Average weight has risen over the past several centuries, but the last few decades have witnessed a particularly rapid rise in weight and obesity that has affected individuals of all ages. While weight growth in children and working-aged adults has received the most attention, the public health consequences of overweight are arguably greatest for the near-elderly and elderly. We propose to study the determinants of weight in this population, and the consequences of weight for their health. To guide the research, we will develop a dynamic economic model of weight gain that illustrates the relationships between food prices, exercise availability, income, and weight. The empirical investigation will begin with some descriptive analysis designed to characterize the changing age-profile of weight and weight gain, by socioeconomic status and other demographic characteristics. We then turn to the determinants of weight. We propose to explore the impact on weight of several key forces: food prices, retirement, income, spousal health, and occupational characteristics. We present some preliminary evidence that both food prices and retirement have substantial effects on weight among the elderly and near-elderly, effects that are in line with the predictions of the theory. Finally, we explore the consequences of weight for health and health expenditures, if any. We begin by characterizing the association between weight and health-related outcomes over the latter part of the life-cycle. We focus in particular on whether the associations between weight and poor health change with age and with socioeconomic status. We then propose to study the causal links between obesity and health. We will use geographic variation in food prices, as well as variation generated by exemptions of food from sales tax; these serve as instruments for weight. We will explore the effect of this variation in weight on health, and the direct effect of price variation on nutritional status. Finally, we will examine whether the impacts of obesity on health can be mitigated by health insurance.
Showing the most recent 10 out of 309 publications