YOU ARE HERE: LAT HomeCollections


Who's Sorry Now? Folks Who Sold, Study Reports

June 03, 1997|THOMAS S. MULLIGAN

NEW YORK — You got to know when to hold 'em and know when to fold 'em, Kenny Rogers sings, but recent research shows that individual investors don't seem to know when to do either.

In a study by financial economist Terrance Odean, stocks that investors sold consistently outperformed the ones they bought, and by a widening margin over time.

The results confirm the conventional wisdom that "people should be investors, not traders," Odean said in an interview. "They should take a long horizon and not jump in and out of stocks."

The study--titled "Why Do Investors Trade Too Much?"--was based on trades made from January 1987 through December 1993 in 10,000 randomly chosen customer accounts at an unidentified large discount brokerage firm. The database included 97,483 transactions--49,948 purchases and 47,535 sales.

The performance of the stocks was charted at intervals of four months, one year and two years, and was compared with a broad index of stocks traded on the New York and American stock exchanges and the Nasdaq Stock Market.

Discount brokerage data is useful because the trading "is not complicated by agency relationships," Odean said.

In other words, since the investors aren't getting a broker's advice, their successes and failures are their own. Moreover, there is no question of "churning," or broker-initiated trading intended primarily to generate commissions.

Of course, people initiate their own trades not only to improve their investment performance but also for tax purposes, to raise cash for emergencies or major expenditures and for other reasons. Such trades may come at inopportune times and could hurt overall results, something Odean took into account.

But it startled him to discover that when he tried to screen out such transactions and focus in on trades that seemed to be motivated purely by a desire to make a profit--times when investors sold out their entire positions in a stock and replaced it with another stock within three weeks--things only got worse.

When all transactions were examined, the "sells" beat the "buys" by margins of 1.45 percentage points after four months, 3.22 points after a year and 3.57 points after two years.

When only the profit-motivated trades were considered, the sells trounced the buys by 2.46 points, 5.07 points and 8.61 points, respectively.

The results of an earlier Odean study--"Are Investors Reluctant to Realize Their Losses?"--using the same database, sound a similar theme: that people tend to hang on to their losing stocks too long and to sell their winners too early.

Does all this mean investors might do better if they bought a clue--that is, paid extra for the advice of a full-service broker?

Not necessarily, experts say.

Although there apparently are no studies strictly comparable to Odean's measuring the trading records of customers at full-service brokerages, research has shown that such customers tend to barely break even before fees and commissions and to lose money after those transaction costs are taken into account.

"Why do people trade at all?" Meir Statman, a professor at Santa Clara (Calif.) University who studies investor behavior, asked in a recent interview. "If you're lucky, you'll find another idiot on the other side of the trade; if you're not, you'll find an insider who knows something you don't know."

"We think they're just overconfident in their ability to find rules that govern the markets," he added.

Odean blames investor overconfidence for the terrible results of trades in his studies, which he undertook as part of his doctoral dissertation at UC Berkeley's Haas School of Business.

Not only did investors apparently think they knew more than they did, but what little they did know they tended to misinterpret, concluded Odean, who will begin teaching next fall at UC Davis' business school.

Bruno Gerard, a finance professor at USC, said other research indicates that investors overreact to what they hear about the market. This can cause them to buy stocks that have received a lot of favorable attention but that may have peaked, or, on the other hand, to sell stocks that were the subject of negative reports but that may have already bottomed out.

That is one reason why stocks that have been losers in one six-month period tend to outperform over the next six months, he said.

Gerard noted that because Odean's study looks only at trades and not at the overall performance of the investors' portfolios, it does not give a full picture of small investors' results between 1987 and 1993.

Odean acknowledges that, saying the people he studied may in fact have done very well for themselves just by being in the stock market--especially in light of what their money might have done for them had it been in real estate or bonds during those years.

Los Angeles Times Articles