Reducing costs to increase school participation

Where school fees do exist, eliminating them can lead to large increases in participation. Most countries now provide free primary education, but in low-income countries, annual public secondary school fees can cost as much as one-third of an average family’s yearly income. A program in Ghana offered full scholarships to academically qualified students who did not immediately enroll in secondary school. Roughly 80 percent of these students enrolled in secondary school after receiving the scholarship compared to only 20 percent enrollment in the comparison group at the beginning of the first academic year. Eight years on, the majority of the scholarship winners had completed senior high school [2].

Conditional cash transfers (CCTs) consistently increase school enrollment and attendance, but are expensive. Results from eighteen high-quality randomized evaluations of CCTs in twelve countries all found positive impacts on school participation.Researchers conducted randomized evaluations of CCTs in Burkina Faso [3], Cambodia [4], China [5] [6] Colombia [7], Ecuador [8], Honduras [9][10], Malawi [11], Mexico [12][13] [14], Morocco [15], Nepal [16], Nicaragua [17][18][19], and Tanzania [20]. Given the expense of CCT programs, they should primarily be viewed as social assistance programs that also increase attendance, rather than the most efficient solutions to increase school participation. Furter discussion of CCT cost-effectiveness can be found in our policy bulletin.

Small changes in the timing of a CCT can affect the ability of families to save and pay for school and can affect school enrollment decisions. The impact of cash transfer programs on education is sensitive to the timing of support because matching the timing of transfers to when large education expenditure takes place makes it easier for families to save the transfers for education expenditure. A CCT program in Colombia [7] included a transfer payout schedule to provide a larger lump-sum payment when re-enrollment fees were due. Compared to a traditional CCT program, the timed transfers reduced drop out and increased enrollment in tertiary schools. Another CCT evaluation in Colombian secondary schools by the same researchers found that providing “graduation bonuses” around the time of enrollment in tertiary education greatly increased subsequent enrollment compared to a traditional CCT program.

Even small incentives, or removing small costs, can have large impacts. If the sole policy objective is to increase enrollment and attendance at school, smaller incentives can be just as effective as the large payments common in CCTs. Smaller incentives have accordingly been more cost-effective at increasing attendance. Four evaluations on reducing small costs by providing free school uniforms or school meals in Kenya [21][22], Jamaica [23], Burkina Faso [24], and Uganda [25] found positive impacts on participation. An evaluation in Malawi [11] found that providing a considerably smaller cash transfer was just as effective and more cost-effective than a larger CCT for increasing participation.

Reducing costs by shortening travel time to school increases school enrollment. Many areas of the world with low school enrollment are remote or affected by conflict. In areas where few schools exist, using existing resources to create new local schools is a very effective way to increase enrollment and attendance. Two evaluations of programs that created local schools in Afghanistan [26] and Pakistan [27] found very large gains in enrollment. Reducing distance to school can be particularly helpful for girls, due to the restrictions on their mobility in these contexts. However, it is important to note that the Afghanistan school creation program was done through low-cost means using existing community resources, making it relatively cost-effective, as opposed to constructing new schools in low population areas, which is often very expensive. These findings are supported by rigorous, non-experimental studies in Burkina Faso [28], India [29], and Indonesia [30].

Reducing the effort cost of attending school by reducing child morbidity leads to large gains in school attendance. Conditions such as anemia and infection by parasitic worms can sap a child’s energy, making regular attendance in school more challenging. Two evaluations in India [31] and Kenya [32][33] found that, in areas where anemia or worm infections are prevalent, addressing these conditions with iron pills and school-based deworming increased school attendance.

The most cost-effective programs to increase student participation referenced above are those that addressed child morbidity (such as intestinal worms and chronic anemia) or reduced the distance to school through the creation of low-cost schools in areas where few schools exist. Some programs may be effective at increasing schooling but may also be expensive. Therefore, where authors have provided J-PAL with cost data, we compare the cost-effectiveness of the programs. The most cost-effective programs to increase student participation addressed health problems or reduced the distance to school by leveraging existing infrastructure to create schools in communities without schools. On average, CCTs are not as cost-effective as these approaches. However, when comparing cost-effectiveness, it is important to recognize that CCTs also provide benefits other than school attendance.

The cost effectiveness of various approaches can depend on local costs and contexts. We group evaluations by region in the graph below to reflect this.