a great academic paper: “Playing Favorites: How Firms Prevent the Revelation of Bad News”

A sidebar next to the “Heard on the Street” column of yesterday’s Wall Street Journal highlighted a research study by Profs. Lauren Cohen and Christopher Malloy of Harvard and Prof. Dong Lou of the London School of Economics titled “Playing Favorites…”.  I found the study itself on Prof. Cohen’s web page (along with a lot of other neat stuff that I’ll be writing about soon).  It’s well worth reading at least the summary at the beginning.

information manipulation

Mssrs. Cohen, Lou and Malloy studied the way in which companies choreograph conference calls by selecting the analysts who are allowed to ask questions.  Their finding:  companies that routinely call only on the analysts who are the most favorably disposed toward them–as measured by how bullish their official buy/sell recommendations are–may have a temporary positive bounce right after the call.  But they end up being significant subsequent stock market underperformers.

how so?

The authors suggest three motivations for companies to engage in this type of information manipulation.  It may want only softball questions because (in my words, not necessarily theirs):

–a company may be trying to “smooth” earnings by making more or less arbitrary changes in its accruals for, say, inventories or tax rates, and doesen’t want these devices subjected to scrutiny, or

–a company may have made its earnings guidance only by the skin of its teeth–and maybe even then only by making dubious accounting choices, which they may not want exposed to the light of day, or

–a company may want to present a glowing, somewhat bookkeeping PED-assisted, quarter because it’s going to soon sell new equity and wants the most favorable price.

why do analysts play ball?

After all, it’s embarrassing to be seen as a sycophant.

CL&M point out that the analysts called on may get to fill the holes in their earnings forecast spreadsheets, and therefore improve their future profit forecasts.  Colleagues who can only listen to the call may have different holes but can’t get them filled, so their forecasts may gain nothing.  In any event, CL&M demonstrate the favored analysts do indeed have better numbers.

As far as “why,” my experience is that the brokers the favored analysts work for garner more investment banking business from the firm in question.  Management returns the follow-up calls of the favored analysts faster, so they have better answers for their money management clients.  And when the CEO or CFO visits institutional holders of the stock, the softball-tossing analysts most often get to arrange the itinerary.   All this means more money for the brokerage houses they work for–and because of this for the analysts themselves.

the best earnings estimates?

How do the CEO’s pets end up with the best earnings estimates?  It may be the way CL&M suggest.

I don’t believe it, though.  Personally, I think selective disclosure still happens, despite its prohibition by Regulation FD.  At the very least, more frequent contact with management gives the favored analyst a greater chance to understand and decipher body language, intonation and word choice–all of which can provide valuable hints about how business is going.

why do these firms subsequently crash and burn?

CL&M observe that the analyst selection behavior is designed to prevent bad news from reaching the market.  But there are limits to how long negative information can be suppressed.  Eventually the companies have to come clean–if for no other reason than the auditors have arrived.  This results in negative earnings revisions and stock underperformance.

Maybe so.  Personally, I have a simpler, though non-verifiable view.  I think the conference call manipulation stems from–and reveals–a character flaw in a firm’s CEO.  The boss doesn’t want to have his ideas questioned and never wants to hear bad news.  Because he always shoots the messenger, subordinates hide unpleasant realities until they’re so big they’ve done serious damage to the firm.

one complaint, sort of

The study has a lot of statistics but no names of offending companies.  They would be worth knowing.  A shrewd investor, however, would keep that information for himself.  He’d also thereby avoid a firestorm of protest from egomaniacal CEOs.