Home / Business Focus
Sun, Dec 29, 2002 - Page 12 News List

Coke forces analysts to (gasp) do their work


Many Wall Street analysts have complained about Coca-Cola's recent decision to stop helping them prepare their quarterly or yearly earnings forecasts. Instead, they should be calling on other companies to follow Coke's lead.

If all corporations stopped feeding earnings forecasts to brokerage firms, analysts would have to rely on their own research. Serious, independent analysis might flourish.

A new study suggests how rare such analysis is these days. The study, by Eric Zitzewitz, an assistant professor of strategic management at Stanford's Graduate School of Business, documents that an increase in what he called "herding" behavior occurred among analysts after August 2000, when new disclosure rules from the Securities and Exchange Commission took effect.

Under those rules, called Regulation FD for ``fair disclosure,'' the SEC prohibits companies from revealing materially important information to analysts in private conversations. It requires them to communicate such information to all investors at once, in a news release or a conference call.

The study found that after companies could no longer freely leak information to favored analysts, those analysts began acting more as a herd, often saying the same things at the same time.

For his study, Professor Zitzewitz examined all earnings forecasts in the database of the Institutional Brokers Estimates System, now part of Thompson First Call, for the period from Jan. 1, 1993, more than seven years before Regulation FD took effect, to June 30, 2001, almost a year after the new rules.

A telling variable is the number of "multiforecast days" -- meaning days when more than one analyst following a particular company issues or revises his forecast. Before Regulation FD, the study found, 30 percent of analyst forecasts for a given company came on multiforecast days. Over the 11 months after that rules took effect, that portion grew to 50 percent, and analysts increasingly just redistributed the kind of corporate earnings guidance that Coca-Cola has announced that it will stop making.

To determine whether an analyst's forecast was repetitive, Zitzewitz looked at a number of factors. They are described in his study, called "Regulation Fair Disclosure and the Private Information of Analysts," which is available at http://papers.ssrn .com/sol3/delivery.cfm/SSRN_ID305219_code020414670.pdf?abstractid=305219.

One of the more revealing factors was the extent to which a forecast diverged from the consensus estimate. The study found that on those increasingly rare occasions after Regulation FD when an analyst issued a report on his own -- on a day when no other analysts following the company also did so -- his opinions seldom veered much from received opinion.

Over all, Zitzewitz concluded that herd behavior approximately doubled after Regulation FD. Before the rules, according to the study, about 65 percent of new earnings information reached investors from forecasts of analysts acting alone. Under the new rules, that portion dropped to just 27 percent.

While all this shows that very little independent analysis is being conducted on Wall Street, it does not mean that much deep thinking went into analysts' reports before Regulation FD. To the contrary, the settlement of almost US$1.5 billion this month between securities regulators and 10 big Wall Street firms makes clear that analysts and corporations have been working hand in hand for many years to distribute self-serving earnings forecasts that they hoped would propel stock prices higher.

This story has been viewed 2992 times.

Comments will be moderated. Keep comments relevant to the article. Remarks containing abusive and obscene language, personal attacks of any kind or promotion will be removed and the user banned. Final decision will be at the discretion of the Taipei Times.

TOP top