You are welcome!

re: Pure Earth: GiveWell notes that its Pure Earth estimates are "substantially less rigorous than both our top charity cost-effectiveness estimates," so I don't want to read too much into it. However, a claim that an intervention is merely 18X better at helping poor people than they are at helping themselves still strikes me as extraordinary, albeit in a way that we become acclimated to over time.

For reference, Pure Earth's estimate of 1 DALY/$ is 5.59 (= 1/(18*0.00994)) times GiveWell's estimate.

I've followed your work a bit w.r.t. animal welfare.

Thanks!

That's 15 chicken DALYs right?

No, it is 15.0 "normal" DALY/$, i.e. 1.51 k times (= $15.0/0.00994$) as cost-effective as GiveWell's top charities.

Hi Seth,

GiveWell's 2021 analysis of Pure Earth links to a sheet according to which its cost-effectiveveness is 18 times that of unconditional cash transfers.

Pure Earth is a GiveWell grantee that works to reduce lead and mercury exposure. In an August 2023 post, they provided a "preliminary analysis" suggesting that their lead reduction program in Bangladesh "can avert an equivalent DALY for just under $1."

I estimated corporate campaigns for chicken welfare have a cost-effectiveness of 15.0 DALY/$.

Thanks for the suggestion, Seth. I am a little wary of adding "disability" to the title, because I have not relied on actual disability of farmed animals:

In agreement with the above [description of how I calculated the annual disability of farmed animals], disability of farmed animals throughout this post refers to the potential for increasing their (affective) welfare up to the level of fully healthy animals. In contrast, the global burden of disease study (GBD) focuses on actual disability.

Interesting! Open Philanthropy (OP) granted 8.3 M$ to THL in 2023, and Animal Charity Evaluators' 2023 review of THL mentioned a funding gap for 2024 and 2025 of 10.5 M$. So I assume OP's last $ going to THL each year is either at or above OP's cost-effectiveness bar (it could be above because OP may not want to provide more than a certain fraction of the total funding of THL). However, in this case, I do not understand why the cost-effectiveness implied by Emily's statement differs from THL's estimate. @Martin Gould or @EmmaTheresa may have thoughts on this.

I clicked on the link over "just $2.63 to spare a hen" on the page from THL you linked, but it is broken^[1].

1. ^{^}

   I sent an email to info@thehumaneleague.org informing THL about it, and asking them if they could share how they obtained their cost-effectiveness estimate.

Thanks for the kind words, CB!

Thanks for the comment, Kaleem.

In a nutshell, both globally and in China:

• Farmed cows and pigs (and other mammals) account for a tiny fraction of the disability of the farmed animals I analysed.
• The annual disability of farmed animals is much larger than that of humans, even under the arguably very optimistic assumption of all farmed animals having neutral lives.
• The annual funding helping farmed animals is much smaller than that helping humans.

I think the 1st point holds for most countries (not only China), and the 2nd and 3rd for basically all countries. I could have a title like "Farmed animals are neglected, both globally and China". However, I think this could be read as farmed animal welfare being more neglected relative to its scale in China than in other countries. I believe this is true^[1], but it is not necessarily implied by the points above.

1. ^{^}

   I estimate China accounts for 33.3 % of the disability of farmed animals, but only 2.20 % of the philanthropic spending on farmed animals. I suppose one could argue these numbers are interesting, and the title could reflect them, but I am wary of communicating that more funding should to China without having looked into the respective cost-effectiveness (i.e. not only scale and neglectedness, but also tractability, which is arguably lower in China).

I think if we prevent an extinction event in the 21st century, the natural assumption is that probability mass is evenly distributed over all other futures, and we need to make arguments in specific cases as to why this isn't the case.

I make some specific arguments:

As far as I can tell, the (posterior) counterfactual impact of interventions whose effects can be accurately measured, like ones in global health and development, decays to 0 as time goes by, and can be modelled as increasing the value of the world for a few years or decades, far from astronomically.

[...]

