The International AIDS Vaccine Initiative recommends targeting resources to research institutions in developing countries in order to accelerate the development of an effective HIV vaccine. In contrast, this paper shows that neither lump-sum nor in-kind transfers are an effective policy. We analyze several financing mechanisms as a means to overcome the lack of depth in HIV-vaccine research in a non-cooperative framework. At first, we point to cases in which financial support is actually counterproductive. Then we analyze whether in-kind transfers are preferable to lump-sum transfers. Even if donors prefer aid in kind because the incentives for moral hazard of recipients can be reduced, we demonstrate that it is effective only if recipients have cost advantages. Copyright © 2013 John Wiley & Sons, Ltd.