This thesis explores interdependence between donor countries’ health aid expenditures. There are two ways in which donors may interact with their peers when deciding how much aid to provide, depending on how the impact of foreign aid is valued by donors. Firstly, the impact might be seen in terms of achieving a common goal of improving standards of living in less developed countries. In this case, foreign aid is essentially an international public good. As such, the aid given by a donor’s peers is considered substitutable for its own aid, and thus the more other donors give, the less a particular donor must give to achieve the common goal. In other words, there are incentives to free ride off the aid given by peers. However, if donors value foreign aid instead for the private benefits it confers, such as international status or recognition, foreign aid given by other donors might instead be seen as complementary, as the more other countries give, the more a particular country must give in order to achieve recognition or gains in international status. There might also exist a positive relationship between donors’ aid supplies due to an international “peer pressure effect” resulting from membership to institutions such as the OECD Development Assistance Committee. There has been little research examining the nature of donor interaction in foreign aid, but a few studies provide empirical evidence to support a “peer pressure effect”. These studies have tended to use only data on foreign aid disbursements and have not looked at whether the results differ for aid commitments. This thesis contributes to the literature on donor aid supply interaction by empirically examining whether there exists a peer effect (positive or negative), using data for both commitments and disbursements of bilaterial health aid. An extension will be made to the existing theoretical framework to distinguish between aid commitments and disbursements. This thesis will also test the hypothesis that instead of a peer effect, there is a leader effect, whereby a single leader country influences the aid expenditures of all other countries. The empirical analysis will be undertaken on a panel of 20 OECD donor countries over the period 1990 to 2009. Based on the theorectical models, models for commitments and disbursements will be estimated using three different methods: Fixed Effects (FE) Two-Stage-Least-Squares, FE Generalised Method of Moments, and FE Limited Information Maximum Likelihood estimation. The principal finding is that donor countries have a tendency to free ride off the health aid commitments and disbursements of their peers, and the effect is stronger for disbursements.