Targeting Remittances to Boost Students’ Education Spending

Remittances—money sent by immigrants to their home countries—constitute a substantial international cash flow. According to the World Bank, remittance payments topped $400 billion in 2012 and are expected to reach $534 billion by 2015. As a result, policymakers are looking toward remittances as a potential development instrument.

Researchers at the Abdul Latif Jameel Poverty Action Lab (J-PAL) recently partnered with the Salvadoran Fundación Empresarial para el Desarrollo Educativo (FEPADE) to subsidize targeted remittance payments from migrants to specific students living in their home country. In their paper, “Subsidizing Remittances for Education: A Field Experiment Among Migrants from El Salvador,” Kate Ambler, Diego Aycinena, and Dean Yang report the results of a randomized control trial (RCT) regarding the effects of remittances and FEPADE matching funds on recipient students’ educational performance. They find that subsidized remittance payments increase investment in recipient students’ education and create spillover effects for other school-age children in recipient students’ households.

In collaboration with FEPADE, the authors created EduRemesa, a program that allows Salvadoran migrants to send remittance payments directly to a student of their choice to subsidize their education. From November 2011 to February 2012, they approached Salvadoran migrants at two Salvadoran Consulate locations in Washington, DC, inviting them to participate in EduRemesa. To avoid any spillover effects, the authors divided interested migrants into treatment and control groups both by date and location. Participants in both groups identified a Salvadoran student they wished to support by entering their information into a lottery. Subsequent randomization allowed the authors to measure any potential effects of EduRemesa by comparing students from the treatment and control groups.

The treatment group included three sub-categories: the first gained access to EduRemesa and received information about the benefits of investing in education; the second group received program access, information, and 1:1 matching funds from FEPADE; the third group received access, information, and 3:1 matching funds from FEPADE. A fourth group, serving as the control, did not gain access to EduRemesa, but still received information about the benefits of educational investment. The total sample size included 734 migrants along with 734 student beneficiaries.

Each student beneficiary received an ATM card to access his or her EduRemesa funds. While these funds were designated for educational expenses, FEPADE did not monitor or enforce students’ spending. To measure the effects of EduRemesa, the authors conducted follow-up surveys between July and October 2012 with students in the treatment and control groups, as well as with an adult in each student’s household.

Ambler, Aycinena, and Yang find that migrants in the treatment group who were not offered matching funds did not participate in EduRemesa. Among migrants who received matching funds from FEPADE, EduRemesa had a positive effect on recipient students’ educational spending, according to their responses to follow-up surveys. Students in the 3:1 matching group spent an average of $301 more annually on their education in comparison to students in the control group. Students in the 1:1 matching group spent $75 more on their education than those in the control group. Notably, the treatment effect was particularly large for girls. Within the 3:1 treatment group, girls’ educational spending increased by $509 annually, or 36 percent more than the control group. The authors assert this increase in education spending is due to the subsidized remittances, as well as increased household investment in children’s education. As a result, the subsidized remittances to recipient students also created positive spillover effects for other school-age children in the household.

Ambler, Aycinena, and Yang acknowledge that migrants’ participation in EduRemesa was relatively low—only 19 percent of migrants who were offered the 3:1 matching funds treatment participated in the program, and take-up rates for the 1:1 matching funds treatment group were only 7 percent. Moreover, it is possible that their findings may be subject to selection bias, given that students had to travel to San Salvador to enroll in the EduRemesa program.

Still, the authors believe their findings illustrate a potential avenue for directing remittances toward development programs in immigrants’ home countries. For instance, public education initiatives in El Salvador, such as conditional-cash transfers, currently focus on primary school enrollment, while secondary enrollment remains quite low. Subsidized remittances may be an effective way to promote continued school enrollment by bolstering investment in education at the household level.

Photo cc: MCAgoldsevenCam