“Beauty makes princes of those who have it” Oscar Wilde
Note: If after reading this blog, you still think that you would never discriminate, please take the test linked at the end of the post and surprise yourself
In August, 2011, The Economist published one of those quirky and interesting articles that make most of us subscribers: The economics of good looks.
The takeaway was simple: beauty gives you a premium in the labor market. Nevertheless, and although there was a clear and strong statistical correlation between beauty and earnings, there was no strong evidence on causality.
So if you have an attractive face, it is more likely that you earn more; but if you earn more it is not clear that your looks got you there.
Do the differences in earnings by ratings on a beauty scale represent causal effects? One possible interpretation is that they represent causal effects of plastic surgery. Such a manipulation would make differences causal, but it appears unclear whether cross-sectional correlations between beauty and earnings in a survey from the general population represent causal effects of plastic surgery.
So what you need is a very clear manipulation of one trait, like the one that shows that Emily and Greg are more employable than Lakisha and Jamal, and received 50 percent more callbacks for interviews.
Do attractive faces receive more job offers?
The authors electronically submitted over 2,500 fictitious resumes of young job seekers in response to real job postings on the most important job search web site in Buenos Aires, Argentina. Each fictitious resume included commonly used names and surnames. Schooling levels, addresses, and other attributes were randomly assigned.
Then a set of 100 pictures (50 men and 50 women) was professionally transformed by mixing up pairs of real pictures, and manipulated by changing length and width proportions in accordance to an ideal beauty “golden ratio”.
Facial attractiveness is maximized when the vertical distance between the eyes and the mouth is approximately 36% of its length; and the horizontal distance between the eyes is approximately 46% of the face’s width. These were the attractive candidates. The unattractive candidates were generated by moving two distances away from the golden ratios.
Then, for each vacancy six applications were sent (three per gender): attractive, unattractive and no picture attached. And for each resume, callbacks were tracked.
Resumes from attractive people received 36% more callbacks than unattractive people. There is a beauty premium, independently of the person’s age (there is little variation here as all applicants are relatively young), gender, or marital status. The results are unaffected by hair or skin color, vacancy name, or firm type.
There is a beauty premium for all occupations, but it is statistically significant only for administrative support and food service jobs. In addition, attractive candidates not only get more callbacks, but they are contacted sooner than less attractive candidates.
We also know that living in the right address also helps in Argentina. Why is all this relevant to a blog that focuses on what works in development economics?
Here we are not talking about blatant discrimination based on gender, ethnicity or sexual orientation. What this evidence shows is no less insidious. Individuals enter job markets – on line or not – and are discriminated against because of attributes that are not so evident – an address, a face, a name. Most of these attributes are required when filling an application in Mexico, Chile, Colombia or the IDB. These attributes effectively provide a signal that does get picked up by the market.
So, what can be done? The solution is simple. Enable anonymous job applications, at least for that first screening. No need for a picture, a name, an address. Nobody will be able to hold against you that you are who you are; after all Swedish schools drop personal pronouns.
Meanwhile, you can try Photoshop.
So let’s take beauty out of the eye of the beholder.
Now, if you read this far, test yourself on your own discrimination score.