The general opinion seems to be that foreign aid has saved lives from disease, holding per capita income fixed, and gains in life expectancy and infant survival have probably increased total (but not per capita) GDP through larger populations, but there is very little evidence of a positive effect on GDP per capita, i.e. reducing poverty.
Analyses tend to find weak effects of aid, which disappear when replicated on new datasets, and effect sizes tend to shrink as sample size and data quality increase. The literature also shows the usual signs of data mining and publication bias like spikes around the significance threshold, disproportionate reporting of positive results, etc. See this article, for instance.
The Gates Foundation, GiveWell, and many others seem to buy the basic model that public goods (agricultural and medical research for Gates, vaccines) work, and public health can generate welfare/save lives, but are not great for economic development.
I don’t want to claim that we now understand how to do aid without making mistakes. But I do want to say that decades of bad aid have not accomplished as much as they might have if they were actually designed to help the poor (rather than win the Cold War, support American farmers, or other political goals.) I think it’s reasonable to expect that evidence-based aid will do better than aid as a whole has so far.
The general opinion seems to be that foreign aid has saved lives from disease, holding per capita income fixed, and gains in life expectancy and infant survival have probably increased total (but not per capita) GDP through larger populations, but there is very little evidence of a positive effect on GDP per capita, i.e. reducing poverty.
Analyses tend to find weak effects of aid, which disappear when replicated on new datasets, and effect sizes tend to shrink as sample size and data quality increase. The literature also shows the usual signs of data mining and publication bias like spikes around the significance threshold, disproportionate reporting of positive results, etc. See this article, for instance.
The Gates Foundation, GiveWell, and many others seem to buy the basic model that public goods (agricultural and medical research for Gates, vaccines) work, and public health can generate welfare/save lives, but are not great for economic development.
I don’t want to claim that we now understand how to do aid without making mistakes. But I do want to say that decades of bad aid have not accomplished as much as they might have if they were actually designed to help the poor (rather than win the Cold War, support American farmers, or other political goals.) I think it’s reasonable to expect that evidence-based aid will do better than aid as a whole has so far.