We use an excludable instrument to test the effect of bilateral foreign aid on economic growth in a sample of 97 recipient countries over the 1974-2013 period. Our instrument interacts donor government fractionalization with a recipient country's probability of receiving aid. The results show that fractionalization increases donors' aid budgets, representing the variation over time of our instrument, while the probability of receiving aid introduces variation across recipient countries. Controlling for countryand period-specific fixed effects that capture the levels of the interacted variables, the interaction provides a powerful and excludable instrument. Making use of the instrument, our results show a positive but insignificant effect of aid on growth. We also investigate the effect of aid on consumption, savings, investments and net exports and investigate heterogeneity according to the quality of economic policy, democracy and the Cold War period. We find that aid increases investment and consumption, while it decreases net exports. In no regression do we find that aid affects growth. However, the coefficients from the instrumental variables regressions are also not statistically different from the positive and significant OLS estimates. Résumé. Aide et croissance : nouveaux éléments de preuve grâce à un instrument exclusif. À l'aide d'un instrument exclusif, nous évaluons l'impact de l'aide étrangère bilatérale sur la croissance économique d'un échantillon de 97 pays bénéficiaires entre 1974 et 2013. Notre instrument met en interaction le fractionnement de l'aide apportée par les gouvernements contributeurs et la probabilité qu'un pays bénéficiaire puisse recevoir de l'aide extérieure. Les résultats suggèrent qu'en matière d'aide, le fractionnement entraîne une augmentation des budgets des pays contributeurs, constituant ainsi la