Animal Charity Evaluators (ACE) is an effective altruism aligned organization which aims to provide information and recommendation to donors and advocates about the effectiveness of various interventions and charities. As one can expect, ACE reviews animal charities based on the effectiveness of their programs and recommends a small number of charities as “top” and “stand-out” charities. ACE also has a Movement Grants program, which provides small grants to various organizations. According to ACE, it has a budget over one million US dollars, and granted and influenced over ten million US dollars worth donations to its recommended charities and promising projects in 2021. This amount (10 million US dollars) is larger than the total amount of Effective Altruism Animal Welfare Fund grants in 2021.
One can say that ACE is positioned to take the role of GiveWell in the Effective Altruism animal welfare cause area. It functions as an expert authority which evaluates different interventions and comes to conclusions about which ones are the most effective and which charities carry them out in the most effective way. One can also say that ACE’s role is even more significant than GiveWell, since it is not just donors who may follow its advice, it is also the animal advocates who may benefit from this expert authority by following its advice on which interventions are the most effective to help animals.
In this essay I will put forward three points of criticism, which I believe can uncover certain defects that ACE has, and can help ACE to overcome them:
1) ACE’s current style of reasoning is somewhat opaque and therefore may mislead ordinary donors and advocates.
2) ACE is omitting from making substantial claims about the most effective ways to help animals, and thus failing in its primary role of evaluating the effectiveness of different interventions and charities.
3) ACE is currently underrating the effectiveness of programs which aim for animal welfare reforms relative to the effectiveness of other interventions.
In the first criticism, I will argue that while ACE makes cost-effectiveness comparisons between charities which carry out similar programs (for example two charities which both engage in animal welfare campaigns), it does not comment how it is comparing charities which carry out different programs (for example, one charity which engage in animal welfare campaigns and another charity which has a vegan pledge program). I will claim that some of the terms that ACE is using such as “relative to other charities” or “highly cost-effective” is likely to be understood by ordinary donors and advocates differently than ACE’s understanding, and ACE is not providing clear context when making these statements.
In the second criticism, I will show that in its current form, ACE seems to accept that almost all programs in farmed animal advocacy may be effective and does not seem to answer fundamental questions which an evaluation organization has to answer, like “which program(s) help animals the most?” or “which programs should an animal charity or an advocate prioritize?” or “which charities should a donor prioritize?”. I will argue that starting from its foundation and in its current form, ACE functions more like a hits based giving “fund”, rather than an evaluation organization, and these multiple roles make it difficult for Animal Charity Evaluators to achieve its primary role: evaluation. I will suggest that ACE should not be timid to put forward its views on the effectiveness of different programs, whatever those views are or will be. I will also have some proposals such as applying stricter and clear cut evaluation criteria, focusing on the past track record of charities rather than their future potential success, and removing research organizations from its evaluations.
In the third criticism, I will argue that given the extremely strong track record of animal welfare corporate campaigns and outreach, and given the (unfortunate) fact that the number of vegans did not budge even though there has been enormous effort in the past, ACE should make its effectiveness evaluations accordingly. In the past, ACE had been criticized for overrating animal welfare reforms. After persistent criticism, ACE seems to move away from its previous position and qualify various interventions in its evaluations as “effective”. Although I agree that some of those criticism was correct and it was the right call for ACE to cut off problematic evaluation methods, I think ACE made an overkill when it put almost all farmed animal advocacy programs in the same league as programs which aim for animal welfare reforms which have the best track record of effectiveness by far. I will show that there is a massive difference in terms of tractability between animal welfare campaigns and other programs, and ACE is not currently acting on this obvious difference.
I think that the most obvious critique is the first one. This is probably also the easiest to fix. I believe that the second critique is the most important one, and probably the most difficult to fix. While the first two criticisms are primarily procedural and structural, the third criticism puts forward a substantial argument about the effectiveness of animal advocacy interventions, and is therefore more controversial. All these criticisms have some links in between and inform one another, but one can also read and judge them separately. One can disagree with one, but agree with another. The validity of each criticism does not depend on the validity of others.
I would like to state that this is a shallow report. I have decent background knowledge and experience in animal advocacy but I could only spend about two weeks preparing this essay. Although I am very confident about the main arguments of this essay which argues that ACE is not fulfilling its evaluation authority role in a substantial, clear manner, and ACE’s evaluations do not reflect the massive differences of tractability between different animal advocacy programs, I am not entirely confident of my understanding of ACE’s evaluation processes and reasoning. So one should not immediately conclude its judgment of ACE based on this critique alone.
I want to say that although this is an essay criticizing ACE, I want the best for ACE and its team members. I am a previous donor to ACE and I always follow their content. I have no intention to discredit their work, their decisions or the work of their recommended charities and grantees. My aim is to present a better and clearer mechanism for ACE to fulfill its mission. I am also aware that the ACE’s task is hard, very hard. Unlike GiveWell, ACE is tasked to evaluate advocacy groups, which makes things much more complicated. What is even more difficult is to compare the cost-effectiveness of different programs in almost totally different fields and/or countries. The problem of “comparing apples and oranges” indeed poses a great burden. I am also aware that ACE has been the target of vicious criticism in the past. This is not one of them and I hope it will not be perceived as such. I will also try to provide as much context as possible below in order to show the difficulties which ACE had faced and is still facing, and how those difficulties are also the reasons for the problems that I am criticizing.
1.ACE’s current style of reasoning is somewhat opaque
1.1. A general description of ACE’s evaluation approach
Following Effective Altruism principles, ACE uses cause prioritization. Accordingly, ACE focuses on farmed and wild animals. There is no mismatch between these areas and ACE’s charity evaluations and grants. There are no top or standout charities in ACE’s recommended charity list which do not work in these areas . Same goes with ACE’s Movement Grants: all groups which have received these grants have at least some programs on farmed or wild animals. It is clear that ACE uses this type of prioritization very strictly. The only exception might be Nonhuman Animal Project which was on ACE’s standout charity list between 2015 and 2019.
ACE has four evaluation criteria: programs (i), room for more funding (ii), cost-effectiveness (iii), leadership and culture (iv).
The most decisive criteria which I want to focus on are programs and cost-effectiveness. For programs, ACE states: “A charity that performs well on this criterion has programs that we expect are highly effective in reducing the suffering of animals. Effective programs: i) work in high-priority cause areas or countries, ii) work toward high-priority outcomes, iii) target high-priority animal groups, and/or iv) pursue interventions that are expected to be highly effective.”
For cost-effectiveness, ACE states: “A charity’s recent cost effectiveness provides an insight into how well it has made use of its available resources and is a useful component in understanding how cost effective future donations to the charity might be. Given the limitations of strictly quantitative approaches, we have moved away from using a fully quantitative model and have instead transitioned to a more qualitative approach that analyzes the resources used and the outputs achieved for each intervention type.”
As for high-priority outcomes, which are mentioned in the programs section, ACE provides a menu of outcomes that “aim to capture the full spectrum of work done by animal advocates to try and produce change for animals.” In this menu of outcomes, there are six main categories: decreased consumption of animal products, direct help, improvement of welfare standards, increased availability of animal-free products, increased prevalence of anti-speciest values and stronger animal advocacy movement. In this menu of outcomes, ACE does not comment on which outcomes or interventions which could work toward those outcomes are more important.
In almost all of its comprehensive reviews, ACE states that: “We categorize the work of animal advocacy charities by their outcomes, broadly distinguishing whether interventions focus on individual or institutional change. Individual-focused interventions often involve decreasing the consumption of animal products, increasing the prevalence of anti-speciesist values, or providing direct help to animals. Institutional change involves improving animal welfare standards, increasing the availability of animal-free products, or strengthening the animal advocacy movement.”
ACE also states: “Currently, we find the arguments for an institution-focused approach more compelling than individual-focused approaches.”
ACE concludes that: “..., when considering charities to evaluate, we prioritize those that work to improve welfare standards, increase the availability of animal-free products, or strengthen the animal advocacy movement. We give lower priority to charities that focus on decreasing the consumption of animal products, increasing the prevalence of anti-speciesist values, or providing direct help to animals.”
