A Quantile Regression Approach to Panel Data Analysis of Health Care Expenditure in OECD Countries

This article investigates the variation in the effects of various determinants on the per capita health care expenditure. A total of 28 OECD countries are studied over the period 1990-2012, employing an instrumental variable quantile regression method for a dynamic panel model with fixed effects. The results show that the determinants of per capita health care expenditure do vary with the distribution of the health care expenditure growth, while the change patterns are dissimilar. Specifically, the lagged health spending growth has a significantly positive effect, with an effect that decreases towards the higher quantiles of growth of per capita health care expenditure. Per capita GDP has a significantly positive effect, both the short and long run income elasticities are smaller than one, and health care is a necessity. The density of physicians only has a significant negative effect at the lower tail of the distribution. The elderly population has the reverse effect at the lower and upper tails, and this shows an upward trend with the increase in health expenditure growth. Life expectancy has an effect similar to the proportion of the old. Variable representing Baumol's model of "unbalanced growth" theory has a significantly positive effect, and the change pattern of its influence shows a marked upward trend. However, one component of "Baumol variable", labor productivity, only shows significant effect in the low half of the distribution. More attention needs to be paid to the influence of determinants in health expenditure study.