Introduction: focus of the paperFollowing the adoption of the MDG, particularly the first one that is to reduce poverty by half between 1995 and 2015, numerous studies have examined how external aid can contribute to their achievement. The formula "doubling aid to reduce the poverty by half" relied on the implicit assumption that aid was an effective instrument for poverty reduction. The formula and corresponding assumption have been debated. Two opposite views clearly appeared, one, represented by Jeffrey Sachs in his End of Poverty, underlining the need for a big push to get low income countries out of poverty traps, the other one, illustrated by the attacks of William Easterly against aid as a support of a big push and the idea of a poverty trap, and also including arguments about a limited absorptive capacity. Elsewhere we have argued that the absorptive capacity of aid depends on aid modalities and can be enhanced by a reform of aid, a way by which big push and absorptive capacity views can be reconciled and to which we come back later (Guillaumont and Guillaumont Jeanneney, 2010).Actually the academic debate on aid effectiveness of the first millennium decade has been dominated by another controversy, relying on cross country regressions and initiated by the highly influential paper of Burnside and Dollar (2000). After so many cross country studies following their paper, supporting or, more often, criticizing it, there seems to be a temptation to consider this research orientation as a deadlock and to switch to micro impact analysis. Whatever the importance and need of impact micro-analyses, we argue that it would be a dangerous dismissal to give up the cross-section approach, for several reasons. First the methodological weaknesses of many studies does not entail that other ones relying on better methodology and data cannot lead to more robust results. Second, since cross section studies on aid effectiveness will never be totally given up, there is a risk to see only the most provocative (and possibly least robust) retaining the attention of media and policy makers, a policy challenge to be kept in mind in the orientation of research. Finally micro studies and impact analyses, while they supply policy makers with useful information in a given context, are not an appropriate tool for assessing the impact of macro-economic features of countries on aid effectiveness.The aim of this paper, relying on results of the cross country literature on aid effectiveness, and drawing only on those we consider as particularly relevant and robust is to examine how aid can contribute to poverty reduction, with a special focus on the way it can address the vulnerability of many developing countries.
