The impact of Obesity on Earnings - Evidence from Swedish Panel Data

Sammanfattning: Obesity is one of the most important public health concerns around the world. Research suggests that obesity has potentially important effects on labour market outcomes. Using longitudinal data, the thesis aims to study the effect of obesity on wage earnings in Sweden. The data set used in this study is the Swedish level-of-living Survey (LNU) from the year 1991 and 2000. Different estimation procedures, namely Pooled Ordinary Least Squares (POLS), Fixed Effects (FE) Model, and Instrumental Variable (IV) approach are used to examine the effect of obesity on wages in Sweden for all individuals, and for males and females separately. The empirical results suggest a strong negative correlation of Body Mass Index (BMI) and wages for females (p<0.01), but not for males. Though statistical significance reduces considerably, however, accounting for unobserved individual effects, the FE estimation also shows that a one unit increase in BMI score is associated with more than 0.6% lower females wage earnings, and an obese woman’s wage earnings is about 6% lower than a normal weight woman’s (p<0.10). Based on the IV estimation, I further find a strong statistically significant wage penalty for an obese female over a normal weight female that is causally about 8% (p<0.01). The findings are robust to alternative specifications and sub-samples examined in the sensitivity analysis. The thesis concludes that higher BMI score or obesity may cause a wage penalty for females but not for males in Sweden.