Dynamic analysis of life-cycle consumption models under health uncertainty

Abstract

This dissertation consists of three studies. In the first two studies, I consider life-cycle consumption models under health uncertainty at the micro-level, while in the third study, I examine a demand function for medical care at the macro-level. In the first study, I examine the effects of health uncertainty on the precautionary behavior of household with a dynamic life-cycle consumption model. I derive testable equations for changes in consumption and medical expenditures. I also provide a measure of conditional health variance by combining the Markov health transition probabilities with the latent health scores. The empirical results using the Korean Household Panel Study show little evidence of precautionary motive to insure against future health risk. In the second study, I provide an overview of health capital models explaining the mechanism through which health uncertainty affects socioeconomic inequalities in health. The changes in the extent of income-related inequalities in self-assessed health, measured by the concentration index, are calculated among Korean adults from 1993 to 1996. The empirical results suggest that the main trends of inequalities in health may be due to the inequalities in income distribution over the sample period. In the third study, I employ the intratperiod first-order condition of a generalized health capital model. From this condition, I derive the cointegration relationship among the log of medical care expenditures, the log of consumption, and the log of relative price of medical care. The empirical results using aggregate U.S. data are consistent with the prediction of the model. While the cointegration test results are similar to those of previous studies using ad hoc demand functions, the microeconomic foundations for this model provide useful insight as well as a theoretical justification for these models.