Abstract:
Human capital is considered critical for economic development in Sub-Saharan Africa. While inputs, such as years of schooling, have increased substantially in Sub-Saharan Africa, the output, such as literacy and numeracy, have improved much less during the post-colonial period. Hence, this thesis traces the development and determinants of human capital in Sub-Saharan Africa to contribute to our understanding of this development.
The first study of the dissertation estimates numeracy in Sub-Saharan Africa at the subnational level for the birth cohorts between 1950 and 1990. By combining data from censuses and household surveys, I estimate the ABCC Index to measure numerical skills. Additionally, I carefully evaluate the suitability of the index and the data to capture basic numeracy. The results show mostly stagnating numeracy, alt-hough in some countries it increased. Moreover, there are strong regional differences, with Southern Africa performing best and Western Africa worst.
The second study focuses on how inputs into the education system are transformed efficiently into out-puts focusing on the role of children’s health. I employ average female height as the proxy for nutrition during childhood. To proxy schooling efficiency, I take the ratio of numeracy (ABCC Index) and years of schooling. The data is at the subnational level and covering the birth decades 1950 to 1999. For causal interpretation I use negative rainfall shocks during early childhood as an instrumental variable. I find that better nutrition increases education efficiency.
The third study investigates the relationship between child labour and educational outputs in Sub-Saharan Africa. I particularly focus on the lowest income groups for whom child labour might be neces-sary to afford education. Based on data from UNICEF’s Multiple Indicator Cluster Surveys, the findings indicate that, on average, child labour is negatively associated with schooling performance. However, for children from the poorest households, this relationship is reversed. This suggest that outright bans on child labour without accompanying compensation mechanisms may inadvertently cause more harm than good.
The last study focuses on how historical education systems contribute to the reproduction of elites in Sub-Saharan Africa, using South Africa as a case study. By spatially linking data from the Cape Colony Education Reports (1881-1905) to results from the 2018 Systemic Tests, I investigate whether the loca-tion or quality of historic segregated schools matter for current educational outcomes. I find no evidence that the mere presence of a historic school predicts present-day performance implying there are no broadly accessible benefits. However, when focusing on school quality, I find a robust positive relation-ship for historic white schools, but no consistent effect for black schools. I argue that this asymmetry reflects the legacy of systematic neglect of black education during the 20th century, in contrast to the institutional continuity afforded to white schools.
For policymakers, identifying effective strategies to improve educational outcomes can be challenging. This dissertation highlights several key areas where progress is possible. First, it is crucial to measure educational attainment accurately. Second, the findings show that targeted interventions, such as im-proving children’s nutrition, reducing financial barriers to schooling, and addressing the legacy of histor-ical inequalities, can meaningfully enhance educational outcomes.