As for where would programs be more effective ACE states that: “We prioritize charities in countries with relatively large animal agricultural industries, few other charities engaged in similar work, and in which animal advocacy is likely to be feasible and have a lasting impact.”.
As for which groups of animals should be prioritized, ACE states: “We prioritize programs targeting specific groups of animals that are affected in large numbers and receive relatively little attention in animal advocacy.”
Finally, ACE states that: “Each charity’s long-term impact is plausibly what matters most. The potential number of animals affected increases over time due to population growth and an accumulation of generations. Thus, we would expect that the long-term impacts of an action would likely affect more animals than the short-term impacts of the same action.”. This point explains why ACE strongly prioritizes programs which aim for a stronger animal advocacy movement.
1.2. ACE’ use of some expressions are somewhat different than their ordinary meaning
1.2.1. “relative to other charities”
In order to select its top and standout charities, ACE has to compare the effectiveness of different animal charities. As a result, some are judged to be more effective than others and therefore earn top or standout status, while others who are judged to be less effective do not earn top or standout status. And when ACE does not qualify a charity as a top or standout charity, ACE can and does justify its decision by claiming the relative ineffectiveness or the relative low outcomes of that charity.
For example, on the overview page of ACE’s evaluation of Compassion in World Farming (CIWF) International, ACE comments that : “Overall, Compassion International’s legal advocacy and corporate outreach work—while potentially impactful—seem to have achieved outcomes that are below average relative to other charities we have reviewed this year.” . Or on the overview page of ACE’s evaluation of Environment & Animal Society of Taiwan (EAST), ACE comments that: “Overall, EAST’s work seems slightly less cost effective than the average charity under review.”
Normally, when an average donor or an animal advocate reads these statements, one can expect that these charities work are less cost effective or have low outcomes relative to all top and standout charities reviewed in those years. A normal interpretation would conclude that all charities which ACE judged to be top or standout charities have higher cost effectiveness and outcomes than Compassion in World Farming International or EAST, thus have relatively more impact per dollar spent.
But this is not the case. While ACE uses the expression “relative to other charities” in a short form in its overview pages, it adds an important context to that expression in the comprehensive report pages which I believe changes the ordinary meaning of the expression.
In those comprehensive report pages, ACE uses the same expression “relative to other charities” again without any additional qualifications at the beginning. But when one goes on reading the comprehensive report, one sees a different meaning.
ACE states that: “This year, we have used an approach in which we more qualitatively analyze a charity’s costs and outcomes. In particular, we have focused on the cost-effectiveness of the charity’s specific implementation of each of its programs in comparison tosimilar programsconducted by other charities we are reviewing this year.” (italics are mine).
ACE also states that: “After accounting for all of their key results and expenditures, we think the cost effectiveness of EAST’s work in their corporate engagement and industry engagement programs seem slightly lower than the average cost effectiveness of other, similar programs working toward improving welfare standards we evaluated this year.” (italics are mine).
In conclusion, when ACE is saying that X charity is less cost-effective or achieved lower outcomes “relative to other charities”, what ACE is meaning is that X charity is less cost-effective or achieved lower outcomes “relative to other charities which are engaged in similar programs”.
Those are two very different things. First, unqualified expression (“relative to other charities”) would normally be understood as the said charity is less effective than all top and standout charities, while the second, true version of the expression (relative to other charities which are engaged in similar programs), would be understood as the said charity is less effective than other charities which carry out similar programs only. And one has to read through the comprehensive reports to understand the difference.
How ACE compares charities which engage in different programs is another topic which I will explain in detail in the next section (1.3. There are some important gaps in ACE’s evaluation reasoning).
1.2.2. “cost-effective”
As one can expect, ACE uses the expression “cost-effective” a lot. Here are some examples:
1- “Compassion USA does not have plans for any significant expansions, so their room for more funding is limited. While we do not have concerns about the cost effectiveness of their program to encourage individuals to reduce their consumption of animal products, we are less confident about the cost effectiveness of their corporate outreach and public engagement programs.”
2- “We believe that xiaobuVEGAN’s work is particularly effective due to the neglectedness of farmed animal advocacy in China. As a growing charity planning to expand their programs and invest in staff training, we think that xiaobuVEGAN has room for more funding, and we do not have concerns about the cost effectiveness of their programs.”
3- “SVB’s work to increase the availability of animal-free products and their work to decrease consumption of animal products seem particularly cost effective.”
4- “We think that Vegetarianos Hoy’s programs that improve welfare standards, decrease consumption of animal products, and increase the availability of alternatives to animal products are highly effective.”
5- “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”
For an ordinary donor or animal advocate, “cost-effective” means more or less something like, the degree to which something is effective or productive in achieving desired outcomes in relation to its costs. In this context, for an ordinary animal charity donor or an animal advocate, a cost-effective intervention in animal advocacy would reduce or remove a high amount of animal suffering for a low cost.
However ACE uses the expression “cost-effective” somewhat differently than its ordinary meaning. Firstly, at least for some interventions, “the effect” or “the outcome” of the intervention is not necessarily the reduction of animal suffering, but some intermediate steps which might lead to that outcome. For example, when ACE claims that a program which involves corporate outreach to restaurants in order to increase the prevalence of plant based options, is cost-effective, the “effect” is not measured by the amount of animal suffering removed from the food supply chains, but rather the number of restaurants which add a vegan option to their menus. Even when ACE states that “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”, ACE does not mean that the effect of the intervention is the reduction of animal product consumption, it is more like reaching out to and participation of more people in these programs.
You can look at the ACE’s cost-effectiveness estimates spreadsheet for public view.
Secondly, when ACE is commenting on the cost-effectiveness of a program, for example: “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”, it is making these comments relative to each intervention type.
ACE states: “A charity’s recent cost effectiveness provides an insight into how well it has made use of its available resources and is a useful component in understanding how cost effective future donations to the charity might be. Given the limitations of strictly quantitative approaches, we have moved away from using a fully quantitative model and have instead transitioned to a more qualitative approach that analyzes the resources used and the outputs achieved for each intervention type.” and states: “The variation in cost effectiveness between charities was more dependent on which interventions the charity used rather than how they were implemented. This suggests that, rather than modeling the cost effectiveness of each charity, it would be more useful to model the average cost effectiveness of each intervention and incorporate that into our discussion of effectiveness in our Programs criterion.” (italics are mine).
For example, when ACE states “While we do not have concerns about the cost effectiveness of their [CIWF’s] program to encourage individuals to reduce their consumption of animal products, we are less confident about the cost effectiveness of their corporate outreach and public engagement programs.”, it means that CIWF’s corporate outreach program is relatively less cost-effective than other corporate programs carried out by other charities, it does not mean that this corporate outreach program is relatively less cost-effective than all other programs, or CIWF’s individual outreach program. It is entirely possible (in my view, very probable) that CIWF’s corporate outreach program is much more cost-effective than its individual outreach program in terms of reducing animal suffering per dollar spent. And at this point, I think an average donor or animal advocate would be unable to understand this important difference since ACE is not providing this important context.
Another example is this statement: “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”. An ordinary person would understand from this claim that each vegan pledge program would convince a considerable number of people to become vegan, vegetarian or flexitarian and decrease their consumption of animal products for a significant amount of time. However, that is not how ACE is using this expression. In ACE’s understanding, a highly effective program which decreases the consumption of animal products, is one which is relatively more effective compared to other programs which aim to decrease the consumption of animal products, such as leafleting or humane education.
1.3. There are some important gaps in ACE’s evaluation reasoning
You might be asking yourself by now how ACE compares the cost-effectiveness of different programs (for example the cost-effectiveness difference between an animal welfare corporate outreach program and restaurant outreach program which aims to popularize vegan options) and the effectiveness of charities which engage in different programs (for example the cost-effectiveness difference between Compassion in World Farming and xiabuVEGAN).
