Cash Transfers and Adolescent Welfare (Malawi)

Wikimedia Commons. Author: John Duffell

Adolescent girls in developing countries are considered to be an important target group for policymakers. Targeted interventions for this group may not only affect their welfare directly, but they also have the potential to bring benefits to future generations. This paper investigates whether one such intervention, the Zomba Cash Transfer Program in Malawi, helped empower adolescent girls in the short-run. Summarizing evidence from multiple papers examining the impacts of this program on a broad range of outcomes and providing some new analysis here, this paper suggests that the answer is a clear ‘yes.’ The program effectively increased access to financial resources, increased schooling outcomes, decreased teen pregnancies and early marriages, improved health, and generally enabled beneficiaries to improve their agency within their households. (p. 21).

That is from the conclusions of a paper by Sarah J. Baird, Ephraim Chirwa, Jacobus de Hoop, and Berk Özler. 

And more

The CCT program changed some common socioeconomic patterns that affect young women in Malawi, as it induced beneficiaries to delay childbearing and marriage. There is some evidence that these changed socioeconomic patterns are accompanied with changed marital and fertility preferences, suggesting that empowering adolescent women may not only increase their bargaining power within future relationships, but it may also affect the type of relationship they enter into in the first place. (p. 22).

The paper also examines Unconditional Cash Transfers (UCT), if you look at page 19 you will see that CT and UCT affect outcomes differently. Both seems very positive, but their effects on education and other indicators (such as pregnancy, etc.) are different. 

In general the evidence I have seen so far is very positive in favor of CT and UCT. And that is a kind of area where negative results are also interesting and important not only academically but from the policy point of view, so I do no think there is much publication bias, although we need consider that possibility.