Creating projects that are maximally cost-effective is now comparatively less valuable; creating projects that are highly scalable with respect to funding, and can thereby create greater total impact even at lower cost-effectiveness, is comparatively more valuable.
I think this framing is wrong, or at best unhelpful because we shouldn’t avoid prioritizing cost-effectiveness. When you stop prioritizing cost-effectiveness, it stops being effective altruism. Resources are still finite. The effectiveness of solutions to dire problems still differs dramatically. And we have only scratched the surface of understanding which solutions are gold and which are duds. I think it’s cost-effectiveness all the way down.
I hope Will doesn’t mean “creating maximally cost-effective projects is now less valuable” when he says “creating maximally cost-effective projects is now less valuable”. I hope Will means “We should use average cost-effectiveness instead of marginal cost-effectiveness because cost-effectiveness often decreases with more funding. This means that some projects which were more cost-effective at small levels of funding will become less cost-effective at larger levels of funding, which will shift our priorities.” I hope he means that, because I think that’s the correct take.
To illustrate, imagine there’s two projects, A and B, and we have to decide to allocate all of our funds to one or the other. Project A is more cost-effective if our funding supply is limited. This could be treating an incredibly painful but rare disease, where the cost to find potential patients quickly rises as you run out of people to treat. Then there’s project B, which is about as cost-effective regardless of how much money you spend. A classic example of a “project B” type project is cash transfers.
The figures below depict this. The first is a figure lent to me by Michael Plant. I also attach my less clear hand-drawn figure.
As EA acquires more funds, creating and funding maximally cost-effective projects is just as valuable. But our heuristics for cost-effectiveness will change. Instead of asking, “what’s the average cost-effectiveness of spending $10,000 on project A and B”, which would favor A, we should ask “what’s the cost-effectiveness of spending $10,000,000 on project A and B”, which would favor project B.
Cost-effectiveness changes with the supply of funding
I think this framing is wrong, or at best unhelpful because we shouldn’t avoid prioritizing cost-effectiveness. When you stop prioritizing cost-effectiveness, it stops being effective altruism. Resources are still finite. The effectiveness of solutions to dire problems still differs dramatically. And we have only scratched the surface of understanding which solutions are gold and which are duds. I think it’s cost-effectiveness all the way down.
I hope Will doesn’t mean “creating maximally cost-effective projects is now less valuable” when he says “creating maximally cost-effective projects is now less valuable”. I hope Will means “We should use average cost-effectiveness instead of marginal cost-effectiveness because cost-effectiveness often decreases with more funding. This means that some projects which were more cost-effective at small levels of funding will become less cost-effective at larger levels of funding, which will shift our priorities.” I hope he means that, because I think that’s the correct take.
To illustrate, imagine there’s two projects, A and B, and we have to decide to allocate all of our funds to one or the other. Project A is more cost-effective if our funding supply is limited. This could be treating an incredibly painful but rare disease, where the cost to find potential patients quickly rises as you run out of people to treat. Then there’s project B, which is about as cost-effective regardless of how much money you spend. A classic example of a “project B” type project is cash transfers.
The figures below depict this. The first is a figure lent to me by Michael Plant. I also attach my less clear hand-drawn figure.
As EA acquires more funds, creating and funding maximally cost-effective projects is just as valuable. But our heuristics for cost-effectiveness will change. Instead of asking, “what’s the average cost-effectiveness of spending $10,000 on project A and B”, which would favor A, we should ask “what’s the cost-effectiveness of spending $10,000,000 on project A and B”, which would favor project B.