Although I tried my best to answer this question, I could not find an answer on ACE’s website.
We have only a partial explanation which was cited above: “We categorize the work of animal advocacy charities by their outcomes, broadly distinguishing whether interventions focus on individual or institutional change. Individual-focused interventions often involve decreasing the consumption of animal products, increasing the prevalence of anti-speciesist values, or providing direct help to animals. Institutional change involves improving animal welfare standards, increasing the availability of animal-free products, or strengthening the animal advocacy movement.”
However, ACE does not comment how they are prioritizing between programs. Firstly, ACE does not indicate how they weigh the interventions which they give high priority. For example, it is not clear how much more weight does ACE put on the effectiveness of a corporate animal welfare campaign over, say, a program which aims to strengthen the animal advocacy movement, both of which are prioritized highly by ACE. Secondly, ACE does not comment on the degree of difference between high and low priority works. For example, it is not clear how much more weight does ACE put on the effectiveness of a corporate animal welfare campaign over, say, a vegan pledge program.
ACE also does not comment on how important it is for a charity to be operating in a country where there are lots of farmed animals and animal advocacy is neglected. For example it is not clear how much more weight does ACE put on an animal charity operating in China, over say, an effective yet European animal charity.
These questions are important and decisive since ACE does make those comparisons and act on them eventually by awarding top and standout status to some charities, while not awarding these to the rest of charities which it is reviewing.
It is not also clear whether there are some categories or quotas by which ACE selects its recommended charities on the basis of the effectiveness within the framework of that category. For example, is there a quota for charities which operate in regions where animal advocacy is neglected? Or is there a quota for charities which focus on research? Or is there a quota for charities which focus on animal welfare reforms or vegan advocacy?
From an outsider viewpoint, for the reasons explained above, how ACE is making these comparisons and reaching its decisions are somewhat opaque and can benefit from clearer reasoning.
Here are some suggestions which can fix these problems:
ACE can consider adding more context to its statements without expecting donors or advocates to read the comprehensive reviews in detail, and can consider using expressions in their ordinary meanings.
ACE can consider providing more reasoning about its comparisons between different programs and charities which engage in different programs and operate in types of countries.
ACE can consider creating categories for different charities which engage in different programs and evaluate their status in each different category. For example: “ charities which engage in programs which aim to improve animal welfare”, “ charities which engage in research programs”, “charities which engage in programs which aim to spread veganism”, “charities which operate in regions where there are lots of farmed animals and animal advocacy is neglected”, “charities which focus on movement building” etc.
In ACE’s defense, it is very difficult to compare the effectiveness of different programs. “Comparing apples and oranges” is indeed a problem for an evaluation organization. However, I think it is fair to say that one should at least say clearly that it is weighing apples when it is evaluating apples and not oranges, and vice versa. I think it is also reasonable to expect from an evaluation organization an explanation of how it is comparing apples and oranges since it is actually comparing apples and oranges. Finally, even if ACE is making these comparisons separately in each category, I think it is better for ACE to state those categories (best apples in the apple category, best oranges in the orange category), and make its recommendations in those separate categories. I can imagine that this may sacrifice some simplicity since there will be multiple lists rather than the current single list, but I think a coherent and clear complexity is preferable to an incoherent and obscure simplicity.
2. ACE is omitting from making substantial claims about the most effective ways to help animals.
2.1. The elephant in the room for funding
On ACE’s website’s opening page it reads: “Have you ever wondered how you can have the greatest impact helping animals? So we have. Animal Charity Evaluators conducts research to answer that very question. Explore our website to learn how you can achieve the greatest good with your efforts to help animals”
This is a perfect opening page for an evaluation organization. The problem is that ACE is not answering that very question. My point here is not that ACE’s evaluations are defective, or its answers are bad (I will make this point later on, but not in this part). It is more fundamental: ACE is omitting from evaluating the effectiveness of different interventions for farmed animals, thus it is not even giving a substantial answer or answers to this very question.
To be fair, ACE makes three points. First, as mentioned above, we should prioritize farmed and wild animals since their numbers are higher and these fields are neglected. Second, as mentioned above, ACE states that it finds the arguments for an institution-focused approach more compelling than individual-focused approaches. Third, as mentioned above, we should consider our long term impacts more carefully, since they will be the most important ones.
But from a practical point of view, these points are either too obvious, too general or too abstract. They do not help an animal advocate who wants to decide which interventions should be prioritized in its organization or a donor who wants to know which charity is impacting the most with additional donations.
In order to give practical guidance to these persons, ACE has to answer this obvious question: Which intervention(s) are the most effective to help animals?
There are many interventions in farmed animal advocacy: corporate outreach for animal welfare reforms, campaigns for animal welfare reforms, vegan pledge programs, protesting for the end of factory farming, restaurant outreach for more vegan options, outreach for meatless monday programs in schools, campaigns for legal change, humane education programs, developing and promoting alternative proteins, building a community of protesters, media campaigns for raising awareness, building a community of alternative protein entrepreneurs etc.
ACE seems to accept that almost all of them can be similarly effective, and do not argue for much prioritization between these programs. For example, ACE does not comment whether animal welfare corporate outreach is more or less effective than restaurant outreach for more vegan options. Nor does it comment whether meatless monday outreach is more or less effective than corporate campaigns for animal welfare.
While ACE does argue for some prioritization of institutional interventions over individual interventions, even this claim is made very softly. There are many recommended charities and movement grantees which focus primarily on individual interventions (mostly vegan outreach). Some examples are standout charities like xiabuVEGAN and Dharma for Animals, and ACE Movement Grantees like Viral Vegans and Gyvi Gali.
So at the end of the day, ACE does steer donors and advocates to farmed and wild animal welfare cause areas, but does not direct them to particular effective interventions. Imagine that GiveWell defends the view that one should donate to charities which operate in Africa but does not comment on which type of interventions most effectively improve the wellbeing of the global poor. Imagine that on GiveWell’s website there are no comparisons of effectiveness between bednet distributions, food aid, malaria medicine distributions, school renovations, deworming, cash transfers, vitamin A supplements, playpumps, vaccine programs and many more. And there are charities in its top charity list which carry out almost all of these programs. Imagine that GiveWell says it prioritizes health charities over education charities but in its top charity list there are some education charities as well. Unfortunately, this hypothetical scenario resembles the current evaluation approach of ACE.
Now let’s look at what GiveWell is doing. First and foremost, GiveWell focuses a lot on programs: “The program an organization chooses to implement is a major factor in the overall impact of its work and thus its performance along our criteria. Because its choice of program(s) has such a strong effect on the amount of good an organization can accomplish, we typically focus on identifying promising programs before we start looking for organizations that are implementing them.” (italics, underlines and bolds are mine). Currently, GiveWell is prioritizing only four types of interventions: medicine to prevent malaria, nets to prevent malaria, supplements to prevent vitamin A deficiency, and cash incentives for routine childhood vaccines. One should note that, in the global health & development field there are hundreds of interventions. But GiveWell prioritizes only four. It has even removed deworming and direct cash transfers, since those are deemed less effective compared to other interventions. I think GiveWell’s prioritization approach, which puts forward a few interventions out of hundreds of interventions which are deemed to be the most effective, by looking at the empirical evidence and theoretical reasons, is a good example of how an evaluation organization should function.
In ACE’s defense, ACE does not have the same level of evidence available to GiveWell, in its own evaluations. ACE states that: “There is little available evidence to support the effectiveness of any given intervention” and that “There is currently no empirical evidence that reviews the effectiveness of movement building in animal advocacy” and that “Due to the lack of research on the extent that animal advocacy research results are used by the movement to prioritize and implement their work, our confidence in the short-term effects of this intervention is low” and that “Currently, there is no peer-reviewed research available about influencing the availability of animal-free products through institutional outreach.” and that: “We believe that increasing the availability of animal-free foods will decrease the consumption of animal products. However, we note that the body of empirical research—especially measuring causal effects—is limited to a few field experiments conducted at universities.” While there are many studies which calculate the effectiveness of different global health and development interventions, the same volume of literature does not exist in animal advocacy cause area.
