Hi Wayne, thatâs fair. I hadnât been including farmed animal welfare in the comparison because I donât think people donating to therapy organizations are doing it for animal welfare reasons.
I donât think it would be practical for givewell to include animal welfare in its evaluations. I think donors who care about both animal and human welfare would have more impact giving to separate projects optimising for each of those goals
Nice post Joey, thanks for laying it out so clearly.
I agree with almost all of this. I find it interesting to think more about which domains /â dimensions Iâd prefer to push towards prioritization vs. pluralism:
Speaking loosely, I think EA could push more towards pluralism for career decisions (where personal fit, talent absorbency and specialization are important), and FAW/âGHD/âGCR cause prioritization (where I at least feel swamped by uncertainty). But Iâm pretty unsure on where to shift on the margin in other domains like GiveWell style GHD direct delivery (where money is ~fungible, and comparisons can be meaningful).
e.g. I suspect Iâm more willing to prioritize than you on bednets vs. therapy. I think you /â AIM are more positive than me about therapy (as currently delivered) on the merits. Sure, thereâs a lot of uncertainty, but having spent a bit of time with the CEAs, I just find it real hard to get to therapy being more cost-effective than bednets in high burden areas.
you could pretty easily imagine a GW-like charity evaluator that ranks income as x4 as important as GW does coming to pretty different but still highly compelling top charities
I agree moral weights are one of the more uncertain parameters, though I think the range of reasonable disagreement given current evidence is a bit less wide than implied here. Iâd love to see someone dive deep on the question and actually make the case that we should be using moral weights for income 4x higher vs. health, rather than theyâre plausible.
I guess a general theme is that I worry about a tendency to string together lots of âplausibleâ assumptions without defending them as their best guess, and that eroding a prioritization mindset. I think youâd probably agree with that in general, but suspect we have different practical views on some specifics.
Hi Midtermist, I think this is a pretty important worry and appreciate you sharing your perspective.
Just speaking for myself and the EA (global health and wellbeing) program I work on (though itâs mostly led by Mel Basnak now).
Here are a few things weâre doing:
We fund Probably Good, who try to empower people to think along impact-focused lines while remaining open-minded about how different people can best help others.
We support orgs like Founders Pledge and Charity Entrepreneurship, who share our core values but who do their own research and might have different views.
Our grants to effective giving orgs have not been conditional on them changing their recommendations (though weâre more likely to support orgs whose recommendations we believe in).
Tbc, I think itâs a tough problem and tradeoffs between respecting autonomy and standing behind your judgment calls as a grantmaker.
Yes, I think thatâs mostly fair.
Air pollution in South Asia has a lot of different sources requiring distinct policy interventions, and often at local levels. Eliminating emissions entirely from any one source category would address a relatively small fraction (5%-10%, say) of the problem in a given city or state.
Some interventions that are relatively scalable across India, and likely to be effective are politically intractable: power plant emission controls (expensive and with costs borne by a small group of influential firms), or subsidies to make cleaner fuels affordable for household cooking (expensiveâneeds an outlay of at least $1B/âyear).
That said, about $5M/â $18M spent so far in the program in India fund organizations that work with government agencies in identifying and executing interventions, leveraging existing government resources.
Some of the promising ideas so far that our grantees are working on include better handling of construction dust by large private developers, and support for city governments to handle dispersed sources (municipal waste burning, resuspension of road dust). We also think that crop residue burning in northern India has seen steady (if slow) improvement, and expect to continue supporting governments in this process.
Thanks ajyl! I think the biggest takeaway for me is that you can do a bit of storytelling without compromising on rigor /â honesty.
The ITN framework is compelling and appeals to people outside EA.
Jenna Forsythâs work in Bangladesh really resonated with people. Combining that story with an estimate of 20,000 lives saved seemed to hit home for a lot of folks.
Comparison is helpful. Lead exposure is estimated to kill 1.5 million people but receives only $15m in funding is compelling to people who can put those numbers in perspective. But lead exposure is estimated to kill more people than HIV/âAIDS and malaria combined, but receives ~1,000x less funding is much more universally powerful.
I think focusing on our plan in the funding proposal (measurement, mitigation, mainstreaming), and setting measurable goals against that was helpful.
Youâre right that was referring to LEAF. Weâve been excited at how well this collaborationâs gone, so weâre thinking about what else we might be able to do in a similar vein, but havenât decided what thatâll look like yet.
