Revenue Estimate On Google Raises Questions On Research
The recent flap over Morgan Stanley analyst Mary Meeker’s latest report on Google raises questions about market research produced by investment banks.
The issue blew up when Meeker published a report on the impact on Google’s new overlay ads placed on its YouTube video site, estimating a potential net revenue boost to Google of $720 million per year. Henry Blodget, a former Meeker rival during his time at Merrill Lynch, immediately pointed out on SiliconAlleyInsider.com that due to a calculation error, Meeker had overestimated the revenue gain by a factor of 1,000.
Meeker quickly corrected the calculation in her report. However, rather than adjusting the total revenue impact figure down proportionately to $750,000, she bumped up a few of the assumptions, coming up with a new revenue estimate range of $75-$189 million.
Blodget, who was barred from the securities industry in 2003 after being charged by the SEC with providing tainted research on companies to win investment banking business, points out this is hardly the first time an analyst has “backed into” the numbers they are projecting. But it’s rare that the public gets visibility into the process.
ResearchRecap recently featured research supporting the view that analysts asociated with investment banks often offer overly optimistic recommendations on companies underwritten by related investment banking units. However, the research also shows that the market has properly discounted these recommendations, even during the bubble period of the late 1990’s.
These were the findings of a new research report “Do Analyst Conflicts Matter? Evidence from Stock Valuations” by Anup Agrawal of the University of Alabama’s Culverhouse College and Mark Chen of Georgia State University.
Another academic study indicates that a reluctance to downgrade stocks has resulted in the buy recommendations of investment bank analysts lagging the performance of independent analysts’ picks.
Four years after the SEC’s historic Global Research Analyst Settlement agreement with ten investment banks this month’s Journal of Financial Economics includes a paper examing how Wall Street research compares with that of the independent research firms. The authors, Brad Barber, of Cal-Davis, Reuven Lehavy of University of Michigan, and Brett Trueman of the UCLA Anderson School, looked at the average investment return on buy recommendations from independent firms, as compared to those from investment banks.
During the period January, 1996 through June, 2003 (prior to the Settlement agreement), the researchers found that buy recommendations from independent firms outperformed those of investment banks by 3.1 basis points per year (nearly 8% annualized). The performance gap was most pronounced during the dot-com boom period of 1999-2000, and was based in part on investment banks’ reluctance to downgrade stocks during the bear market.
Taken as a whole, these results suggest that at least part of the underperformance of investment bank buy recommendations is due to a reluctance to downgrade stocks whose prospects dimmed during the early 2000’s bear market, as claimed in the SEC’s Global Research Analyst Settlement.
Additional analyses find that the underperformance of investment bank buy recommendations extends not only to the ten investment banks sanctioned in the research settlement but to the nonsanctioned investment banks as well.
The paper, Comparing the stock recommendation performance of investment banks and independent research firms, is available for free download.
You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
October 29th, 2007 at 8:15 am
[...] blogosphere would be well-served to double-check their figures before publishing. As Internet guru Mary Meeker learned a few weeks ago, the online community will quickly identify and challenge any potential discrepancies in their [...]