| dc.description.abstract | While there are many studies of how official development aid (ODA) affects economic growth, there are far fewer studies of how aid affects health outcomes. Also, most of the studies of aid effectiveness have been cross-country studies. These studies have been criticized for lacking country specificity and a growing number of influential voices are questioning their usefulness for aid evaluations. There is clearly a lack of systematic studies of aid effectiveness below the country level. In this paper, I aim to fill a gap in the literature by researching how ODA affects infant mortality at the subnational level in Uganda. By matching geocoded data on the placement of aid projects with information on infant mortality from geocoded Demographic Health Surveys, and using a quasi-experimental difference-in-differences strategy, I am able to analyze if geographical proximity to active aid projects reduces infant mortality. The unit of analysis is 124 100 children born by 30 550 mothers. The results show that geographical proximity to active aid projects reduces infant mortality in most of the models. The finding is however surrounded by some uncertainty since the significance disappear in the most conservative test of aid. I also find evidence that projects are placed in areas that on average have lower infant mortality than non-aid locations. This suggest that aid projects do not reach those who need them the most. The various mechanisms studied in this paper all have the direction we would expect from the theory. This indicates that the intermediate factors suggested to be important in explaining infant mortality in the theory section, are in fact important explanatory factors for infant mortality. | en_US |
