Did you ask ACE to review this before publishing? It seems like the kind of thing that would be worth getting feedback on before publishing. I didn’t look at this for more than a couple minutes, but I saw immediately that there might be some conceptual disagreements between you and ACE—for example, I noticed that in your first example, you assume in your example (I believe), that if LIC didn’t spend 200k on the lawsuit against Costco, they wouldn’t spend it on anything else. It’s unclear to me that this is the counterfactual, or how ACE is conceptualizing those funds. There might be reasoning behind their decisionmaking that would be useful to your critiques they could share.
I also felt like this felt pretty politically motivated. Not sure if that is your intention, but paragraphs like this:
ACE’s recommendations determine which animal charities receive millions of dollars in donations.[1] Thus far, we have reviewed 5 of ACE’s “Top 11 Animal Charities to Donate to in 2024” and only one of them (Shrimp Welfare Project) appears to be an effective charity for helping animals. ACE’s poor evaluation process leads to ineffective charities receiving recommendations, and many animals are suffering as a result.
Without any evidence feels pretty intense. ACE is kind of low hanging fruit to pick on in the EA space, so this read to me like more of that, without necessarily the evidence base to back it. Reading your report, I felt kind of like “oh, there are interesting assumptions here, would be interested to learn more”, and not “ACE is doing an extremely bad job.”
E.g. I think the questions that would be good to ask in a critique of ACE might be:
If ACE didn’t exist, how would the funds the direct be spent otherwise? Would that be better or worse for animals?
Is historical track record / cost-effectiveness the only lens on which to evaluate charities?
If the answer is yes, seems very hard to start new things!
I don’t know if the LIC legal case is this, but celebrating the potential impact of promising bets that didn’t pan out seems good to me.
I also think getting feedback on statements like this would be really helpful:
The correct formula for calculating cost-effectiveness is simply impact divided by cost. Rather than using this simple formula, ACE has elected to create a methodology that does not properly account for impact or cost.
I think ACE has wanted to do this at points in their history — my impression is just that it is incredibly difficult, so they’ve approached it from other angles instead. I also don’t think it’s clear to me that ACE’s goal is to report cost-effectiveness. I think clarifying this with them, and getting a sense of why they don’t do what you see as the simple approach would be useful for making this critique stronger. And, I don’t think people should make giving decisions based only on historic cost-effectiveness—just because an opportunity was impactful doesn’t mean the organization needs more funds to do that work, that it will scale, work in the future, etc.
I don’t disagree that ACE might be directing funds to ineffective charities! I don’t really think non-OpenPhil EA donors should give to farmed animal welfare, for example. But, I don’t think it is obvious to me that ACE going away means money going to more effective charities—I expect it would mostly be worse—people giving to animal charities with basically no vetting.
That being said, critique of critical organizations is great in my opinion, so appreciate you putting this out there!
“I don’t really think non-OpenPhil EA donors should give to farmed animal welfare, for example.” Wow, this is interesting! I would love to know what you mean by this?
Having read your reflections, I’m still curious as to why you don’t think non-OpenPhil donors should give to farmed animal welfare, if you feel comfortable sharing it publicly. I guessed four options, ordered from most to least likely, but I might have misunderstood the post
We should donate to wild animal welfare instead, as it’s more cost-effective
There are no donation opportunities that counterfactually help a significant amount of farmed animals
There is no strong moral obligation to improve future lives, and donations to farmed animal welfare necessarily improve future lives, as farmed animal lives are very short
Tomasik-style arguments on the impact of animal farming on the amount of wild animal suffering
Is it a combination of these? As a concrete example, I’m curious if you believe that the Shrimp Welfare Project shouldn’t be funded, should be funded by “non-EA” donors, or will be funded anyway and donors shouldn’t worry about it.
By the way, thank you for nudging towards sharing evaluations with the evaluated organization before posting, I think it’s a really valuable norm.
Thanks! My wording in the above message was imprecise, but I mean something like farmed vertebrates. SWP is probably among the two most important things to fund, in my opinion.
Basically I think the size of good opportunities in farmed animal advocacy is smaller than OpenPhil’s grantmaking budget and there are few scalable interventions, though I don’t think I want to go into most the reasons publicly. Given that they’ve stopped funding many of what I believe are more cost-effective projects, and that EA donors are basically the only people willing to fund those, EA donors should be mostly inclined to fund things OpenPhil can’t fund instead.
So some combination of 1+2 (for farmed vertebrates) + other factors
I also felt like this felt pretty politically motivated. Not sure if that is your intention, but paragraphs like this:
ACE’s recommendations determine which animal charities receive millions of dollars in donations.[1] Thus far, we have reviewed 5 of ACE’s “Top 11 Animal Charities to Donate to in 2024” and only one of them (Shrimp Welfare Project) appears to be an effective charity for helping animals. ACE’s poor evaluation process leads to ineffective charities receiving recommendations, and many animals are suffering as a result.
Without any evidence feels pretty intense. ACE is kind of low hanging fruit to pick on in the EA space, so this read to me like more of that, without necessarily the evidence base to back it. Reading your report, I felt kind of like “oh, there are interesting assumptions here, would be interested to learn more”, and not “ACE is doing an extremely bad job.”
What claims did we make that we did not provide evidence for?
…we have reviewed 5 of ACE’s “Top 11 Animal Charities to Donate to in 2024” and only one of them (Shrimp Welfare Project) appears to be an effective charity for helping animals. ACE’s poor evaluation process leads to ineffective charities receiving recommendations, and many animals are suffering as a result.
I understand these are forthcoming, but no evidence is provided for this entire part—part of the reason I pushed on this is I think seeing your alternative evaluations would be very helpful for interpreting the strength of the critique of ACE. Without seeing them, I can’t evaluate the latter half of the quoted text. And in my eyes, if these are similar to the evaluation here of LIC, it’s pretty far from demonstrating that ineffective charities are receiving recommendations, etc. And, given that you’ve only evaluated <50% of their charities so far, it seems preemptive to make the overall claim. I think the overall claim is very possibly true, but again, I think to make the argument that animals are directly suffering as a result of this, you’d have to demonstrate that those charities are worse than other donation options, that donors would give to the better options, etc.
Note: this reply addresses everything Abraham claims we did not provide evidence for.
