If you’re like most people, you probably support the concept of disclosure, wherein professionals, such as physicians, lawyers, financial advisers, accountants, and many others are required to disclose their conflicts of interest to consumers. After all, knowing about conflicts of interest will help consumers steer clear of advice that benefits the adviser more than the consumer, right?
Not necessarily, says Sunita Sah, Johnson Graduate School of Management. Sah has spent much of the last eight years studying conflicts of interest and the effects of disclosure on both advisers and advisees, and she’s found many unintended consequences. “An underappreciated problem with disclosure is that it can make matters worse,” she says.
One of Sah’s earliest studies examined what happens during face-to-face disclosure of conflicts of interest. “The adviser tells the consumer, ‘You can choose option A or option B, but I should let you know that I will receive a bonus if you choose option B,’” Sah says. “How does the recipient of this information react to the disclosure? Consumers not only hear that the adviser has a conflict of interest, they also hear that if they don’t choose option B, they will deprive the adviser of a bonus. Instead of being a warning, disclosure has become an implicit favor request. The advisee now feels pressure to help satisfy the adviser’s self-interest.”
To further understand the psychological processes of consumers, Sah ran a series of studies and found that when advisers disclosed their conflict of interest, advisees reported feeling less trust in the adviser but at the same time great pressure to comply with the request. The pressure to comply was often so great, it increased compliance with the advice. “Advisees feel pressure to help satisfy their adviser’s self-interest but would prefer not to have that pressure,” Sah says. “We call this the panhandler effect.”
The panhandler effect is most common in situations where advisers are commonly known to have conflicts of interest, such as when salespeople work on commission. Sah identified another effect, called insinuation anxiety, which is more likely to be present in situations in which the adviser holds more power in the relationship, such as when a patient consults with a doctor about treatment. Imagine a doctor presenting a patient with two treatment options—the standard drug or a clinical trial—then disclosing that she will get a referral fee if the patient chooses the clinical trial.
“If the doctor had not disclosed this, the patient may have rejected the clinical trial for any number of reasons,” Sah says. “But now that the doctor has disclosed the conflict of interest, the disclosure is salient. If the patient declines the clinical trial, it could insinuate that they don’t trust the doctor and that the conflict of interest has corrupted her good advice. The patient now feels pressure to comply with the doctor’s advice to avoid signaling distrust to her.”
“Advisees feel pressure to help satisfy their adviser’s self-interest but would prefer not to have that pressure. We call this the panhandler effect.”
An Unintended Consequence of Disclosure—An Expertise Cue
Sah recently documented another unintended consequence of disclosure that occurs when doctors admit their specialty bias. She asked male subjects to imagine they had prostate cancer. They then viewed a video where a surgeon explained the two possible treatments—surgery or radiation oncology. Sah discovered that the men were more likely to choose surgery when the surgeon added a disclaimer in the form of a statement acknowledging that, as a surgeon, he was biased toward recommending surgery. “Disclosures of specialty bias increased trust in the doctor and the likelihood the men would choose a treatment in accordance with the doctor’s specialty,” Sah says, “This occurred even though the men acknowledged that the surgeon was biased toward surgery.”
To explore this phenomenon in more depth, Sah turned to disclosures in a different context—online blogging. Bloggers sponsored by the brands whose products they review or promote are required to disclose this sponsorship. Sah and her collaborators examined consumers’ reactions to disclosures from bloggers and found they actually increased consumers’ perception of a blogger’s expertise. “If there was a disclosure statement, the subjects were more likely to rate the blogger as more credible and were more persuaded by the blogger. For example, they were more likely to take the blogger’s advice or share the blog with other people,” Sah says. “Disclosure worked as an expertise cue, and thus, subjects felt greater trust in the blogger’s expertise. This was really fascinating because most bloggers would hypothesize the opposite—that disclosing their sponsorship would have a negative impact. If the goal of the disclosure was to make consumers more skeptical of the advice, it failed. So, we have to be very careful about how we use conflict-of-interest disclosure and test it in the field to see if it’s working as intended.”
When Professionals Succumb to Conflict-of-Interest Behavior
Sah has also turned her attention to why advisers accept conflicts of interest in the first place. Her experience, first as a medical doctor receiving offers of gifts from pharmaceutical companies and later as a consultant to a pharmaceutical company, led her to hypothesize that doctors use their feelings of personal struggle and sacrifice to justify accepting gifts that lead to potential conflicts of interest. This was born out in her study of new doctors just out of medical school. When new doctors were reminded of their personal sacrifices, such as lack of sleep and high debt, they were much more likely to say it was acceptable to take gifts from drug companies.
“If the doctors were asked whether poor working conditions or stagnant salaries justified accepting gifts from drug companies, the majority rejected this rationalization,” Sah says. “But when reminded of their personal sacrifices, they were still more likely to say gifts were acceptable than were those not reminded of the same sacrifices. This demonstrates that even rationalizations that are explicitly rejected can still have an effect on attitudes at an unconscious level.”
Professionalism and Favorability Bias
In a further series of studies, Sah investigated whether people’s sense of their own professionalism might protect them from potential conflicts of interest. Subjects were asked to rate themselves on a variety of traits that contribute to professionalism, such as their degrees of objectiveness and impartiality. Then they engaged in a series of tasks designed to, first, provide an opportunity for them to be influenced by a small gift and, later, allow them to react to the gift giver in a consequential situation. Subjects who rated themselves high in professionalism were more willing to accept a gift and also to show a favorability bias toward the gift giver.
“Those high in professionalism were more likely to think they can regulate their bias down the line,” Sah says. “However, they actually demonstrated greater favoritism toward the gift giver. Perhaps because they were so confident of their professionalism and believed their decisions were objective and fair, they did not adjust for any potential bias.”
Conversely, those who rated themselves low on professionalism showed less preference toward the gift giver. “We think that’s because they were more on guard,” Sah says. “They knew they’d accepted a gift from the gift giver, and they realized they were probably going to be biased toward that person, so they adjusted for that. We were surprised. They showed much less bias than we expected.”