This is a problem I’ve been thinking about for awhile in a broader EA context.
It’s claimed fairly widely that EA needs a lot more smallish projects, including ones that aren’t immediately legible enough to be fundable by large institutional donors (e.g., because the expected value depends on assessments of the competence and value alignment of the person running the project, which the large institutional funders can’t assess). It’s also claimed (e.g., by Nick Beckstead of OpenPhil at EA Global San Francisco 2017) that smallish earning-to-give donors’ best bet to do the most good is to use their local knowledge to find and fund promising opportunities that the big institutional donors aren’t already covering.
This creates a seemingly obvious opportunity for an EA org to make it easier for donors to crowdfund these kinds of projects. E.g., by being a 501(c)(3) they can funnel donations from DAFs, which individuals can’t accept. (For me, at least, this is a bigger deal than tax deductibility; my DAF is overprovisioned relative to my personal savings right now, so I’d rather make donations from there.)
The two obvious hypotheses for why nobody’s already doing this are 1) all the EA meta-orgs are too constrained on staff time to set it up, and 2) it doesn’t actually work because the level of oversight required to avoid undue legal and/or reputational risk would destroy the efficiency gains. I would very much like to know to what extent each of these is the case.
I think to some extent we have this, but it depends upon local trust networks that aren’t always legible to newer people. (For example, LessWrong is funded through CFAR)
LessWrong seems like a bit of a weird example since CFAR’s senior leadership were among the people pushing for it in the first place. IIRC even people working at EA meta-orgs have encountered difficulties and uncertainty trying to personally fund projects through the org.
This is a problem I’ve been thinking about for awhile in a broader EA context.
It’s claimed fairly widely that EA needs a lot more smallish projects, including ones that aren’t immediately legible enough to be fundable by large institutional donors (e.g., because the expected value depends on assessments of the competence and value alignment of the person running the project, which the large institutional funders can’t assess). It’s also claimed (e.g., by Nick Beckstead of OpenPhil at EA Global San Francisco 2017) that smallish earning-to-give donors’ best bet to do the most good is to use their local knowledge to find and fund promising opportunities that the big institutional donors aren’t already covering.
This creates a seemingly obvious opportunity for an EA org to make it easier for donors to crowdfund these kinds of projects. E.g., by being a 501(c)(3) they can funnel donations from DAFs, which individuals can’t accept. (For me, at least, this is a bigger deal than tax deductibility; my DAF is overprovisioned relative to my personal savings right now, so I’d rather make donations from there.)
The two obvious hypotheses for why nobody’s already doing this are 1) all the EA meta-orgs are too constrained on staff time to set it up, and 2) it doesn’t actually work because the level of oversight required to avoid undue legal and/or reputational risk would destroy the efficiency gains. I would very much like to know to what extent each of these is the case.
I think to some extent we have this, but it depends upon local trust networks that aren’t always legible to newer people. (For example, LessWrong is funded through CFAR)
Hmmm. Do you think that’s a bug, or a feature?
LessWrong seems like a bit of a weird example since CFAR’s senior leadership were among the people pushing for it in the first place. IIRC even people working at EA meta-orgs have encountered difficulties and uncertainty trying to personally fund projects through the org.