Reducing poverty through cash transfers for African children

Conditional cash transfer programmes have had some success in reducing poverty in developing countries, notably in Latin America. In sub-Saharan Africa, where poverty levels are high and government resources limited, cash transfers must have significant impact at low cost. Selecting appropriate target groups for cash transfers is key to achieving this.

A paper from the United Nations Development Programme(UNDP) assesses the potential impact of cash transfer programmes in 15sub-Saharan countries. Such programmes, if implemented, would provide cashtransfers to poor families, possibly under the condition that the cash be usedto access state health or schooling facilities (conditionality). The paperlooks at the effect of a simple cash transfer on current poverty levels, andthe impact of this extra money on school attendance under different targetingcriteria and transfer levels.

In the 15 countries studied, almost 46 percent (29 million) of childrenaged five to ten do not attend school: they are on average 17.4 percent poorerthan the national average. The effectiveness of cash transfer programmes forthese children depends on appropriate target group selection and monitoringlevels, but this also increases administrative costs.Inappropriate or non-existent targeting may reduce administrative costs butbenefit non-poor households, weakening programme effectiveness.

The authors therefore seek to determine thetarget groups and transfer levels best suited to poverty reduction insub-Saharan Africa. Key findings include:

• Transferring 0.5 percent ofGross Domestic Product (GDP) to school-age children would have a small impacton reducing the actual number of people living in poverty, but a high impact onimproving the situation of extremely poor people.
• Targeting children from poorhouseholds leads to much greater poverty reduction, as the transfers receivedare likely to be higher under targeted programmes than under universal schemes.
• Poverty reduction is greaterif the transfer targets only rural children rather than all children: an optionif the administrative costs of targeting poor children are too high.
• Progressive cash transfers(increasing with age) are effective for poor children, though there is littledifference between progressive and equal transfers when all children aretargeted.
• Transfers proportional tothe average national poverty line (targeting 20, 30, or 40 percent of thoseliving in poverty) have a far greater impact on national poverty reduction thantransferring 0.5 percent of GDP.
• Proportional transfers maynot be affordable for most countries in Africa: targeting 40 percent of thoseunder the poverty line in Ethiopia would cost 8.31 percent of its GDP.

The authors also find that a cash transferprogramme without conditionality is not enough to achieve a substantialincrease in school attendance. Bearing in mind the data limitations of thestudy, the authors recommend the following:

• sizeable cash transfers(between 2-8 percent of GDP)
• combining cash transferswith conditionality, along with an improvement in school quality
• a broad targeting design (for example, by geographical regions): thiscould be sufficient given the widespread poverty in the countries studied, andwould avoid the administrative costs of detailed targeting (for example, byincome).