In response to this, I have three replies. Firstly, there is evidence of tractability for at least some interventions, for animal welfare campaigns in particular. National statistics show a steep reduction in the share and the number of chickens in caged systems after the success of animal welfare corporate outreach and campaigns. While I will make this point in detail in the last part, I have to point out that I disagree with ACE’s claim that “there is little available evidence to support the effectiveness of any given intervention” (italics are mine). There might be little evidence for some interventions (for example restaurant outreach for vegan options), but that does not mean there is little evidence for all possible interventions.
Secondly, if there is not enough evidence for making substantial evaluations, then I think ACE should make this its first priority to conduct research to fill this gap. As a matter of fact, ACE claims that it is already doing that: “Have you ever wondered how you can have the greatest impact helping animals? So we have Animal Charity Evaluators conducts research to answer that very question.” (italics are mine). However when one looks at ACE’s research page, one cannot find research which answers that question. Although ACE has some research briefs on some interventions (but not all of them), there is no research which compares the effectiveness of different interventions. ACE needs to conduct research which can be instrumental to compare different programs, but currently there is no such completed research by ACE. This is also evident since ACE itself admits that there is not enough research, conducted by others or ACE. One should also note that ACE provides many useful resources such as interview series and roundtable series with leading animal advocates, and research briefs. But those are not instrumental for solving the main problem.
Thirdly, while I do not think this is an optimal solution, if ACE sincerely believes that “there is little available evidence to support the effectiveness of any given intervention” and the situation is so futile that one simply cannot make substantial claims about the effectiveness of different interventions, I think it is best for ACE to restructure and rebrand as a hits based “fund” rather than an evaluation organization. I am suggesting this not because I disagree with ACE’s recommendations (I have my disagreements, but those are not relevant at this point), but because ACE is seeming to imply that evaluation is not possible, at least not in a substantial and practical way. If ACE holds the view that it cannot meet the challenge of evaluating the effectiveness of animal advocacy programs, then it should adopt a different role (like a fund).
Currently, ACE on one hand claims to be taking on the challenge of evaluating the effectiveness of different animal advocacy interventions, on the other hand however it omits this challenge by declining to answer the fundamental question that an evaluation organization has to answer: Which intervention(s) are the most effective to help animals?
2.2. Animal Charity Evaluators is acting more like a fund than an evaluation organization.
In its current and past recommendations ACE has a number of charities that do not have a track record of directly impacting animal welfare. These are recommended because of their potential future impact. One can categorize these into two groups. First group consists of charities which engage in research which might impact animal welfare if they reveal important truths and charities which try new (promising) interventions such as alternative protein development or legal reform which might impact animal welfare if they succeed. Some examples are Nonhuman Rights Project, Good Food Institute (at least in 2016), Faunalytics, Wild Animal Welfare Initiative.
Second group consists of charities which operate in regions where animal advocacy is neglected. These are also prioritized for their potential future impact. Some examples are, xiabuVEGAN, Good Food Fund, Dharma Voices of Animals in 2021. These two groups of organizations which are awarded top or standout charity status constitute the majority of ACE’s recommended charities.
ACE’s Movement Grants program functions almost exclusively like a fund. ACE states that:
“Movement Grants is our grant program aimed at building and strengthening the global animal advocacy movement. We are interested in funding groups working on various approaches to animal advocacy, especially those that are underfunded, target large numbers of animals, and are in regions with a relatively small animal advocacy movement. We provide Movement Grants for three main reasons:
We believe that a broad, pluralistic animal advocacy movement is more likely to be resilient—and hence more impactful—than a narrow, monistic animal advocacy movement.
There is little available evidence to support the effectiveness of any given intervention, and we think that animal advocacy is more likely to be successful by continuing to fund a wide range of interventions.
Many countries do not yet have a well-established animal advocacy movement—by funding projects in those countries, we can build up the movement in neglected regions and address animal suffering as a global problem.” (italics are mine)
To be clear, I have no objections against the existence of funds, fund management and grant making. However, ACE’s primary role is evaluation, not fund management or grant making. These are secondary to its role. In fact, these are so secondary that they are not even necessary. An evaluation organization can perfectly fulfill its role of evaluating interventions and charities without engaging in any fund management or grant making at all.
In ACE’s defense, when ACE was founded in 2015, Effective Altruism animal welfare cause area needed both an evaluation organization and a fund. So I think it is understandable that ACE assumed both roles at the beginning.
However, currently there are animal welfare funds other than ACE: Open Philanthropy Farm Animal Welfare Program and Effective Altruism Animal Welfare Funds. So it is not a necessity for ACE to assume this role, next to its evaluation role.
There are two problems with ACE’s role as a fund.
Firstly, ACE mixes these roles. This would not be a huge problem if ACE had one top charity list evaluated by their track record à la GiveWell and a seperate hit-based giving list. But this is not the case. Some of ACE’s top and standout charities are recommended because of their potential future impact, some of them are recommended because they have good track records and engage in tractable programs. One cannot use the same criteria for evaluation and hits based giving. As a result, it is not clear how ACE is making reliable evaluation when executing both charity evaluation and fund management all at once. And it makes the “comparing apples and oranges” problem even worse: “%80 probability of having X kilos of apple pies in 2-3 years vs. %10 probability of having Y liters of orange juice in 20-30 years”. As explained above, ACE is currently unable to answer fundamental questions about the effectiveness of existing programs. Adding the potential future impacts to these calculations renders ACE’s task almost insurmountable.
Secondly, ACE assumes its fund manager role at the expense of its evaluator role. Like all organizations ACE has limited staff. In fact, it has only five research staff. Fund management is hard work. It involves research, evaluation (taking a hits-based approach), grant making, and follow up assessment. So currently ACE is using its limited and precious resources (human and financial), for its fund management work, rather than its primary role: evaluation. This could be fine if ACE was perfectly executing its evaluation role and was not mixing both roles as pointed out above. However, as explained above, ACE has shortcomings in its evaluation methods and desperately needs more empirical and/or theoretical research for answering fundamental questions about effectiveness of different programs. Under these circumstances, it is not an optimal use of ACE’s resources to focus on “building and strengthening the global animal advocacy movement”.
I do not mean to diminish ACE’s position, on the contrary, I think ACE could have a much stronger position and higher impact in Effective Altruism animal welfare cause area if it could double down on its primary evaluator role. A good example is the relationship between GiveWell, Open Philanthropy, and Effective Altruism Funds. GiveWell is an evaluation organization, while Open Philanthropy and Effective Altruism Funds are grant making organizations. Open Philanthropy and Effective Altruism Funds, benefits greatly from GiveWell’s work as an evaluator. Both of those funds channel a significant portion of their grants to GiveWell’s top charities and use some of their resources for low probability-high impact opportunities. If GiveWell did not exist or did not prioritize effective charities in its current form, these funds would be in a very difficult position. They would had to assume evaluator role next to their funder role with limited expertise and experience .This is a good example of effective cooperation and division of labor: GiveWell is better positioned (has more experience and expertise) to make better judgments about the most cost-effective solid funding opportunities, while Open Philanthropy in particular, has a better understanding and experience of hits-based giving.
I think a similar cooperation and division of labor could be much more effective than the current situation where we have three funds instead of two but do not have an evaluator which makes substantial prioritization.
2.3. A framework for a new approach
So how should ACE evaluate? Frankly, I do not know for sure. I do not have the competence or the experience to comment on this question authoritatively. For that reason, I am fairly uncertain about the validity of my suggestions in this part and I am sure that someone with more competence and experience (most probably ACE staff) would give a much better answer to this question. But since this essay is not an open thread, I will put forward my suggestions, for the purpose of showing at least in a crude way, how things could be different and more coherent if ACE adopts a different approach to evaluation. I will take GiveWell’s approach as a model and argue that ACE can take a similar approach.