Here are some intuition pumps for why reducing the nearterm risk of human extinction says practically nothing about changes to the expected value of the future. In terms of:

• Human life expectancy:
  • I have around 1 life of value left, whereas I calculated an expected value of the future of 1.40*10^52 lives.
  • Ensuring the future survives over 1 year, i.e. over 8*10^7 lives (= 8*10^(9 - 2)) for a lifespan of 100 years, is analogous to ensuring I survive over 5.71*10^-45 lives (= 8*10^7/(1.40*10^52)), i.e. over 1.80*10^-35 seconds (= 5.71*10^-45*10^2*365.25*86400).
  • Decreasing my risk of death over such an infinitesimal period of time says basically nothing about whether I have significantly extended my life expectancy. In addition, I should be a priori very sceptical about claims that the expected value of my life will be significantly determined over that period (e.g. because my risk of death is concentrated there).
  • Similarly, I am guessing decreasing the nearterm risk of human extinction says practically nothing about changes to the expected value of the future. Additionally, I should be a priori very sceptical about claims that the expected value of the future will be significantly determined over the next few decades (e.g. because we are in a time of perils).
• A missing pen:
  • If I leave my desk for 10 min, and a pen is missing when I come back, I should not assume the pen is equally likely to be in any 2 points inside a sphere of radius 180 M km (= 10*60*3*10^8) centred on my desk. Assuming the pen is around 180 M km away would be even less valid.
  • The probability of the pen being in my home will be much higher than outside it. The probability of being outside Portugal will be negligible, but the probability of being outside Europe even lower, and in Mars even lower still^[5].
  • Similarly, if an intervention makes the least valuable future worlds less likely, I should not assume the missing probability mass is as likely to be in slightly more valuable worlds as in astronomically valuable worlds. Assuming the probability mass is all moved to the astronomically valuable worlds would be even less valid.
• Moving mass:
  • For a given cost/effort, the amount of physical mass one can transfer from one point to another decreases with the distance between them. If the distance is sufficiently large, basically no mass can be transferred.
  • Similarly, the probability mass which is transferred from the least valuable worlds to more valuable ones decreases with the distance (in value) between them. If the world is sufficiently faraway (valuable), basically no mass can be transferred.

Thanks for the context, MHR.

Their ratings (higher = better) for their recommended charities are

Is there a single page with all the scores, or did you check the cost-effectiveness sheet of each recommended charity?

My personal advice would be that I think the EA Funds Animal Welfare Fund is probably the expected value maximizing option, while The Humane League is probably the best option if you're somewhat risk-averse.

I used to prefer the Animal Welfare Fund (AWF) too, but now think THL may well be the best option. It looks like AWF pays to little attention to cost-effectiveness. From Giving What We Can's evaluation of AWF (emphasis mine):

Fourth, we saw some references to the numbers of animals that could be affected if an intervention went well, but we didn’t see any attempt at back-of-the-envelope calculations to get a rough sense of the cost-effectiveness of a grant, nor any direct comparison across grants to calibrate scoring. We appreciate it won’t be possible to come up with useful quantitative estimates and comparisons in all or even most cases, especially given the limited time fund managers have to review applications, but we think there were cases among the grants we reviewed where this was possible (both quantifying and comparing to a benchmark) — including one case in which the applicant provided a cost-effectiveness analysis themselves, but this wasn’t then considered by the PI in their main reasoning for the grant.

Thanks, Joris, and welcome to the EA Forum!

What I'm worried about is the error bars - multiplying errors can cause wild differences between the estimated and actual numbers. If the error bars of the two funds (CCF and TCF) overlap significantly, it might be too soon to judge which one is best.

Agreed:

• I estimated the cost-effectiveness of CCF is:
  • 3.28 times that of TCF, with a plausible range of 0.175 to 30.2 times. So it is unclear to me whether donors interested in improving nearterm human welfare had better donate to GiveWell’s funds or CCF.

Thanks SummaryBot!

Results are sensitive to the distribution type, but focusing on the far right tail is most relevant for extinction risk.

I guess reasonable distribution types will lead to astronomically low extinction risk as long as one focusses on the rightmost points of the tail distribution.

Extraordinary evidence would be needed to justify a meaningfully higher risk estimate.

To clarify:

• Extraordinary evidence would be required to move up sufficiently many orders of magnitude for an AI, bio or nuclear conflict to have a decent chance of causing human extinction. I think underweighting the outside view is a major reason leading to overly high risk.