It is generally claimed that obesity adversely affects wages. This article is devoted to identifying the extent to which the claim is consistent with data. Drawing upon the National Longitudinal Survey of Youth 1997 (NLSY97), it examines the effects of obesity on wages by gender and ethnicity. First, an ordinary least squares model is estimated. Second, a fixed-effects (FE) model is used to remove time invariant unobserved heterogeneity. Finally, the FE specification is further estimated after replacing contemporaneous weight variables by one-year lags of weight variables to avoid reverse causality. Body mass index (BMI) is used as a continuous measure of weight and BMI splines (BMI ≥ 30 for obese, 30 > BMI ≥ 25 for overweight, 25 > BMI ≥ 18.5 for healthy weight and BMI<18.5 for underweight) are used as binary measures of weight. Lots of variables related to human capital, demographics, family background and personal attitude are controlled for. Findings provide evidence that white males receive a wage premium for higher BMI. Wages of all other ethno-gender groups seem to remain unaffected by obesity.