Trustworthy and dominant-seeming men: access to corporate management. Dominant-seeming women: not so much.
For all the effort put into convincing young children that what matters most is who you are on the inside, research confirms what we all grow to understand: looks matter. Not just in terms of personal attraction, but on the job as well.
Candidates whose faces exude a sense of trustworthiness are more likely to win elections. Same goes for someone bent on climbing the corporate ladder. Teachers whose faces read as attractive have higher ratings, and lawyers deemed attractive win more legal cases.
Now comes evidence that how corporate C-suites size up a stock analyst’s face seems to impact the analyst’s access to management, which in turn impacts the quality of analyst earnings forecasts.
Baruch College’s Lin Peng, UCLA Anderson’s Siew Hong Teoh, Chinese University of Hong Kong (Shenzhen)’s Yakun Wang and Cornell’s Jiawen Yan analyzed facial traits of nearly 800 sell-side stock analysts working between 1990 and 2017 who also had a LinkedIn profile photo as of 2018.
The researchers deployed artificial intelligence and machine learning models to score faces for characteristics that connote trustworthiness (are your intentions to help or hurt me?), dominance (do you exude the confidence and ability to get the job done?) and attractiveness. Prior research found these three main traits explain the bulk of what matters to us when we are forming visual impressions of someone.
Facial impression data on a given analyst was then combined with an analysis of their earnings forecast accuracy dating as far back as 1990.
In a paper published in the Journal of Accounting Research, the researchers found that analysts whose faces scored highest on trustworthiness and dominance made more accurate forecasts than analysts with the lowest scores for those facial traits.
The researchers surmise that corporate management grants more access to analysts whose faces they read as being highly trustworthy and capable.
That hypothesis is buttressed by further analysis of attendance at analyst/investor conferences. Through attendance databases, the researchers were able to ascertain meetings that both an analyst and management were present for, suggesting potentially valuable access.
Showing Up and Looking Good
Among analysts who showed up in person, and whose faces scored highest for trustworthiness and dominance, the incremental accuracy of their forecasts in the six months after attending a conference improved by more than 15%. That is significantly higher than the 5.1% average accuracy bump prior research found for conference attendees.
Moreover, among stocks with high institutional ownership, the advice of analysts whose faces scored as being highly trustworthy/dominant was followed more intently. When an analyst whose face signals high trust and high dominance revised an earnings forecast, the price movement for that stock was incrementally more than when less trustworthy/dominant-seeming analysts changed their call. That dynamic was strongest among more volatile stocks, suggesting institutional investors are even more dependent (subconsciously, or not) on analysts in the face of uncertainty.
Attractiveness was only a fleeting early-career help for analysts. The researchers found that the earnings accuracy for attractive-seeming analysts with less than two years of experience was superior to the accuracy of new analysts who had lower attractiveness scores. But when seasoned analysts are considered, the researchers do not find a significant beauty effect. The implicit message being that looks might help someone get initial access to management, but over time, trustworthiness and dominance are what matters more.
The Gender Double Standard
Wall Street has long been Exhibit A for a boy’s club, a point driven home by the fact that less than 13% of the subjects in this study were female. Among that small subset, the researchers found a glaring gender bias in how facial traits seem to be internalized by corporate management.
Dominance is an accepted norm for men, but for women, not so much. And that appears to affect analyst access to management.
The overall forecast accuracy for analysts (male and female) with the highest dominance scores was 7.9% better than the average. But among women whose faces exuded the strongest dominance traits, their forecast accuracy was 8.5% worse than women with the lowest dominance traits.
To be clear, skill doesn’t seem to be an explanation. The researchers’ analysis found women’s overall forecasting accuracy was on average 2% better than their male counterparts. What may be more at play is a well-known double standard that punishes women for the same trait — dominance — admired in men. In this instance, women with high dominance scores seem to be punished with less access to management.
A similar gender bias based on perceived facial traits might also impact career advancement. Institutional Investor’s annual All-Star analyst list is an industry popularity contest of sorts voted on by (predominantly male) asset managers and buy-side analysts.
The researchers calculate men with a high dominance score are 1.8% more likely to make it onto the list relative to men with a low dominance score (which is a 25.4% increase relative to the male sample mean). Women with the highest dominance scores were 8.4% less likely to become an All-Star relative to women with the lowest dominance scores (which is a 67.7% reduction relative to the female sample mean).
The authors stress that more research on the perceptions of female analysts’ facial traits is needed to better understand what is going on. And they acknowledge that their research is based only on how facial traits are perceived, not the actual characteristics of a given subject.
That’s a shortcoming of all decision making that uses facial recognition technology for more than identification. This research provides fresh data that in one field, those perceptions appear to have a meaningful impact on information flow and may be another factor playing into career trajectories for women.
Siew Hong Teoh
Lee and Seymour Graff Endowed Professor; Professor of Accounting
About the Research
Peng, L., Teoh, S.H., Wang, Y., Yan, J. (2022). Face Value: Trait Impressions, Performance Characteristics, and Market Outcomes for Financial Analysts. Journal of Accounting Research, 60(2), 653-705.