Thanks for flagging out! Have fixed it.

Hey Vasco, the adjustment is specific to GiveWell vs us (or indeed, non-GW CEAs), since GiveWell probably is the most rigorous in discounting, while other organizations are less so, for various reasons (mainly time - that's true for us, and why we just use a rough 10x GW threshold; and it's true of FP too; Matt Lerner goes into detail here on the tradeoff between drilling down vs spending researcher time finding and supporting more high EV opportunities instead).

Relative to every other organization, I don't find CEARCH to be systematically overoptimistic in the same way (at least for our deep/final round CEAs).

For our GWWC evaluation, I think the ballpark figure (robustly positive multiplier) probably still holds, but I'm uncertain about the precise figure right now, after seeing some of GWWC's latest data (they'll release their 2023-24 impact evaluation soon).

Our grantmaking always aims at maximizing DALYS averted (with income and other stuff translated to DALYs too).

In terms of cost-effectiveness, it's nominally 30-50x GW, but GiveWell is more rigorous in discounting, so our figures should be inflated relative to GW. Based on some internal analysis we did of GW's greater strictness in individual line-item estimation and in the greater number of adjustments they employ, we think a more conservative estimate is that our estimates may be up to 3x inflated (i.e. something we think is 10x GW may be closer to 3x GW, which is why we use a 10x GW threshold for recommending GHD causes in the first place - to ensure that what we recommend is genuinely >GW, and moving money to the new cause area is +EV).

So my more conservative guess for our grantmaking is that it's closer to 9-15x GW, but again I have to emphasize the high uncertainty (and riskiness, which is the inherent price we pay for these ultra high EV policy interventions).
 

I do think RTSL's salt policy work (and other salt policy projects, particularly ImagineLaw in the Philippines) are reasonably good bets for maximizing life years saved. That said, I don't an individual donation to RTSL would help insofar as smaller donors can't purpose restrict it (see their donation button at https://resolvetosavelives.org/).

In practice, I would suggest donating to CEARCH's GHD policy regranting budget (via https://exploratory-altruism.org/work-with-us/, or just email me and I'll put you in touch with our fiscal sponsor), making a note on purpose-restriction if you wish, and then your donation goes out as part of a broader consolidated package (e.g. that 63k grant we made on SSB tax enforcement was me personally and 5 other EA donors pulling together).

On nuclear/volcanic winter - won't the direct effect just be straightforwardly mass extinction of wild animals, which eliminates their suffering? And in contrast, a lot of currently valuable farmland may just not be usable when temperatures shift, so there may not be an offset. A lot of uncertainty regardless, and reasonable people can disagree.

Hi Vasco,

(1) We've generally looked at DALYs (and not just deaths/YLL averted), but given the high cost-effectiveness of both hypertension/salt & diabetes/SSB in DALY terms (with the former being somewhat less cost-effective but having deaths make up like 90% of the burden), they're plausible candidates (CEAs linked in the cause evaluation result spreadsheet). Trans fat/tobacco/alcohol are other plausible candidates - given the clear scientific evidence on mortality + it being difficult to beat policy ideas for cost-effectiveness. You'll probably also have more speculative stuff like funding development of new vaccines or doing biological control of mosquitoes, but we haven't done any deep research there.

Nuclear/volcanic winter famine mitigation is another candidate (CEA in the spreadsheet), though obviously there's a strong self-defeating element from a WAW perspective.

(2) GiveWell's grantmaking criteria include not just cost-effectiveness but also evidence of effectiveness (which means excluding those high-uncertainty high-EV stuff), though I would say that there is a distinction between their public facing recommendations (which do need to work within the constraint of retail donor risk aversion) and some of what GiveWell funds through other means (e.g. the explicitly more maximization-oriented All Grants Fund or via recommendation to OP). Some riskier stuff GiveWell/OP has funded include alcohol policy and pesticide suicide prevention.

(3) Chris Smith and his team are great, but extremely limited in their time, so I don't think there's much ability to expand beyond lead and air pollution right now, even if they wanted to. Also, it's always important to keep in mind that OP isn't any different from other research/grantmaking organizations insofar as the researchers/programme officers are constrained by donor preference and risk aversion (specifically GV's).

Hi Vasco, I made a consolidated comment here: https://forum.effectivealtruism.org/posts/k7NjuGEKdRSrrJHmZ/deep-report-on-hypertension?commentId=weQHRcDt53BkCuLct

Hi Vasco,

(1)   In hypertension/salt reduction policy, CEARCH (in collaboration with the donors we advise) has made 150k in grants (specifically, projects advocating for - and assisting governments in implementing - reformulation policies to reduce sodium in food).

