Berkeley - Stock analysts consistently underestimate the future revenues of Internet companies even though the data to make more accurate predictions are accessible, according to a new study by three accounting researchers at the Haas School of Business at the University of California, Berkeley.

Moreover, the study has found, the growth rates of Internet companies' Web traffic and past revenues can predict the analysts' margins of error.

The study was produced by Haas School accounting professors Brett Trueman, M. H. Franco Wong, and Xiao-Jun Zhang.

Despite the importance of revenue predictions to the valuations of companies, 90 percent of analysts' revenue forecasts fell short of realized revenues, according to the study's authors. For portal and content/community providers, analyst forecasts were 9 percent below actual revenues, on average, while for electronic retailers, known as e-tailers, they were 13 percent below.

"The findings of this study can potentially have a significant impact on the way investors interpret and use analysts' forecasts," said Trueman.

The Haas School team found that the higher the growth in Web usage - as measured by growth in unique, or individual, visitors, the number of pages they view, and the time they spend on the firms' sites - the greater the degree to which analysts underestimate future revenues, particularly for the e-tailers. Furthermore, the greater a firm's prior revenue growth, the greater the margin of the analysts' underestimate, especially for the portal and content/community providers.

With profits and price-to-earnings ratios being all but useless in valuing Internet companies, analysts have turned to revenues and the price-to-revenue ratio to evaluate them. As a result, revenue forecasts have become a critical and ever more important tool in investment circles.

"In light of the emphasis on revenues in the valuation of Internet stocks, we felt it was particularly important to understand the roles played by analysts, Web traffic growth, and past revenue growth in the forecasting of future revenues," said Zhang.

The researchers also examined a number of statistical forecasting models that are based solely on past financial data. They found that growth in Web traffic can predict errors even in the best of these forecasts.

"This finding is particularly important when valuing firms with little or no analyst coverage, such as small firms or those that have recently gone public, as they, of necessity, rely on historical financial numbers for forecasting purposes," said Wong.

The study was motivated by previous work on the valuation of Internet stocks conducted by these professors. In "The Eyeballs Have It: Searching for the Value in Internet Stocks," they found that investors pay particular attention to an Internet firm's revenues, cost of revenues, and Web traffic, in determining the market value of these firms.

Media Metrix provided the data on unique visitors as well as the data used by the researchers to calculate page views and minutes spent at a firm's Web sites. The study sample included 95 publicly traded e-tailers and portal and content/community providers over the period from October 1998 to December 1999.

###

Links:

Trueman's Web site
www.haas.berkeley.edu/~trueman/index.htm