What happens to aid fungibility when the recipient government takes control? Effects of aid ownership in Rwanda

Abstract

Motivation: Aid fungibility and aid ownership have largely been discussed separately in the development literature. This is unfortunate because donors have often aimed to reduce the former and to increase the latter, without fully understanding the relationship between them. Purpose: We analyse what happens to aid fungibility when the recipient government takes ownership of its development process by examining the case of Rwanda. Methods and approach: We use a mixed-methods approach for our analysis. We start with an econometric model to determine if aid fungibility is present in Rwanda. To complement and explore quantitative insights, we interviewed key informants in Rwanda. We used the principal–agent framework to help interpret our findings. Findings: We find that aid is fungible, albeit in a U-shaped relationship. Initially an increase in development assistance causes government development expenditure to fall, but subsequent increases in assistance cause government development expenditure to increase as well. We conclude that the government took over its development process through: (1) stimulating the donor to align their spending to its priorities; and (2) urging donors to switch from project aid to budget support. Policy implications: While government taking ownership of its development process should reduce concerns about fungibility, it could also contribute to marginalization of certain sectors or themes if this ownership is not inclusive.