Abstract
A recent series of papers attempting to estimate the fundamental or deep determinants of cross-country income levels finds that the quality of institutions 'trumps' all other possible deep determinants. This dissertation employs a similar methodology to this literature in order to estimate the fundamental determinants of health status, another important measure of economic development. Two-stage least squares is used to estimate a static, parsimonious, cross-country regression model developed to evaluate the relative importance of geographical characteristics, institutions, religion, and ethnic diversity as deep determinants of three health indicators: life expectancy, infant mortality and child mortality. Institutional structure is found to be the most important deep determinant of cross-country health status. Countries that have a strong system of law, adequate property rights and sufficient checks against government expropriation, in general, have higher life expectancies and lower rates of child and infant mortality. This result appears to be robust to including religion and ethnic diversity variables in the regressions, to changes in the proxy variables for institutional quality and geography, and to alterations in the set of instrumental variables.