The Dangers Of Full Disclosure
Dan Ackman, 06.08.05, 9:31 AM ET

Conflicts of interest are everywhere. In finance, analysts tout shares
in companies as their colleagues down the hall sell investment banking
services; in accounting, the same firms (until very recently) provided
lucrative consulting services to the same companies they audited. Drug
companies supply all manner of perks to doctors who then prescribe
their medicines.

Some of these conflicts are inevitable--others are not. But the
solution when they occur is often said to be full disclosure. As long
as the guest on CNBC tells the home viewer that his bank may be in bed
with the stock he's picking, all is well. Your doctor who prescribes
you Lipitor just got Laker tickets courtesy of Pfizer, but he told you
about it. So what if Goldman Sachs is representing both the New York
Stock Exchange and Archipelago Holdings --everybody knew, so it's cool,
right? (see: "The FDA's Conflict Of Interest Problem", and "The NYSE's
Conflict Of Interest Olympics").

Maybe not. It turns out that disclosing conflicts doesn't always solve
the problem, and it may even do more harm than good. That's the
conclusion of a recent study by George Loewenstein, Don Moore and
Daylian Cain, "The Dirt on Coming Clean: Perverse Effects of Disclosing
Conflicts of Interest," published in the Journal of Legal Studies.

Most people think that if they know that their adviser--whether a
doctor, a lawyer or a stock broker--has a conflict, they can adjust,
take the advice with a grain of salt, as the saying goes. But it turns
out they cannot, according to Loewenstein, an economist at Carnegie
Mellon University.

The reason in part is that most people don't understand what the
conflict of interest does in the first place. A doctor may rightly say
that he wouldn't sell his advice for the price of a basketball ticket;
an analyst will argue that he gets paid for being right, not for
generating investment banking fees (at least recently). Goldman Sachs
will say that no one deal fee is worth risking its reputation for
probity. They are all right; but that's not how it works, Loewenstein
says.

"People have this very limited model of conflicts of interest,"
Loewenstein said in an interview. "They think in terms of corruption,
but the larger problem is unconscious bias." It's not that the
professional accepts a quid pro quo and consciously or intentionally
misrepresents his advice, "[The] bias is more frequently the result of
motivational processes that are unintentional and unconscious," he says
in his study.

In other words, a doctor might over time shade his advice to conduct
procedures that profit him, or prescribe drugs from companies that have
funded his research or whose rep has simply bought his medical students
a few lunches. It happens without the adviser knowing it. People have a
way of choosing the courses that serve their interests. Loewenstein's
experimental studies testify to the point.

Full disclosure is often though to be a solution. But it can backfire.
First, people can't really judge the effect of the conflict of interest
on the advice they are receiving. Second, because most people are
inclined to trust their advisers (after all, they picked them because
they are trustworthy) when they have been informed about conflicts,
they feel sure that their advisers can overcome them. When the adviser
lays bear the conflict himself, the recipient of the disclosure trusts
him all the more.

The fact of the disclosure may also affect the advice. Disclosure might
deter advisers from giving biased counsel. "On the other hand, advisers
might be tempted to provide even more biased advice, exaggerating their
advice in order to counteract the diminished weight" the recipients
will give it, Loewenstein writes.

Some professionals might even feel a "moral license" to give bad
advice. This license may even have a legal component. For years, and to
this day, tobacco companies like Altria and Reynolds American cited the
warning labels on cigarette packages as a key aspect in their defense
of smoker lawsuits.

But some conflicts can be avoided--drug companies giving freebies to
doctors is an obvious candidate, Loewenstein says. And assuming that
disclosure wipes the problem away is a serious mistake.