Key takeaways

• A study in Ethiopia found conditional cash transfers increased secondary school enrollment. Youth offered transfers were more likely to enroll and stay in school.
• Earlier notification delivered bigger impacts. Students informed more than a year in advance saw larger enrollment gains and higher exam pass rates.
• Benefits extended beyond schooling. The program reduced early marriage and improved household well-being, including increased school enrollment among younger siblings.

Conditional cash transfers (CCTs) are among the most widely studied and successful interventions targeting school enrollment and attendance in low- and middle-income countries. Cash payments defray the costs of enrollment and transportation that systematically discourage many families from sending children to school; they may also better enable families to meet their needs without labor income generated from children or adolescents, who can reallocate their time to school.

Though much of the literature on CCTs focuses on primary school, a robust body of evidence also suggests that they can effectively lower the financial barriers that prevent poor and rural youth from enrolling in secondary school. This in turn enables youth to accumulate skills that can be applied in the job market or in further education, improving their long-term economic trajectory.  

On World Youth Skills Day (July 15), it’s important to highlight not only strategies to support youth in building their skills in secondary school, but also the evidence gap around when it is best to intervene. Should a transfer be offered to students on the cusp of the transition to secondary school, when the enrollment decision is imminent? Or does notifying younger students earlier—well before that decision must be made—generate anticipatory effects, giving families time to plan, prepare, and invest in the steps required to continue their education?  

A recent randomized controlled trial conducted by IFPRI researchers in collaboration with World Vision seeks to address this question in the context of rural Ethiopia, a setting where secondary school enrollment remains persistently low. Results show the program is an effective and relatively low-cost way to increase enrollment, and has other positive impacts on students and their families.

In the two regions where we worked, Amhara and Oromia. primary schools outnumber secondary schools nearly ten to one, and more than 80% of rural households live in a community without a secondary school. For adolescents in extremely poor households, continuing past primary school often means moving away from home and paying for room, board, and transport—costs that are prohibitive for many families. Accordingly, around two-thirds of youth do not enroll in secondary school.

The trial included 2,005 youth from households participating in the Productive Safety Net Program (PSNP), Ethiopia’s flagship social protection program, and thus by definition targeted youth in the poorest 15% of rural households. This group faces substantial barriers to educational attainment: sampled youth were on average nearly two years behind their target grade for their age, and 40% reported that they did not know a single person who had completed secondary school.

Working through SPIR II, a U.S. government-funded consortium led by World Vision, we offered a cash transfer of $300 per year (for up to two years) to families whose children enrolled in secondary school. The transfer amount was benchmarked to fully cover the estimated costs of accommodation, transport, and other expenses. The offer was randomized across 116 subdistricts, none of which had a secondary school within their borders, and we measured outcomes about two years later using both a follow-up survey (with an attrition rate of only 6%) and enrollment records compiled from schools.

Crucially, the sample included three distinct cohorts in order to allow us to explore the effects of transfer timing:

Eighth graders were informed of the transfer roughly three months before the primary school leaving exam that determines eligibility for secondary school.

Seventh graders were enrolled of the trial and informed of the availability of the transfer for future secondary school enrollment more than a year before their potential matriculation.

The third subsample, dropouts, had already passed the primary school leaving exam (within the previous two years), but had dropped out without continuing into secondary school.

Positive impacts on enrollment

Our findings suggest that the transfer had a substantial effect on enrollment. Pooling across cohorts, youth offered the transfer were 21 percentage points more likely to ever enroll in secondary school (relative to a control-group mean of 43%) and 15 percentage points more likely to be enrolled at follow-up (relative to a mean of 24%); estimates using school records are, if anything, somewhat larger. The transfer also generated a small but meaningful decline in the probability of marriage before age 18—slightly under two percentage points, relative to a control mean of 6%—an effect observed for both girls and boys.

The most striking pattern, however, concerns timing. The largest absolute effects on enrollment were observed for seventh graders—precisely the group given the longest advance notice. Their probability of current enrollment rose by 21 percentage points, an 80% increase relative to the control-group mean of 26%, meaningfully larger than the effect for eighth graders.

Notably, seventh graders offered the transfer were also more likely to pass the primary school leaving exam (a 13 percentage point increase, relative to a mean of 45%), a margin on which eighth graders—informed only shortly before the exam—showed essentially no response. This pattern is consistent with the hypothesis that early notification allows families and students to plan and invest in the run-up to the transition, including preparation for the exam itself, amplifying the transfer’s ultimate effect on enrollment.

The case of dropouts

By contrast, prior dropouts showed a very large initial response—a 41 percentage point increase in the probability of ever re-enrolling—but the majority dropped out again, leaving a much smaller nine percentage point increase in enrollment at follow-up.

To understand why, we used a machine learning method (a generalized random forest) to examine which baseline characteristics best predict where the transfer worked. The analysis suggests that the weaker effects for dropouts are driven not by dropout status itself but by other observable characteristics correlated with it—in particular, lower prior academic performance, lower socioeconomic status, and older age. Financial constraints, which dropouts themselves most often cited as their reason for leaving school, appear to be only part of the story.

Broader impacts

We also observed positive effects that extended beyond the targeted youth. Households receiving the transfer reported significantly lower food insecurity and higher asset holdings, and the number of younger siblings enrolled in school rose by roughly 14%. One plausible explanation is that families were able to economize on schooling costs: the transfer-induced increase in educational spending was considerably smaller than the transfer itself, leaving resources to meet other household needs.

Finally, we assessed cost-effectiveness relative to comparable transfers in the literature. At an estimated cost of $3,258 (in 2021 purchasing power parity or PPP terms) per additional year of secondary schooling induced, this program ranks among the more cost-effective secondary-school transfers evaluated to date, broadly comparable to recent trials in Niger and Malawi.

Taken together, our findings suggest that conditional cash transfers can meaningfully expand secondary school enrollment and reduce early marriage among some of the poorest rural households. Crucially, timing matters: notifying younger students well before the transition to secondary school appears to unlock an important channel of anticipatory planning. For program designers weighing how to allocate scarce resources, this suggests that reaching students earlier—rather than waiting until the enrollment decision is imminent—may be a low-cost way to increase the return on a given transfer.

Jessica Leight is a Senior Research Fellow with IFPRI’s Poverty, Gender, and Inclusion (PGI) Unit. This post is based on research that is not yet peer-reviewed. Opinions are the author’s.

This work was funded by World Vision under the SPIR II Program, funded by the U.S. Government, with additional support from the CGIAR Science Program on Policy Innovations.

Reference:
Leight, Jessica; Gilligan, Daniel O.; Mulford, Michael; and Zafar, Sarim. 2026. The early bird gets the cash: Early notification and conditional cash transfers for secondary school. IFPRI Discussion Paper 2421. Washington, DC: International Food Policy Research Institute. https://hdl.handle.net/10568/183379