2.3.1. Stage I: Deciding which programs are effective
First and foremost, as explained above, ACE should start research projects which aim to identify programs which are evidence based and cost effective. This is crucially important since as GiveWell indicates: “organization’s choice of which program to implement is one of the most important factors affecting the impact it can have” (italics, underlines and bolds are mine. I considered adding capital letters too, but decided that that was too much).
ACE can conduct shallow reviews of a larger number of programs and then conduct more intensive reviews of the programs that seem promising. ACE can also partner with other research organizations like Faunalitics or Rethink Priorities. ACE can also ask leading charities which conduct certain programs to make their case, and later critically review their evidence. In fact, ACE seems to have a good template for this process on this page. But the outcomes of this process seem to be lacking.
While making cost effectiveness calculations, ACE should qualify only direct effects on animals as “effect”. It should not count intermediary steps like more vegan options in restaurants, or movement building or number of participants in a pledge program as “effect”. If the effects of an intervention is not tractable enough to be able to show tangible effects on animals, then it should not be qualified as cost effective, at least for the time being. But it could qualify as cost effective if more evidence becomes available in the future.
At the end, ACE can also make judgments on the effectiveness of different interventions and can finally make some prioritizations. I think this is the most important stage, and it can fix ACE’s main shortcomings.
2.3.2. Stage II: Determining charities which have a track record of engaging in effective programs, executing them relatively cost-effectively than other charities, have room for funding, and are transparent. (Top charities)
After establishing the effectiveness of programs, selecting top charities is relatively easier.
ACE would select its candidates and evaluate them on the basis of their track record of engaging in effective programs, executing them relatively cost-effectively than other charities, having room for funding, and transparency. These are more or less similar to GiveWell’s criteria for evaluating charities.
The major difference between ACE’s current approach and this alternative is that this approach is insistent on the track record of engaging in effective programs which are supposedly decided by the research process explained above. Currently ACE does not apply this form of prioritization.
Another minor difference is that I removed ACE’s “leadership and culture” from this evaluation criteria since the impact of this criterion is redundant in the track record (past success) of the organization. I think this criteria is useful and necessary if one is taking the hits based giving approach like a fund (which explains why ACE use it since it is assuming that role as well), but it is not necessary for an (effectiveness) evaluation organization (which explains why GiveWell does not use this criteria). Let me be clear, I think ACE’s intention of promoting healthy organization culture practices in the animal advocacy movement is valuable and laudable. I also believe that diversity, equity, inclusion, and protection from harassment and discrimination are also very important. However, I think ACE is assuming more roles than necessary, again. I think these important issues can be addressed and assessed by another organization which has the resources and the expertise (for example one organization which focuses on racial inequalities is Encompass). I think since ACE’s capacity is already overstretched, it is not able to make the best of this issue. In order to have a real insight about the leadership and the culture of the organization, one needs to conduct in depth research, such as participant observations or open-ended interviews with multiple members of the organization. Currently, ACE is mainly conducting an online survey among the members of the organizations. And as ACE points out “...surveying staff and volunteers could lead to inaccuracies due to selection bias, and that surveys may not reflect employees’ true opinions as they are aware that their responses could influence ACE’s evaluation of their employer”. Therefore it may not be a good criterion to take into consideration while evaluating the current effectiveness of the organization, while it may be a good data point for a hit based giving decision.
Another difference of this approach is the scope of candidates for evaluation. ACE should not evaluate charities which carry out research, like Faunalitcs or Wild Animal Initiative. ACE should not also evaluate charities which do not engage in tractable effective programs or in promising programs which will prove its effectiveness in the future. This is also in line with GiveWell’s model. GiveWell does not consider research organizations or organizations which engage potentially impactful programs which yet lack tractability, as candidates for its top charities. Providing funding for these organizations may also be impactful and even more impactful, from a hit-based giving perspective. But these decisions should be considered and taken by “funds” like Open Philanthropy and Effective Altruism Animal Welfare Fund or simply by individual donors. As pointed above, ACE should not mix evaluation with hit-based fund management.
2.3.3. Stage III: Determining charities which do not have a long track record but are engaging in effective programs (Incubation Grants)
ACE can also consider evaluating charities which engage in effective programs but which currently do not have a long track record, and/or are not currently very cost-effective but can overcome these if given the possibility of scaling. Examples of these kinds of charities may be organizations which engage in effective programs in regions where animal advocacy is neglected. ACE should evaluate these charities in a different category since these are operating in a different context.
For example, let’s assume that The Humane League is a top charity which engages in animal welfare campaigns which is considered by ACE as an effective program. The Humane League achieves many commitments from corporations and affects millions of animals with a budget of millions dollars. Now let’s imagine that another organization from Chile (say Vegetarianos Hoy), which engages in the same effective program (animal welfare campaigns), but due to lack of scale and unfavorable socioeconomic conditions it does not have the same level of cost effectiveness as The Humane League, yet. I think ACE can evaluate Vegetarianos Hoy and other charities which engage in tractable, effective programs in different regions, in a separate category and recommend those who are relatively cost-effective in this category. These organizations should at least have the potential to be top charities of ACE if they are provided enough funding for scaling. This is in line with GiveWell’s now inactive Incubation Grants. GiveWell, currently does not use this name to better reflect that these grants are not exclusively used to incubate early-stage organizations.
One may object to this “stage III” since it may seem contradictory to one of the points that I made earlier: separation of evaluation and fund management roles. In response, I think there is a significant difference between evaluating and funding organizations which engage in intractable interventions, and evaluating and funding organizations which are engaging in the same interventions that already worked elsewhere but just need some funding to scale them in another region. While the former is a form of hit based giving, the latter seems more like a safer bet.
One may also object to this framework by pointing out that ACE cannot function like GiveWell because it is evaluating something entirely different: advocacy. While I agree that ACE cannot function exactly like GiveWell and advocacy evaluations involve larger intervals of uncertainty, I think the main reasons which makes ACE look so different from GiveWell at the moment, have less to do with the things it is evaluating, but more to do with its design of evaluation. As explained above, I think the main issues that separate ACE from GiveWell are firstly, ACE’s reluctance of making claims about the effectiveness of different interventions, i.e. prioritization, and secondly, ACE’s role of fund management which goes hand in hand with its evaluator role. If ACE could make those prioritizations and give up its fund management role, it could very much function like GiveWell, with perhaps larger uncertainty intervals but nevertheless similarly. Sure, in its models ACE will have to make lots of assumptions and grapple with uncertainties, but one should also note that GiveWell also makes lots of assumptions and operates under uncertainty.
Until now, my criticism and suggestions were mainly formal, procedural or structural. In the next part I will make a substantive argument about ACE’s preferences between different interventions.
3. ACE is currently underrating the effectiveness of programs which aim for animal welfare reforms.
As explained above, ACE is not prioritizing heavily between existing interventions to help farmed animals. I think this is a major error not just from a formal point of view, but from a substantive point of view as well. One set of interventions is significantly more tractable than the rest of the interventions: interventions which aim to improve animal welfare standards by corporate outreach and campaigns. I will call them “animal welfare campaigns” in short. I will also focus on cage-free campaigns in particular in order to keep it short, and comment on broiler and fish welfare campaigns later on.
3.1. Massive differences in tractability
Animal welfare campaigns have a short history: almost two decades. The first major cage-free commitment was made by Whole Foods in 2004. But the real momentum began after 2015 (only 7 years ago) when a number of animal charities focused on this intervention and Open Philanthropy provided major grants to organizations which carry out these programs. This is a good example of why prioritization and picking effective programs are crucial in doing the most good for animals.
These campaigns have been very successful to convince or to coerce companies to adopt cage-free standards in their egg supply chain. Although there was some uncertainty about whether corporations would actually stick to their commitments, most of them did implement the reforms. The change can also be seen in government statistics.
Cage-free campaigns are also very successful internationally. You can see hundreds of commitments in Chicken Watch. European Union statistics also show a steep reduction in caged eggs (source).
