Weights on the Rise: Where and for Whom

Abstract

Using over 70 nationally representative surveys, I find evidence of a dramatic increase in the share of women who are overweight in developing and transition economies, especially in Latin America and the Middle East. Urban rates of overweight, measured by the body mass index, are also far higher than those in rural areas. In examining the inter-temporal changes in the entire standardized weight distribution, there are nonetheless many instances where it is not possible to reject the null of non-dominance, especially in rural areas. It is also clear that the distribution of women's BMIs in most countries is becoming markedly less equitable, and that this increase in univariate inequality is driven largely by the increase in BMIs among overweight and obese women. This is somewhat analogous to the often discussed rich get richer story. That is, the increasing concentration of weight, standardized by heights, among the overweight is driving a significant share of the overall increase in BMI inequality. A related finding is that when I decompose the changes in the prevalence of overweight into the effect of shifts in the mean versus changes in the distribution, in many countries, even holding mean BMI constant, there would be a marked increase in the prevalence of clinically overweight women due to changes in the shape of the BMI distribution.