CAN MARKET REGULATORS

tell when a bubble has developed and stop it in progress? Alan Greenspan, the former Fed chief, and Ben Bernanke, his unlucky successor, have argued to the contrary. That view underlines the debate today over whether Washington was negligent in failing to restrain excesses in the housing market.

The regulators, steeped in the academic theory of efficient markets (the notion that markets are as right as they can be), say no one is smart enough to know when a market is wrong. Many Wall Street practitioners say this is nonsense. After all, if they could never spot a mispriced security, none could consistently beat the market (though in a given year, some investors would always get lucky).

Warren Buffett has been arguing that the efficient-market view is wrong longer and more persuasively than anyone. Buffett has had an uncanny record of avoiding bubbles he liquidated his investment partnership in 1969, at the height of the madness for "go-go" stocks, and avoided dot-com issues in the '90s. And in a remarkable though little-noticed series of investments disclosed recently, he has made his most emphatic declaration yet that he can identify a bubble when it is happening. As we will see, this is exactly what Bernanke says is impossible.

Let's give the Federal Reserve chief his say first. Bernanke was a highly esteemed scholar before his government career. He specialized in the Great Depression, and he has seemed to doubt that the market was wrong even in 1929, when investors mistakenly believed the U.S. had obtained permanent prosperity. Actually, the Depression lay right around the corner. But Bernanke has referred to the "speculation" of that era in quotation marks as though the market's astronomical rise in 1929 was not necessarily a case of speculation after all.

He has told me that the Fed's action to prick the market then was a great mistake one that made the Depression longer and more serious. We can't deal here with that more complicated issue whether pricking bubbles is helpful or not only with the more limited question of whether bubbles can even be identified.

For more SmartMoney Magazine features, turn to the June issue.

Bernanke held they cannot be, even in 1999, when dot-com stocks without any earnings (or prospects thereof) were going public and doubling on their very first day. In 2002, when the country was suffering the fallout of the high-tech crash, Bernanke, by then a Fed governor, allowed in a speech in New York: "If the Fed had had the foresight to prick the bubble at an early stage, the argument goes, the economy might have been spared needless trauma."

But he quickly rebutted this notion. "First the Fed cannot reliably identify bubbles in asset prices.... To declare that a bubble exists, the Fed must not only be able to accurately estimate the unobservable fundamentals underlying equity valuations; it must have confidence that it can do so better than the financial professionals whose collective information is reflected in asset-market prices. I do not think this expectation is realistic, even for the Federal Reserve." Bernanke has stuck to that view since.

Now for Buffett. In the 2007 annual letter of

Berkshire Hathaway

The contracts run for either 15 or 20 years, and they expire between 2019 and 2027. This suggests that Berkshire (in which this columnist owns stock) inked the deals between 2004 and 2007. Technically, the contracts are option-like derivatives, but they work like this: Berkshire, as noted, pocketed a total of $4.5 billion. In return the counterparties will be entitled to large payments from Berkshire but only on the date each contract expires and only, in each case, if the index in question is lower than on the date the contract was issued. Thus Berkshire is betting very large sums that certain stock markets will not be lower in either 15 or 20 years' time.

Usually, this is a pretty safe bet. Most markets do advance over time. But there have been notable instances in which markets have stagnated over decades. The U.S. stock market did not break fresh ground from 1929 to 1954, nor again from 1969 to 1982. Japan's stock market has yet to come close to its 1989 high of 38957. And in the U.S. the S&P 500 remains below the high it set in March 2000.

Why would Buffett, an investor famously adverse to speculation, risk billions of Berkshire's dollars on market calls? Look again at the above-cited periods, especially at their beginning points: 1929, 1969 and 2000 in the U.S. and 1989 in Japan. Each was the peak of a speculative frenzy. But absent such unusual circumstances, markets go up over time. The Oracle of Omaha is betting he can recognize when markets are in a bubble or at the very least, that he can tell when markets are not in one. And judging by his bet on the S&P, he doesn't think the U.S. market is extraordinarily overpriced now (or was in the recent past).

This has profound significance for ordinary investors, too. If Bernanke is right and one can never tell when the market is too high, then it makes sense to always be invested particularly in market indexes. If Buffett is right, then in those rare (and identifiable) moments of gross speculation, it's better to keep your powder dry. It all depends on which sage you believe: Bernanke or Buffett?

Also See:

• Looking Up to Buffett

• SmartMoney's 2008 Broker Survey

• Lowenstein: Is the Fed Really Helping Us?