One of the complaints often heard about the investment industry is lack of disclosure about compensation. It is up to clients to ask their financial adviser how they are compensated, and even then it might be difficult to verify if the adviser is telling the truth. The reality is that compensation has a huge impact on the investment recommendations by advisers.
It would seem that more disclosure is the obvious answer, but according to one study I read, it might not make much of a difference in the actions of clients and might make the advisers even more biased.
My wife is currently reading the book Why We Make Mistakes by Joe Hallinan. This book is similar to quite a few other books I’ve read that analyze why we make the decisions we do. Books like Your Money Or Your Brain, Nudge, Sway, Paradox of Choice and many others look at various situations we face and try to figure out why we make consistently irrational decisions and what we can do about it.
She pointed out one section of the book to me that refers to a study done by George Loewenstein from Carnegie Mellon. He wanted to evaluate the effects of conflict of interests disclosure from advisers, on the decision making of their clients. The study is call “The Dirt On Coming Clean – Perverse Effects of Disclosing Conflicts of Interest“.
His study used volunteers who played one of two roles – the “Adviser” and the “Estimator“. Estimators had to estimate how much money was in a glass jar and were rewarded for accuracy. Advisers were provided with more information than the estimators and were instructed to give advice to the estimator in order to help them estimate more accurately. Each Adviser provided an estimate to help the Estimators.
The Advisers were divided up into two groups – one set of Advisers were compensated according to how accurate the “client” estimated the amount of money and the second group was compensated according to how high the “client” estimated – accuracy didn’t matter for that second group.
Another test was that the conflict of interest that the second type of Adviser faced was disclosed to some, but not all of the Estimators.
The actual mean jar value was $18.16.
- Advisers who were compensated on estimator accuracy estimated an average of $16.48.
- Advisers who were compensated on high Estimator estimates, but conflict of interest was not disclosed estimated an average of $20.16.
- Advisers who were compensated on high Estimator estimates and disclosed their conflict of interest estimated an average of $24.16.
It’s fairly obvious from the results that compensating the advisers for encouraging a higher estimate influenced their behaviour. What was more surprising is that disclosing the conflict of interest actually increased the bias even more.
Lowenstein says that “moral licensing” is one of the reasons this happens. Basically this theory says that an adviser with an undisclosed conflict of interest will feel guilty enough about it that they will try to “do the right thing” to some degree. By disclosing the conflict of interest, it allows the adviser to do whatever they want since they have admitted the conflict and therefore don’t have to feel guilty about it anymore.
On the Estimator side, Lowenstein showed that although the Estimators did discount the advice from the Adviser when the conflict of interest was disclosed, they underestimated the severity of the conflict and the estimates were less accurate compared to the estimates provided where there was no conflict of interest.
Lowenstein concludes that conflict of interest disclosures may not have much benefit, and can even backfire and produce more distorted estimates as a result. He concludes that the best way to deal with a conflict of interest is to remove it.
In Britain, financial advisers are not allowed to receive commissions. This doesn’t mean they don’t get paid – just that they have to charge their clients directly instead of being paid by a third party, such as a mutual fund company.
What do you think? Should financial advisers be more open about disclosure or should commissions be banned like in Britain?