Are better looking people more easily promoted and do they tend to make higher wages then their less aesthetically appealing co-workers?
These are the questions that I set out to answer by following a methodology similar to what Hamermesh and Binddle (1994) used in their research paper titled “Beauty in the Labor Market” . Using data that contains fewer explanatory variables, but more observations (1260), a multiple regression analysis with qualitative (beauty) information will be estimated. I find that a persons earnings are not affected by above average looks in a statistically or economically significant way. While being good looking might not be an advantage, having below average looks does harm a persons earning potential. Surprisingly, the harm is of being plain looking is larger for men than for women. A woman rated as below average in physical appearance can be expected to earn about 11.5% less an hour . Men also earn about 14.5% less if they are viewed as not good looking by society’s standards. This is consistent with the findings in Hamermesh and Binddle’s research.
Data and Summary Statistics
The people rated as the best looking were given a 5 rating while those deemed not so attractive where given lower ratings. A rating of 3 was described as average looking and about 57% of the population fell into this category. The attractive ratings by sex are summarized below and they appear to have a similar distribution.
In order to control test for attractiveness and its affect on wages one has to control for other variables that affect wages. There may be a genetic component relating ability, intelligence, and physical symmetry (which people find attractive). I believe that controlling for education, experience, tenure, marital status, and other demographics capture some of genetic contributions to a persons ability. Maybe further research can test and extend this hypothesis to cross reference the results found in this post.The list abbreviation and description of these variables are below:
One can’t really see the relationship from the tightly packed points on the graph. This is merely illustrative and one cannot draw a conclusion from simple two variable graphs without controlling for other variables, but it does appear to have some pattern. It looks like there is a positive (negative) relationship between income and attractiveness between the least (average) attractive to (most)average attractiveness. What can be observed in the male sex?
The relationship between good looks and wages appears to follow a similar patter to that of women. All of these graphical depictions are interesting, but in order to gain a more scientific estimate one must control for other variables that may affect wages. In order to accomplish this, a regression relating looks and wages is estimated below.
Multivariable Regression: Controlling for Other Factors and Statistical Significance
The regression above represents helps explain the differences in a man’s hourly wage. The column “Coef.” represents the percentage in a man’s wages from the variables in the first column. The box in blue highlights the attractiveness variable and ever category below that box represents variables that the regression is controlling for.
Being rated as less attractive than most reduces a man’s wages by about 14% after controlling for other things such as education, experience, union membership, race, marital status, geographical location, industry, and size of home city. This result is statistically significant as evidenced by a t-statistic whose absolute value is larger than 1.96. Men rated as having above average looks did not posses any advantage when it came to earnings.
Women’s Physical Attractiveness and Their Wages
Being rated as a below average looking woman reduces ones wage by about 11.5 percent and this result is statistically significant at the 10% level. Women with above average looks don’t posses an advantage over their not so attractive co-workers when it comes to wages.