Evaluating CTs’ impacts

Results from evaluations of Africa’s CT

Results from Evaluations of Cash Transfer Programs in Sub-Saharan Africa


Burundi UNHCR Cash Grants

  • Cash transfers encouraged refugees’ return to Burundi.
  • More than 50% of returnees used cash to invest in agriculture or construction.
  • Some refugees used cash for investment in assets, transportation costs, and the purchase of food or medicine.
  • Some refugees invested cash in businesses.
Source: UNHCR n.d.

Ethiopia Productive Safety Net Program

  • Most of the transfers went to purchase necessary food. Households also used cash to invest in education, agriculture, debt repayment, health care, and small businesses.
  • Increased households’ months of food security and livestock ownership.
  • Program improved households’ self-reported welfare.
  • Many of the positive effects found in the impact evaluation were severely dampened in households that received low or irregular transfers. Results also were muted if households did not have access to other food security programs.

Malawi Social Cash Transfer

  • After six months of transfers:
  • Beneficiary households went less than one day per month without enough food, as compared to five days for nonbeneficiary households.
  • Asset ownership in beneficiary households increased more than that of nonbeneficiary households.
  • Beneficiary households’ food consumption and diversity improved more than that of nonbeneficiary households.
  • Beneficiary households reported improvements in 62.7% of their children’s health in the past 6 months. (The corresponding amount for nonbeneficiary households was 5.5%.) The corresponding percentage for beneficiary household heads was 71%; the amount for nonbeneficiary household heads was 5.5%.
  • School enrollment in nonbeneficiary households declined, whereas it did not decline in beneficiary households.

Malawi Zomba Cash Transfer

  • After two years of transfers:
  • Improvement in school enrollment in the group receiving unconditional cash transfers was less than half (43%) of the improvement for the conditional cash transfer group. School attendance was also higher for conditional cash transfer than for unconditional cash transfer beneficiaries.
  • The conditional cash transfer improved English test scores and cognitive ability and marginally increased Trends in International Mathematics and Science Study math scores. The improvement in conditional cash transfer English scores was statistically above that of the improvement in English scores for unconditional cash transfer beneficiaries.
  • The unconditional cash transfer significantly decreased the probability that girls would become pregnant or marry. The effect was primarily due to the unconditional cash transfer’s impact on adolescent girls that dropped out of school. No effects of the unconditional cash transfer or conditional cash transfer were found on girls who remained students.

Mozambique Food Subsidy Program

  • Program increased the proportion of expenditures on food by 22%. Positive effects on food share, as well as the probability of consuming flour, were even larger in female-headed households.
  • Program increased the likelihood that women ate additional meals daily, and it marginally increased daily additional meals for boys age 5 to 9.
  • Adults in beneficiary households increased their probability of working (17% for male adults and the elderly and 24% for adult women, although this was only marginally significant), whereas boys age 5 to 9 were 29% less likely to work.
  • The number of hours that adults in beneficiary households spent in their own fields decreased, indicating additional labor time was being spent outside the household.
  • One indicator showed that children’s acute malnutrition (wasting) decreased by 30%, but the study concluded this result may have been an anomaly given no other results supported such strong nutritional impacts.
Child Support

South Africa Child Support Grant

  • Using instrumental variables regressions, an evaluation found that the grant decreased poverty and child hunger and increased food consumption and school attendance.
  • Using exogenous variations in grant eligibility or duration, studies have determined that the grant improved school attendance, labor force participation, self-reported measures of children’s hunger, and children’s height-for-age reports.
  • Using propensity score matching and difference-in-differences, an evaluation found that the grant decreased children’s hunger by 4% to 7% and increased school attendance by 6% to 8%. The grant also helped beneficiary households continue agricultural activities and increased mobile phone use among beneficiary households. The grant was not found to impact children’s labor activities, household employment, or likelihood of using a social worker.
Old Age

South Africa Old Age Pension

  • Most evaluations rely on extension of pension benefits to all South Africans.
  • One study suggested the pension affected household composition (notably, by adding children under age 5 and young women of child-bearing age), whereas older working-age women departed. Other research suggests no effect of the pension on household composition.
  • Female pension receipt was associated with increases in girls’ nutritional outcomes, but not boys’ nutritional outcomes.
  • Pension eligibility of a male in South African households was associated with increased school attendance and decreased market labor among children over age 5.
  • Impacts of the pension on households with orphans were mixed.
  • There were mixed results in studies of the impact of the pension on labor supply. Some studies found pension receipt was associated with lower labor supply in certain household adults; other studies found that pensions were associated with increased adult labor supply, often through migration; some studies suggest the pension has not impacted labor supply and migration.

Tanzania Rewarding STI Prevention and Control in Tanzania (RESPECT)

  • After two years of transfers:
  • Treatment group receiving the larger cash transfer size showed a 25% reduction in sexually transmitted infection prevalence.
  • No impact was found among those receiving the smaller transfers.

Zambia Chipata Social Cash Transfer, Zambia Kalomo Social Cash Transfer, Zambia Kazungula Social Cash Transfer

  • Results are from evaluation of Chipata, Kalomo, and Kazungula cash transfers using retrospective data as well as propensity-score-matching and odds-weighted regressions.
  • Positive effects of transfers were found on consumption, especially nonfood consumption.
  • Greatest impacts were seen in areas with the highest vulnerability (for example, consumption effects were 150% higher in the poorest district, Kazungula, than in the other districts).
  • Wealthier households appeared to be able to use the transfers to increase household assets, whereas households with fewer beginning assets were not able to do so.

The programs’ impacts on education were mixed. School attendance increased across the board only for Chipata district, which imposed a soft conditionality of school attendance and an extra payment for school enrollment.

Notes: UNHCR = Office of the United Nations High Commissioner for Refugees. Results from Ethiopia’s Productive Safety Net Program (PSNP) describe the entire PSNP and not just the Direct Support (cash transfer) component.