University of Hannover, Development and Agricultural Economics, Germany
The paper examines the degree to which food aid (food-for-work and free food distribution) is targeted to the poorest and most vulnerable groups according to household income and asset ownership using cross section data of 149 households in Northern Ethiopia. By doing so, it is examined whether resource related indicators played an overriding role in the targeting process or whether there is a significant leakage to asset rich households. I used Heckman two-step econometric estimation procedure in an attempt to account for sample selection bias.
Food-for-work participation does not appear to be self-targeted with relatively wealthier households less likely to participate. The probability of participation was found to be mainly related to household demography like age and marital status of household head whereas resource related covariates do not appear to influence it. Households with higher farm income and oxen holding were more likely to take part in food"=for-work programmes pointing to leakage in targeting. However, off"=farm income is negatively related. The findings do not support the commonly held notion that female"=headed households are more food insecure and should be targeted for food"=for-work. The intensity of participation also doesn't seem to depend on poverty related factors, however, households with large farm size found to have spent less number of days in food"=for-work programme. On the other hand the probability of participation in free food distribution programme increased significantly with increasing age and off"=farm income. Households were more likely to receive free food in the current year, if they had participated in this programme during the four previous years while households that received free food only once or twice in this period were less likely to have participated in the current year.
Keywords: Ethiopia, food aid, food-for-work (FFW), free food distribution (FD), Heckman model, targeting