Oct. 18, 2001

Heavier white female workers earn less than their slimmer colleagues, Cornell researcher finds

The heavier white working women are, the less money they make, a Cornell University researcher's study concludes.

Women who weighed 65 pounds more than other women in a sample of 1,442 white female workers earned an average 7 percent less than their slimmer colleagues, when other factors were controlled for statistically. That difference in income is roughly equivalent to the wage effect of one year of education, two years of continuous employment at one job or three years of work experience.

However, the same relationship between weight and income did not hold true for Hispanic and African American working women, says John Cawley, a health policy scholar, economist and an assistant professor of policy analysis and management at Cornell.

Cawley found only weak evidence that overweight Hispanic women earn less, and no evidence that the wages of overweight black women are affected. Cawley also found no evidence that weight affects the probability of employment for white, black or Hispanic women, or the probability of holding a white-collar job.

Cawley, who presented his findings at the Robert Wood Johnson Conference on the Politics of Obesity in Burlington, Vt., in June, analyzed data in the National Longitudinal Survey of Youth on 2,843 white, black and Hispanic females over as many as 12 years up to 1998. Cawley's analysis controlled for weight in alternate ways: weight both in pounds (holding constant height in inches) and body mass index, a measure of weight for height that is the standard measure of fatness in medicine and epidemiology. He also controlled for many other factors that affect wages, such as education, measures of intelligence, number of years at the current job and local unemployment rates.

The average weight of women in the sample was 148.6 pounds. The difference between women at the average weight and those who weighed more than 95 percent of the sample was 65 pounds. Previous studies of this question also have found that some heavier women earn less. However, those studies were unable to determine whether the correlation was the result of weight lowering wages, low wages raising weight, or perhaps some unobserved third factor causing both high weight and low wages. The innovation of Cawley's work is to use the econometric method known as instrumental variables to determine whether weight truly lowers wages; the results of this method suggest that weight does have such an effect, but only for white women. Cawley also attempts to improve on previous studies by using a sample that was surveyed over a longer period of time, and by correcting for reporting error in weight and height.

"It should be stressed, however, that the finding that weight lowers wages is not conclusive evidence of workplace discrimination," Cawley points out. "Another hypothesis also consistent with these findings is that heavier workers are less productive at work. It has repeatedly been found, for example, that obese workers are more likely to miss work due to illness. However, this explanation is complicated by the fact that the analysis finds no evidence that weight lowers wages for black women."

Nevertheless, he points out that the finding that weight lowers certain women's wages has become increasingly important in the last two decades. The percentage of Americans who meet the clinical definition of obesity has risen from 15 percent in 1980 to 23 percent in 1990 to 26 percent in 1999.

The study currently is a working paper of the National Bureau of Economic Research and is available for downloading free of charge at http://www.nber.org/papers/w7841.