Beyond Learning and Accountability: Comparing the Political Use of Evaluation in International Organisations

The vast majority of international organizations (IOs) conduct evaluation as an integral part of their organizational practices. Traditionally, the purpose of evaluation is organizational learning and accountability. Yet, building on a rationalist-power perspective, this paper recognizes evaluation as a political tool in organizational stakeholders’ struggle for interests and power. While evaluation itself is often thought to be a scientific process of expertise and reflection, the politics of evaluation acknowledge that political interests guide the selection and formulation of evaluation research questions as well as the presentation and framing of findings. Our central claim is that who controls structural resources of evaluation systems (budget and agenda-setting), affects what kind of evaluation is conducted in an IO. We scrutinize this hypothesis empirically for six UN system IOs which differ depending on whether member states, the international public administration, or both (two IOs per group) control evaluation resources. We systematically compare how high level evaluation stakeholders answer general questions on evaluation processes. First, we find that the behavioral orientation of evaluation units (whom they perceive as sponsor and primary user) differs depending on who controls evaluation resources. We also find the use of evaluation to follow this pattern: Depending on member state or administrative control of resources (or a mix), evaluation results are used in line with typical member state or bureaucratic evaluation interests (collective principal and agent control dynamics vs. collective agent and bureaucratic influence dynamics). This paper is the first to offer comparative empirical insights that unmask the political use of evaluation in IOs. Findings point to a latent variable in the relation between member states and bureaucracy that explains why in a similar group of UN system IOs, which subscribe to the same UN evaluation principles, evaluation systems manifested in such a different way.