Reducing World Poverty by Improving Evaluation of Development Aid

This article argues that given its structural conditions, development aid bears a particularly heavy burden of learning and accountability. Unfortunately, however, the organization of evaluation guarantees that evaluations will be inconsistent and it creates incentives for positive bias. This article presents evidence from organizational studies of aid agencies, from the public choice literature, from eight development projects in Africa, and from one in India, that demonstrates positive bias and inconsistency in monitoring and evaluation. It proposes that the evaluation function should be professionalized through an approach titled “monitoring and evaluation for cost effectiveness,” and it compares this approach to the World Bank's results-based monitoring and evaluation, the Development Assistance Committee's five evaluation criteria, and evaluations based on randomized trials. This article explains the analytics of the proposed approach and suggests directions for further research.