Citi Proves That Contrary Opinion Works More than We Thought

Contrary opinion doesn’t always work, which is why it’s only part of what I look at. As I’ve stated many times, it only works when the opinion gets one sided to an extreme.

The folks at Citi (C) decided to take a look at the performance of analysts recently. They wrote about it Friday morning. It’s not pretty, as FT/Alphaville summarizes:

Anyway, the point according to Citi, is that analysts have a lot of difficulty predicting or catching up to turning points in the market - hence their lagging behind in the recent bull market. A contrarian, they say, could build a portfolio based on analysts’ most hated stocks that might perform reasonably well.

I don’t disagree. And, although the shabby performance of the analyst community is a story in and of itself, here’s something that the Citi and FT folks didn’t mention, if they’re even aware of it. The database that Citi uses is rife with fraudulent changes in ratings! That is, people sneak into the database that tracks analysts’ ratings on stocks, and they rewrite history. I’ll let the abstract speak for itself:

We document widespread ex post changes to the historical contents of the I/B/E/S analyst stock recommendations database. Across a sequence of seven downloads of the entire I/B/E/S recommendations database, obtained between 2000 and 2007, we find that between 6,594 (1.6%) and 97,579 (21.7%) of matched observations are different from one download to the next. The changes, which include alterations of recommendation levels, additions and deletions of records, and removal of analyst names, are non-random in nature: They cluster by analyst reputation, brokerage firm size and status, and recommendation boldness. The changes have a large and significant impact on the classification of trading signals and back-tests of three stylized facts: The profitability of trading signals, the profitability of changes in consensus recommendations, and persistence in individual analyst stock-picking ability.

In other words, if an analyst has a bad call, there is a tendency for that call to be modified in some way.

Yet even after those changes get made, according to Citi’s research, analysts STILL STINK at key turning points! Imagine how bad analysts would have looked (and how good contrary opinion would have appeared) if nobody cooked the books!!

Comments

[DmytrenkoArtem:]

nice article.. its a way how i am trieng to look on market

May 31 06:31 PM

[Tracy Alloway:]

Hi - This is Tracy Alloway from FT Alphaville. We do mention the problems of the I/B/E/S database in the article. This earlier FT Alphaville post is linked to in the article:

ftalphaville.ft.com/bl.../

Jun 01 02:43 AM

[reluctantQuant:]

The aggregation process for large, multi contributor databases is always akin to making sausage.

Some "clean up" entries are legit, some are not. The person who gives the data to the colletor / aggregator is not always the sharpest tool in the shed, and don't expect that the head of research is watching that outflow too closely,

If a firm left Bill's name on a recommendation six months after Bill left by mistake, should that be corrected post-facto?

If the firm fired the lowly admin who sends that data to Bloomberg/Ibes/ThomReut six months ago and no one sent any data since, should that be corrected post-facto? The vendors don't "blank it out" just because there's no update.

The article here is quite valid - I'm not defending any of these goofballs, but the stats utilized here are a little squirrelly.

Whether data is good or bad, at the end of the day, at least half of the analysts are below average anyway. Take it all with a very large grain of salt...

--rq

Jun 01 08:03 AM