Incentive and crowding out effects of foodassistance: Evidence from randomizedevaluation of food-for-training project inSouthern Sudan

Food assistance is one of the most common forms of safety net programs in postconflictsituations. Besides the humanitarian and promotional roles, there arewidespread scepticisms of food assistance regarding its possible influence ondisincentive to work and on crowding out of private transfers. While there is arelatively large amount of empirical research on social protection in stable context, itis less researched in post-conflict situations. Based on randomized evaluation of afood-for-training program implemented in Southern Sudan, this paper estimatesthese effects. We observe a significant negative impact (about 13%) on per capitahousehold income. However, there is no effect on the hours of work or the type ofthe economic activities of the adult members. The decline in income mostlyhappened through a reduction in child labor. There is also a positive effect on schoolenrolment for girls (about 10 percentage points) and an improvement in theirhousing status. We also do not find any indication of crowding out of privatetransfers for the participants. This is most likely due to the extent of private transfersbeing very low to begin with. However, there is a small but significant impact of thetransfers given out by the participants.