Thanks for posting this. My comment is on the viability of donor screening; I’m looking for places to reduce the level of burden to make this more viable.
I’m not going to focus on the specific details of the CC policy, but will mention one big issue that I think makes it a weak model in general. It comes across too close to one-size-fits-most, rather than being heavily risk stratified. That stratification should depend on the amount, the presence or absence of yellow/orange/red flags in a more cursory screen, and so on.
For at least some organizations, I like the $10K threshold for performing some sort of donor screening. That’s in part based on my gut that third parties may see a five-figure donation as significant. I could be persuaded to go higher depending on the size of the organization and the nature of its work. Maybe I’m showing my own biases, but I think it would be hard for the general public to get or at least stay mad at a bednet-distribution charity that took a problematic $50K donation as long as it promptly disgorged it upon detection, and those orgs’ effectiveness is less dependent on public esteem anyway. Tolerances for policy or meta work would probably be less.
But for orgs on the relatively higher screening side of things: in the $10K to perhaps $25K or $50K range, that should probably look more like 15-30 minutes of time from a junior staffer / contractor / intelligent college student intern looking for yellow, orange, or red flags, not a full-fledged review. If no flags are found, clear the donor for three years (unless their donations go up a level or the organization happens to learn of concerns). If there are flags, send the file to the development director (“DD”) or other mid-level staff. The DD could clear yellow flags or decline red flags, or do a medium-to-full fat writeup if necessary.
In my view, the primary point of vetting one’s midsize donors for most orgs is to stop donations from red-klaxon donors—e.g., SBF’s parents, most people who have been convicted of crimes involving moral turpitude or fraud, people who are readily known to have engaged in pretty shady behavior with crypto initial coin offerings, etc. It is not realistic to scope midsize-donor investigation to catch everyone from whom one wouldn’t take a donations after a bespoke custom investigation. At this level, I don’t think it is necessary to consider an individual donor’s employer unless it presents a conflict of interest, the donor is a senior executive, or the employer is really problematic/controversial.
Moving to the $50K-$250K range, I predict that a lot of donors will still be low risk and should be able to screen OK even at stricter orgs without involvement of senior staff or a lot of paperwork. Anyone whose source of income is being a lawyer at a typical law firm, a physician doing typical physician things, a trader at somewhere like Jane Street, etc. shouldn’t be too hard to clear efficiently. One potential crux is that I’m relatively more concerned that the donations are cleanly obtained and relatively less concerned about the donor’s non-financial activities (as it were). So I don’t necessarily think most orgs would have many donors who need the full-dress vetting treatment.
When does a donor become more than “midsize” for screening purposes? One non-exclusive indicator: when the org cannot afford to disgorge a donor’s money going back several years immediately and without grave impacts on the viability of the org’s continuing work, then that donor is not midsize. In other words, perhaps post-acceptance screening (cf. post-claims underwriting in insurance) can be a viable part of the overall vetting strategy . . . but only if the org can immediately spit back the money that it wouldn’t have taken had it conducted a full investigation upfront.
I’ll check feasibility with GiveWell’s donor numbers because I know where to find those. There were 138 100K+ donors through GiveWell in 2022, plus an unclear number of anonymous donors (p. 12). A few hundred screened donors seems appropriate for an org moving as much money as GiveWell, especially since most of its donors are repeat donors and I do not think full reinvestigations would be necessary each year.
What to do with the anonymous money is likely to be a major issue here, though. If an org is going to do that, I guess the only mitigation may be healthy reserves for post-acceptance screening if the donor’s identity becomes known and is problematic?
Thanks for posting this. My comment is on the viability of donor screening; I’m looking for places to reduce the level of burden to make this more viable.
I’m not going to focus on the specific details of the CC policy, but will mention one big issue that I think makes it a weak model in general. It comes across too close to one-size-fits-most, rather than being heavily risk stratified. That stratification should depend on the amount, the presence or absence of yellow/orange/red flags in a more cursory screen, and so on.
For at least some organizations, I like the $10K threshold for performing some sort of donor screening. That’s in part based on my gut that third parties may see a five-figure donation as significant. I could be persuaded to go higher depending on the size of the organization and the nature of its work. Maybe I’m showing my own biases, but I think it would be hard for the general public to get or at least stay mad at a bednet-distribution charity that took a problematic $50K donation as long as it promptly disgorged it upon detection, and those orgs’ effectiveness is less dependent on public esteem anyway. Tolerances for policy or meta work would probably be less.
But for orgs on the relatively higher screening side of things: in the $10K to perhaps $25K or $50K range, that should probably look more like 15-30 minutes of time from a junior staffer / contractor / intelligent college student intern looking for yellow, orange, or red flags, not a full-fledged review. If no flags are found, clear the donor for three years (unless their donations go up a level or the organization happens to learn of concerns). If there are flags, send the file to the development director (“DD”) or other mid-level staff. The DD could clear yellow flags or decline red flags, or do a medium-to-full fat writeup if necessary.
In my view, the primary point of vetting one’s midsize donors for most orgs is to stop donations from red-klaxon donors—e.g., SBF’s parents, most people who have been convicted of crimes involving moral turpitude or fraud, people who are readily known to have engaged in pretty shady behavior with crypto initial coin offerings, etc. It is not realistic to scope midsize-donor investigation to catch everyone from whom one wouldn’t take a donations after a bespoke custom investigation. At this level, I don’t think it is necessary to consider an individual donor’s employer unless it presents a conflict of interest, the donor is a senior executive, or the employer is really problematic/controversial.
Moving to the $50K-$250K range, I predict that a lot of donors will still be low risk and should be able to screen OK even at stricter orgs without involvement of senior staff or a lot of paperwork. Anyone whose source of income is being a lawyer at a typical law firm, a physician doing typical physician things, a trader at somewhere like Jane Street, etc. shouldn’t be too hard to clear efficiently. One potential crux is that I’m relatively more concerned that the donations are cleanly obtained and relatively less concerned about the donor’s non-financial activities (as it were). So I don’t necessarily think most orgs would have many donors who need the full-dress vetting treatment.
When does a donor become more than “midsize” for screening purposes? One non-exclusive indicator: when the org cannot afford to disgorge a donor’s money going back several years immediately and without grave impacts on the viability of the org’s continuing work, then that donor is not midsize. In other words, perhaps post-acceptance screening (cf. post-claims underwriting in insurance) can be a viable part of the overall vetting strategy . . . but only if the org can immediately spit back the money that it wouldn’t have taken had it conducted a full investigation upfront.
I’ll check feasibility with GiveWell’s donor numbers because I know where to find those. There were 138 100K+ donors through GiveWell in 2022, plus an unclear number of anonymous donors (p. 12). A few hundred screened donors seems appropriate for an org moving as much money as GiveWell, especially since most of its donors are repeat donors and I do not think full reinvestigations would be necessary each year.
What to do with the anonymous money is likely to be a major issue here, though. If an org is going to do that, I guess the only mitigation may be healthy reserves for post-acceptance screening if the donor’s identity becomes known and is problematic?