I think this is an important question more people should be thinking about (particularly if one believes that there is going to be a significant increase in philanthropic funding due to AI money).
Rather than refer to research I can write about my personal reasons/takes:
I primarily work in global health (and AI safety), and my main reason for doing so is: measurability: it’s possible to measure the impact of global health projects fairly rigorously, compared to democracy promotion or economic growth. Additionally, many global health problems are probably just more tractable compared to various issues you noted (e.g. tackling corruption or improving international coordination or improving democracy). Nonetheless, something being less tractable and measurable doesn’t mean you shouldn’t work on it. Still, if you care about impact attribution, global health is likely to be better than many other “near-termist” areas you can compare it to (due to an existing high-quality evidence base, ease of organizing RCTs etc.)
That said, I believe global health delivery (e.g. charities delivering a particular drug, or service) projects seem to dominate EA’s global development portfolio.
My impression is that, the primary reason for this overconcentration is global health delivery being more measurable compared to economic growth, health systems strengthening and other things.
I believe that (influential) EAs overindex on measurability in global development, and there should be more funding, labour and attention going into global development projects that have a “worse measurability profile”. But I think this is part of a broader issue of EA community having surprisingly weak infrastructure (besides Probably Good) for helping/incentivizing people to think about career/issue prioritization within global development (or more politics-flavored issues).
I think this is an important question more people should be thinking about (particularly if one believes that there is going to be a significant increase in philanthropic funding due to AI money).
Rather than refer to research I can write about my personal reasons/takes:
I primarily work in global health (and AI safety), and my main reason for doing so is:
measurability: it’s possible to measure the impact of global health projects fairly rigorously, compared to democracy promotion or economic growth. Additionally, many global health problems are probably just more tractable compared to various issues you noted (e.g. tackling corruption or improving international coordination or improving democracy). Nonetheless, something being less tractable and measurable doesn’t mean you shouldn’t work on it. Still, if you care about impact attribution, global health is likely to be better than many other “near-termist” areas you can compare it to (due to an existing high-quality evidence base, ease of organizing RCTs etc.)
That said, I believe global health delivery (e.g. charities delivering a particular drug, or service) projects seem to dominate EA’s global development portfolio.
My impression is that, the primary reason for this overconcentration is global health delivery being more measurable compared to economic growth, health systems strengthening and other things.
I believe that (influential) EAs overindex on measurability in global development, and there should be more funding, labour and attention going into global development projects that have a “worse measurability profile”. But I think this is part of a broader issue of EA community having surprisingly weak infrastructure (besides Probably Good) for helping/incentivizing people to think about career/issue prioritization within global development (or more politics-flavored issues).
Thank you!