The welfare gains by cage-free standards are also real and significant. Layer hens in cage-free systems can step on soil, walk, turn around, flap their wings, sleep on high perches, make dust baths, lay eggs in nesting boxes. All those things are very important for their welfare and they are unable to do these things in cage systems. One can look at the Welfare Footprint Project’s website for more information.
The good thing about animal welfare reforms is that one can track the effect of the campaigns on the animals. Another important thing is that animal welfare reforms have a “lock in” effect. Corporations and consumers are forced to buy and consume cage-free rather than caged eggs since the latter option is removed from the menu of options.
Other interventions in the farmed animal advocacy lack this kind of tractability.
Interventions to spread veganism for example, aim to convince people to become vegan through education seminars, pledge programs or leafleting etc. Although more people becoming vegan is a good thing, the question we should ask ourselves is whether these interventions work effectively to bring about the change we want to see. These interventions are intractable since one cannot tract participants and check whether they have become vegan after these interventions. After participating in these events and reading materials, participants can simply not become vegan or reduce their animal product consumption. Although ACE points out that “Some empirical studies suggest that self-monitoring—which is part of taking a veg*n pledge—reduces meat consumption, at least in the short run. Other studies that measure the impact of veg*n pledges suggest that some participants adopt a more plant-based diet for several months after engaging with a pledge.”, as ACE again says: “Previous studies mostly rely on self-reported consumption data, which can be subject to misreporting and biases.” I think the most important data (which is missing in ACE’s reports and reviews) that one should take into account is that after all these years and efforts, the number of vegans is not increasing. One can look at YouGov data for the UK and Gallup data for the US. By the way, in these surveys participants are asked about their self-identity and diet preference, but many people do eat meat despite their answers in surveys (probably due to social desirability bias), thus one can suspect that the real numbers can be even lower.
Same issues with tractability also apply to institutional outreach to influence the availability of animal free products. A restaurant can put some vegan options in its menu as a result of the outreach, but customers may simply not choose them and we may never know whether this intervention reduced the number of animals used and in particular how much it reduced animal product consumption. ACE’s research brief only makes a weak claim that this type of intervention can succeed in making vegan options available, but it doesn’t even claim that the availability of animal free options would lead to lower animal product consumption which is the main objective and counts as the real “effect”. So it is fair to say that this intervention is not tractable at all. Another weakness of this intervention in my view is that one cannot say that it is very neglected. Market forces already play their role here. If there is a sufficient demand in the market for the availability of vegan or plant based options in a restaurant or a market, it is fair to expect that they would do this themselves without the outreach. At least some of the restaurants would do this proportionate to market demand, and the rest would follow if the first movers make decent profit. On the other hand, if there is not enough demand for these goods, then restaurants would eventually drop them from their menus. I can understand that there might be some cases where outreach may accelerate or improve this process but one should not expect a radical change from the existing supply and demand outcomes, since this intervention neither supplies a new good to the market (as alternative protein intervention does) or nor change the demand (as vegan outreach does).
Same issues with tractability also apply to “movement building”. ACE, states that it “Each charity’s long-term impact is plausibly what matters most”, and argues that long term impacts should be given high priority while evaluating the effectiveness of animal charities. This can also explain the foundation of ACE’s Movement Grants whose aim is “building and strengthening the global animal advocacy movement”.
I have no objection to the idea that long term impacts and movement building are important. But I find it odd to isolate and decouple movement building impacts of a charity from its programs (which may be animal welfare campaigns, vegan advocacy or increasing the availability of plant-based options, etc.) which actually aim to impact animals. Isn’t that the point of movement building? And if let’s say a charity builds a bigger movement, the begging questions are: “What is that movement for? What will these more numerous or better organized advocates do to help animals?”
Well, probably they will primarily continue their existing programs since movement building activities would also promote these programs and emphasize their importance. One cannot build a movement without promoting and emphasizing the importance of the very thing that movement is doing. For example, if an animal charity’s main programs are about vegan advocacy or increasing the prevalence of plant-based options, the “movement” that this charity builds would almost surely focus on these programs in the future as well and/or the movement in question can be mobilized for these programs. So even if one agrees that movement building is important, one has to see that it functions as a “multiplier”, multiplying the impact of the organizations’ main program(s) which directly affect animals. As a consequence, if a charity carries out programs which have low tractability (and probably are ineffective), one cannot conclude that due to “movement building” those programs will have much higher impact in the future.
As for alternative protein development and promotion, I think one can make the theoretical argument that if one can develop price and taste competitive plant-based products, consumers may substitute their animal product consumption with these new goods. One can also make the case that factory farming is inherently inefficient since it is first feeding the animal to consume it later, which wastes resources, while plant based food production may use all its raw materials to produce its final product without or with little waste. Since this intervention is novel (do not have a track record of success or failure) and has a decent set of theoretical reasons to back up its case, I agree that funding these interventions from a hits based giving point of view makes sense. However, empirical evidence is not yet clear. Although plant-based foods’ growth in the market is observable, we do not have clear evidence that these goods are substituting animal products, which is our final aim. It is also unclear how much and which type of animal products are being substituted. If alternative proteins only substitutes dairy and beef, then they would hardly remove much animal suffering from the food system, since those animals are large and provide so much animal products. In order for this intervention to be tractable and impactful, one needs to show not only that these goods are in many restaurants and retailers, or that their market share is growing, or that they received a lot of investment, but also that these products substitute significant quantities of animal products which are mostly responsible for high animal suffering (mainly chicken, eggs and fish). As far as I know, we do not have that evidence yet. In fact, there is a study which shows that substitution effect is limited.
Finally, there is one other intervention which some animal charities (which are also in ACE’s recommendation list) engage in: Meatless Monday campaigns. Meatless Mondays campaigns usually target public institutions like schools and universities. The good thing about Meatless Monday campaigns is that it doesn’t just provide an option which can be ignored by the consumers, it removes the option of animal products for lunch meals on mondays. One can therefore make the inference that at least for those meals, animal consumption is reduced. That is how ACE seems to calculate the effectiveness of Meatless Mondays: “We estimate that switching from a typical diet to a vegetarian diet for one year spares approximately 30 land animals. We treat one meatless meal as the equivalent of between ⅓ and ½ of a vegetarian day”.
However there are a lot of moving parts here. Firstly, the institution can simply distribute meals with animal products to other meals of the week. Second, the students can compensate for not eating meat for that meal, with eating more meat at home. Third, it is not clear which type of meat is eliminated, the effect could be much lower if it is just minced beef. Fourth, if the institution or the participants compensate the eliminated meat with eggs, chicken or fish, it could have a negative effect. Fifth, students or workers can bring their own meal or choose to have lunch elsewhere. Finally, while there is an expectation that Meatless Mondays campaigns will convince people to decrease their animal product consumption further and have an impact beyond the cafeteria, given the constant rise of animal consumption and stagnant vegan-vegetarian share of population, this expectation is not supported by evidence. One can also read the comments to this forum post, from which I developed most of these points.
In conclusion, animal welfare reforms massively outperforms other interventions in terms of tractability.
3.2. ACE’s current position with regards to animal welfare reforms and the implications of this critique
As explained above, ACE is not making substantial claims about the relative effectiveness of different interventions. While ACE is stating that it highly prioritizes institutional focused interventions, it does not comment about how much it highly prioritizes the institutional focused interventions relative to individual interventions and it does not also comment how it prioritizes between institutional interventions. Finally, one can interpret that in ACE’s view many charities engage in movement building if they engage with other advocates or organizations even if they do not have a seperate movement building program, so in ACE’s view, an animal charity which has a individual focused intervention can also have high scores since it also engages in movement building which is considered as institutional focused intervention by ACE. If you look at the comprehensive reports of recommended charities, a significant number of charities are considered to be movement building, even if they do not have a seperate program.
