THE RELATIONSHIP BETWEEN SCHOOLING AND HEALTH: INVESTIGATION INTO AN OCCUPATIONAL LINKAGE (HEALTH CARE, MEDICAL ECONOMICS, INJURIES, ILLNESSES, CHOICE)

This dissertation's main objective is to determine if, how, and the extent to which occupational factors may account for the positive correlation which is empirically observed between schooling and health. To this end a theoretical model is developed and estimated which allows schooling to affect health directly and indirectly, through choice of work environment and other market inputs in health production. The model describes individuals' behavior in the labor market. It marries the health economics literature with labor economic studies of occupational choice, by allowing individuals to trade health for higher wage income associated with more hazardous jobs. Thus, the model assumes that individuals determine their health and occupation with reference to each other and not independently as the bulk of the health economics literature implies.; The empirical analysis is based on data from the 1980 Health Interview Survey, combined with occupational information from the Dictionary of Occupational Titles. Demand functions for job characteristics and other inputs in health production are estimated, in addition to the health production function. These functions are then used to decompose the total effect of schooling on health into direct and indirect effects. A decomposition analysis is also performed for the total effects of sex and race on health. Based on the OLS estimates of the health production function there is evidence of a small occupation effect in the schooling-health relationship, but this effect is conservatively estimated and biased downward given the presence of unobserved differences in individuals' initial health status. A two stage estimation procedure is unsuccessfully tried to obtain consistent estimates. OLS estimates indicate a substantial occupation effect in the sex-health relationship and a small occupation effect in the race-health relationship. In addition, it is found that the effect of occupational hazards on health is not constant, but varies with the length of job tenure. Due to data problems the study is inconclusive with respect to the validity of the theoretical model. The general findings highlight the importance of investigating the health effects of occupational hazards using longitudinal data and comprehensive measures of hazards which include stress at work.