Thanks Rafael! A few theories:
There just arenât many people whose job it is to look for important, tractable and neglected causes. Itâs striking how much of the recent upsurge in interest has been related to EA.
While there are some individual cases of very severe exposure, most of the burden is caused by relatively small risks of harm spread across entire populations, so victims arenât identifiable.
A lot of global health funding is allocated by disease category. Lead exposure cuts across disease categories, so it just doesnât naturally fit into a lot of funding mechanisms.
The elimination of leaded gasoline had a huge positive impact globally. The US, and a few other countries, also do regular testing and surveillance to identify other sources of exposure. But not many other countries do that. So I think thereâs a story that with the decline in US blood lead levels, people assumed the problem was fixed by eliminating leaded gasoline when other countries, particularly LMICs, still have a lot of exposure from other sources.
The most direct focus of LEAFâs source-specific mitigation work is paint and spices, but thatâs largely because of tractability: these are both products where there are only weak economic incentives to use lead, and where production is fairly consolidated. That makes them easier to regulate. For other sources (batteries, cookware, cosmetics etc.), we want to fund more exploratory work: piloting and testing regulatory interventions rather than scaling them.
I think figuring out what to do on informal ULAB recycling is really important. ~80% of global lead is used in lead acid batteries, and informal recycling is responsible for particularly severe cases of exposure for people living nearby.
A benefit of all the recent attention on lead exposure is I expect weâll increasingly see requests for assistance from governments, so it would be very helpful to have âbest practiceâ playbooks we could apply to ULAB regulation. To my knowledge, those donât currently exist (though there are some resources here and here). Intuitively, Iâm excited by market-based solutions like tax breaks or even subsidies for well-regulated formal sector battery recycling to make it cost-competitive with unsafe informal recycling. But thatâs pretty weakly held. Pure Earth have been thinking about this a lot, as have the US EPA and others.
Thanks Parth, I appreciate it and thank you for your support along the way!
How did you first learn about lead as an issue? I know you were looking into public health regulation broadly at that time (e.g. pesticides) and in your conversation notes from 2017 you cite various statistics (e.g. WHO statistic that 10% of children globally have >20 micrograms of lead NYU study on $1 trillion of economic costs). What first rang the alarm bells for you that this was an area worth investigating?
IIRC, I first learned about it as a development issue from some very early work my colleague Andrew did at GiveWell. At the time, GiveWell wasnât funding policy work, so itâd been deprioritized. But Iâd got interested in public health policy from investigating this grant on pesticide suicide when I was at Giving What We Can, so I was pretty keen to dig into it.
But it was a pretty slow burn and I only got really excited about it between 2019 and 2021. Things I found particularly persuasive were this systematic review suggesting the median kid in an LMIC had elevated blood lead by US standards, Pure Earth telling me about Jenna Forsythâs work in Bangladesh to remove lead from spices, and this comparative analysis highlighting its relative neglectedness, even compared to other hugely neglected issues like tobacco.
How was Pure Earth so far ahead of everyone on this? They published their first âWorldâs Most Polluted Places Reportâ in 2008 and wrote in it: âRelative to other public health interventions, pollution remediation can be very cost effective..⌠projects cost between $1 - $50 per year of life gained. This compares favorably to the $35 to $200 per year of life gained for World Bank estimates on interventions related to water supply, improved cooking stoves and malaria controls.â
Iâm not sure! When I first started talking to Pure Earth, theyâd recently started to focus more narrowly on lead exposure, having previously worked across a bunch of different pollutants. They were also starting to explore focusing on more regulatory interventions when before they were mostly cleaning up individual toxic waste sites. It was pretty cool to see an organization really shifting and following the burden like that. Fwiw I havenât reviewed the $1-$50 estimate, but Iâm skeptical of the claim if itâs referring to remediating individual toxic waste sites (though so much is down to how many years of âspeedupâ to count that itâs hard to know without digging in).
Besides your $250K grant to IPEN in July 2019, why did it take you nearly four years from when you started investigating the topic in late 2017 to start funding work in the space? Relatedly, why did it take you so long to fund LEEP, and with relatively small grants to-date?
Yeah, I think about this a lot. If we believe the GBD estimates, 6 million people died from lead exposure in the four years it took me to recommend substantial grants in the space. So even though that certainly wasnât the only bottleneck, itâs a source of personal angst.
