By MARK HULBERT

Published: January 1, 2006

AT the level of investment theory, a debate continues to rage between active stock pickers and those who advocate passive indexing. At the level of everyday practice, however, the debate appears largely over. A new study finds that stock picking accounts for a small and declining proportion of the market’s overall trading volume.

The study, Is Stock Picking Declining Around the World?,papers.ssrn.com/sol3/papers.cfm?abstract_id=849627.

Measuring the prevalence of stock picking would seem difficult, if not impossible. One would have to know each investor’s intent when buying or selling a particular stock. Was the investor’s transaction part of a passive indexing strategy, intended merely to recreate the market’s portfolio and to match its performance, or was the objective to pick only a few stocks in the hope of outperforming the overall market? And investors themselves may change their minds along the way, further complicating the measuring.

The report’s authors sidestep these thorny problems by approaching the question from a much different angle. They calculated what stocks’ trading volume would be in a hypothetical world in which all investors abstained from active stock picking. Armed with the answer, they could estimate the prevalence of stock picking by comparing each stock’s actual volume with its hypothetical extreme.

What trading would occur in a world where no one was a stock picker? According to the study, it would take place only when investors wanted to put more money into the entire stock market, or to take money out. All stocks would be bought or sold in tandem on such occasions, with no one stock bought or sold more than any others.

Note that this does not mean that the same dollar amount of each stock would be traded. In a world of universal indexing, the authors note, there would be greater dollar trading volume of larger-capitalization stocks than smaller-cap ones. What should be the same for all stocks in this hypothetical world is the ratio of trading volume to a stock’s total market capitalization.

In the real world, the researchers say, these ratios vary widely from stock to stock. That’s hardly surprising. But the study did find that they vary far less today than they did in earlier decades. In the 1960’s, for example, the earliest decade in the study, the ratios differed so much that the authors estimate that stock picking accounted for as much as 80 percent of all trading volume. In the current decade, by contrast, they estimate that it accounts for just 24 percent.

They repeated these calculations for 42 foreign stock markets, though back just to 1995, because there was less data for them than for the American market. They found that stock picking had declined in almost all those markets as well, accounting for less volume from 2000 through 2004 than it did from 1995 through 1999.

The study’s authors did find, however, that stock picking was more prevalent in emerging countries than in developed countries. For example, they say, it accounts for some 80 percent of the trading volume in the Chinese stock market.

IF stock picking has become so rare in the United States, shouldn’t it be easier for stock pickers to find market-beating stocks? After all, the market becomes less efficient in setting stocks’ prices as fewer investors are trying to beat it. In other words, is it possible that indexing has become too prevalent?

The researchers say they think that the answer is no. According to a complex model that they developed, the stock market becomes remarkably efficient even when relatively few investors are trying to beat it — as few as 1 in 10, in fact. That means that even while stock picking has declined sharply in recent decades, enough investors are still engaged in the pursuit that the rest of us should avoid it ourselves.

Mark Hulbert is editor of The Hulbert Financial Digest, a service of MarketWatch. E-mail: strategy