|Does Money Buy Better Health? Three essays examining the causes of socioeconomic differentials in health after midlife
|Year of Publication
|University of Michigan
|Demographics, Health Conditions and Status
Does money buy better health? This dissertation seeks to more clearly discern whether social policy aimed at reducing health disparities is likely to be effective through programs that transfer income and wealth to needy adults in their latter working age years. First, I discuss explanations for why money might matter for health. Competing explanations emphasize patterned responses to social structural inequalities and individual health investments. The assumptions underlying these perspectives can be expressed in the language of experiments. Differences reduce to assumptions about the importance of unobserved heterogeneity bias, and the relative importance of fixed, time-varying, and health-related factors as sources of this bias. Second, I investigate the importance of money for health using a sample of 7,400 respondents and their age eligible spouses nearing or undergoing the transition into retirement. Four waves of data from the Health and Retirement Study (HRS) are used to estimate the influence of individual income, wealth, and education on three health outcomes: self-rated health, depressive symptoms, and functional limitations. Chapter Two seeks to more accurately compare the true effect of education with money on self-reported health. More so than education, income and wealth over the work career are influenced by health and other factors. Adjusting for family background and feedbacks from prior health to income reduces estimates of income influence by 75% or more and wealth by about 50%. The effect of education is also diminished, but only by 20 to 30%. These findings suggest that the long-term influence of psychological skills and social environments derived from education outweigh the health benefit of greater economic resources in the peak earning years of middle age. Chapter Three focuses on minimizing unobserved heterogeneity bias due to fixed confounders. Results suggest that income and wealth boosts over several years will have only small influences on self-reported health. Chapter Four examines unexpected life events as sources of income variation, to set up natural experiments. Unexpected job loss, union membership, and occupation are instrument variables. Compared to unadjusted estimates, these alternative estimates suggest only a very small impact of income and wealth on self-reported health.