“ACE’s poor evaluation process leads to ineffective charities receiving recommendations”
Our review covered how under ACE’s evaluation process:
Charities can receive a worse Cost-Effectiveness Score by spending less money to achieve the exact same results.
Charities can have 1,000,000 times the impact at the exact same price, and their Cost-Effectiveness Score can remain the same.
The most important factor in determining the impact of an intervention is decided before the intervention even begins.
This is clear evidence that ACE uses a poor evaluation process. Is the fact that ACE’s evaluation process rewards inefficiency, and punishes efficiency, “no evidence” for ACE recommending ineffective charities?
If you’d like me to get even more specific, let’s look at Problem 1 of our review:
We go on to detail how if LIC had spent less than $2,000 on the lawsuit (saving over $200,000) and achieved the exact same outcome, ACE would have assigned LIC a Cost-Effectiveness Score of 1.8. The lowest Cost-Effectiveness Score ACE assigned to any charity in 2023 was 3.3. This means if LIC had spent less than $2,000 on the lawsuit, LIC’s Cost-Effectiveness Score would have been significantly worse than any charity ACE evaluated in 2023.
Instead, LIC spent over $200,000 on the lawsuit, and LIC rewarded them for this inefficiency by giving them a Cost-Effectiveness Score of 3.7, and deeming LIC a top 11 animal charity.
As we noted in our review, these Cost-Effectiveness Scores are defined by ACE as “how cost effective we think the charity has been”. LIC achieved no favorable legal outcomes despite receiving over a million dollars in funding. As we also noted in our review, every lawsuit LIC filed was dismissed for failing to state a valid legal claim.
If I provided evidence that a Law Firm Rating Organization rewards law firms for losing lawsuits and wasting money, and punishes law firms for winning lawsuits and saving money, would this be no evidence that the Law Firm Rating Organization is recommending ineffective law firms?
ACE’s poor evaluation process leads to ineffective charities receiving recommendations, and many animals are suffering as a result.
Our review details how ACE’s recommendations direct the flow of millions of dollars. Are you asking for evidence that directing millions of dollars toward ineffective animal charities, rather than effective ones, leads to animal suffering?
we have reviewed 5 of ACE’s “Top 11 Animal Charities to Donate to in 2024” and only one of them (Shrimp Welfare Project) appears to be an effective charity for helping animals.
Imagine a film critic watches 5 of the 11 films that received a ‘Best Films’ award and writes, “Of the five films I’ve seen, only one appears to deserve the award. I plan to release my reviews of the films shortly.” Does this statement by the film critic require evidence?
I don’t know what you’re saying is inaccurate. My reply addressed every single word from the section you claimed I didn’t provide evidence for.
My claim in the comment above was that you haven’t provided any evidence that:
5 / 11 (or more) ACE top charities are not effective
We never made this claim.
That animals are suffering as a result of ACE recommendations
I’ll ask again. Our review details how ACE is rewarding charities for inefficiency (and punishing them for efficiency), and how LIC was rewarded for their inefficiency with the designation “Top 11 Animal Charities to Donate to in 2024.” Our review also details how ACE’s recommendations direct the flow of millions of dollars. Are you asking for evidence that directing millions of dollars toward ineffective animal charities, rather than effective ones, leads to animal suffering?
You straw manned us, and now you claim that “This is starting to feel pretty bad faith”.
Here is the quote of what we said:
we have reviewed 5 of ACE’s “Top 11 Animal Charities to Donate to in 2024” and only one of them (Shrimp Welfare Project) appears to be an effective charity for helping animals.
Here is the quote of what you said we claimed:
5 / 11 (or more) ACE top charities are not effective
Notice that we said that only one of the 5 appears to be effective (meaning 4did not appear to be effective), and you changed this claim to 5are not effective.
Is the claim “4 did not appear to be effective” the same as “5 are not effective”?
I didn’t look at this for more than a couple minutes
Thank you for reading some of the article. I hope that you find some time to read the rest.
Did you ask ACE to review this before publishing?
No, I did not ask ACE. I hope that this article inspires a public discussion.
I noticed that in your first example, you assume in your example (I believe), that if LIC didn’t spend 200k on the lawsuit against Costco, they wouldn’t spend it on anything else. It’s unclear to me that this is the counterfactual, or how ACE is conceptualizing those funds.
What do you mean by conceptualizing funds? In this hypothetical, they simply spend $200k less on the lawsuit. LIC did not spend their entire budget, and charities oftentimes do not. Under ACE’s methodology, LIC’s cost-effectiveness would worsen if they spent $200k less and achieved the exact same total outcomes as a charity. The calculations we’ve done are 100% objective, and if you can find an error that we made, please let us know. You can find those calculations here:
Reading your report, I felt kind of like “oh, there are interesting assumptions here, would be interested to learn more”
What assumptions are you referring to?
If ACE didn’t exist, how would the funds the direct be spent otherwise? Would that be better or worse for animals?
If ACE didn’t exist, I would hope more funds would go to effective charities instead of ineffective ones. I hope to take part in this positive change.
Is historical track record / cost-effectiveness the only lens on which to evaluate charities?
If the answer is yes, seems very hard to start new things!
I don’t know if the LIC legal case is this, but celebrating the potential impact of promising bets that didn’t pan out seems good to me.
LIC has a historical track record, and it is a bad one. People should have the opportunity to start something new. However, they shouldn’t be rated a top 11 animal charity after receiving over a million dollars in funding and failing to achieve any positive legal outcomes.
I also felt like this felt pretty politically motivated.
I saw that you call into question the integrity of the article. I want to be clear in saying that we have no relationship with any charity. However, I noticed that you co-founded the Wild Animal Initiative, which is a charity endorsed by ACE. Still, I don’t question your feedback on the article. I hope that going forward you will evaluate our reviews for what they are, rather than suggest something “political” is going on.
That being said, critique of critical organizations is great in my opinion, so appreciate you putting this out there!
Thank you for reading some of the article. I hope that you find some time to read the rest.
To be clear, I read the whole thing—I meant that I think the fact that a pretty important issue jumped out to me within a few minutes of starting reading struck me as a reason that getting feedback from ACE seems really important.
Did you ask ACE to review this before publishing?
I really think you should! I also really think you should ask for feedback from other people who have done charity evaluations, and the charities you evaluate. You should definitely still publish them, but they’ll be better critiques for having engaged with the best case for the thing you’re critiquing!
