Identifying Poor Refugees for Targeting of Food and Multi-Sectoral Cash Transfers in Niger

Both the number of forcibly displaced people and the demand for financing for humanitarian response have increased over the last five years. While financing for humanitarian needs has reached a historic high in 2016 with3.9 billion committed to UNCHR, the needs continue to grow at an even faster pace with a total budget of 8.2 billion requested in 2018 (UNHCR, 2019b). In light of this financing gap, UNHCR and other humanitarian partners are forced to prioritize among the many assessed needs. Targeting of assistance to refugees, at the extreme, could be the difference between life and death, yet little evidence exists on effective targeting strategies in a refugee context. In Niger, humanitarian actors—UNHCR and WFP—were faced with a severe downsizing of total assistance for the protracted Malian refugees, resulting in a drastic change of the targeting approach; from assisting nearly all refugees in the same way to targeting food and multi-sectoral cash assistance to the most vulnerable. To assess the most efficient method of targeting households, we undertook an assessment of three different methods to estimate household welfare, including: (i) Proxy Means Testing (PMT), (ii) adapted participatory wealth ranking (PWR), and (iii) Principal Components Analysis (PCA). The models are tested for two refugee communities in Niger—Intikane and Tabareybarey. The PMT method outperforms the PWR method and the PCA, especially for the Intikane site where the R2 of the PMT is two-thirds more efficient than the PWR method and 50 percent more efficient than the PCA. In the Tabareybarey refugee camp, the R2 of PMT is 11 percent more efficient than the PWR method and 48 percent more efficient than the PCA