Socioeconomic, institutional, environmental, and demographic determinants of population health in 90 developing countries: panel data evidence

Abstract

Anchored in Grossman’s health production framework, this study examines how socioeconomic, institutional, environmental, and demographic conditions are associated with population health in developing countries. Population health is measured through three complementary indicators—life expectancy, infant mortality, and maternal mortality—to capture broader survival conditions, early-life vulnerability, and maternal health risks. Using panel data for 90 developing countries from 2002 to 2018, the study estimates country fixed-effects models and uses Driscoll–Kraay standard errors to strengthen inference in the presence of heteroskedasticity, autocorrelation, and cross-sectional dependence. Additional robustness checks are conducted using Panel-Corrected Standard Errors, Feasible Generalized Least Squares, and winsorized specifications. The results show that human capital, government effectiveness, and population density are consistently associated with higher life expectancy and lower infant and maternal mortality, and that these coefficient signs remain stable across robustness checks. Economic misery is associated with lower life expectancy and higher mortality in the baseline models, although its significance weakens in some sensitivity tests. Government health expenditure and educational attainment are generally associated with more favorable health outcomes, but their estimates are less stable after winsorization. Ecological footprint yields mixed associations across the three health indicators, while political stability produces unstable and, in some cases, counterintuitive coefficients. Overall, the findings suggest that population health in developing countries is associated more consistently with human capability and institutional effectiveness than with health expenditure alone.