Category: Evidence-based Management

The Wall Street Journal had a column today by Jason Zweig on the role that bad group dynamics have likely played in much of the current mess and a short but useful list of ways to avoid bad financial decisions.  I was quoted as saying, essentially, that groups bring out the best and worst in people, a theme that goes back to Freud and is bolstered by studies of group decision-making. 

I especially liked two of Jason's points:

• Reframe the question. Committees considering
  an important decision should break into a "pro" and "con" group, each
  developing the best arguments supporting its side. Individuals can do
  something similar by asking not only how much they will make if they
  are right but also how much they could lose if they turn out to be
  wrong. Try coming up with three reasons against an investment as well
  as three for it.

When we discussed the above approach during the interview, I also suggested that authority can cause a real problem as people are sometimes afraid to contradict the leader. One way to deal with authority issues is to either simply leave the room of you are the boss, and better yet, leave and divide the the team into two smaller teams to each develop their own point of view — this is exactly what President John F. Kennedy did with executive committee during the Cuban Missile Crisis, as I wrote in this post on Management by Getting Out of the Way.  

A related point that Jason and I discussed is group size. It didn't make in the article, so I will explain a bit here.  One of the things I worry about when I see the groups making all these investment decisions is that they are simply too big.  J.Richard Hackman — the world's leading group effectiveness researcher — argues and presents evidence that once groups get over five or six people, productivity and group process begins to suffer (see his explanation here or in his book Leading Teams) — in fact, one of Hackman's studies showed that the optimal team size was 4.6 people! Of course, the exact best number depends on the task and on the how experienced and skilled the team is at working together.

To return to the article, I also especially liked the advice Jason relayed from Harvard's Max Bazerman, as it is so strongly supported by evidence — but something that few investors or investment committees seem to be able to bring themselves to do:

• Define the default position. Max Bazerman,
  an expert on decision-making at Harvard Business School, suggests that
  investors start with the assumption that the ideal portfolio is a
  diversified basket of low-cost index funds. Any deviation from that
  strategy should require extraordinarily compelling evidence.

If you are interested in learning more about how to avoid the worst decision traps, Max has written a wonderfully useful book: Judgment in Managerial Decision Making.

P.S. On a much different subject, I am also grateful to Jason for telling me one of the best asshole revenge stories I have ever heard.

In my post yesterday, I talked about seating arrangements, and the notion that men –although not necessarily women — at the head of the table are seen as leaders.  Cindy asked about standing vs. sitting and power dynamics.  My hunch, is that standing is taken as a sign of power, and I bet there is some research out on the question. Indeed, given the struggles that women have to be recognized as leaders, especially in mixed-gender groups, I would suggest both going to the head of the table AND standing if you want to run the show.

But power is different than compassion.  I just ran into an interesting study suggesting that oncology (i.e., cancer) doctors who sit rather than stand are seen as more compassionate, and that is what patients much prefer them to do during consultations.  To generalize beyond these data, it strikes me that anytime that it is an emotionally sensitive meeting (rather than one where efficiency is the main goal), sitting down is good idea.

Here is the reference:

F.Strasser, J.Palmer, J.Willey, L.Shen, K.Shin, D.Sivesind, E.Beale, E.Bruera.  "Impact of Physician Sitting Versus Standing During Inpatient Oncology
Consultations: Patients' Preference and Perception of Compassion and
Duration. A Randomized Controlled Trial. Journal of Pain and Symptom Management, Volume 29, Issue 5, Pages 489-497,2005.

Here is the abstract:

The purpose of this study was to determine the impact of
physician sitting versus standing on the patient's preference of physician
communication style, and perception of compassion and consult duration.
Sixty-nine patients were randomized to watch one of two videos in which the
physician was standing and then sitting (video A) or sitting and then standing
(video B) during an inpatient consultation. Both video sequences lasted 9.5
minutes. Thirty-five patients (51%) blindly preferred the sitting physician, 16
(23%) preferred the standing, and 18 (26%) had no preference. Patients
perceived that their preferred physician was more compassionate and spent more
time with the patient when compared with the other physician. There was a
strong period effect favoring the second sequence within the video. The
patients blinded choice of preference (P =0.003),
perception of compassion (P =0.0016),
and other attributes favored the second sequence seen in the video. The
significant period effect suggests that patients prefer the second option
presented, notwithstanding a stated preference for a sitting posture (55/68,
81%). Physicians should ask patients for their preference regarding physician sitting
or standing as a way to enhance communication.

When I was in graduate school at The University of Michigan, the group dynamics researchers used to talk about how seemingly innocent things could have a big effect on how much influence that people had.  One of the things they talked about a lot (as I recall from old studies done at Harvard) was that just putting people at the head of the table — randomly assigning them — would lead them to talk more, to order other people around, and to be seen as the leader. Indeed, a lot of this research is summarized in this great story called "You are Where You Sit" that was in BusinessWeek in 2007 — especially in the above graphic (click on to see a larger version). 

