The Impact of Zambia's Unconditional Child Grant on Schooling and Work: Results from a Large-scale Social Experiment

"Since the mid 1990s, and following the successful implementation of large scale programs in Brazil, Mexico and South Africa, cash transfers have become an important part of the poverty alleviation toolkit in developing countries..."

In Africa, the social protection movement has been primarily driven by poverty alleviation within a human rights framework, which, along with the relatively low supply of services such as health and schooling, has led to unconditional cash transfers (UCTs) as the standard approach in this sector. This paper reports on the impact on child schooling and work of the Government of Zambia's Child Grant Program (CGP), a UCT programme for households with children under age 3 years in 3 remote districts of the country. The study looks to see if the programme has an impact on the schooling and work of school-age children who in principle are not the main focus population of the programme.

CGP households receive a flat (i.e., irrespective of household size) 70 kwacha (KW) a month (equivalent to US$12) transfer, an amount deemed sufficient to purchase one meal a day for everyone in the household for one month. However, the unconditional nature of the programme allows households the freedom to spend the transfer as they wish and expands the potential range of outcomes the programme may affect, including schooling of "non-target" children (age 7-14).

The CGP impact evaluation is a multisite cluster randomised controlled trial (RCT). Within each of the 3 geographically targeted districts, selected based on their high poverty and child malnutrition rates (Kaputa in Northern Province, and Kalabo and Shangombo in Western Province), communities designated by Community Welfare Assistance Committees (CWACs) were randomly assigned to either the treatment condition to start the programme in December 2010 or to the delayed control condition to start the programme at the end of 2013. There are 2,519 households and 14,565 people in the evaluation study, including 4,793 children ages 5 and under.

The study finds that the CGP has no discernible impact on school enrollment of children age 7-14. However, when breaking the sample by older (11-14) and younger (7-10) children - based on the grade structure of the Zambian schooling system - it finds a significant impact among children age 11-14, which coincides with the exact age range where sharp dropout begins to occur in Zambia, with point estimates in the range of 7-8 percentage points.

The study also shows a significant impact of the CGP in reducing the probability of paid work by 4-5 percentage points. For older children, a key barrier across the developing world is the time-cost of school attendance in terms of income foregone. The CGP's reduction of work-for-pay among children 11-14 suggests that alleviating the opportunity cost of schooling is another pathway through which the programme facilitates school enrollment.

Per the article, the positive impacts of the CGP on older, secondary age children are consistent with previous literature on conditional cash transfers that finds larger impacts at older age groups.

In reflecting on the findings, the researchers note that financial barriers to schooling remain important in Zambia, even at primary levels, despite the elimination of formal school fees. Financial costs include formal or informal school fees or other costs such as transportation, uniforms and shoes, or school equipment (stationary), and these costs tend to be higher for secondary school age children. Barriers to school access tend to affect girls disproportionately. The provision of a UCT has the potential to alleviate some of these barriers. In fact, results indicate that school expenditures are 48% higher for treated children than for their control peers after 2 years among children 7-14 enrolled; even though lower, the difference is still large and significant (at the 10 percent level) at 36 months (29%).

In conclusion, "the CGP's objectives are focused on very young children, and so schooling effects, particularly among older children, would be considered somewhat secondary level or 'spillover' effects facilitated in part by the unconditional nature of the program; the potential for such wider impacts in programs that are not tightly conditioned may not be fully appreciated" in the literature - until now.