This dissertation broadly examines the economic and health-related consequences of individual behaviors, and their interaction with government programs. It is divided into three distinct chapters. The first chapter examines how changes in family income contribute to increasing obesity among low-income families. In the general adult female population, the prevalence of obesity decreases substantially as family income increases. However, this relationship is not necessarily causal, as numerous other factors could be driving this negative correlation. I make use of the expansion of the New York State Earned Income Tax Credit (EITC) program over the course of the 1990s as a source of exogenous variation in family income in order to estimate the causal effect of family income on obesity. I show that increasing family income has a positive effect on weight and obesity prevalence among the sample population. This effect is concentrated among those who are already obese. The second chapter simulates the effect on New York State residents of an expansion of the EITC on employment, hours worked, income, and poverty, and compares these results to a simulation which excludes labor supply effects. Relative to estimates excluding labor supply effects, the preferred behavioral results show that an expansion of the New York State EITC increases employment by an additional 14,244 persons, labor earnings by an additional $95.8 million, and family income by an additional $84.5 million; decreases poverty by an additional 56,576 persons; and increases costs to the State by $29.7 million. The third chapter is a co-authored work with Richard Burkhauser and John Cawley. It returns to the topic of obesity, and investigates which measures of fatness most accurately predict application for Social Security Disability (DI) benefits. Although the social science literature has wholly embraced the use of body mass index (BMI) as a measure of fatness, many medical researchers argue that BMI is a poor measure of a person?s true fatness. Our results indicate that despite the limitations of BMI, it is consistently a significant predictor of future application for DI, although more accurate measures of fatness occasionally perform better as predictors of application.