This work analyses the targeting performance and the impact evaluation of three projects. The first paper is about the water subsidies in the Republic of Congo. The study shows that the self-targeting scheme, using an Inverse Block Tariff structure has poor performance, indeed only those households who are connected can benefit from it, and they are usually non-poor. The self-targeting mechanism used in the Public Work program in Liberia works better since the proportion of poor involved in the program is high. This project has no entry barrier which partly explains the good result. The targeting performance of the Modernization of Agriculture project in the DRC is also poor, most of the beneficiaries being non-poor. The impact in terms of poverty reduction is important in both projects involved in this type of analysis (Modernization of Agriculture in the DRC and Public Work in Liberia), at least for the beneficiaries. The case of the DRC project, however, shows that it is important to lift some other constraints like access to credit and infrastructures in order to improve the impact of the program.