What do you mean by conceptualizing funds? In this hypothetical, they simply spend $200k less on the lawsuit. LIC did not spend their entire budget, and charities oftentimes do not. Under ACE’s methodology, LIC’s cost-effectiveness would improve if they spent $200k less and achieved the exact same total outcomes as a charity. The calculations we’ve done are 100% objective, and if you can find an error that we made, please let us know. You can find those calculations here:
Yep, this seems right, but it’s also the case that if they did something else with that funding, the effectiveness of that action would be rated much more highly, which also seems correct. I think the issues you point to are interesting, but they strike me as intentional decisions, which ACE may have internal views on, and for which I think getting their feedback might be really important. You are correct about a mathematical fact, but both you and ACE seem to have different goals (calculating historic cost-effectiveness vs marginal impact of future dollars), and there are assumptions underlying your analysis that if changed, might change the output.
What assumptions are you referring to?
I meant ACE’s assumptions—I thought your post raised some really good questions. They are issues that if I saw, I’d email to ACE and ask why they made the choices they made, then choose whether or not to publicly publish them based on their response. Maybe these choices are reasonable, and maybe they aren’t—you raised some really good points I think. But it just seems hard to evaluate in a vacuum.
LIC has a historical track record, and it is a bad one. People should have the opportunity to start something new. However, they shouldn’t be rated a top charity after receiving over a million dollars in funding and failing to achieve any positive legal outcomes.
Again, I don’t really see good evidence for this—what is the typical track record for legal campaigns? How much do they cost? How long do they take to work? These all would be important questions to answer before claiming cost-effectiveness or lack thereof. In this case, I could easily be persuaded to agree with you, but not for any of the reasons in your analysis — the fact that they spent some money some lawsuits and it didn’t work isn’t the only evidence I’d want to think about whether or not donations to them will be useful.
Yep, this seems right, but it’s also the case that if they did something else with that funding, the effectiveness of that action would be rated much more highly, which also seems correct.
Our hypothetical in Problem 1 of the review is about two scenarios:
LIC spending $204,428 on the Costco lawsuit and achieving outcome X. (this is what actually happened)
LIC spending $1,566 on the Costco lawsuit and achievement outcome X. (this is what happened in the hypothetical)
Note that both scenarios achieve the exact same result, which is outcome X. ACE would rate LIC less cost-effective for spending $1,566 (saving $202,862) to achieve outcome X.
The hypothetical is not about something else they could do with the funding. The hypothetical is about assessing what happens to a charity’s Cost-Effectiveness Score if they save money to achieve the same outcome.
it’s also the case that if they did something else with that funding, the effectiveness of that action would be rated much more highly, which also seems correct.
How do you know that if they did something else with that funding the effectiveness of that action would be rated much more highly? According to ACE, the Costco lawsuit was a particularly cost effective intervention in spite of the fact that the lawsuit was dismissed for failing to state a valid legal claim:
“We think that out of all of Legal Impact for Chickens’ achievements, the Costco shareholder derivative case is particularly cost effective because it scored high on achievement quality.”
I’m also not sure how it would even be possible to evaluate a hypothetical in which LIC does something else with the funding. Could you explain how this hypothetical would work, and how it would be evaluated? All of the examples in our review are 100% objective and based on ACE’s own methodology. There is no subjectivity in our examples, and this was done intentionally.
You are correct about a mathematical fact, but both you and ACE seem to have different goals (calculating historic cost-effectiveness vs marginal impact of future dollars)
I’m not sure how ACE’s goals could align with the principles of effective altruism if they intentionally created a methodology that contains the problem described above.
Again, I don’t really see good evidence for this—what is the typical track record for legal campaigns? How much do they cost? How long do they take to work?
Imagine there is a law firm that has received over a million of dollars in funding, existed for multiple years, and failed to secure any favorable legal outcomes. Also imagine that their most cost-effective lawsuit was one in which they spent over $200,000, and the lawsuit was dismissed for failing to state a valid legal claim.
If this law firm were rated one of the top 100 law firms in the world, what would you think of the organization that assigned this rating? Would you say there is not good evidence for this being an incorrect rating?
ACE rated LIC as one of the Top 11 Animal Charities to Donate to in 2024. Prior to being reviewed, LIC received over a million dollars in funding, existed for multiple years, and failed to secure any favorable legal outcomes. According to ACE, LIC’s most cost-effective intervention was one in which they spent over $200,000, and the lawsuit was dismissed for failing to state a valid legal claim.
Are these factors poor evidence for LIC not being one of the 11 best animal charities to donate to?
I don’t really have a strong view about LIC—as I’ve mentioned elsewhere in the comments, I’m skeptical in general that very EA donors should give to farmed vertebrate welfare issues in the near future. But I don’t find this level of evidence particularly compelling on its own. I think I feel confused about the example you’re giving because it isn’t about hypothetical cost-effectiveness, it’s about historic cost-effectiveness, where what matters are the counterfactuals.
I broadly think the critique is interesting, and again, seems like probably an issue with the methodology, but on its own doesn’t seem like reason to think that ACE isn’t identifying good donation opportunities, because things besides cost-effectiveness also matter here.
But I don’t find this level of evidence particularly compelling on its own.
You don’t find these facts particularly compelling evidence that LIC is not historically cost-effective?
LIC’s most cost-effective intervention was one in which they spent over $200,000, and the lawsuit was dismissed for failing to state a valid legal claim.
LIC received over a million dollars in funding prior to being reviewed
LIC existed for multiple years prior to being reviewed
LIC failed to secure any favorable legal outcomes, or file any lawsuit that stated a valid legal claim?
What would be compelling evidence for LIC not being historically cost-effective?
I think I feel confused about the example you’re giving because it isn’t about hypothetical cost-effectiveness, it’s about historic cost-effectiveness, where what matters are the counterfactuals.
ACE does 2 separate analyses for past cost-effectiveness, and room for future funding. For example, those two sections in ACE’s review of LIC are:
Cost Effectiveness: How much has Legal Impact for Chickens achieved through their programs?
Room For More Funding: How much additional money can Legal Impact for Chickens effectively use in the next two years?
Our review focuses on ACE’s Cost-Effectiveness analysis, not on their Room For More Funding analysis. In the future, we may evaluate ACE’s Room For More Funding Analysis, but that is not what our review focused on.
