This dissertation consists of three empirical essays on topics in public finance and labor economics. The first chapter explores how homebuyers respond to information about high-risk sex offenders in the neighborhood, whose locations were widely publicized as a consequence of Megan’s Law. I assemble a unique dataset and analyze mobility patterns when arriving at causal estimates of the impacts of sex offender proximity. Sex offenders tend to sort into deteriorating neighborhoods with higher crime rates. In this sense, Megan’s Law does provide information that corresponds to the general qualities of a neighborhood. However, while homebuyers do respond negatively to the presence some offenders, the market does not reflect their distaste for the majority of offender locations, which are home to more transient and mobile individuals.

The second chapter examines the link between youth obesity and education. Using a nationally representative longitudinal dataset, I document a negative relationship between obesity and years of education for young adult females, and especially for white and Hispanic females. The adverse association primarily occurs during post-secondary education and is less relevant among males and black females. There is little evidence that adolescent obesity determines high school achievement, though obesity does predict more absences from school. Health and emotional problems associated with body weight account for some of the perceived effects of obesity. Results suggest that the true role of obesity in academic outcomes is weak.

The final chapter, joint with Cecilia E. Rouse, evaluates a learning communities experiment, which had goals of fostering student engagement and collaborative learning among at-risk students at a diverse two-year college. We find that the program improved academic achievement during the semester it was implemented, but most advantages disappeared in the following terms. The last section of the essay considers the role of motivational factors in explaining the program’s impacts and, more directly, their effects on academic outcomes. We observe that perceiving a more engaging experience in the first semester of college does not necessarily translate to better performances afterwards.