I run the Centre for Exploratory Altruism Research (CEARCH), a cause prioritization research and grantmaking organization.
Joel Tan
Karma: 1,488
Hi Jaime, I’ve updated to clarify that the “MEV” column is just “DALYs per USD 100,000″. Have hidden some of the other columns (they’re just for internal administrative/labelling purposes).
Thanks for the thoughts, Jaime and Nick!
For what it’s worth, CEARCH’s list of evaluated causes (or more specifically, top interventions in various causes) and their estimated cost-effectiveness is here: https://docs.google.com/spreadsheets/d/14y9IGAyS6s4kbDLGQCI6_qOhqnbn2jhCfF1o2GfyjQg/edit#gid=0
I think Nick is fundamentally correct that because uncertainty is so high, sorting isn’t particularly useful. Most grantmaking organizations, to my understanding, prefer to use a cost-effectiveness threshold/funding bar, to decide whether or not to recommend/support a particular cause/intervention/charity.
For ourselves, we use 10x GiveWell for GHD, as (a) most of the money we move is EA and the counterfactual is GiveWell (so to have impact we the ideas we redirect funding/talent to be more cost-effective than GiveWell in expectation, and (b) we have such an aggressive bar because GiveWell is very robust in their discounting relative to us (which takes a lot of time and effort). An aggressive bar helps ensure that even if your estimated cost-effectiveness estimate is too optimistic relative to GiveWell, it can eat a lot of implicit discounts while still ensuring that the true cost-effectiveness is >GiveWell. (so when we say something is >=10x GiveWell it’s not literally so, more of a reasonably high confidence claim that it’s probably more cost-effective (in expectation).
If I’m understanding this concern correctly, it’s along the lines of: “they’re not making a financial sacrifice in shutting down, so it’s less praiseworthy than it otherwise would be”.
Just to clarify, charity founders (at least CE ones) take a pay cut to start their charity—they would earn more if working for other EA organizations as employees, and much more if in tech/finance/consulting/careers that typical of people with oxbridge/ivy/etc education levels. The financial sacrifice was already made when starting the charity, and if anything, quitting is actually better for you financially.
Disclosure: Sarah and Ben are friends, and we came out of the same CE incubation batch, so I’m not unbiased here.
I think it’s speaks well of a person’s integrity, objectivity, and concern for impact that they’re able to make a clear eyed assessment that their own project isnt having the desired impact, and then going ahead to shut it down so as to not burn counterfactually valuable resources.
It’s something that’s worth emulating, and I do try to apply this myself—via regular CEAs and qualitative evaluations of CEAECH’s expected impact (especially as a meta org with a more indirect path to impact). We’re only wasting our own time otherwise!
Hi Vasco,
The GiveWell team handling the nutrition portfolio reached out to me to discuss salt policy, and SSB taxes are on their longlist, iirc. Of note is the fact that Vital Strategies (which have gotten GiveWell grants for their alcohol policy work, also does SSB tax advocacy).
I’ve generally moved to the view that geomeans are better in cases where the different estimates don’t capture a real difference but rather a difference in methodology (while using the arithmetic makes sense when we are capturing a real difference, e.g. if an intervention affects a bunch of people differently).
In any case, this report is definitely superseded/out-of-date; Stan’s upcoming final report on abrupt sunlight reduction scenarios is far more representative of CEARCH’s current thinking on the issue. (Thanks for your inputs on ASRS, by the way, Vasco!)
Hi Nick!
Yep, that’s definitely a concern for governments (same with other policy interventions for nutrition). For funders—to be fair, that’s not much different from direct delivery (e.g. for vaccinations or contraception, we can’t really know the impact until we finish our M&E and see the uptake rates/disease rates change)
Hi Mo,
I don’t think I read that part of Michael’s thesis before, but it does look interesting!
In general, I think it’s fairly arbitrary what a cause is—an intervention/solution can also be reframed as a problem (and hence a cause) through negation (e.g. physical activity is a preventative solution to various diseases like cardiovascular disease or diabetes, and in a real sense physical inactivity is a problem; having an ALLFED-style resilient food supply is a mitigatory solution to nuclear winter—even if we can’t prevent nuclear exchange, we can perhaps stop billions from dying from famine—and in that sense lack of foods capable of growing in abrupt sunlight reduction scenarios is a problem).
Yep, that’s what Jeroen submitted, and he posted it to the forum after. I think it’s a really useful perspective to have. In GHD, we already have various risk factors that we can address to solve multiple diseases (e.g. hypertension for coronary heart disease, stroke, kidney disease etc), and it makes sense to apply this perspective more broadly.
Thanks for the thoughtful comment!
(1) On the evidence base for mass media interventions: Our choice to prioritize mass media (and so attempt more detailed modelling) was based on it (a) looking good in cost-effectiveness at a a shallower research stage with a less; (b) being recommended by the WHO as the most cost-effective intervention for promoting physical activity; and (c) generally, mass media interventions being 2nd only to policy in cheap scalability.
We relied on the study you cited (Abioye, Hajifathalian & Danaei), and took into consideration all three meta-analyses (impact on sedentary behaviour, sufficient walking, and sufficient physical activity), while discounting based on the degree to which measured outcome is dissimilar to GBD/WHO definitions of sufficient physical activity (>= 600 METs minutes per week). We did not exclude the sedentary meta-analysis results based on the fact that its reliant on a low quality outlier, but did perform our own analysis, to discount it (and the other meta-analyses) on the basis of underlying study quality (especially with respect to endogeneity) and publication bias.
