Background: Budget support became increasingly popular towards the second half of the 2000s as the most consequential aid instrument for implementing the principles of effective aid formulated in the 2005 Paris Declaration. Although evaluations found that budget support contributed to improved development outcomes in several countries, the instrument has come under criticism in recent years and many bilateral donors have partly or fully stopped using it.

Objectives: The synthesis on the effectiveness of budget support aims at making lessons learnt from implementing, ending and evaluating budget support available to decision makers in the current aid environment where related financing instruments are used, e.g. for achieving the SDGs and combatting climate change.

Methods: The synthesis follows a systematic-review process as defined by the Campbell Collaboration as closely as possible. Since for the field of budget support, counterfactual-based impact evaluations are not available, the evaluation team took a qualitative approach to conduct the synthesis, thus contributing to the debate on how to transfer systematic-review standards to qualitative, complex and methodological versatile areas of research and evaluation.

Results: The synthesis finds convincingly broad evidence that budget support is indeed an effective instrument to promote important developmental outcomes, such as improvements in public financial management, budget processes, and provision of public goods and services.

Yet, not only the political debate on budget support largely ignored solid empirical evidence on the effectiveness of the instrument; in hindsight, most empirical work on the aid modality equally turned a blind eye on politically disputed aspects of budget support, namely the risks and unintended effects.

Conclusions: With key elements of the instrument’s intervention logic severely under-researched, future empirical work should focus more clearly on potential risks of budget support, on the effects and causal mechanisms of specific inputs and on the consequences of donor withdrawal.