The short (and probably unsatisfying) answer is I was working on a lot of other things. I think over the ~5 years I spent at GiveWell, ~25% of my time was on public health policy, and lead was only a fraction of that. With hindsight, I wish Iâd advocated more effectively for GiveWell to put more capacity towards that work (though itâs also worth noting that a lot of the most compelling data points werenât available until ~2020).
On LEEP specifically, we were pretty close to making a grant at GiveWell, but Iâd got hung up on whether I believed the evidence for lead paint being a significant source of exposure (I still find that evidence tricky to interpret, but lead paint regulation has also been far more tractable than I expected such that I think it should be a priority anyway). When I moved to Open Phil in 2022, it was to launch the EA program on global health and wellbeing. We also agreed to transition the lead portfolio and other public health policy grants from GiveWell to Open Phil, but it took a bit of time for us to get to launching a program. I agree with Alexanderâs assessment of our mistakes here. Fwiw this has all made me quite appreciative of other funders in the EA space who funded LEEP. I admire their work, and itâs been fun working with Lucia and Clare (LEEPâs co-EDs) in the build up to the launch.
Open Philâs main role was organizing a group of donors to commit most of the funding which was announced alongside the Partnership for a Lead-Free Future. The Lead Exposure Action Fund comprises $104m of the $150m that was announced, and Open Phil staff will manage the allocation of those funds. The work to launch a collaborative fund was quite far along when we formalized a partnership with USAID, so itâs an interesting coming together of two strands.
Weâve also been working very closely with USAID and UNICEF to help design the PLF (and weâre still working together to flesh it out), and have pledged to co-fund a portion of the PLFâs operating costs.
Overall, itâs been a really positive (and fun!) experience working with USAID. Smoother than I wouldâve guessed. I think thereâs a few things that have helped with that.
First, lead exposure is just really compelling on impact grounds; Samantha Power and Atul Gawande are convinced about it for the same reasons we are. Two generalizable takeaways for me were that the ITN framework is convincing outside narrow EA circles, and people at the very highest levels of government resonate with cause prioritization. That makes sense: theyâre the people who have to make decisions at that level. Both those make me optimistic about the potential for more collaboration.
Second, weâve been working with some really cool people who share our values. Samantha Powerâs senior adviser, Garrett Lam, is the person who first brought lead exposure to Samantha Powerâs attention, and I believe that was partly due to various touchpoints with this community (but wouldnât want to speak too much for him).
Thanks Nick, thatâs very kind of you to say. Iâve also really appreciated your contributions to the forum.
Yes! Iâve updated on how quickly smart and driven people can build networks and expertise to make an impact on policy, especially in neglected areas.
I do think lead exposure (and particularly paint) was a wise choice for LEEPâs founders who were starting out with less previous policy experience. Itâs a fairly technocratic regulatory intervention without much opposition, the areaâs neglected so itâs easier to build your network, and there are already model laws and various international agreements in place. And I also think LEEPâs team is unusually capable!
Other kinds of policy advocacy (e.g. global aid policy in countries like the U.S.) are pretty different in my experience. Speaking loosely, itâs a more crowded space with many established players, and often involves navigating complex political landscapes and competing priorities. I wouldnât want to write off less experienced people having an impact here, but coming in cold does seem tough. The Partnership for a Lead-Free Future shows we can have an impact there as well, but I think that story relied at least in part on strong networks and subject matter expertise.
Yes! Giving What We Can is how I first got involved with effective altruism, and Iâve given at least 10% of my income since graduating. This year, I gave to the Lead Exposure Elimination Project, Effektiv Spenden, and the GiveWell All Grants Fund. I think theyâre all awesome organizations, and I feel confident in personally standing behind them, given how well I know their work.
In 2017, I split my donation across cause areas and part of the reason was to signal that I believed in the importance of GCR and animal welfare work. I also thought giving might motivate me to learn more about those areas. But I donât think the signaling really mattered, and as I became more convinced health policy was an area I could make a difference in, forming a more concrete view on high level cross-cause comparisons didnât seem like a priority.
So since then Iâve mostly donated to orgs I know well through my work. As well as feeling like I could stand behind those decisions, I also see it as a gesture of appreciation to grantees and putting my money where my mouth is.
Weâre limiting ourselves to a fraction of GWWCâs funding. At the moment, thatâs in the region of 70%, but GWWC and OP are aligned that we expect to dial it down substantially over time, both to enable a more diverse funding base for GWWC and to free up programmatic budget for other organizations.