However, I would like to pose a question to you: Given the ACE often gives charities a worse historic cost-effectiveness rating for spending less money to achieve the exact same outcomes (see Problem 1), how confident do you feel in ACE’s ability to analyze future cost-effectiveness (which is inherently more difficult to analyze)?
I don’t find that evidence particularly compelling on its own, no. Lots of projects cost more than 1M or take more than a few years to have success. I don’t see why those things would be cause to dismiss a project out of hand. I don’t really buy social movement theories of change for animal advocacy, but many people do, and it just seems like many social movement-y things take a long time to build momentum, and legal and research-focused projects take forever to play out. Things I’d want to look at to form a view on this (though to be clear, I plausibly agree with you!):
How much lawsuits of this type typically cost
What the base rate for success is for this kind of work
How long this kind of work typically takes to get traction
Has anyone else tried similar work on misleading labelling or whatever? Was it effective or not?
Has LIC’s work inspired other lawsuits, as ACE reported might be a positive side effect?
I don’t think we disagree that much here, except how much these things matter — I don’t really care about ACE’s ability to analyze cost-effectiveness outside broad strokes because I think the primary benefits of organizations like ACE is shifting money to more cost-effective things within the animal space, which I do believe ACE does. I also don’t mind ACE endorsing speculative bets that don’t pay off — I think there are many things that were worth paying for in expectation that don’t end up helping any animals, and will continue to be, because we don’t really know very many effective ways to help animals so the information value of trying new things is high.
But to answer your question specifically, I’d be very skeptical of anyone’s numbers on future cost-effectiveness, ACE’s or yours or my own, because I think this is an issue that has historically been extremely difficult to estimate cost-effectiveness for. I’m not convinced that’s the right way to approach identifying effective animal interventions, in part because it is so hard to do well. I don’t really think ACE is making cost-effectiveness estimates here though—it seems much more like trying to get a rough sense of relative cost-effectiveness, which, putting aside the methodological issues you’ve raised, seems like the right approach to me, but only a small part of the information I’d want to know where money should move in animal advocacy.
What the base rate for success is for this kind of work
How long this kind of work typically takes to get traction
The Nonhuman Rights Project provides a possible point of comparison. From 2013 to 2023 they raised $13.2 Million. As far as I know, they have never won a case.
I don’t find that evidence particularly compelling on its own, no. Lots of projects cost more than 1M or take more than a few years to have success. I don’t see why those things would be cause to dismiss a project out of hand.
The question I asked was: “You don’t find these facts particularly compelling evidence that LIC is not historically cost-effective?”
The question was not about whether these facts are compelling evidence that LIC won’t be successful in the future, or if the project should be dismissed.
Wait, those are related to each other though—if we haven’t seen the full impact of their previous actions, we haven’t yet seen their historical cost-effectiveness in full! Also, you cite these as reasons the project should be dismissed in your post—you have a section literally called “Legal Impact for Chickens Did Not Achieve Any Favorable Legal Outcomes, Yet ACE Rated Them a Top Charity” which reads to me that you believe that it is bad they were rated a Top Charity, and make these same arguments (and no others) in the section, suggesting that you think this evidence means they should be dismissed.
Wait, those are related to each other though—if we haven’t seen the full impact of their previous actions, we haven’t yet seen their historical cost-effectiveness in full!
No, they are not. Historical cost-effectiveness refers to past actions and outcomes—what has already occurred.
All of LIC’s legal actions have already been either dismissed or rejected. What are you suggesting we need to wait for before we can analyze LIC’s historical cost-effectiveness in full?
You are conflating the issue of past cost-effectiveness with future potential.
Also, you cite these as reasons the project should be dismissed in your post—you have a section literally called “Legal Impact for Chickens Did Not Achieve Any Favorable Legal Outcomes, Yet ACE Rated Them a Top Charity” which reads to me that you believe that it is bad they were rated a Top Charity, and make these same arguments (and no others) in the section, suggesting that you think this evidence means they should be dismissed.
Did I claim that I don’t think LIC “should be dismissed”?
both you and ACE seem to have different goals (calculating historic cost-effectiveness vs marginal impact of future dollars)
ACE states (under Criterion 2) that a charity’s Cost-Effectiveness Score “indicates, on a 1-7 scale, how cost effective we think the charity has been [...] with higher scores indicating higher cost effectiveness.”
Would you mind clarifying what you believe ACE’s goal is, and what you believe my goal is?
The analysis in my review is entirely about calculating historic cost-effectiveness. ACE’s Cost-Effectiveness Scores are also entirely about calculating historic cost-effectiveness.
From this post, it seems like you’re trying to calculate historic cost-effectiveness and rate charities exclusively on that (since you haven’t published an evaluation of an animal charity yet I could be wrong here though). My understanding of what ACE is trying to do with its evaluations as a whole is identify where marginal dollars might be most useful for animal advocacy, and move money from less effective opportunities to those. Cost-effectiveness might be one component of that, but is far from the only one (e.g. intervention scalability might matter, having a diversity of types of opportunities to appeal to different donors, etc.). It’s pretty easy to imagine scenarios where you wouldn’t prefer to only look at cost-effectiveness of individual charities when making recommendation, even if that’s what matters in the end. It’s also easy to imagine scenarios where recommending less effective opportunities leads to better outcomes to animals—maybe installing shrimp stunners is super effective, but only some donors will give to it. Maybe it can only scale to a few M per year but you influence more money than that. Depending on your circumstances, a lot more than cost-effectiveness of specific interventions matters for making the most effective recommendations.
My understanding is also that ACE doesn’t see EAs as its primary audience (but I’m less certain about this). This is a reason I’m excited about your project—seems nice to have “very EA” evaluations of charities in addition to ACE’s. But, I also imagine it would be hard to get charities to participate in your evaluation process if you don’t run the evaluations by them in advance, which could make it hard for you to get information to do what you’re trying to do, unless you rely on the information ACE collects, which then puts you in an awkward position of making a strong argument against an organization you might need to conduct evaluations.
My understanding is ACE has tried to do something that’s just cost-effectiveness analysis in the past (they used to give probability distributions for how many animals were helped, for example). But it’s really difficult to do confidently for animal issues, and that’s part of the reason it’s only a portion of the whole picture (along with other factors like I mention above).
