School Fees, Seasonality, and Under-investment in Human Capital

Policy Context

Although 87% of children globally complete primary school, only 63% do so in sub-Saharan Africa (United Nations, 2023). In Uganda, the problem is particularly acute: between 50% and 70% of children drop out before completing primary school (Etukuri and Kwesiga, 2023). This high dropout rate persists despite Uganda’s Universal Primary Education policy, which eliminated formal tuition fees in public schools. In practice, schools continue to charge informal dues and other indirect costs that remain a major barrier for low-income families. National survey data consistently cite school expenses as the main reason for non-enrollment and dropout (Uganda National Panel Survey, 2019-2020). Yet pilot evidence from Eastern Uganda reveals a puzzling pattern: when offered a choice between a direct school-fee subsidy and an equivalent – or even larger – unconditional cash transfer, the vast majority of caregivers (87%) preferred the subsidy, with two-thirds maintaining this preference even when offered a 50% cash premium. These results suggest that financial constraints alone do not explain low enrollment; instead, caregivers may value the commitment that subsidies provide to ensure educational investments. This study seeks to provide the first experimental evidence on caregivers’ willingness to pay for commitment in education.

Study Design

This randomized controlled trial includes 2,000 students entering Primary 5 and Primary 7 across 30 randomly selected government schools in Jinja District, Uganda. Within each school, students are randomly assigned to one of three groups: a control group (no intervention), a subsidy group (T1), and a cash group (T2). T1 students receive a school-fee subsidy paid directly to the school at the start of the academic year, while T2 caregivers receive an unconditional cash transfer of equivalent value, delivered at the same time. This design tests whether school fees act as a binding constraint on enrollment, and whether the form of financial support – cash versus subsidy – affects school participation. Before treatment assignment, each child’s primary caregiver completes an incentive-compatible multiple price listing (MPL) to reveal their preference between a subsidy and varying amounts of cash. After the survey, caregivers are randomly assigned to one of the treatment groups (with a small subset receiving the outcome tied to a randomly selected MPL choice). This cross-randomization allows us to test whether caregivers who express a preference for commitment – by choosing a subsidy – experience larger impacts when they receive the subsidy, compared to when they receive cash instead.

The primary outcomes include school participation (enrollment and attendance) measured using administrative data, school records, survey responses, and unannounced enumerator spot checks. Academic performance is assessed through school test scores and standardized assessments administered by the study team. Importantly, because this experiment induces an exogenous shift in educational attainment based on the random assignment of treatment and control, the design allows for causal identification of the returns to education. The study is conducted in close partnership with Uganda’s Ministry of Education and Sports and is designed to directly inform national education policy.

Results and Policy Lessons

Preliminary results from the pilot and the ongoing main study indicate a strong and persistent demand for commitment in educational investments. Across both samples, a large majority of caregivers prefer a direct school-fee subsidy over an equivalent cash transfer delivered at the same time. Nearly half of caregivers are willing to forgo a substantially larger cash amount in order to receive a subsidy that is paid directly to the school, suggesting that commitment value plays an important role beyond pure liquidity.

Caregivers also display a strong preference for delayed liquidity that aligns with the school-fee deadline. Many prefer receiving cash at the start of the school term rather than immediately, and a sizable share are willing to accept a smaller transfer if it is timed to coincide with fee payments. These patterns indicate that concerns about self-control, diversion of funds, or competing household demands may shape financial preferences around schooling.

While outcome data on enrollment, attendance, and learning are still being collected, early evidence from the pilot suggests that even modest school-fee subsidies generate meaningful short-run increases in school participation. Ongoing analysis will assess whether caregivers’ stated preferences for commitment and timing translate into improved educational outcomes, and whether aligning program design with these preferences can reduce underinvestment in children’s education.