Economic perspectives on the obesity epidemic

This thesis takes its starting point in the large spread and rise in obesity prevalence that have been observed around the world, making obesity an increasing public health concern. The thesis consists of an introductory chapter, which justifies the study of obesity from economic perspectives and puts the subsequent studies in context, and four studies that address obesity from different economic perspectives. The first study follows a cohort of Swedish women, aged 20-68 in 1980/81, to analyze income-related inequalities in obesity. Using a version of the concentration index, it finds that the risk of obesity is concentrated among the poorer, but over time, as the cohort ages, inequality declines, because in absolute terms obesity prevalence increases uniformly across the income distribution. The study uses a long-run income measure, and shows that long-run inequality differs substantially from short-run inequality. The second study uses U.S. cross-sectional data between 1960 and 2008 to study time trends in obesity, severe obesity, and body-mass index for different educational, income, and racial/ethnic groups. It shows that, with some exceptions, the greatest part of the increases in body-mass index and obesity over time are shared by individuals in all subgroups of society, despite well-known baseline disparities (particularly among women). This result has implications for our understanding of the kinds of societal changes that have really had an impact on individual decision making and thereby caused the obesity epidemic. Plausible explanations for the obesity epidemic must be consistent with universal trends and should not affect particular socioeconomic groups only. The third study has more of a methodological nature and examines whether measures of socioeconomic disparities in body-mass index and obesity are biased because self-reported weight and height tend to have errors (misreporting), and because body-mass index does not take body composition into account (misclassification). Among women, different educational groups misreport differently, leading to underestimation of the education disparity when using self-reported information. Male disparities are more sensitive to the definition of obesity. Estimating the risk of obesity defined by using waist circumference gives rise to an educational gradient, which is not present when using body-mass index to classify men. Finally, the fourth study continues the discussion about decision making and explanations for the obesity epidemic. It takes the view that institutions may shape individual behavior through norms and habits, and explores the role of economic freedom at the macroeconomic level as one such institution. The main empirical analysis uses a panel of 31 high-income countries and data for the period 1983 to 2008, and finds a positive and statistically significant relationship between the level of economic freedom and both the level of, and five-year change in, BMI