From this post, it seems like you’re trying to calculate historic cost-effectiveness and rate charities exclusively on that (since you haven’t published an evaluation of an animal charity yet I could be wrong here though)
This is not what we are trying to do. We simply critiqued the way that ACE calculated historic cost-effectiveness, and how ACE gave Legal Impact for Chickens a relatively high historic cost-effectiveness rating despite have no historic success.
My understanding of what ACE is trying to do with its evaluations as a whole is identify where marginal dollars might be most useful for animal advocacy, and move money from less effective opportunities to those.
ACE does 2 separate analyses for past cost-effectiveness, and room for future funding. For example, those two sections in ACE’s review of LIC are:
Cost Effectiveness: How much has Legal Impact for Chickens achieved through their programs?
Room For More Funding: How much additional money can Legal Impact for Chickens effectively use in the next two years?
Our review focuses on ACE’s Cost-Effectiveness analysis, not on their Room For More Funding analysis. In the future, we may evaluate ACE’s Room For More Funding Analysis, but that is not what our review focused on. We wanted to keep our review short enough that people could read it without a huge time investment, so we could not include an assessment of every single part of ACE’s evaluation process in our review.
It is also less reasonable to hold ACE accountable for their Room For More Funding analysis, since this is inherently more subjective and difficult to do. It is far easier for ACE (or any charity evaluator) to analyze historic cost-effectiveness than to analyze future cost-effectiveness. However, I would like to pose a question to you: Given the ACE often gives charities a worse historic cost-effectiveness rating for spending less money to achieve the exact same outcomes (see Problem 1), how confident do you feel in ACE’s ability to analyze future cost-effectiveness?
My understanding is ACE has tried to do something that’s just cost-effectiveness analysis in the past (they used to give probability distributions for how many animals were helped, for example).
ACE responded to this thread acknowledging that the problems listed in our review needed to be addressed, and that they changed their methodology (to a cost-effectiveness calculation of simply impact divided by cost) to do so:
This is not what we are trying to do. We simply critiqued the way that ACE calculated historic cost-effectiveness, and how ACE gave Legal Impact for Chickens a relatively high historic cost-effectiveness rating despite have no historic success.
FWIW this seems great—excited to see more comprehensive evaluations. Yeah, I agree with many of your comments here on the granular level — it seems you found something that is a potential issue for how ACE does (or did) some aspects of their evaluations, and publishing that is great! I think we just disagree on how important it is?
By the way, I’m ending further engagement on this (though feel free to leave a response if useful!) just because I already find the EA Forum distracting from other work, and don’t have time this week to think about this more. Appreciate you going through everything with me!
Did you ask ACE to review this before publishing? It seems like the kind of thing that would be worth getting feedback on before publishing. I didn’t look at this for more than a couple minutes, but I saw immediately that there might be some conceptual disagreements between you and ACE—for example, I noticed that in your first example, you assume in your example (I believe), that if LIC didn’t spend 200k on the lawsuit against Costco, they wouldn’t spend it on anything else. It’s unclear to me that this is the counterfactual, or how ACE is conceptualizing those funds. There might be reasoning behind their decisionmaking that would be useful to your critiques they could share.
I also felt like this felt pretty politically motivated. Not sure if that is your intention, but paragraphs like this:
Without any evidence feels pretty intense. ACE is kind of low hanging fruit to pick on in the EA space, so this read to me like more of that, without necessarily the evidence base to back it. Reading your report, I felt kind of like “oh, there are interesting assumptions here, would be interested to learn more”, and not “ACE is doing an extremely bad job.”
E.g. I think the questions that would be good to ask in a critique of ACE might be:
If ACE didn’t exist, how would the funds the direct be spent otherwise? Would that be better or worse for animals?
Is historical track record / cost-effectiveness the only lens on which to evaluate charities?
If the answer is yes, seems very hard to start new things!
I don’t know if the LIC legal case is this, but celebrating the potential impact of promising bets that didn’t pan out seems good to me.
I also think getting feedback on statements like this would be really helpful:
I think ACE has wanted to do this at points in their history — my impression is just that it is incredibly difficult, so they’ve approached it from other angles instead. I also don’t think it’s clear to me that ACE’s goal is to report cost-effectiveness. I think clarifying this with them, and getting a sense of why they don’t do what you see as the simple approach would be useful for making this critique stronger. And, I don’t think people should make giving decisions based only on historic cost-effectiveness—just because an opportunity was impactful doesn’t mean the organization needs more funds to do that work, that it will scale, work in the future, etc.
I don’t disagree that ACE might be directing funds to ineffective charities! I don’t really think non-OpenPhil EA donors should give to farmed animal welfare, for example. But, I don’t think it is obvious to me that ACE going away means money going to more effective charities—I expect it would mostly be worse—people giving to animal charities with basically no vetting.
That being said, critique of critical organizations is great in my opinion, so appreciate you putting this out there!
“I don’t really think non-OpenPhil EA donors should give to farmed animal welfare, for example.” Wow, this is interesting! I would love to know what you mean by this?
(I responded privately to this but wrote up some related reflections a while ago here).
Having read your reflections, I’m still curious as to why you don’t think non-OpenPhil donors should give to farmed animal welfare, if you feel comfortable sharing it publicly. I guessed four options, ordered from most to least likely, but I might have misunderstood the post
We should donate to wild animal welfare instead, as it’s more cost-effective
There are no donation opportunities that counterfactually help a significant amount of farmed animals
There is no strong moral obligation to improve future lives, and donations to farmed animal welfare necessarily improve future lives, as farmed animal lives are very short
Tomasik-style arguments on the impact of animal farming on the amount of wild animal suffering
Is it a combination of these? As a concrete example, I’m curious if you believe that the Shrimp Welfare Project shouldn’t be funded, should be funded by “non-EA” donors, or will be funded anyway and donors shouldn’t worry about it.
By the way, thank you for nudging towards sharing evaluations with the evaluated organization before posting, I think it’s a really valuable norm.
Thanks! My wording in the above message was imprecise, but I mean something like farmed vertebrates. SWP is probably among the two most important things to fund, in my opinion.
Basically I think the size of good opportunities in farmed animal advocacy is smaller than OpenPhil’s grantmaking budget and there are few scalable interventions, though I don’t think I want to go into most the reasons publicly. Given that they’ve stopped funding many of what I believe are more cost-effective projects, and that EA donors are basically the only people willing to fund those, EA donors should be mostly inclined to fund things OpenPhil can’t fund instead.
