Academic commentary on law, business, economics and more

February 26, 2006

Whisper Numbers

posted by Bill Sjostrom at 12:19 pm

A “whisper number� once meant the consensus Wall Street unofficial and unpublished earnings-per-share forecast for a public company. Wall Street analysts historically derived a whisper number by ferreting out non-public information from company personnel. The number was then shared with top clients. Studies indicated that trading on whisper numbers could result in abnormal returns (see here). The SEC viewed the selective disclosure reflected in whisper numbers as problematic because, among other things, it “leads to a loss of investor confidence in the integrity of our capital markets.� Hence, in 2000 it adopted Regulation FD prohibiting selective disclosure and essentially putting an end to traditional whisper numbers.

However, a new type of whisper number has emerged at web sites such a According to the site: is an independent financial research firm that collects sentiment and market expectations from the investment public. The “whisper number” found on is derived from an average of individual investors’ expectations regarding earnings for the most recent quarter. The whisper number is available to the general public and is considered an alternative to analyst estimates (or consensus numbers) for quarterly earnings.

This article describes a recent study on the impact of post-Reg FD whisper numbers:

Although a collection of random investor opinions may seem a dubious source of financial wisdom, whisper numbers do appear to be a factor in the movement of share prices, according to a study conducted by finance professors at San Jose State University. “We were somewhat surprised that the stock market as a whole reacted significantly to whispers,” says Prof. Hanis Zaima, co-author with Prof. Maretno Harjoto of “Conflict in Whispers and Analyst Forecasts: Which One Should Be Your Guide?

The study, based on data from from January 1999 to April 2002, revealed a number of statistically significant findings. Most surprising for the authors was that when earnings per share met or exceeded analyst forecasts — but not the whisper number — on average the share price experienced significant losses after the earnings release. That indicates that the market gave more weight to the whisper number than to the analyst estimate, says Zaima; she adds that “different information is being captured by the forecasts given by analysts and whispers.”