Furthermore, although ACE has many charities in its recommendations which engage in animal welfare campaigns, ACE expresses some concerns about animal welfare reforms in the footnotes of its comprehensive reviews: “On average, our team considers advocating for welfare improvements to be a positive and promising approach. However, there are different viewpoints within ACE’s research team on the effect of advocating for animal welfare standards on the spread of anti-speciesist values. There are concerns that arguing for welfare improvements may lead to complacency related to animal welfare and give the public an inconsistent message—e.g., see Wrenn (2012). In addition, there are concerns with the alliance between nonprofit organizations and the companies that are directly responsible for animal exploitation, as explored in Baur and Schmitz (2012).” (italics are mine).
I think this comment in the footnotes should be explained in more detail by ACE. First of all, the cited article, Wrenn (2012) argues that “...abolitionism, as developed by Francione, is the only morally consistent approach for taking the interests of nonhuman animals seriously. Further, it is suggested that the newness of the abolitionist movement and the mainstream nonhuman animal welfare movement’s dismissal of abolitionism has thus far prevented any substantial abolitionist success.” (italics are mine). Given the negative outlook of Francione towards animal welfare reforms, one could expect that ACE might be discounting animal welfare reforms a lot, even if on average it still sees it as “a positive and promising approach”. Because, according to Francione, “Animal welfare reforms are not [even] ‘baby steps’. They are big steps in a backward direction. The problem is use, not treatment. The goal is to abolish animal use, not to regulate treatment. The means to the goal? Go vegan and educate others about veganism.” (italics are mine). Another issue I would like to point out is that while ACE voices concern about animal welfare reforms following a critique à la Francione, it does not mention the same critiques for other interventions. This is interesting because according to Francione almost all of the interventions which ACE is evaluating and approving as effective, do not meet the “moral baseline” of Francione. I think if ACE is discounting animal welfare reforms for Francionist arguments, then the same arguments should lead to the discount of other interventions like Meatless Mondays, vegan pledges, diet change advocacy which aims for “reduction” of animal products, any institutional outreach, or the use of graphic images which show animal suffering in factory farming. Those who are familiar with Francione’s literature know perfectly well that Francione is not just against animal welfare reforms, he is pretty much against most of the interventions mentioned above, including vegan pledges.
Second issue that I am somewhat concerned about in this detail is that Francione’s viewpoints are almost antithetical to current effective animal advocacy approaches. Of course, an organization should be open to diverse viewpoints, and we should never try to suppress them. However, if these diverse viewpoints lead to dysfunctionality, where members do not even agree about the fundamentals of their task, and mainly operate by uneasy concessions and/or opaque reasoning in order to reach a joint resolution, this may not be an optimal framework for both of the viewpoints. If this is the case (which may not be, I am just speculating about a probability), it might be better to present these different viewpoints in separate dissenting votes, rather than in joint decisions.
While from my understanding ACE is underrating animal welfare campaigns, due to the reasons explained above, animal welfare campaigns should be rated and prioritized much highly (in absolute and relative terms). The implications of this approach would probably result in animal charities which primarily engage in animal welfare campaigns occupy all of top and standout charity statuses, ranked according to the number of animals they affect by the commitments and the implementation of the commitments they achieved, the scope of the welfare gains of the reforms, moral weight of the animals relative to their sentience or intelligence (differences say between fish, chickens or shrimps), and their operation costs. This approach would probably result in leaving out animal charities which primarily engage in programs which are not tractable.
As a matter of fact, one can say that Open Philanthropy and Effective Altruism Welfare Funds are already prioritizing animal welfare campaigns, even if they are funds which are functioning on a hits based giving approach. When you look at their grant reports, one can see that the overwhelming majority of their grants go to animal welfare interventions. They do also fund novel interventions which are not yet tractable such as alternative protein development and promotion, and wild animal welfare research (which, as explained above, makes sense from a hits based giving perspective). However, apart from very few exceptions, neither Open Philanthropy nor Effective Altruism Welfare Funds, seem to provide grants to organizations which engage in restaurant outreach for more plant-based options, vegan pledge programs, vegan outreach, or meatless monday campaigns, even from a hits based giving perspective. ACE on the other hand, seem to be in equal distance from each type of programs and seems to be engaging in hits based giving more freely than Open Philanthropy and Effective Altruism Welfare Funds, by attributing top and standout status and providing grants to a wider range of charities, even if it is proclaiming to be evaluator organization, not a hits based fund.
3.3. Concerns and replies
Is animal welfare campaigns all that we should be doing according to this approach?
Yes and no.
First, an unapologetic yes.
If animal welfare campaigns are relatively more tractable and impactful than other interventions according to the existing evidence and reasons, then of course we should be prioritizing animal welfare campaigns over other interventions. Not doing that would result in more animal suffering. Given that we have so much work to do in this field, I think not prioritizing according to evidence and reason has a high opportunity cost.
Prioritization is the key to doing the most good and not acting on evidence and reason on this issue would contradict with the basic principles of effective altruism. I think animal advocates should invest more of their efforts and resources into animal welfare campaigns rather than other interventions. Donors should also channel more of their resources into these projects.
Second, a qualified no.
Firstly, of course, everyone is free to disagree and do what they think is right. My entire point rests on the assumption that the argument made above is supported by evidence and reason, and therefore is convincing for the reader. If one thinks the opposite, then of course one should not be following this advice.
Second, prioritization of animal welfare campaigns is constrained by room for funding. And if animal welfare campaigns are supported to the point of having no more room for funding (and staff), one may look for the second most effective option. However, given that there is a lot of room for funding and there is a lot to be done, this prospect is not likely.
Third, another likely prospect may be to work on and fund promising and novel interventions such as alternative protein development and research for finding new or better ways of helping animals (with a focus on wild animal welfare) or other novel projects supported with reasons. As I mentioned above, these novel interventions may also be deemed effective (even more effective) when more evidence becomes available in the near future.
But these exceptions does not mean that we should not be prioritizing tractable interventions over intractable ones.
I think this answer is similar to the answer which can be given to the question “are bed nets all there is?” First, an unapologetic yes: one should be donating to the most effective funding opportunity available. Second, a qualified no. First, you can do whatever you want with your money, no one can tell you what to do. Second, there could be instances where the Against Malaria Foundation does not have more room for funding and one should look for the second best opportunity and finally, there can be hits based opportunities which can also be effective like funding research for gene drive in mosquitos. But these exceptions should not lead us to the conclusion that we should not be prioritizing between different programs and put bed net distribution programs in the same category with other intractable programs.
Shouldn’t we aim higher?
Yes, but only if we can achieve higher.
Some advocates are not very passionate about animal welfare campaigns because these reforms are incremental as they do not end animal exploitation or lead directly to animal liberation.
While I share this frustration myself and find the desire for more progress for animals laudable, one has to see that aiming higher does not immediately result in achieving higher. There are many aims such as “ending factory farming”, “banning slaughterhouses”, “cutting the global production and consumption of animal products by %50 by 2040”. I would be happy to see these changes, but when deciding about what to prioritize we should not be just judging options by our preference for the end goals, we should also look for the tractability of these interventions. And unless we have evidence that these are in fact tractable, it may not be an effective strategy to always aim high, since that would have an opportunity cost of not carrying out projects which are tractable.
I think this answer is similar to the answer which can be given to the question “shouldn’t we aim higher for improving the welfare of the global poor?”. Yes, but only if we can achieve higher. Some advocates may be less passionate about health interventions in Africa because these reforms are incremental, they do not end the poverty of the global poor. They may be willing to aim for more such as “comprehensive economic development”, “large scale institutional reforms”, “recognition of colonialism and reparations”, “overthrow of global capitalism”. Although there might be some probability to achieve these goals, and everyone has the right to pursue whatever project they like, it is hard to say that the effective strategy is to always aim high, since that would have an opportunity cost of not carrying out projects which are tractable.
Finally, I think these concerns are underestimating how high they are in fact aiming, and overestimating our capacities to reach these goals. Global animal product consumption is at an all time high and increasing (over 330 million metric tons). “The global farmed animal advocacy movement”, only has about 200 million dollars in support and consists of (at most) thousands of people where it has the most power, and only hundreds or even tens of advocates in the rest of the world. We should also not ignore the unfavorable socioeconomic conditions worldwide which limit the public support for animal rights.