So some combination of 1+2 (for farmed vertebrates) + other factors
What claims did we make that we did not provide evidence for?
I understand these are forthcoming, but no evidence is provided for this entire part—part of the reason I pushed on this is I think seeing your alternative evaluations would be very helpful for interpreting the strength of the critique of ACE. Without seeing them, I can’t evaluate the latter half of the quoted text. And in my eyes, if these are similar to the evaluation here of LIC, it’s pretty far from demonstrating that ineffective charities are receiving recommendations, etc. And, given that you’ve only evaluated <50% of their charities so far, it seems preemptive to make the overall claim. I think the overall claim is very possibly true, but again, I think to make the argument that animals are directly suffering as a result of this, you’d have to demonstrate that those charities are worse than other donation options, that donors would give to the better options, etc.
Note: this reply addresses everything Abraham claims we did not provide evidence for.
“ACE’s poor evaluation process leads to ineffective charities receiving recommendations”
Our review covered how under ACE’s evaluation process:
Charities can receive a worse Cost-Effectiveness Score by spending less money to achieve the exact same results.
Charities can have 1,000,000 times the impact at the exact same price, and their Cost-Effectiveness Score can remain the same.
The most important factor in determining the impact of an intervention is decided before the intervention even begins.
This is clear evidence that ACE uses a poor evaluation process. Is the fact that ACE’s evaluation process rewards inefficiency, and punishes efficiency, “no evidence” for ACE recommending ineffective charities?
If you’d like me to get even more specific, let’s look at Problem 1 of our review:
We go on to detail how if LIC had spent less than $2,000 on the lawsuit (saving over $200,000) and achieved the exact same outcome, ACE would have assigned LIC a Cost-Effectiveness Score of 1.8. The lowest Cost-Effectiveness Score ACE assigned to any charity in 2023 was 3.3. This means if LIC had spent less than $2,000 on the lawsuit, LIC’s Cost-Effectiveness Score would have been significantly worse than any charity ACE evaluated in 2023.
Instead, LIC spent over $200,000 on the lawsuit, and LIC rewarded them for this inefficiency by giving them a Cost-Effectiveness Score of 3.7, and deeming LIC a top 11 animal charity.
As we noted in our review, these Cost-Effectiveness Scores are defined by ACE as “how cost effective we think the charity has been”. LIC achieved no favorable legal outcomes despite receiving over a million dollars in funding. As we also noted in our review, every lawsuit LIC filed was dismissed for failing to state a valid legal claim.
If I provided evidence that a Law Firm Rating Organization rewards law firms for losing lawsuits and wasting money, and punishes law firms for winning lawsuits and saving money, would this be no evidence that the Law Firm Rating Organization is recommending ineffective law firms?
Our review details how ACE’s recommendations direct the flow of millions of dollars. Are you asking for evidence that directing millions of dollars toward ineffective animal charities, rather than effective ones, leads to animal suffering?
Imagine a film critic watches 5 of the 11 films that received a ‘Best Films’ award and writes, “Of the five films I’ve seen, only one appears to deserve the award. I plan to release my reviews of the films shortly.” Does this statement by the film critic require evidence?
(Responding because this is inaccurate): My claim in the comment above was that you haven’t provided any evidence that:
5 / 11 (or more) ACE top charities are not effective
That animals are suffering as a result of ACE recommendations
Which remains the case — I look forward to you producing it.
I don’t know what you’re saying is inaccurate. My reply addressed every single word from the section you claimed I didn’t provide evidence for.
We never made this claim.
I’ll ask again. Our review details how ACE is rewarding charities for inefficiency (and punishing them for efficiency), and how LIC was rewarded for their inefficiency with the designation “Top 11 Animal Charities to Donate to in 2024.” Our review also details how ACE’s recommendations direct the flow of millions of dollars. Are you asking for evidence that directing millions of dollars toward ineffective animal charities, rather than effective ones, leads to animal suffering?
This is starting to feel pretty bad faith, so I’m actually going to stop engaging.
You straw manned us, and now you claim that “This is starting to feel pretty bad faith”.
Here is the quote of what we said:
Here is the quote of what you said we claimed:
Notice that we said that only one of the 5 appears to be effective (meaning 4 did not appear to be effective), and you changed this claim to 5 are not effective.
Is the claim “4 did not appear to be effective” the same as “5 are not effective”?
Hi Abraham,
Thank you for reading some of the article. I hope that you find some time to read the rest.
No, I did not ask ACE. I hope that this article inspires a public discussion.
What do you mean by conceptualizing funds? In this hypothetical, they simply spend $200k less on the lawsuit. LIC did not spend their entire budget, and charities oftentimes do not. Under ACE’s methodology, LIC’s cost-effectiveness would worsen if they spent $200k less and achieved the exact same total outcomes as a charity. The calculations we’ve done are 100% objective, and if you can find an error that we made, please let us know. You can find those calculations here:
Original: https://docs.google.com/spreadsheets/d/1BzLUSefsd5K2uhGw81UPAe_v0E_JJR8Kdgh8zsBtxsU/edit?gid=0#gid=0
Hypothetical: https://docs.google.com/spreadsheets/d/1IZbSfk4eNukmUU7ntRT0wCgvrLG6_ZcyXt3uANmFYbE/edit?gid=1427660869#gid=1427660869
What assumptions are you referring to?
If ACE didn’t exist, I would hope more funds would go to effective charities instead of ineffective ones. I hope to take part in this positive change.
LIC has a historical track record, and it is a bad one. People should have the opportunity to start something new. However, they shouldn’t be rated a top 11 animal charity after receiving over a million dollars in funding and failing to achieve any positive legal outcomes.
I saw that you call into question the integrity of the article. I want to be clear in saying that we have no relationship with any charity. However, I noticed that you co-founded the Wild Animal Initiative, which is a charity endorsed by ACE. Still, I don’t question your feedback on the article. I hope that going forward you will evaluate our reviews for what they are, rather than suggest something “political” is going on.
Thank you, and I appreciate your feedback!
To be clear, I read the whole thing—I meant that I think the fact that a pretty important issue jumped out to me within a few minutes of starting reading struck me as a reason that getting feedback from ACE seems really important.
