AFD’s policy

EVALUATION

AFD’s policy

I - What is AFD’s policy on impact evaluations?

AFD has been financing and conducting impact evaluations for several years now. Indeed, this new tool provides one solution to the deficit in terms of knowledge on the impacts of development actions. Moreover, it makes it possible to build links between research and evaluation. This is fully in line with AFD’s strategic orientations and helps to make evaluation more rigorous and research more relevant.

In this respect, AFD is a financial and technical stakeholder in a number of these exercises in partnership with the relevant actors in the South, development operators and scientific teams. By being directly involved, AFD can get a better understanding of, and subsequently control, the complexity of these operations and therefore provide a clear vision of the extent of results achieved. Moreover, as with all evaluations, the regular participation of all stakeholders in the process is a favourable condition for the use of results in future programmes.

In addition, AFD participates in and contributes to – either regularly or periodically - international networks on impact evaluations, particularly the NONIE and 3IE. These networks provide an opportunity to share experience and knowledge of these evaluations and are a way to promote experimentation in development policies.

Finally, these evaluations come as additions, not as replacement to all the other evaluation operations (thematic, programmatic, instrument and project evaluations) and operational knowledge production (capitalisation)

II - What type of impact evaluation does AFD seek to contribute to?

Impact evaluations must provide responses to relevant development issues in contexts adapted to their analysis. These evaluations do not seek to measure all the impacts of a project financed by AFD, they focus on certain aspects considered as decisive for understanding the effects of development actions in certain contexts (sustainability of health insurance systems, type of microfinance, type of agricultural advice, role of participative planning for public equipment, etc.).

The choice of interventions to measure impacts is also closely related to a general objective of producing knowledge on development, which requires understanding the mechanisms of the impact – in terms of the behaviour of beneficiaries or the project’s interaction with its physical, economic, social and institutional environment. These evaluations are consequently global public goods in themselves, since they aim to provide operational recommendations for the intervention studied.

All impact measurement methods, whether they are based on matching, random allocation or other methods considered as robust, are worth taking into account depending on the nature of contexts and questions asked, and the time and resources available. In addition, and to ensure the impact evaluations go further than simply measuring impact, it is generally imperative to conduct and analyse qualitative studies in parallel to the quantitative process.