Beauty and the Worker - Part 6

This does not mean that each good-looking male will earn 4 percent more than each average-looking male. We have seen that there are many other factors that affect earnings, and these will differ between men whose looks are viewed as the same. Even more important, there is tremendous randomness in earnings that is unrelated to looks or any of the other things we can measure and that affect earnings. Among a randomly chosen group of male workers, or female workers, at least half of the differences in earnings are due to things that we can't measure; and among those that we can measure, looks account for only a small fraction of the differences. Looks do matter a lot; but other things matter much more.

Because so few people are classified as beautiful (rated 5) or homely (rated 1), it is not possible to distinguish statistically the impact of being beautiful from being above-average (rated 4), or of being homely from being plain (rated 2). Despite that, and even though the differences are not statistically meaningful, additional analyses of these same data show that the beautiful man or woman earns more than the above-average, and the homely earn less than the plain. Extreme looks are uncommon, but they generally produce extreme effects on success in labor markets.

The word “generally” is key here. Many people believe that a “bimbo effect” exists — that extremely good-looking women are somehow penalized in labor markets. In my own research I have found only one bit of evidence for this effect: In a study of attorneys, the very best-looking female attorneys were less likely to achieve partnership before their fifth year after graduation from law school than average-looking women attorneys. Like their brethren, though, their extreme beauty did give them higher earnings. There may be bimbo effects in some instances, but they are pretty rare.

There have been many efforts to measure the effect of beauty on earnings using data on individuals in other countries. Interest in the topic is hardly limited to the United States. All of these have tried to adjust for many of the same determinants of earnings that I have used to isolate the effects of beauty in the United States. The availability of information on all these measures differs across countries and sets of data, so that the studies are neither entirely comparable to those from the United States nor to each other. They are also not comparable for another crucial reason: We saw that there are international differences in the willingness of raters of beauty to classify people as being below-average in looks.

Add comment

Security code