I really think you should! I also really think you should ask for feedback from other people who have done charity evaluations, and the charities you evaluate. You should definitely still publish them, but they’ll be better critiques for having engaged with the best case for the thing you’re critiquing!
Yep, this seems right, but it’s also the case that if they did something else with that funding, the effectiveness of that action would be rated much more highly, which also seems correct. I think the issues you point to are interesting, but they strike me as intentional decisions, which ACE may have internal views on, and for which I think getting their feedback might be really important. You are correct about a mathematical fact, but both you and ACE seem to have different goals (calculating historic cost-effectiveness vs marginal impact of future dollars), and there are assumptions underlying your analysis that if changed, might change the output.
I meant ACE’s assumptions—I thought your post raised some really good questions. They are issues that if I saw, I’d email to ACE and ask why they made the choices they made, then choose whether or not to publicly publish them based on their response. Maybe these choices are reasonable, and maybe they aren’t—you raised some really good points I think. But it just seems hard to evaluate in a vacuum.
Again, I don’t really see good evidence for this—what is the typical track record for legal campaigns? How much do they cost? How long do they take to work? These all would be important questions to answer before claiming cost-effectiveness or lack thereof. In this case, I could easily be persuaded to agree with you, but not for any of the reasons in your analysis — the fact that they spent some money some lawsuits and it didn’t work isn’t the only evidence I’d want to think about whether or not donations to them will be useful.
Our hypothetical in Problem 1 of the review is about two scenarios:
LIC spending $204,428 on the Costco lawsuit and achieving outcome X. (this is what actually happened)
LIC spending $1,566 on the Costco lawsuit and achievement outcome X. (this is what happened in the hypothetical)
Note that both scenarios achieve the exact same result, which is outcome X. ACE would rate LIC less cost-effective for spending $1,566 (saving $202,862) to achieve outcome X.
The hypothetical is not about something else they could do with the funding. The hypothetical is about assessing what happens to a charity’s Cost-Effectiveness Score if they save money to achieve the same outcome.
How do you know that if they did something else with that funding the effectiveness of that action would be rated much more highly? According to ACE, the Costco lawsuit was a particularly cost effective intervention in spite of the fact that the lawsuit was dismissed for failing to state a valid legal claim:
“We think that out of all of Legal Impact for Chickens’ achievements, the Costco shareholder derivative case is particularly cost effective because it scored high on achievement quality.”
I’m also not sure how it would even be possible to evaluate a hypothetical in which LIC does something else with the funding. Could you explain how this hypothetical would work, and how it would be evaluated? All of the examples in our review are 100% objective and based on ACE’s own methodology. There is no subjectivity in our examples, and this was done intentionally.
I’m not sure how ACE’s goals could align with the principles of effective altruism if they intentionally created a methodology that contains the problem described above.
Imagine there is a law firm that has received over a million of dollars in funding, existed for multiple years, and failed to secure any favorable legal outcomes. Also imagine that their most cost-effective lawsuit was one in which they spent over $200,000, and the lawsuit was dismissed for failing to state a valid legal claim.
If this law firm were rated one of the top 100 law firms in the world, what would you think of the organization that assigned this rating? Would you say there is not good evidence for this being an incorrect rating?
ACE rated LIC as one of the Top 11 Animal Charities to Donate to in 2024. Prior to being reviewed, LIC received over a million dollars in funding, existed for multiple years, and failed to secure any favorable legal outcomes. According to ACE, LIC’s most cost-effective intervention was one in which they spent over $200,000, and the lawsuit was dismissed for failing to state a valid legal claim.
Are these factors poor evidence for LIC not being one of the 11 best animal charities to donate to?
I don’t really have a strong view about LIC—as I’ve mentioned elsewhere in the comments, I’m skeptical in general that very EA donors should give to farmed vertebrate welfare issues in the near future. But I don’t find this level of evidence particularly compelling on its own. I think I feel confused about the example you’re giving because it isn’t about hypothetical cost-effectiveness, it’s about historic cost-effectiveness, where what matters are the counterfactuals.
I broadly think the critique is interesting, and again, seems like probably an issue with the methodology, but on its own doesn’t seem like reason to think that ACE isn’t identifying good donation opportunities, because things besides cost-effectiveness also matter here.
You don’t find these facts particularly compelling evidence that LIC is not historically cost-effective?
LIC’s most cost-effective intervention was one in which they spent over $200,000, and the lawsuit was dismissed for failing to state a valid legal claim.
LIC received over a million dollars in funding prior to being reviewed
LIC existed for multiple years prior to being reviewed
LIC failed to secure any favorable legal outcomes, or file any lawsuit that stated a valid legal claim?
What would be compelling evidence for LIC not being historically cost-effective?
ACE does 2 separate analyses for past cost-effectiveness, and room for future funding. For example, those two sections in ACE’s review of LIC are:
Cost Effectiveness: How much has Legal Impact for Chickens achieved through their programs?
Room For More Funding: How much additional money can Legal Impact for Chickens effectively use in the next two years?
Our review focuses on ACE’s Cost-Effectiveness analysis, not on their Room For More Funding analysis. In the future, we may evaluate ACE’s Room For More Funding Analysis, but that is not what our review focused on.
However, I would like to pose a question to you: Given the ACE often gives charities a worse historic cost-effectiveness rating for spending less money to achieve the exact same outcomes (see Problem 1), how confident do you feel in ACE’s ability to analyze future cost-effectiveness (which is inherently more difficult to analyze)?
I don’t find that evidence particularly compelling on its own, no. Lots of projects cost more than 1M or take more than a few years to have success. I don’t see why those things would be cause to dismiss a project out of hand. I don’t really buy social movement theories of change for animal advocacy, but many people do, and it just seems like many social movement-y things take a long time to build momentum, and legal and research-focused projects take forever to play out. Things I’d want to look at to form a view on this (though to be clear, I plausibly agree with you!):
How much lawsuits of this type typically cost
What the base rate for success is for this kind of work
How long this kind of work typically takes to get traction
Has anyone else tried similar work on misleading labelling or whatever? Was it effective or not?
Has LIC’s work inspired other lawsuits, as ACE reported might be a positive side effect?
