This post focus on the interesting findings of Ester Duflo, described in the published paper “Grandmothers and Granddaughters: Old-Age Pensions and Intrahousehold Allocation in South Africa”. In the beginning of the 1990s South Africa, the benefits and coverage of the national social pension programme were expanded to include the black population, reflecting the end of the apartheid era. This event created an interesting subject to explore how cash transfers redistributed to adults accrues to children’s health and physical development. It is widely recognized that low levels of investments in child health have consequences for the economic growth, distribution and welfare in the society.
The literature has suggested that benefit transfers to women are associated with greater improvements in child health since larger shares of the household income is spent on nutrients and health as compared to transfers falling in the hands of male beneficiaries. As one of relatively few effective cash transfer programs in developing countries, the South African old age pension program serves as a natural experiment of its impact on children’s nutritional status conditional on the recipient’s gender. In rural areas, the benefits amounted to about twice the median per capita income by 1993 and more than a quarter of black South African children under age five lived in a household with adult(s) eligible for pension. Using weight-for-height and height-for-age as anthropometric indicators of child nutrition, the impact on child health of children living in households with member(s) eligible for pension is studied.
The estimates suggest that pensions received by women significantly impacted the anthropometric status of girls but had little effect on that of boys. No similar effect was found when the pension recipient was male. In conclusion, it is suggested that public transfer efficiency programs may depend on the gender of the recipient. In critique, the improvement in child health is not likely to be the objective of an old age pension scheme. However, the study may still have relevant implications for public policy design, namely: should the objective of a policy be to transfer resources to children to improve child nutrition and health, it may be preferable that the state transfers target women rather than men. Although the evidence from this study cannot be generalized to other (developing) countries and, of course, would be hard to replicate due to the scope of the old-age pension program expansion in post-apartheid South Africa, it suggests that the impact of a public transfer program may vary depending on how it is administered. Moreover, even though this study focuses on the impact on child nutrition, it would also be interesting to study other investments in child human capital such as education. For example, maybe pensions received by men have a higher propensity to affect grandchildren’s education level?
Thea Sand 2292