John Maluccio (Middlebury College), August 2015
Social cash transfer programs (both conditional and unconditional) have become the anti-poverty program of choice in many developing countries. Their approach, combining short-term poverty reduction with the potential for enhanced investment in human capital, has widespread policy appeal. Numerous evaluations, many based on rigorous experimental designs, leave little doubt that such programs can reduce current poverty and increase health and children’s school attainment––in the short term. But evidence is notably lacking on whether these short-term gains eventually translate into the types of longer term benefits ultimately needed to fully justify these programs. For example, do health or achievement gains persist even after the transfers end? Do the increased investments in human capital improve the welfare of the next generation? As the policy discussion shifts towards considering exit strategies for some programs, there is a demand from policy makers for establishing whether such longer term gains exist.
Fortunately, we are well positioned to begin addressing these questions since so many social cash transfer programs have been accompanied by rigorous evaluations, like those carried out under The Transfer Project. Such impact evaluations lay the necessary groundwork for the much needed longer-term assessment, particularly if there is an intact control group. This is not always the case, however, as in many evaluation designs, often for ethical reasons, the original control group eventually receives the intervention. Nevertheless, even in the absence of the original control group, much can be learned from continued assessment. In those situations, we can still learn about longer-term effects, by focusing on the timing of the intervention during an individual’s life. For example, to understand the lasting impacts on schooling or schooling related outcomes (such as the labor market), one can contrast children who received the intervention near the typical “drop-out” age in treatment areas with children of the same age in original control areas, i.e., those who would have received the intervention at a later age, possibly already having dropped out. (For an example of an approach along these lines using the evaluation of a conditional cash transfer program in Nicaragua, see http://www.povertyactionlab.org/evaluation/long-term-effects-conditional-cash-transfer-program-nicaragua.) Of course, longer-term evaluations face their own specific challenges, including the potential for substantial selective attrition (e.g. if sufficient resources are not dedicated to tracking) which can invalidate the results.
Continuing to monitor and assess the short-term impacts of new and innovative social transfer programs in different contexts is important. At this point, however, it is arguable that the returns to research are higher for longer-term assessment.
For more on the impacts of social cash transfers, see reports from The Transfer Project: https://transfer.cpc.unc.edu/publications/reports