The use of cash, food, or vouchers as a transfer modality is heavily contested. At this policy seminar, John Hoddinott and Annalisa Conte will present results from a three-year IFPRI-World Food Programme study that assessed the comparative performance of these transfer modalities on household food security in Ecuador, Uganda, Niger, and Yemen. In all countries, an experimental design was used with modalities randomly assigned at a locality level. The timing, frequency, and value of transfers were equalized to the extent possible across modalities, thus ensuring that differences in outcomes were attributable to the modality and not to other confounding factors.

The findings indicate that there is no one “right” transfer modality. The relative effectiveness of different modalities depends heavily on contextual factors such as the severity of food insecurity and the functioning of markets for grains and other foods. In three countries (Ecuador, Uganda, Yemen), cash had a relatively larger impact on improving dietary diversity as did vouchers in Ecuador, but in the fourth country (Niger), food had a larger impact on dietary diversity. By contrast, in two countries, food had a relatively larger impact in terms of increasing quantity of calories available for consumption at the household level. These studies also point to the need to pay increased attention to delivery costs: cash transfers were always significantly cheaper to deliver than food.

Work on this study was made possible in part by the generous support of the CGIAR Research Program on Policies, Institutions and Markets