“From a historical perspective, it makes sense,” said Culverhouse Assistant Professor of Accounting Dr. Chez Sealy, frowning as he considered the idea. “If I need a leader for my tribe, I’m going to look for the guy who looks like the most macho, masculine leader in order to protect us.”
“George Washington looked best on a horse,” agreed Culverhouse Professor of Finance Dr. Doug Cook.
Sealy and Cook weren’t disclosing their personal views on who should be in charge. Instead, they were gathered with two other Culverhouse professors (Professor of Finance Dr. Shawn Mobbs and Associate Professor of Accounting Dr. Quinn Swanquist) to discuss how and why personal appearance seems to matter in business situations.
Like it or not, it is generally agreed that appearance does matter. Conventionally attractive people tend to fare a little better in business and in life in general. For instance, it has been observed that good looking people tend to enjoy benefits that are less available to average looking people, like more prestigious work, higher pay, and more professional respect and influence. Still, researchers are continuing to discover how and where appearance matters in business, and perhaps most importantly, whether it should matter at all. Does it merely reflect a possibly outdated bias like the “macho leader” one, or does personal appearance actually predict important outcomes, like executive performance?
Earlier this spring, the four researchers met over Thai food to discuss their research on how executive appearance affects career outcomes. Over curry and sushi, the professors first described their respective projects.
Facial Traits and Audit Partner Success
In recent research, Sealy and Swanquist investigated how facial traits relate to career outcomes for audit partners in large accounting firms. They found that partners whose facial appearance violated common gender stereotypes tended to be less successful. For instance, male partners whose faces appeared warmer (a stereotypically female trait), and female partners whose faces appeared more professional and competent (a stereotypically male trait) tended to have smaller client portfolios. This was harder on the men: males with traditionally female facial traits were less likely to work for Big 4 firms.
Attractiveness and CEO Success
In their research, Mobbs and Cook focused on how facial attractiveness correlates with CEO succession. Using a unique, objective measure, they found that higher attractiveness among executives correlated with promotion to CEO, influence on shareholders, and faster promotion to board chair once appointed as CEO. They also found a significant increase in compensation for executives that were one standard deviation more attractive. “All things being equal, two executives, one is objectively, by this (measure), more attractive than the other, they are more likely to be selected (as CEO),” Mobbs said.
It’s Not What You Know. It’s How You Know It
In academic research, methods matter, and business research is no exception. While all four researchers found support for the general idea that facial features predict executive career outcomes, they arrived at these insights from different directions. For instance, Sealy and Swanquist used human raters to assess various facial features of their subjects: After collecting photographs of audit partners, the researchers asked raters to make inferences about the partner in a given photo’s attractiveness, competence, professionalism, likability, trustworthiness, and so on. They then compared this information to the partner’s client portfolio, with attention to factors like portfolio size and largest client. To ensure reliability, they used a huge sample: over 2,400 photos, each rated ten times: more than 24,000 artifacts.
Mobbs and Cook, on the other hand, noticed that many measures of attractiveness relied on surveys, but surveying the actual decision-makers for CEO hires—the boards of directors—was not feasible. They also worried about “noise” that human rating introduced into system. For example, Mobbs and Cook hired Mechanical Turk raters (crowdsourced, remote workers) to rate CEO photos from 1-10. One rater scored every photo one out of ten and when Cook inquired of the rater, who was in his twenties, he responded “All of these people are like, 60. They are all ugly to me!” Yet many Mechanical Turk raters are in their twenties, and most of the CEO candidates being rated are in their late fifties or early sixties.
Mobbs and Cook wanted a more objective measure, so they used a mask developed by plastic surgeons, based on the Fibonacci ratio, that seeks to describe the ideal human face (faces of people like Tom Cruise and Angelina Jolie fit the mask perfectly). The researchers hired a graphic designer to overlay the mask on 100 CEO photos, and measure the distance between the photo and the mask in 25 different nodes. They compared their mask measure to a survey and found them to be highly correlated suggesting that the objective measure is useful in reducing noise is assessing attractiveness.
Face the Consequences
Job searchers know that employers now routinely stalk their social media profiles before extending offers for interviews. That’s why LinkedIn is filled with images of confident, warm, professional-looking people who carefully craft their online personas. And on the surface at least, this just seems like good business sense. It’s the old saying: “Dress (and look) like the job you want.”
So yes, we can all dress professionally, take care of ourselves, and practice good grooming. But some things are simply beyond our control. Barring large expenditures for plastic surgery, some of us must simply live with less attractive faces. The real question is, does facial attractiveness actually mean anything beyond a hardwired urge from other people to prefer one face over another?
For instance, if facial attractiveness correlates with charisma, and charisma correlates with successful leadership, we might expect more attractive people to be better leaders. At this point, the slope gets slippery. What is to stop us from simply selecting people for important roles based largely on their looks, something we are likely prone to do anyway? Is executive attractiveness a bias we need to control for in job selection, in promotion?
At this point, we simply do not know, but Swanquist believes that artificial intelligence could be a help in this area: “AI is a lot easier, probably, to deprogram for biases than a person.”
Culverhouse Conversations is an initiative to bring small groups of Culverhouse faculty from different departments and disciplines together, over a meal, for a structured but casual conversation about a given research topic.