UNCOVERING THE CAUSAL PATHWAYS OF HEALTH AND EDUCATION CHOICES
Open Access
- Author:
- Rice, Andrew B
- Graduate Program:
- Economics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 28, 2010
- Committee Members:
- Edward James Green, Dissertation Advisor/Co-Advisor
Edward James Green, Committee Chair/Co-Chair
N Edward Coulson, Committee Member
Coenraad Arnout P Pinkse, Committee Member
Duane Francis Alwin, Committee Member - Keywords:
- life expectancy
education
structural estimation
health
time preference - Abstract:
- This dissertation concerns the relationship between education and health. While this is not a new topic, the focus of this research is to explicitly model the education and health decision process and estimate such a model. Compared to current literature which focuses on statistical modeling as opposed to economic modeling of the decision, this project allows for deeper intuitive and policy analysis. Chapter one reviews the Economics literature on the relationship between health and education. It then discusses the current issues in the literature and proposes a solution to some of those problems. This chapter also provides a simple economic model of health and education choice from which a more complicated dynamic model is derived in chapter two. Chapter two proposes a dynamic model of health and education choice allowing for unobserved heterogeneity. This model is then estimated using data from the Health and Retirement Study. Results from the estimation suggest that an exogenous increase in education only increases expectation of life by approximately one third of the amount typically reported in the current literature. Chapter three performs various robustness checks on the structural estimation described in chapter one. One test is to simulate data from the estimated model and perform statistical analysis analogous to what the current literature would perform on real data. This analysis shows that while the model predicts a far less mortality decline from a year of school, simulated data still generates the same results as real data when put to the standard analysis. The other major test performed in this chapter is to vary the amount of unobserved heterogeneity allowed in the model to see how sensitive it is to this number. Chapter four performs various policy experiments using the model estimated in chapter one. It examines the cost and benefits of compulsory education policy, college cost policy, and smoking policies. While all of these policies have various effects, this chapter only looks at the health benefits of these levers and compares the various cost benefit ratios. Ultimately the conclusion is that even with the reduced benefit of education compared to the current literature, education investment as a health policy is most likely effective. However, policies that target health decisions, especially smoking are superior in terms of cost benefit ratios.