Carnegie Mellon Press Releases

Back to Press Releases

Carnegie Mellon News Service Home Page

Carnegie Mellon Today

8 1/2 x 11 News

News Clips

Web News Stories

Calendar of Events



Press Release

Contact:
Michael B. Laffin
412-268-2900

For immediate release:
January 6, 2005

Disclosing Conflicts of Interest Can Have Damaging Effects

Carnegie Mellon study suggests that fixes to address conflicts may actually make matters worse

PITTSBURGH—Disclosing conflicts of interest may clear the conscience of those who face them, but contrary to conventional wisdom and accepted business practices, startling new research at Carnegie Mellon University suggests that such confessions may actually harm those they are intended to protect.

The study, "The Dirt on Coming Clean: Perverse Effects of Disclosing Conflicts of Interest," by Daylian Cain, George Loewenstein and Don Moore, reveals how seemingly ethical behavior - full disclosure of potential conflicts of interest by an advisor (e.g., a realtor, physician, lawyer or stock analyst) - can compromise advisor-client relationships and can harm the client. The paper is published in this month's Journal of Legal Studies (www.journals.uchicago.edu/JLS) and is part of a larger body of research by the authors about how attempts to regulate ethical behavior in this post-Enron era can actually backfire.

According to the researchers, disclosure gives those with conflicts of interest a moral license and a strategic incentive to further skew information made public. "Once an advisor has disclosed his potential conflict, he may feel that it is fair game to offer biased advice since he has made his motives known. Also, with the disclosure, the advisor feels an incentive to try even harder to manipulate his client who will be less likely to take the advice at face value," said Cain, a doctoral candidate in organizational behavior at Carnegie Mellon's Tepper School of Business and lead author of the study.

"In other words, while disclosure may warn a client, 'Cover your ears and do not listen to an advisor,' the disclosure may also encourage advisors to counteract the warning by yelling even louder," Cain said. Loewenstein, professor of economics and psychology at Carnegie Mellon's College of Humanities and Social Sciences, added: "The warning effect achieved by the disclosure often will not offset the increased bias. And while we know that disclosure has further problems, the main conclusion in our study is that disclosure left those receiving advice worse off for having been warned."

Inspecting the Money Jars

For the study, involving nearly 150 undergraduates at Carnegie Mellon, "estimators" (study participants serving as the client), were shown several jars of coins, briefly and from a distance, and asked to guess the total value of the money in each jar. The more accurate the estimate, the more the estimators were paid.

Groups of "advisors" inspected the jars more closely and gave the estimators advice via written suggestions of each jar's worth. The first group of advisors was paid more when their clients responded with accurate estimates. Second and third groups, however, were paid according to how high their clients' estimates were compared to the actual jar-values.

The third group of advisors also had to clearly disclose their conflicts of interest to their estimators, warning the estimators of their incentives to give biased advice. Not surprisingly, the conflicted advisors (in the second and third groups) gave more biased advice - i.e., their advice was higher compared to what they actually thought - than those in the first group who did not have a conflict of interest.

Even more telling was the fact that advisors in the third group who were required to disclose their incentives gave even more biased advice than their counterparts in the second group who did not provide disclosure. While disclosure caused the estimators to slightly discount the advice they received, this discounting was insufficient to offset the increased bias the advice contained. The result was that estimators receiving the disclosure ranked the worst in overall performance, even against those who were not informed about their advisors' incentives to give biased advice.

Full Disclosure No Remedy

"Disclosure is not always bad," said Moore, assistant professor of organizational behavior and theory at the Tepper School. "Future research should attempt to determine when disclosure works and when it fails. However, previous research suggests that bad advice will influence your decisions even if you try to ignore it."

Moore and Cain said the study should serve as a warning to policymakers that seemingly straightforward regulations intended to help consumers may have unintended consequences.

"My previous research suggests that forcing people to share information may only encourage them to avoid those circumstances where the information has to be disclosed. We see that forcing people to come clean may only encourage them to play even dirtier, either in giving increasingly biased advice when their motives are out in the open, or in doing business as usual from behind increasingly closed doors," Cain said.

The research on disclosure spawned a Carnegie Mellon conference and a forthcoming book ("Conflict of Interest: Challenges and Solutions in Business, Law, Medicine, and Public Policy") with Cambridge University Press, edited by Moore, Cain, Loewenstein, and Harvard Business School's Max Bazerman.

###


Other Carnegie Mellon News || Carnegie Mellon Home