One of the most powerful distortions that besets the decision making process is conflict of interest. When a person has conflicting interests in a situation it becomes very difficult to avoid the type of conscious and unconscious biases . Why wasn’t Arthur Andersen doing an effective job of auditing Enron’s books? The obvious answer is greed. As a for-profit organization, Andersen did not necessarily place full and accurate auditing of their clients as a primary motivation. Auditing firms want repeat business from their clients, and an audit that threatens a client’s financial picture is not likely to make for a happy customer. At the individual level, it is common for auditors to be offered positions with an audit client after making connections during the course of an audit. The potential for a lucrative job offer can bias auditors against providing the most accurate audit.

It’s clear that conflicts of interest are a long-standing source of bias in decision making situations. One of the primary tools that companies and regulatory bodies have used to fight it is disclosure. The idea of disclosure recognizes that totally eliminating conflicts of interest is a challenging goal, and offers that a public declaration of these conflicts gives professionals an incentive to “be on their best behavior”, and gives the public notice that certain things should be taken with a grain of salt. Is disclosure an effective way to prevent conflict of interest? New results from experimental economics suggests the answer is no.

Daylian M. Cain, George Loewenstein, and Don A. Moore present evidence in a new paper — The Dirt on Coming Clean: Perverse Effects of Disclosing Conflicts of Interest. They found that not only did people not discount the advice from biased advisors enough to account for their conflict of interest, but that, perversely, making the disclosure sometimes caused advisors in experiments to bias their advice even more, presumably on the basis that making a disclosure reduces the moral imperative to provide accurate and unbiased advice.
From the abstract:

Conflicts of interest can lead experts to give biased and corrupt advice. Although disclosure is often proposed as a potential solution to these problems, we show that it can have perverse effects. First, people generally do not discount advice from biased advisors as much as they should, even when advisors’ conflicts of interest are disclosed. Second, disclosure can increase the bias in advice because it leads advisors to feel morally licensed and strategically encouraged to exaggerate their advice even further. As a result, disclosure may fail to solve the problems created by conflicts of interest and may sometimes even make matters worse.

Read more: The Dirt on Coming Clean: Perverse Effects of Disclosing Conflicts of Interest

Columbia professor of statistics and political science Andrew Gelman has posted a review of Fooled by Randomness by Nassim Taleb to his blog. Gelman notes in his comments he notes that he has done research of his own on the statistics of low-probability events, which of course is one of Taleb’s favorite topics. From the abstract:

Researchers sometimes argue that statisticians have little to contribute when few realizations of the process being estimated are observed. We show that this argument is incorrect even in the extreme situation of estimating the probabilities of events so rare that they have never occurred. We show how statistical forecasting models allow us to use empirical data to improve inferences about the probabilities of these events. Our application is estimating the probability that your vote will be decisive in a U.S. Presidential election, a problem that has been studied by political scientists for more than two decades…

Read more: Estimating the Probability of Events That Have Never Occurred: When Is Your Vote Decisive?

Modern life in America brings with it a degree of personal choice which is unprecedented in history. A typical American faces an abundance of options for most decisions they face throughout the day. We have scores of stores to shop at, restaurants of all varieties to eat at, hundreds of channels of cable or satellite television to watch, and millions of songs to listen to on iTunes. A visit to Starbucks alone allows us to choose from over 38 million potential coffee drinks. In our culture it is almost always assumed that more choice makes us better off. Choice allows people to match their own tastes and preferences more closely. If my favorite coffee drink is a decaf soy sugar-free vanilla latte, it is unlikely that I was very happy with my cup of straight black coffee at the local coffeehouse before Starbucks came to town. Yet in important and overlooked ways, abundant choice can sometimes leave us less happy than we would have been with a more modest set of options.

Sheena Iyengar, a professor at Columbia Business School, researches the situations in which more choice can make us less well off, or where we prefer to have our choices artificially restricted. Her research has yielded counterintuitive results about the relationship between choice and satisfaction. Here she describes a few of her studies:

To explore consumer responses to extensive options, we conducted a field investigation in an upscale grocery store, Draeger’s, in Menlo Park, Calif. On two consecutive Saturdays, a tasting booth for Wilkin & Sons exotic jams was arranged. As consumers passed the tasting booth, they encountered a display with either six or 24 different jams. We observed and calculated the number of people who stopped at the tasting booth as well as the number of people who chose to purchase the jam in question.

The results are striking. They demonstrate that although extensive choice is initially more enticing than limited choice, limited choice is ultimately more motivating. In fact, 60 percent of the passersby approached the table in the extensive-choice condition, as compared to only 40 percent in the limited-choice condition. However, 30 percent of the consumers who encountered the limited selection actually purchased a jam, whereas only 3 percent of those exposed to the extensive selection made a purchase.

In subsequent studies we found that people are actually less satisfied with the choices they make if selected from a larger set of options. For instance, the same Godiva chocolate chosen from a set of 30 chocolates is considered to be less delicious than if it is chosen from a set of six. Moreover, we found that the negative consequences of too much choice extend beyond consumer contexts to work contexts. An examination of individuals completing a task chosen from a larger range of options as compared to a smaller set of options revealed that people performed better at their chosen activities if they have chosen the activity from a smaller range of options.