Unfortunately, as I was looking for those old Harvard studies — which I am still searching for — I ran into a disturbing series of experiments on sex differences by Natalie Porter and her colleagues.  These studies show that when research subjects were shown pictures of people sitting around a table, that when a man was sitting at the head, he was nearly always identified as the leader, but when a woman was sitting at the head of the table, she was only  identified as the leader consistently when it was an all female group. Note that both male and female research subjects made more or less identical ratings — so the women held the negative stereotypes just as strongly as the men,  Here is the abstract if you are curious:

This study shows that women are unlikely to be seen as leaders. Subjects (n =448) rated each member of a five-person group (shown in a photograph) on leadership attributes and also chose one of the five as "contributing most to the group. "Eight different stimulus slides were used. In two slides the "head-of-the-table" cue to group leadership was pitted against sex-role stereotypes. A man seated at the head of the table in a mixed-sex group was clearly seen as leader of his group, but a woman occupying the same position was ignored. The head-of the-table cue identified women as leaders only in all-female stimulus groups. The data were consistent with the hypotheses that sex stereotypes still control social judgments, and that discrimination operates nonconsciously and in spite of good intentions.

This study is old, but not that old — 1983.  I wonder if it would still hold today. Based on how few women are CEOs of Fortune 500 companies, I fear that they still would. What do do you think?

Note the reference is:  Porter, N., Geis, F.L., & Jennings, (Walstedt), J. (1983). Are women invisible as leaders? Sex Roles, 9, 1035-1049.

OK, they didn't use red pumps. They used red shirts.  And this research may fit under the category of psychologists demonstrating what everyone already knows.   But I was still pretty amused to see that Andrew Elliot and Daniel Niesta published an article called "Romantic Red: Red Enhances Men's Attraction to Women" in the Journal of Personality and Social Psychology (2008, vol 95:1150-1164). Note this is not a marginal journal, it is among the most prestigious journals in psychology. 

In a series of five studies, starting with subjects looking at a picture of woman against red versus white backgrounds, the researchers found that heterosexual men were more attracted to a woman pictured against the red background.  Other colors had no effect and the "romantic red" effect did not affect how female subjects rated the attractiveness of the woman in the picture.  In another experiment, the researchers selected a picture of a moderately attractive woman from the website "hotornot.com" and used photoshop to produce two (otherwise identical) pictures of her in a red and a blue blouse. 

The sample strikes me as mighty small in this final study (12 in the red condition, 11 in the blue condition), but the effects were large.  The young men who viewed the woman in red blouse reported that they found her more attractive in general, would be more likely to ask her on a date, found her more sexually desirable, and would spend more money on a date with her compared to the same woman woman in blue.  This research also found that color had no effect effect on men's attributions of a woman's kindness, likeability, or intelligence. And it showed that the subjects in the study were unaware how color was affecting their preferences. 

Here is part of the the researchers' justification for their "red-sex" hypothesis:

"As such, it is likely that women, like other female primates, display red more often and more prominently when nearing ovulation. We also think it reasonable to posit that men, like their more primitive male relatives, are predisposed to interpret a display of red by a female conspecific as a sexual signal and to respond accordingly." (page 1151).

We human beings sure are weird, huh?

I've always thought that people who believed that motivation was affected by the colors of walls and so on were deluding themselves.  Science has a new study that suggests I may be wrong.  Researchers found that people doing tasks with a blue background on their computer monitors were more creative. And, as reported on the fantastic web site BPS Research Digest:

In yet another experiment, participants were given twenty "parts" from
which to design a child's toy. Participants given red parts designed
toys that independent judges rated to be more practical and
appropriate, but less original and novel. By contrast, participants
given blue parts came up with more creative toy designs.

Check out the BPS link for other nuances of how color affects performance.

I remain a bit skeptical, but — given my commitment to evidence-based management — I am less skeptical than before I read this research.  And I am not going to change the blue background that I've had on my computer monitor for several years just in case!

Here is the reference:

Ravi Mehta, Rui (Juliet) Zhu (2009). Blue or Red? Exploring the Effect of Color on Cognitive Task Performances. Science. In Press.

My colleague and close friend Jeff Pfeffer is now writing some stuff at BNET.  You might not always agree with Jeff (I sure don't, our motto when we write together is "when two people always agree, one of them is unnecessary" or "the more we fight, the better we write"), but Jeff is deeply smart and loves being irreverent.  I will try to figure out how to link to the set of columns, but I really liked this one called "When Will We Ever Learn."  Jeff describes three (evidence-based) reasons that people are so bad at learning from past mistakes: Lack of focus on understanding failure, over-reliance on compensation as a management tool, and hubris.  I might add confirmation bias (people see, remember, and believe evidence that fits their biases much better than evidence that clashes) and "rosy retrospection," the tendency to forget the bad and remember the good about the past (even good things that didn't ever really happen).

Check out this interesting Fortune story — it talks about 9 of their 100 Best Places to Work that have never done layoffs. I know it is a tough thing to do, but it is worth knowing that when companies do it last and least, there is evidence to suggest that they bounce back faster when things turnaround.  Something worth keeping in mind, although I know how tough it is and that many companies have no choice.

