Theses Doctoral

Essays on the Economics of Education in Ghana

Awadey, Amanda Aku Ahornam

This dissertation consists of three essays examining the impact of large-scale education policies changing high school duration in a sub-Saharan Africa context, specifically Ghana. These policies vary critical factors, namely instruction time and age at test, competition, and incentives to enroll in high school, which have implications for educational and labor market outcomes. I exploit these exogenous time series variations created to study impacts on learning, college application decisions, educational attainment, and labor market outcomes. A common thread across chapters is the examination of differential impact by gender or socio-economic vulnerability.

In Chapter 1, I interrogate a nationwide reform in Ghana that reduced curricular pace by extending high school duration by a year without changing the breadth of the curriculum and the grade structure at lower levels of schooling. Maintaining the curriculum while increasing duration was aimed at reducing the curricular pace to one more suited to students' learning pace. This is key because, despite enormous strides in increasing school enrollment in developing countries, widespread low and stagnant learning outcomes remain a concern for policymakers and an active line of inquiry for researchers.

Exploiting time-series variation from this quasi-random experiment, I implement a cohort analysis and a regression discontinuity design that leverages a compulsory school start law to examine the impact on learning at the end of high school, collegiate attainment, and labor market outcomes. I find significant positive effects on learning, with females benefiting more. Varying impact sizes across baseline ability levels suggest an extreme curricular gap in this setting. This reform has some positive impact on the decision to apply to college. Furthermore, young individuals who gain from increasing the depth of human capital but do not further their education beyond high school are more likely to be engaged in paid employment outside their household, although males drive these gains.

I continue to investigate this setting in Chapter 2 by examining another nationwide reform that reduced high school duration by a year, three years after it was increased. This removal resulted in two cohorts graduating high school at the same time - the last before the removal of the extra year and the first after the removal. An immediate implication is a sharp increase in the number of students who graduated high school and could apply to university in a given year, exogenously increasing graduating cohort size. A larger high school graduating cohort size may affect human capital formation by impacting accumulated knowledge at the end of high school and collegiate attainment. Fewer high school resources per student and changes in student effort are possible mechanisms through which this cohort size can affect knowledge accumulation. On the college education market, a larger graduating cohort may signal a fall in admission probability through increased college demand if there is no expected commensurate increase in supply. In response, students may change their application strategy on the extensive and intensive margin depending on their revised admission probability estimate.

I test these hypotheses by combining a cohort analysis and a regression discontinuity design leveraging a compulsory schooling law to isolate causal effects. Focusing on students with the extra year, I find a notable fall in end-of-high school performance for students who faced a larger cohort. These individuals are less likely to obtain a college degree. First, they are less likely to apply immediately after graduation, which persists for two years, suggesting many forgo applications altogether. Even when they do, universities face supply constraints and cannot absorb the increased demand. Second, they reduce the selectivity of their applications' field choices. Analogous results hold for females and economically vulnerable students.

In Chapter 3, I assess a possible tradeoff between increased years of schooling and the likelihood of graduating high school in a setting where high school is neither free nor compulsory from the nationwide policy considered in Chapter 1 that increased high school duration by a year. This policy creates an exogenous shock to high school duration for the universe of middle school graduates at this transition point. Bunching in the number of students who complete the highest grade of a schooling level and then drop out is a common phenomenon. It suggests factors that change incentives to enroll at the next level may affect their continuation decisions.

Using a multiple linear regression model with fixed effects to estimate the impact from this exogenous shock, I find that those who expect to be affected by the policy spend more years in pre-tertiary schooling, but this comes at a statistical and economically significant cost to graduating. This finding aligns with policymakers' concerns that some students would be precluded from obtaining a high school education. It is worth noting that sub-groups in the population who may be disadvantaged - females, students from low socio-economic backgrounds, and lower ability students are not more likely to bear the brunt. If anything, they have significantly lower costs.

Geographic Areas


This item is currently under embargo. It will be available starting 2025-04-18.

More About This Work

Academic Units
Thesis Advisors
Pop-Eleches, Cristian
Urquiola, Miguel S.
Ph.D., Columbia University
Published Here
April 19, 2023