We also considered whether the studies being from high-income countries biases the result (n.b. on the one hand, ageing high income countries are older and more sedentary, which implies a larger group of potential beneficiaries and greater population-level effect size; on the other hand, as Rosie Bettle of FP notes in her report on mass media interventions, its potentially the case that basic healthcare knowledge is more lacking in poorer countries, so you can—as with vaccine uptake—increase uptake there in a way).
Overall, we do think there’s is an effect, but it’s highly uncertain, and there’s a real chance it’s zero. For more details, see the ultra-long (and fairly tedious) discussion in this cell (Tractability:B12) of the CEA.
(2) Built environment changes: We did take a look at this option (seeAnnex A in our CEA for a qualitative discussion of the available intervention options, and the relevant evidence base/expert feedback). We ultimately prioritized mass media (for reasons discussed above), but I do think such environment changes will be substantially more impactful but also much, much harder.
For one, if we’re talking about not just high-income countries but LMICs, the situation is very different. Zoning is just much less a factor than in the rich world (they may not have much on the books, and what’s on the books may not be enforced), so the problem they face isn’t the one that the US/UK etc face and which YIMBYs are trying to solve. Their major cities are also extremely dense, and people have low vehicle ownership rates (but worse congestion and air pollution). Making these cities more walkable isn’t just a matter of allowing dense housing, but spending a lot of money to improve public transport, solving air pollution etc, and that’s a fundamentally much harder ask for poor countries.
I used Abioye, Hajifathalian & Danaei on mass media, and Kang et al on pedometers. We also looked at a bunch of other interventions (e.g. built environment measures, digital interventions, point-of-decision prompts like posters by stairs, etc) - do take a look at Annex A in our CEA for a qualitative discussion of the available intervention options, the relevant evidence base/expert feedback, and why we prioritized government public education campaigns.
You can always reach out via the forum!
https://forum.effectivealtruism.org/posts/W93Pt7xch7eyrkZ7f/cause-area-report-antimicrobial-resistance
Thoughts about EGI vs policy advocacy, Joey? Odds of success will should be lower but expected impact higher than either DD/EGI. For learning value I imagine it’s something in between—strong emphasis on communications, but you’re also working in a local LMIC context, and forced to adapt, unlike EGI attempts based in your home country.
It depends on whether you take a person-affecting view or totalist view. On the former, the baby isn’t a person (and won’t be, if you don’t save him/her), and so does not and never will have actual preferences in not dying (since that requires fairly sophisticated views of self over time etc). Hence, despite the fact that saving the baby will achieve comparatively more QALYs, there’s nothing wrong in saving the mother instead. I think this makes sense, though not everyone accepts PAC (e.g. Richard, who also posted here, and has an excellent philosophy blog worth reading, tends towards a more totalist view in population ethics).
For what it’s worth, it’s also important to think about the implications that your position here will have on abortion, since any totalist view here has repugnant conclusions there.
You might want to reach out to Akhil Bansal, who was a Schmidt officer on AMR and now runs an AMR funding circle. He’ll be very up to date on this issue.
While the idea of moral licensing makes sense to me in theory, I’m not too persuaded by the empirical evidence, at least from the cited meta-analysis—the publication bias is enormous, as the authors note.
GiveWell has funded Vital Strategy’s alcohol work, OP has their global health policy focus area (inclusive of alcohol) and CE has incubated the Centre for Alcohol Policy Solutions (though I have limited visibility on their success since incubation a few years ago).
Check out CE’s report on alcohol and tobacco for a short primer; you can also compare their assessment of success rates and neglectedness.
(1) I think Joey’s right, and I’ll phrase the issue in this way—a lot of EAs underrate the impact of habit-formation and overrate the extent to which most of your choices even require active willpower. Your choices change who you are as a person, so what was once hard becomes easy.
I’ve always given at least 10% to effective charities, and now it’s just something I do; it’s barely something I have to think about, let alone require some heroic exertion of will. And while I’m not vegan, I am successfully eating less meat even on a largel6 keto diet, and what surprised me is how much easier it is than I thought it would be.
(2) Let’s accept for the sake of argument that there is a lot of heterogeneity, such that for some people the impact of habit formation is weak and it is psychologically very difficult for them to consistently adhere to non-job avenues to impact (e.g. donating, being vegan, etc). Even so, how would one know in advance? Why not test it out, to see if you’re in the group for which habit formation impact is high and these sacrifices are easy, or if you are in the other group?
Surely it’s worth doing—the potential impact is significant, and if it’s too hard you can of course stop! But many people will be surprised, I think, at just how easy certain things are when they become part of your daily routine.
GiveWell is available (search Clear Fund)!
I currently do direct work—my organization CEARCH researches cost-effective ideas in GHD/longtermism/meta, and works to direct resources in support of particular promising ideas (e.g. via grantmaking, donor advisory, working with Charity Entreneurship and other talent orgs).
However, for most of my career, I was doing a non-EA job (policy work in government and as a consultant), and I engaged with EA simply by giving money to GiveWell. I’ve been a GWWC pledger since 2014, and that to me is classic EA, and the furthest thing from being less engaged or less EA (than someone who does direct work but doesn’t donate).
Edit: And beyond having impact via your donations, you can always attend events (particularly EAGxs) - I think it’s super valuable for younger EAs to get advice from older folks who primarily live and work in non-EA environments, since younger EAs can get stuck in a social and professional environment that is unadulterated EA, the end result of which is adopting a bunch of norms and behaviours that may leave them less effective at achieving impact (e.g. unprofessional workplace or organizational norms, since they literally haven’t worked in a non-EA organization before; or not being used to persuading and engaging non-EA folks, including in government or in corporate environments etc).