The highest-earning women's wages are penalized by obesity as much as five times that of the lowest earners

On the Distributional and Evolutionary Nature of the Obesity Wage Penalty. Christian Brown, P. Wesley Routon. Economics & Human Biology, https://doi.org/10.1016/j.ehb.2017.10.001

Highlights
•    We examine the obesity wage penalty across the wage distribution and career.
•    Quantile and fixed-effect quantile regression control for unobserved heterogeneity.
•    We find an increasingly severe penalty across the wages distribution for women.
•    The obesity wage penalty has grown over time and may slow wage growth.

Abstract: The economics literature supports a link between labor market measures, such as earnings, and health conditions, such as obesity. There is reason to believe the effects of obesity on wages may vary for high- and low-earning individuals and that obesity wage effects may evolve over a lifecycle or from generation to generation. Drawing on data from two longitudinal surveys, we estimate quantile and fixed effect quantile regressions, among others, to further examine the obesity wage effect. Results suggest an increasingly severe penalty across the wage distribution for females. Specifically, the highest-earning women may be penalized as much as five times that of the lowest earners. Results for males suggest that penalties may be present at select wage levels, while prior research has generally found no male obesity penalty. We also provide evidence that the obesity penalty has increased across generations and limited evidence that it may slow earnings growth over one’s lifetime.

JEL: I10; I14; J31
Keywords: Obesity; earnings; wage penalty; longitudinal quantile regression; NLSY79; NLSY97

My commentary: What we imagine is the life of the obese persons? We intuitively believe that getting up in the morning is more slow, that with age the knees will suffer more due to the excessive weight of the body above them, and in some way productivity diminishes.

For those earning less, which are not penalized, the more physical nature of work needs a greater BMI than those of higher earnings.

"Alternatively, individuals with higher BMI may be self-selecting into lower wages by reducing their involvement in the labor market. This is possible if individuals with higher BMIs expect to work less, work for fewer years, or experience attendant health conditions that will limit their ability to work and their employee lifecycle. Further, if higher BMI individuals expect to face discrimination, this may reduce job search time and desire to seek out training, education, and promotion opportunities.

Conversely, obesity penalties are found for the higher key quantiles of the distribution generally associated with more sedentary and competitive forms of work. If BMI is not an important determinant of low wages, and is a major determinant of higher wages, this may suggest that discrimination occurs in the labor market. In an asymmetrical information context, employers discriminate against obese individuals by paying them less or denying opportunities [...] because obesity is seen as an observable proxy for unobservable negative characteristics. It could be that employers are expecting an increased likelihood of healthcare costs and are preemptively selecting obese employees out of employment opportunities which would result in an increasing lifetime earnings profile."