ABSTRACT
Introduction With the rapid expansion of cash transfer (CT) demonstrations across SubSaharan Africa (SSA), implementation issues such as targeting and payment delivery have become extremely important. Targeting in particular is necessary, at least in the short-and medium-term, because of limited state resources and high levels of poverty (often over 40 percent) in the region. In SSA, the policy discussion has focused on social protection for the ultra-poor, typically defined as 10 percent to 15 percent of the poorest in the population. However, in situations of high poverty and low capacity, implementation of efficient targeting mechanisms can be challenging, leading some to advocate against means-tested selection in favor of demographic or geographic selection, or both, as an easier alternative. This article describes the targeting mechanism and provides evidence of the targeting performance of Kenya’s Cash Transfer Program for Orphans and Vulnerable Children (CT-OVC). The CT-OVC program is a useful case study because it incorporates two of the features found in the demonstrations across SSA: it targets the ultra-poor and incorporates community-based identification mechanisms to select program recipients. The CT-OVC program is also a useful case study because it is wholly administered by the state and so it confronts the typical challenges of human and other resource capacity constraints faced by the public sector in SSA, and it utilizes multiple selection criteria to limit the leakage of benefits to non-eligible households. Specifically, the CT-OVC program employs a unique blend of a (decentralized) community-based selection that is common in SSA, along with a centralized, proxy-means test-selection procedure that is standard in the CT programs in Latin America and the Caribbean. It is therefore of policy and programmatic interest to present the CT-OVC targeting mechanism as a potential tool that other countries could mimic and to see how this mechanism performs against common targeting benchmarks. The analysis in this chapter is based on the targeting system used by the program in the first half of 2009. These procedures were slightly adjusted for households that entered the program in the second half of 2009 and these adjustments are described at the end of the chapter.
