Ensaios sobre a relação entre estado de saúde e mercado de trabalho

Abstract

This thesis consists of three essays that deal with the relationship between individual health status and the job market. The first essay examines the effect of individual health status on the decision to participate in the multiple job market. The data used are from the National Health Survey (PNS) 2013 and the empirical strategy makes use of probability models, which seek to correct bias in sample selection and endogeneity. About 3.78% of the sample of employed workers have more than one job. The direction of the results varies according to the health measure, when self-reporting of poor health is used, a lower probability of having more than one job is observed, since the measures associated with chronic diseases were not relevant, in most cases. In some cases, having a poor state of health is associated with a greater probability of participating in this market, a result that may result from the income effect, in which the agent seeks additional sources of income to take care of health problems. The second essay aims to measure the income differential between obese and non-obese for Brazil and understand which components explain this difference. The data used come from the National Health Survey (PNS) 2013. The decomposition method is based on unconditional quantile regressions. The results confirm the existence of the wage gap between obese and non-obese men and women. In the case of men, the difference was favorable to the obese in all quantiles and in the case of women, favorable to the non-obese. The biggest differences were observed at the top of the distribution. This difference is mostly explained by the composition effect, which indicates that they are observable characteristics that cause the wage gap between groups. The wage structure effect, which may have elements associated with discrimination in its composition, was not relevant in most quantiles, thus, there are other factors that were not captured by the method and that would explain part of the difference. Finally, the third essay examine the relationship between health status and retirement decisions. The data come from the Longitudinal Study of Health of the Elderly Brazilians (ELSI-Brasil). The tobit and probit models are the basis for the analyzes and the health index is built from an ordered probit model. The results show the importance of health status as a determinant of the choices of older and older adults. Having good health implies a positive relationship with the expected retirement age, indicating that healthy people intend to retire later. With the likelihood of being retired, the results indicate that individuals with good health are less likely to be retired.