I examine the allocation and impact of foreign health aid at the sub-national
level in Malawi. The literature remains divided over the impacts of health aid— some
scholars fail to find significant relations between health aid and health outcomes, while
others praise notable impacts. Moreover, the approaches scholars use to examine impacts
are as polarized as their results— aid impacts are primarily examined using cross-national
analyses or at the project level. However, the emergence of geocoded aid data allows for
a new analytical approach, one of examining aggregate health aid within a country. I use
an AIC-based hierarchical model averaging approach to determine the best predictor
variables of health aid in four time periods, examining how health aid is allocated
according to socio-economic factors, health conditions, and ethnic preferencing. In
addition, I use propensity score matching methods to examine the causal impacts of
health aid. Results show that aid is generally not allocated to the poorest individuals, but
results are mixed over allocation according to health conditions. In addition, only one
year, 2010, shows evidence of possible ethnic preferencing influencing aid allocation.
Despite mixed results of allocation, propensity score matching methods show health aid
causing statistically significant improvements in health conditions in 2008, 2009, and
2010, causing a reduction of 0.3 to 5 million cases of illness annually. Results highlight
notable aggregate health aid impacts, despite potential inefficiencies or negative
consequences of aid.