I don’t think we disagree that much here, except how much these things matter — I don’t really care about ACE’s ability to analyze cost-effectiveness outside broad strokes because I think the primary benefits of organizations like ACE is shifting money to more cost-effective things within the animal space, which I do believe ACE does. I also don’t mind ACE endorsing speculative bets that don’t pay off — I think there are many things that were worth paying for in expectation that don’t end up helping any animals, and will continue to be, because we don’t really know very many effective ways to help animals so the information value of trying new things is high.
But to answer your question specifically, I’d be very skeptical of anyone’s numbers on future cost-effectiveness, ACE’s or yours or my own, because I think this is an issue that has historically been extremely difficult to estimate cost-effectiveness for. I’m not convinced that’s the right way to approach identifying effective animal interventions, in part because it is so hard to do well. I don’t really think ACE is making cost-effectiveness estimates here though—it seems much more like trying to get a rough sense of relative cost-effectiveness, which, putting aside the methodological issues you’ve raised, seems like the right approach to me, but only a small part of the information I’d want to know where money should move in animal advocacy.
How much lawsuits of this type typically cost
What the base rate for success is for this kind of work
How long this kind of work typically takes to get traction
The Nonhuman Rights Project provides a possible point of comparison. From 2013 to 2023 they raised $13.2 Million. As far as I know, they have never won a case.
The question I asked was: “You don’t find these facts particularly compelling evidence that LIC is not historically cost-effective?”
The question was not about whether these facts are compelling evidence that LIC won’t be successful in the future, or if the project should be dismissed.
Wait, those are related to each other though—if we haven’t seen the full impact of their previous actions, we haven’t yet seen their historical cost-effectiveness in full! Also, you cite these as reasons the project should be dismissed in your post—you have a section literally called “Legal Impact for Chickens Did Not Achieve Any Favorable Legal Outcomes, Yet ACE Rated Them a Top Charity” which reads to me that you believe that it is bad they were rated a Top Charity, and make these same arguments (and no others) in the section, suggesting that you think this evidence means they should be dismissed.
No, they are not. Historical cost-effectiveness refers to past actions and outcomes—what has already occurred.
All of LIC’s legal actions have already been either dismissed or rejected. What are you suggesting we need to wait for before we can analyze LIC’s historical cost-effectiveness in full?
You are conflating the issue of past cost-effectiveness with future potential.
Did I claim that I don’t think LIC “should be dismissed”?
ACE states (under Criterion 2) that a charity’s Cost-Effectiveness Score “indicates, on a 1-7 scale, how cost effective we think the charity has been [...] with higher scores indicating higher cost effectiveness.”
Would you mind clarifying what you believe ACE’s goal is, and what you believe my goal is?
The analysis in my review is entirely about calculating historic cost-effectiveness. ACE’s Cost-Effectiveness Scores are also entirely about calculating historic cost-effectiveness.
From this post, it seems like you’re trying to calculate historic cost-effectiveness and rate charities exclusively on that (since you haven’t published an evaluation of an animal charity yet I could be wrong here though). My understanding of what ACE is trying to do with its evaluations as a whole is identify where marginal dollars might be most useful for animal advocacy, and move money from less effective opportunities to those. Cost-effectiveness might be one component of that, but is far from the only one (e.g. intervention scalability might matter, having a diversity of types of opportunities to appeal to different donors, etc.). It’s pretty easy to imagine scenarios where you wouldn’t prefer to only look at cost-effectiveness of individual charities when making recommendation, even if that’s what matters in the end. It’s also easy to imagine scenarios where recommending less effective opportunities leads to better outcomes to animals—maybe installing shrimp stunners is super effective, but only some donors will give to it. Maybe it can only scale to a few M per year but you influence more money than that. Depending on your circumstances, a lot more than cost-effectiveness of specific interventions matters for making the most effective recommendations.
My understanding is also that ACE doesn’t see EAs as its primary audience (but I’m less certain about this). This is a reason I’m excited about your project—seems nice to have “very EA” evaluations of charities in addition to ACE’s. But, I also imagine it would be hard to get charities to participate in your evaluation process if you don’t run the evaluations by them in advance, which could make it hard for you to get information to do what you’re trying to do, unless you rely on the information ACE collects, which then puts you in an awkward position of making a strong argument against an organization you might need to conduct evaluations.
My understanding is ACE has tried to do something that’s just cost-effectiveness analysis in the past (they used to give probability distributions for how many animals were helped, for example). But it’s really difficult to do confidently for animal issues, and that’s part of the reason it’s only a portion of the whole picture (along with other factors like I mention above).
Thank you for your response!
This is not what we are trying to do. We simply critiqued the way that ACE calculated historic cost-effectiveness, and how ACE gave Legal Impact for Chickens a relatively high historic cost-effectiveness rating despite have no historic success.
ACE does 2 separate analyses for past cost-effectiveness, and room for future funding. For example, those two sections in ACE’s review of LIC are:
Cost Effectiveness: How much has Legal Impact for Chickens achieved through their programs?
Room For More Funding: How much additional money can Legal Impact for Chickens effectively use in the next two years?
Our review focuses on ACE’s Cost-Effectiveness analysis, not on their Room For More Funding analysis. In the future, we may evaluate ACE’s Room For More Funding Analysis, but that is not what our review focused on. We wanted to keep our review short enough that people could read it without a huge time investment, so we could not include an assessment of every single part of ACE’s evaluation process in our review.
It is also less reasonable to hold ACE accountable for their Room For More Funding analysis, since this is inherently more subjective and difficult to do. It is far easier for ACE (or any charity evaluator) to analyze historic cost-effectiveness than to analyze future cost-effectiveness. However, I would like to pose a question to you: Given the ACE often gives charities a worse historic cost-effectiveness rating for spending less money to achieve the exact same outcomes (see Problem 1), how confident do you feel in ACE’s ability to analyze future cost-effectiveness?
ACE responded to this thread acknowledging that the problems listed in our review needed to be addressed, and that they changed their methodology (to a cost-effectiveness calculation of simply impact divided by cost) to do so:
FWIW this seems great—excited to see more comprehensive evaluations. Yeah, I agree with many of your comments here on the granular level — it seems you found something that is a potential issue for how ACE does (or did) some aspects of their evaluations, and publishing that is great! I think we just disagree on how important it is?
By the way, I’m ending further engagement on this (though feel free to leave a response if useful!) just because I already find the EA Forum distracting from other work, and don’t have time this week to think about this more. Appreciate you going through everything with me!
No problem. Thank you for your replies!