Thanks Vasco, this is a fair critique.
On marginal cost-effectiveness, you’re right that the factors I listed are exactly what determines marginal CE.
So the framing in that paragraph is imprecise.
What I was trying to point at was that the typical “one more dollar at the margin” thinking works well for funding ongoing work where each extra dollar buys more of the same thing.
But it doesn’t work as well in two situations.
The first is contexts where the ongoing work itself hasn’t been built out yet, so there isn’t a stable field at the margin to fund. The second is one-off events like passing a regulation, where extra funding may not change the probability of winning at all, or might only at a specific moment, but still maybe worth taking the bet given high CE.
And these are exactly the judgments the EA ecosystem is less equipped to make at arm’s length, which is part of why fewer such grants get made overall.
I should have written the paragraph that way.
On scoring did the rough back-of-the-envelope calculations which informed the subjective scoring of 1-10 in an area. For some data/evidence poor interventions we just did subjective scoring.
On your offer to review CEAs, we’d love help on that for the next round and I’ll DM you to follow up.
Both fair, and I think we’re mostly agreeing on substance.
Regarding your second point, the “high CE” in my original sentence referred to the CE of solving the policy issue itself, not the marginal CE of the next dollar. So, funding may still shift the probability of winning, just deferred to when a window opens rather than at the moment of funding, because it may fund prep work and exploratory research on a high-suffering issue.