Human Capital and Institutions in Health: Evidence on Outcomes and Behaviours

Abstract

Health outcomes and health-related behaviours vary markedly across and within countries globally. Some populations achieve long, healthy lives, while others continue to face enduring disadvantages in health, development, and survival. This dissertation contributes to the empirical evidence on the effects of human capital and institutions on health outcomes and behaviours. It examines how historical and contemporary dimensions of human capital and institutional development influence population health across sub-Saharan Africa and beyond. The first study investigates whether elite numeracy during the seventeenth to nineteenth centuries continues to affect contemporary health outcomes. Using subnational data from 44 sub-Saharan African countries and an instrumental variable approach based on proximity to historical centres, the analysis shows that regions with higher historical elite numeracy show higher life expectancy and lower malaria prevalence today. These effects persist after controlling for historical, geographic, and contemporary factors, highlighting the path-dependent influence of early human capital on modern health outcomes. The second examines how contemporary individual-level health literacy and health numeracy relate to malaria prevention and treatment behaviours in Gabon. Drawing on a cross-sectional household survey and validated psychometric scales, the results show that health literacy and numeracy are associated with behaviours. Numeracy supports treatment-seeking decisions, while literacy promotes correct dosing and engagement with healthcare providers. Mediation analysis demonstrates that literacy influences behaviours indirectly through information-seeking, emphasising the role of knowledge in shaping health actions. The third study analyses the role of institutional quality and development aid in shaping the malaria burden across 38 sub-Saharan African countries (2010 to 2022). Using spatial econometric methods, the study finds that stronger institutions, higher health worker density, and increased aid are associated with higher reported malaria burden, likely reflecting improved detection and reporting. Malaria outcomes are driven more by local conditions, with limited spillovers revealing a need for localised health system strengthening alongside coordinated regional interventions. The fourth study evaluates the effect of compulsory education laws on health inequality, using height inequality as a biological marker of nutritional and developmental disparities across 62 countries from 1810 to 2000. Employing a staggered difference-in-differences design, the results indicate that compulsory education reduced height inequality on average. The findings suggest that education policies can meaningfully narrow health disparities over time, but their success hinges on both the institutional environment and the timing of adoption.