Why sell-side analysts are wrong

By Felix Salmon
June 21, 2010
Aaron Pressman has a good report on the sell-side failure to perceive the risks surrounding BP in the wake of the Deepwater Horizon catastrophe. Why was Wall Street so very bullish on the stock even as it was being pummeled in the markets?

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Aaron Pressman has a good report on the sell-side failure to perceive the risks surrounding BP in the wake of the Deepwater Horizon catastrophe. Why was Wall Street so very bullish on the stock even as it was being pummeled in the markets?

Pressman points to a number of factors, including banks’ desire for a healthy relationship with BP, and analysts’ tendency to cluster together for comfort. Joshua Brown reckons it’s deeper than that, and that sell-side analysts tend to see everything in terms of value and discounted cash flow, which are not particularly useful tools when companies get hit by unexpected tail risks.

And more generally, as Lynn Thomasson shows, sell-side ratings are a pretty good contrary indicator:

Wall Street’s lowest-rated stocks have turned into this year’s best performers…

Huntington Bancshares Inc., the regional lender in Columbus, Ohio, had twice as many “sell” ratings as “buys” in December and jumped 66 percent this year for the fourth-largest advance in the Standard & Poor’s 500 Index. Eastman Kodak Co. and Sunoco Inc. have gained more than 20 percent after more than 30 percent of the analysts covering them at the start of the year recommended getting rid of the shares…

Coca-Cola Co., the world’s largest soda maker, had 14 “buys” and 1 “sell” rating in December and has lost 8.2 percent this year…

Pfizer Inc., the world’s largest pharmaceutical company, is down 16 percent this year even after 81 percent of analysts covering the New York-based firm said investors should buy shares.

The one part of all of this I disagree with is in Pressman’s piece:

Wrong-way Wall Street calls are more than just an academic problem. They cost real people real money.

This was the theory behind Eliot Spitzer’s war on sell-side analysts; it was dubious then, and it’s even more dubious now. Investors — even retail investors — aren’t sheep who simply do what they’re told by sell-side analysts. Institutional investors, in particular, use sell-side analysts as a source of ideas, and even more as a source of access to the company. Big investors are a little bit like bloggers, actually, but instead of linking to articles and then saying what they think, they give out soft-dollar commissions and put on their own trades. Which, like blog entries, can either agree or disagree with the original idea.

But Brown’s point is important as well. Sell-side analysts live in mediocristan, and are prone to being blindsided by the unexpected; they almost never, for instance, recommend negative-carry trades. Investors, if they’re any good, know this. No one ever made money by blindly following sell-side advice, and so we should hardly be surprised that people whose position coincided with the sell-side consensus ended up losing a lot.

9 comments

Comments are closed.