, in the long run we're all dead.) They frequently are efficient in the short run, even if sometimes they don't function well at all. And they should be efficient enough for the court's purposes here. Limiting small shareholders' rights to seek damages from those that have defrauded them would be a far worse result.

(Jonathan Weil is a Bloomberg View columnist. Followhim on Twitter @jonathanweil.)

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.