Value for money in humanitarian assistance: How does cost efficiency vary across cash and voucher programmes?

Motivation Cash and voucher assistance has become a key part of humanitarian response over the last decade as it is able to meet people's basic needs during a crisis and it has high cost efficiency relative to traditional in‐kind assistance. Donors have been introducing cost‐efficiency benchmarks—set as cost‐transfer ratios, the ratio of costs of delivery to the value of cash or goods provided to beneficiaries—for cash and voucher programmes. These benchmarks function as price ceilings for humanitarian agencies providing cash and voucher assistance. The welfare effects of these price ceilings, however, are unclear. They could induce greater efficiency, depending on the cost function for cash and voucher assistance programmes, but they could also have widely different consequences across contexts and could even undermine equity. Purpose We ask what determines the cost‐transfer ratios of cash and voucher assistance programs? How do cost‐transfer ratios vary by size of benefit delivered, local cost levels, scale of programme, and region? What do the results imply for setting cost‐efficiency benchmarks and humanitarian programming? Approach and methods We use a novel set of cost‐efficiency data for 31 humanitarian cash and voucher assistance programmes to examine the variation in delivery costs for humanitarian cash and voucher programmes, and to understand what causes these delivery costs to vary. Regression analysis is used to see the impacts of different characteristics of the programmes on their cost‐transfer ratios. Findings We find substantial variation in cost‐transfer ratio efficiency across cash and voucher assistance programmes based on their design and context. Programme scale and local price levels explain the largest fraction of variation: larger programmes and contexts with high local prices tend to push down cost‐transfer ratios. Policy implications Benchmarks for cost‐efficiency should be specific to context. Global price ceilings for cash and voucher assistance are likely to undermine quality delivery in contexts where local price levels are low and programmes are small, without fully capturing possible efficiency gains in other contexts. This finding, intuitive to economists familiar with economies of scale, runs counter to views common within humanitarian agencies, where limited budgets—budgets insufficient to meet needs—are thought to have the effect of improving cost‐efficiency. Programmes designed around data‐driven benchmarks offer at least as much leverage as price ceilings for improving the value for money of cash and voucher assistance programmes.