This paper criticises the World Bank as overly optimistic with respect to its ability to fine-tune development aid and to focus it on countries with ‘good’ policies rather than on countries with ‘poor’ policies in order to raise its effectiveness. It is shown that recipient regions showed very different patterns of aid inflows and economic growth in the past and that aid flows yielded the highest correlation to growth when their magnitudes shrank. It is furthermore argued that categorising countries by quality of domestic policies is not only questionable at a given point of time especially in countries with failing governmental institutions and open borders as in many African countries. It suffers also from incentive problems so that countries receiving more aid can become victims of changes in their domestic policies which are more permissive and etatist to the disadvantage of private agents. The paper instead pleads for a shift to aid policies decoupled from country-specifics and more oriented to fundamental ‘beyond border’ problems of the well-being of the poor. An international endowment fund under supranational law should finance research and implementation of research findings related to common international goods as it was suggested by Sachs concerning aids and tropical disease research.