Cash versus in-kind transfer programs in humanitarian operations: An optimization program and a case study

Humanitarian organizations, mandated with responding to emergencies, generally provide food assistance via in-kind and/or cash transfers. Although cash and in-kind transfers have had varying effects across different regions, the superiority of one over the other has been debated. This study considers a mixed strategy that includes both cash and in-kind transfer in the presence of finite budget, given a slow-onset disaster such as famine or drought. Importantly, it proposes an evidence-based framework based on the given data. Specifically, a two-stage stochastic program with recourse is proposed, where uncertainty stems from the slow-onset disaster that has non-uniform impact across a given geographical region. The proposed program is first used to study the slow-onset disaster situation in Kenya, and then to evaluate the performance of the cash versus in-kind transfer programs. We also solved larger size problem instances using the sample average approximation (SAA) algorithm, and the resulting analyses underscore the deductions of the case study that although cash transfer is more efficient than in-kind transfer, however, the latter is inevitable due to local unavailability of certain commodities.