All this talk about executive compensation reminded me of one of my favorite academic studies of CEO compensation.  It was published in the late 1980's by my Stanford colleague Charles O'Reilly and two co-authors, one of whom — Bud Crystal — was a consultant who specialized in executive compensation. 

They tracked 105 large firms and compared numerous predictors of how much the CEO was paid — I am looking at the key tables in the article, and see these included number of employees, sales, return on equity, and assets.  The interesting twist, however, was they hypothesized — regardless of a firm's performance and size — that the amount that outside directors on the compensation committee were paid on their own jobs would be a strong predictor of CEO pay.  This is based on research on social comparison and anchoring — the idea is that members of the compensation committee would use their own pay as a guide to help determine how much to pay the CEO — and would be excessively swayed by this vivid information. 

The results still amaze me: After controlling for traditional size and performance measures, the amount of money made outside directors, especially those on the compensation committee, had a huge effect on CEO pay.  O'Reilly and his colleagues report that for every $100,000 that the average member of the compensation committee is paid, the CEO's pay goes up another $51,000 per year.  Remember, these effects are independent of firm performance and size!

There are two lessons here.  The first, as is well-documented, that there is little relationship between what CEOs get paid and firm performance, other — less rational — factors overwhelm it. See Pay Without Performance for the gory details.  The second is that, if you want to make a lot of money, pick the richest and most highly paid people you can find to set your salary. 

This is part of the problem that all those financial services CEO's now face. Obama's $400,000 salary is being bandied-about as the anchor, rather than the pay of the often stunningly wealthy people who have been sitting on their compensation committees for so long.   Not only have their reputations suffered, the basic group dynamics and psychological principles they have to deal with are against them too –  they got merely well-to-do rather than fabulously wealthy people setting their pay now.

The reference to the article is:

O’Reilly, C.A. III, B.G. Main, & G.S. Crystal. 1988. CEO Compensation as
Tournament and Social Comparison: A Tale of Two Theories. Administrative
Science Quarterly. 33: 257-274.

P.S. Charles also has a more recent study that seems to show that the more a firm pays the people on the compensation committee, the more the CEO gets paid!  I quote "On the average,for every $1,000 more in fees that is given to the chair of the compensation committee,the cash compensation of the CEO is $1,746 higher." How is that for a rate of return?  Talk about pay for play, or I guess it is pay for pay.  This is from: O’Reilly, C.A. III and B.G. Main. 2007. Setting the CEO's Pay: It's More Than Simple Economics.  Organizational Dynamics. 36(1):1-22

Dan and Chip Heath of Made to Stick fame have a brilliant column in this month's Fast Company, called Why Incentives are Irresistible, Effective, and Likely to Backfire.   I've written about perverse incentives here and here, and how part of the problem with them is that they sometimes work to well, as people focus on doing the thing they are paid for at the expense of all else.  With Merrill Lynch's bonus system so much in the news, it is interesting that the Heath's use an example from an older book about Merrill to show the downside of how yearly bonuses were handed out.  If this little snippet is a reflection of their culture and work practices, it may also help explain why executives who grew up in this system suffer from tunnel vision –  this strikes me as a powerful way to breed a narrow and self-centered world view.

Don't miss the rest of Heath's article, but here is the little gem of a paragraph that inspired this post:

"Take Merrill Lynch. In the book Riding the Bull, author Paul
Stiles describes his experience as a new trader at the venerable
investment bank. Merrill wanted Stiles, then 29, to trade complex
international bonds in volatile markets. He tried asking advice of the
seasoned traders, but they ignored him — a minute spent helping Stiles
was a minute spent not adding to their monthly bonuses. They kept
barking into their phones for hours at a time and yelled at Stiles
every time his shadow fell across their computer screens. Eventually,
Stiles was reduced to silently observing their behavior from a
distance, like a rogue MBA anthropologist. It surely never dawned on
the person who set up Merrill Lynch's incentive system that the
traders' bonuses would make training new employees impossible."

This story also reminds me of related research that shows when people just think more about money, they are less likely to give help, ask for help, and put more physical distance between themselves and others.

Dave Livingston sent me a link — and some of his usual sharp insights — as well to a Washington Post article published this Sunday about mistake that the Bush Administration made that the Obama Administration can learn from.  I hope, for everyone's sake, that Obama is able to learn from others' mistakes so he doesn't make them himself.  This list also strikes me because, essentially, this is a call for applying the logic of evidence-based management to the job. Here is a link to the article and Dave's summary of the list:

1. Presidents set the tone. Don’t be passive or tolerate virulent
divisions.

2. The president must insist that everyone speak out loud in front of the
others, even — or especially — when there are vehement disagreements.

3. A president must do the homework to master the fundamental ideas and concepts
behind his policies.

4. Presidents need to draw people out and make sure that bad news makes it
to the Oval Office.

5. Presidents need to foster a culture of skepticism and doubt.

6. Presidents get contradictory data, and they need a rigorous way to sort
it out.

7. Presidents must tell the public the hard truth, even if that means
delivering very bad news.

8. Righteous motives are not enough for effective policy.

9. Presidents must insist on strategic thinking.

10. The president should embrace transparency.