Fwiw I disagree that OPâs bar is 2x the cost-effectiveness of GiveWellâs top charities in practice. In my view (not an official OP position or anything) differences in modeled cost-effectiveness between GW and OPâs bar are indistinguishable from noise.
Hi Evan, thank you for the kind words!
I wanted to clarify one point:
>It could exceed the value of all the money that has ever flowed through EA to any charity GiveWell has ever recommended.
LEAF and the Partnership for a Lead-Free Future wouldnât have happened without GiveWell. They were the first major funder in the EA space to recommend grants for lead exposure, and much of the recent progress comes from those early investments, such as grants to Pure Earth and CGD. I also believe (though Iâm less certain) that Charity Entrepreneurshipâs recommendation of lead paint regulation and its incubation of LEEP were based largely on GiveWellâs research.
I know that wasnât really what you were saying, but felt compelled to mention because Iâve heard lead exposure used as an example of what GiveWell gets wrong, when, in fact, GiveWell was earlier to lead exposure than almost anyone, and Iâm grateful to former colleagues who put in a lot of hard work.
(For transparency: I was at GiveWell during this time and have since moved to Open Phil where I work on LEAF. I also think GiveWell made mistakes on lead exposure, many of them mine)
You might find this article helpful for context: https://ââundark.org/ââ2023/ââ07/ââ19/ââthe-vice-of-spice-confronting-lead-tainted-turmeric/ââ
Fwiw Iâm sympathetic to your general point
Implicitly, yes. Though donât use that exact formulation. money vs daly comparison is based on reported preference not swb. Daly vs swb comparison implicitly writes off time spent asleep where I assumed 1 daly = difference between 40->100 on swb scale.
If didnât exclude sleep in botec, would make alcohol look worse as happiness bump from alcohol would be for lower % of time. (Set row 21 to 24)
>Another thoughtâyou measure the effects of alcohol on subjective wellbeing as a fraction of someoneâs waking hours. This seems right from a subjective wellbeing perspective. But is that also the way you think about the value lost by a death? By consistency, you would also need to implicitly downweight the disvalue a death by a third for the time people spend asleep. Or do you already do that in your moral weights?
Oh thatâs interesting. Itâs been a while now since I did this, but I think I was implicitly doing that with this calc
>Iâm not sure I follow the claim that if you assume that alcohol taxation merely shifts the tax burden, there arenât strong reasons to think the deadweight loss will be greater from alcohol taxation vs other forms of taxation. The subjective wellbeing study found that drinking increases peopleâs wellbeing by almost as much as spending time with friends. It seems unlikely to me that if the tax were instead eg on income that the benefits of the income would be as large as this. Intuitively, this seems off.
Interesting. That doesnât seem off to me. If Iâm understanding correctly, the implication of your view is that people would generally be better off if they consumed more alcohol and less of other goods on the margin. Is that right?
To put it another way: increasing taxes on alcohol has two effects on consumer surplus: (i) deadweight loss (ii) a transfer from consumers to the government. I think (ii) is probably positive. Almost all taxes involve some amount of deadweight loss, but we do them anyway because we think public goods and redistribution are worth it.
TBC, Iâm not claiming that higher excise taxes on alcohol relative to other goods merely shifts the tax burden. If we assume perfect rationality (which I believe would be mistaken), having unequal marginal taxes between goods does result in some additional deadweight loss. But it is a counterveiling factor.
>On your botec on the benefits of alcohol, a lot rides on you assuming that a death from alcohol accounts for 40 units of value
A unit of value (in GiveWellâs terms) is equivalent to doubling consumption for a person for a year. A DALY is 2.3 units of value. So you want to be dividing your estimates by 2.3.
The Global Burden of Disease estimates ~30 YLLs and ~10 YLDs per death (I didnât include YLDs in the BOTEC and I underestimated YLLs which makes it conservative, though I also didnât discount the GBD estimates for imperfect evidence quality and black market consumption not addressable through policy which makes it optimistic. Iâd guess these ~cancel).
Edit: didnât see your second comment when writing this where you saw this
>On the other hand, none of this considers hangovers.
Yeah, although interestingly (IIRC) the Baumberg study didnât find any effect on SWB the day after drinking (though Iâm skepticalâmaybe people didnât feel like inputting how sad they were on their phone when they were hungover!)
Any thoughts on the impact multiple of funding them then vs. funding them now?