Social cash transfers in Africa: conditional or unconditional?

The Latin American experience with conditional cash transfers has been generally successful, and there is some support for their introduction to low-income African countries. However, there is little evidence to suggest such programmes will work in Africa. Serious questions concerning the differences between the two continents need to be answered first.

A paper from Overseas Development Institute, UK,considers whether cash transfers to poor households in Africashould be conditional upon their accessing social services such as health andschooling facilities, as in Latin America. The authorargues that the differences between Africa and LatinAmerica, where conditional cash transfers (CCTs)have had success, mean that conditionality may be inappropriate in low-incomeAfrican countries.

In the 1990s, a number of Latin American countries,including Brazil,Mexico and Nicaragua,set up schemes providing (mostly) cash to poor households. In Africa,countries such as Mozambique,Zambia, Malawiand Ethiopiahave implemented similar schemes, but with a focus on achieving food securityfor the poorest households rather than their long-term development by, forexample, increasing school attendance. None of the African schemes haveintroduced conditionality.

Some agencies, including the World Bank, feel that Africancountries should also attach conditions to social cash transfer programmes. Byapplying conditions, programme designers try to influence the behaviour andattitudes of target households in a way that they consider suited to povertyreduction. Conditionality also tends to be more acceptable to policymakers andtaxpayers.

There are four types of concerns as to whether the LatinAmerican experience with CCTs can be applied tolow-income African countries:

• Service delivery agencies (such as health andeducation) may not be able to meet increased demand (in terms of quantity andquality): in Chipata,Zambia, the number ofchildren applying for school enrolment exceeds school capacity by 20 percent.
• Low-income African countries may not have theadministrative skills and resources to implement large-scale cash transferschemes further complicated by conditionality.
• Conditionality could impose administrative costshigher than its potential benefits.
• There may be socio-cultural, ethnic andpolitical differences between African countries and Latin Americathat would make conditionality inappropriate: donors may also fail tounderstand these differences.

In September 2006, the Zambian Social Welfare Department andCARE International started to study the costs and benefits of conditionality inChipata. Such research is essential to establishingwhether conditionality is suited to the African context.

The authors recommend the following to policymakersdesigning social cash transfer schemes for Africa:

• keepingthe administrative costs as low as possible, taking into account thebudget constraints of low-income African countries
• improvingthe capacity of social welfare services, and keeping the organisation fortransfer schemes as simple and as undemanding as possible through a cleardefinition of objectives and target groups, effective targeting andreliable delivery
• further research on the pros and cons of CCTs versus unconditional schemes under differentconditions, in partnership with national governments, donor agencies andsocial researchers.