Are you the sort who stands at attention when your broker calls with news of a recommendation
change on a stock? Do you buy or sell based on the information? Or do you imagine Wall Street stock
analysts as animals that move in homogeneous herds like sheep and lemmings, a timid bunch with
little original thinking to back up their ratings and earnings estimates?
The latter is a sentiment that's gaining in popularity. A story in the
New York Times business section on
December 31 brought the issue to light, and for several years now, CNBC has used the penguin as the
quasi-official mascot of Wall Street, airing images of the lock-stepping birds whenever several
analysts rush to cut their ratings only after a company announces bad news.
But that comparison is probably unfair -- at least to the penguins. Yale finance professor Ivo
Welch recently looked at nearly 54,000 analyst recommendations from 1989 through 1994. In "Herding
Among Security Analysts," published in the December issue of the Journal of Financial
Economics, Welch shows that between 13 and 16 percent of earnings estimate revisions are
explained simply by the change in estimates of all the other analysts. In other words, instead of
spending all their time studying stocks, analysts spend a surprising amount of time studying each
other. (You can read Welch's paper at the Journal of Financial Economics'
website,
or under the Academic Research Papers section of Welch's own
site.)
Welch found several other things worth knowing:
The bulk of the herding occurs soon after the most recent revision. Within a few days
after one analyst adjusts his earnings estimate, the consensus comes closer together as other
analysts flock to follow the leader.
Analysts are more likely to move in a herd when they're bullish than when they're
bearish.
The analysts gather in stronger packs when stocks have generated hot returns over the
prior 60 days.
So what can you do with all this? You can track analysts' earnings revisions at websites like
Zacks, Thomson Invest (
which has estimates from First Call) and
Multex. Welch's research suggests that
if a stock's price has been on a tear, if most analysts are bullish, and if estimates are rising,
then estimates are likely to keep rising. That's good news for the stock price, at least in the
short run. (Welch does warn in his paper that the analytical herd is not a particularly good
predictor of long-term returns. And it's clear from his work that if you're reading analysts'
reports days or weeks after they come out, it's probably too late to profit from them.)
This research helps explain why momentum investing -- buying stocks with rising prices and rising
earnings estimates -- can work well in certain markets. (Among cynics, this is also called "the
greater fool theory": If you buy, you're a fool, but you can always hope to sell to an even greater
fool.)
Personally, I'd never invest this way -- or even dignify it with the name of "investing." But if
you want to gamble, it helps to know how the game is played. Actually, reading Welch's paper made
me wonder if we should be kinder to lemmings and sheep. Maybe we could better symbolize Wall Street
analysts with
amoebas,
which clump together for safety and rotate like a spinning donut, chasing each other in an endless,
mindless dance.
|