These results have important implications for the business world, that we are just beginning to understand. Marketers in particular need to be careful to structure consumer choice to prevent regret from eroding the satisfaction people get from their purchasing decisions. Ms. Iyengar has done choice research in many other contexts as well, which I’ll discuss in later posts.

Read more: Hermes Magazine – Choice and its Discontents

Previously:

• Too Many Choices: Who Suffers and Why?
• The Paradox of Choice

In honor of Valentine’s Day, I thought I’d take a look at some interesting research done on gender differences in decision making in the markets.

Brooke Harrington of Brown University studied the performance of investment clubs over an 11 year period (1986-1997). She split the clubs into three categories based on gender composition: male-only, female-only, and mixed gender. Mean male club performance lagged the S&P 500 by 0.56% a year, while the female clubs beat the S&P by 0.28% a year. Even more interesting is the fact that mixed gender clubs performed best, beating the S&P by 1.98% a year on average.

What would account for the difference?

Ms. Harrington, who also spent 10 months observing the behavior of investment clubs first hand, says she may have nailed down two key reasons for the diverging performances.

First, Ms. Harrington said, the two sexes display different biases in picking stocks, with men focusing on stocks related to their work experiences and women, including professional ones, tending to pick consumer-product stocks. So the mixed clubs tended to pick a more diversified group of stocks.

And, Ms. Harrington said, the investment performance of single-gender clubs suffered from a tendency for these clubs — often formed by friends — to be more “socially cohesive,” with members afraid to hurt the feelings of others or knock down their investing ideas.

“There was group think going on,” she said referring to those of the single-gender variety. “In mixed clubs, there’s less at stake socially.” Mixed clubs are more likely to be formed by colleagues, who are used to disagreeing about work-related issues.

So, apropos of St. Valentine’s Day, it looks like we have an uplifting moral: men and women do better together!

Read more: The New York Times – Mars and Venus Do Better Together

More recently, Stefan Ruenzi at the University of Cologne analyzed mutual funds managed by either a single man or woman (no team managed funds). He found that there were significant style differences between the men and women; consistent with earlier research, the men made more aggressive and risky bets and were more likely to change the style of their fund over time, or “drift”. In addition, men were more overconfident in their abilities to time the market, leading to more frequent trading. Women made more conservative investment choices and were more consistent in following their fund’s stated investing style. In the end, the pros and cons of both camps seemed to even out: the funds managed by men profited from their more aggressive bets, but these profits were diluted from overactive trading. The funds managed by women missed out on the riskier profit opportunities but were more diligent in keeping turnover low and sticking to their professed management style. The risk-adjusted returns of the male- and female-led funds were almost equivalent. So from a performance perspective, there does not seem to be reason to prefer a male manager over a female manager. However, because women on average stay more true to their stated investing style, those who are carefully allocating their assets to optimize a portfolio might have reason to prefer a female managed fund.

Here’s their abstract:

To shed some light on the sophisticated relationship between women, men and money, we investigate gender differences among US mutual fund managers. Based on findings from the existing literature on gender differences, we hypothesize that female fund managers take less risk and follow less extreme investment styles that are more consistent over time. Furthermore, we expect female fund managers to be less overconfident and therefore to trade less. Our empirical results support all of these hypotheses. We then turn to the consequences that arise for investors and fund companies, but find no evidence that behavioral differences between female and male fund managers are reflected in fund performance. The more surprising appears our finding, that female managed funds have significantly lower inflows. As fund families earn their fee income on their assets under management, we search for compensating incentives for fund families to employ female fund managers despite their low fund flows. We find that firms with a high probability of being sued for discrimination, i.e. large and well-established firms, are most likely to employ women. Furthermore, female fund managers are more likely to be employed in less conservative states of the U.S. We conclude with implications of our findings for investors and fund management companies.

Read more: Sex Matters: Gender and Mutual Funds

After that last post on the Columbia Collective Dynamics Group I noticed that Nassim Taleb is giving an informal seminar to the group next Friday, Feb. 17th. I won’t be in New York then unfortunately, otherwise I’d love to go.

Nassim Nicholas Taleb
Dean’s Professor, Sciences of Uncertainty, UMASS at Amherst

Mild vs. Wild Randomness
The talk is on the nontrivial difference between Mild (Gaussian) and Wild randomness (non-Gaussian), its consequences for knowledge, prediction, and social fairness, and how it renders much of the statistical tools ineffectual. Related papers can be found on Taleb’s website: http://www.fooledbyrandomness.com/. Of particular relevance are /epistemologyfattails.pdf, /knolwedge.pdf, and /amherstclass/blackswanexcerpts.pdf (the last of which requires a username and password which will be included in the email announcement).

Anyone want to share access to that last file, which I assume from the filename is excerpts from Taleb’s next book, The Black Swan? Comments are open!

Previously: Columbia Collective Dynamics Group