In diabetes/soda taxes, CEARCH has made 63k in grants (specifically, technical assistance to improve enforcement of and compliance with SSB taxes).

(2)   For the the bigger GW grantmakers, I'm unsure how much I can share given confidentiality, and I don't want to falsely give the impression that these grantmakers have already developed any specific views/positions/recommendations in this area, but I think I can broadly share that:

(a) FP previously asked us to help evaluate two large global NGO that worked on salt policy, with a specific focus on trying estimate the counterfactual advocacy success rate of salt policy advocacy campaigns (it's about 10%). We ended up making a positive recommendation, particularly for RTSL and its salt reduction work. Note that FP already supports RTSL's trans fat reduction work.

(b) GiveWell is currently considering making a salt grant to RTSL, but I understand it's exploratory in nature (see how this goes, then follow-up from there). They have also done some internal CEAs of SSB taxation projects; I think their major concern (a frustration shared by us) relates to high uncertainty over the existing GBD estimates of the SSB burden (n.b. the estimates changed wildly from one iteration of the GBD to the next, and it's not clear to us how or why the methodology changed). FWIW, I don't see any evidence that the GBD estimates are systematically biased (particularly upwards, which would be the main concern), so we're happy to go ahead.

(3)   Broadly speaking, I'll say that while there is very good reason think that health policy to prevent NCDs is extremely cost-effective (NCDs are a big and growing problem + policy offers large scale of impact at low cost), it's also very risky, and very much hits-based EV-maximizing grantmaking, which is not something many grantmakers or donors are comfortable with. Correspondingly, we've only been able to move about 100k per annum so far in this area (compared to something like mental health, where we helped a partner move 10x that).

Hi Vasco,

Apologies if I didn't explain clearly. Yes, the 10% estimate from GWWC was used as a sense-check, against our own calculation based on the assumptions laid out above (i.e. 1st decile of funding is 10 units of impact out of 55, 2nd decile is 9/55, 3rd decile is 8/55 ... 8th decile is 3/55, 9th decile is 2/55, and 10th & final decile is 1/55 - and since 70% of the budget is already funded, the remaining 30% is 3+2+1=6 units of impact out of 55).

Definitely not scientific, but I wanted to model a smooth decline across each decile of funding, and I ended up not worrying too much as Sjir's own subjective assessment converging with ours.

Hi Vasco,

Just copying over the analysis from within the spreadsheet:

• We expect that the typical funding decision will be made in a context where GWWC has already raised - or will likely go on to raise - a meaningful portion of its budget anyway from other donors.
• Hence, I discount for diminishing marginal returns, based on (a) the assumption that the initial 10% of baseline funding has 10 units of marginal returns, and marginal returns decline by 1 unit for every 10% increase in baseline funding; and (b) the assumption that GWWC's funding situation at the end of 2023 (about 70% of baseline budget expected to be secured) is representative of how much funding will usually be left after institutional and other reliable individual donors have made their contributions or have had their expected contributions factored in. In other words, we are assessing the value specifically of filling the last 30% of GWWC's baseline budget.
• Note that we have gotten an independent estimate from GWWC as to the likely marginal value of the last 30% of their baseline budget, and they ballpark it at 10%.
• However, the uncertainty is radical; and both we and GWWC expect better calibration once GWWC trials out different pledge outreach strategies over the next year and measures their returns in terms of pledges secured.

Hope that helps!

Hi Vasco,

We ended up deprioritizing this perhaps 2 years ago - I can't remember the precise reason, but it was something alone the lines of being concerned with tractability and suspecting that the estimates would probably fall, potentially even by a couple of magnitudes, especially given how sensitive they are to even small methodological changes.

Right now, CEARCH thinks that food security policies to mitigate nuclear & volcanic winter (i.e. ALLFED-style work but with more of a policy bent) is more promising; particularly more common-sense asks related to trade, crop relocation, redistribution (rather than technology-focused asks).

As a matter of fact, early this year we finished evaluating project proposals from ALLFED and some other GCR policy organizations for food security/nuclear winter projects, and have been recommending grants to some of our GCR-inclined donor partners. Unfortunately, there's been no concrete interest so far, and we've not been able to move any money in this area, compared to GHD (where we've funded or advised the funding of stuff in nutrition policy and mental health). All things considered, nuclear is definitely more neglected within EA compared to AI and bio.

Hope this is helpful, and thanks for the new tags!


Joel