I think this concern also underestimates the progress animal welfare campaigns have achieved. The welfare of hundreds of millions of animals improved thanks to these campaigns and there is still much to do: cage-free campaigns in Asia, Africa, Latin America, broiler campaigns worldwide, and new fish and wild animal welfare campaigns which have the potential to impact billions of animals.
Shouldn’t we aim for a broader and pluralistic animal advocacy movement?
While I think diversity and pluralism are central values, I do not think we should necessarily value diversity and pluralism of programs. I think ACE and the Effective Altruism community (or the part of the community which is engaged in animal welfare cause area) should affirm moreprioritization rather than more diversity or pluralism of programs if we are to become more effective and help more animals.
The need for prioritization is evident where ACE highly prioritizes farmed and wild animals. But if we follow the pluralism of programs logic to its conclusions we should not be stopping at farmed and wild animals, we should also include programs which aim to help companion animals or lab animals or captive animals etc. If one agrees that we need prioritization, not pluralism, in the case of selecting animal groups, one should also agree that we also need prioritization, not pluralism, in the case of selecting programs.
Finally, neither this essay nor ACE possess the power to radically change the advocacy landscape. A lot of advocacy groups would continue their existing programs regardless of what ACE or my essay claims.
What if this critique is not correct in its substantive claims?
The main critique of this essay is structural rather than substantive. And even if you did not find the substantive claims in the third part of this essay convincing, it should not devalue the claims made in the first and second part of the essay. I am more concerned with ACE’s omission from making substantive claims, whatever those claims will be. I am ready to update my views if more evidence and reason is presented. I think animal advocates, effective altruist animal advocates in particular, need to have more discussions about the effectiveness of different interventions. And ACE has a distinctive role to enrich and validate the arguments in these debates, by using its resources and expertise.
Conclusion
In conclusion, I would like to underline that this essay is aimed to be a constructive criticism, not a destructive one. I would also like to express clearly that although I believe ACE has some defects, those defects are not related to a lack of hard work, good intentions, talent or honesty. My main criticism is more about what ACE is not doing, not about what ACE is doing. As explained above, these issues can be solved if ACE is willing to act more decisively and assume its primary role rather than doing many things all at once.
A final note in defense of ACE: animal advocacy is very controversial. If ACE or anyone else makes substantive claims about the effectiveness of a certain program which even implicitly implies that other programs are less effective or even ineffective, many advocates would take these claims personally. In comparison, if one makes a substantive claim about the effectiveness of a certain program in global health and development cause area, which even explicitly argues that other programs are less effective, most people would be more like “oh, I liked that [other] program, but okay, I will consider my donations accordingly”. In animal advocacy on the other hand, many advocates have a strong emotional and moral attachment to their prefered intervention and even make this part of their identity. You can find a lot of people who would put “vegan” in their bio in social media or in this forum, but you would be unlikely to find a lot of people who would present themselves as “dewormer”, “ethical bed net donor”. I am not saying that this is a bad thing, in fact, it is a good thing that people take these issues so seriously that they make it part of their identity. But my point is rather that the effectiveness debates occur in a much different way in animal advocacy. And I think it is understandable when an organization is somewhat cautious about making bold claims which is very likely to cause a lot of reaction which might lead to divisions and conflict within the animal advocacy movement.
However, I think ACE should take on this challenge and deal with it, and leave those concerns around movement unity to other actors in the movement to resolve (another point about division of labor). Effective Altruism animal welfare cause area needs effectiveness discussions more than ever. We will need to prioritize even more in the near future when fish welfare asks and wild welfare asks mature. Since there will be so many programs, we cannot afford to not prioritize between different programs. I believe ACE can do much more good if it can be more instrumental to help and inform those prioritizations in the animal welfare cause area.
An Evaluation of Animal Charity Evaluators
Introduction
Animal Charity Evaluators (ACE) is an effective altruism aligned organization which aims to provide information and recommendation to donors and advocates about the effectiveness of various interventions and charities. As one can expect, ACE reviews animal charities based on the effectiveness of their programs and recommends a small number of charities as “top” and “stand-out” charities. ACE also has a Movement Grants program, which provides small grants to various organizations. According to ACE, it has a budget over one million US dollars, and granted and influenced over ten million US dollars worth donations to its recommended charities and promising projects in 2021. This amount (10 million US dollars) is larger than the total amount of Effective Altruism Animal Welfare Fund grants in 2021.
One can say that ACE is positioned to take the role of GiveWell in the Effective Altruism animal welfare cause area. It functions as an expert authority which evaluates different interventions and comes to conclusions about which ones are the most effective and which charities carry them out in the most effective way. One can also say that ACE’s role is even more significant than GiveWell, since it is not just donors who may follow its advice, it is also the animal advocates who may benefit from this expert authority by following its advice on which interventions are the most effective to help animals.
In this essay I will put forward three points of criticism, which I believe can uncover certain defects that ACE has, and can help ACE to overcome them:
1) ACE’s current style of reasoning is somewhat opaque and therefore may mislead ordinary donors and advocates.
2) ACE is omitting from making substantial claims about the most effective ways to help animals, and thus failing in its primary role of evaluating the effectiveness of different interventions and charities.
3) ACE is currently underrating the effectiveness of programs which aim for animal welfare reforms relative to the effectiveness of other interventions.
In the first criticism, I will argue that while ACE makes cost-effectiveness comparisons between charities which carry out similar programs (for example two charities which both engage in animal welfare campaigns), it does not comment how it is comparing charities which carry out different programs (for example, one charity which engage in animal welfare campaigns and another charity which has a vegan pledge program). I will claim that some of the terms that ACE is using such as “relative to other charities” or “highly cost-effective” is likely to be understood by ordinary donors and advocates differently than ACE’s understanding, and ACE is not providing clear context when making these statements.
In the second criticism, I will show that in its current form, ACE seems to accept that almost all programs in farmed animal advocacy may be effective and does not seem to answer fundamental questions which an evaluation organization has to answer, like “which program(s) help animals the most?” or “which programs should an animal charity or an advocate prioritize?” or “which charities should a donor prioritize?”. I will argue that starting from its foundation and in its current form, ACE functions more like a hits based giving “fund”, rather than an evaluation organization, and these multiple roles make it difficult for Animal Charity Evaluators to achieve its primary role: evaluation. I will suggest that ACE should not be timid to put forward its views on the effectiveness of different programs, whatever those views are or will be. I will also have some proposals such as applying stricter and clear cut evaluation criteria, focusing on the past track record of charities rather than their future potential success, and removing research organizations from its evaluations.
In the third criticism, I will argue that given the extremely strong track record of animal welfare corporate campaigns and outreach, and given the (unfortunate) fact that the number of vegans did not budge even though there has been enormous effort in the past, ACE should make its effectiveness evaluations accordingly. In the past, ACE had been criticized for overrating animal welfare reforms. After persistent criticism, ACE seems to move away from its previous position and qualify various interventions in its evaluations as “effective”. Although I agree that some of those criticism was correct and it was the right call for ACE to cut off problematic evaluation methods, I think ACE made an overkill when it put almost all farmed animal advocacy programs in the same league as programs which aim for animal welfare reforms which have the best track record of effectiveness by far. I will show that there is a massive difference in terms of tractability between animal welfare campaigns and other programs, and ACE is not currently acting on this obvious difference.
I think that the most obvious critique is the first one. This is probably also the easiest to fix. I believe that the second critique is the most important one, and probably the most difficult to fix. While the first two criticisms are primarily procedural and structural, the third criticism puts forward a substantial argument about the effectiveness of animal advocacy interventions, and is therefore more controversial. All these criticisms have some links in between and inform one another, but one can also read and judge them separately. One can disagree with one, but agree with another. The validity of each criticism does not depend on the validity of others.
I would like to state that this is a shallow report. I have decent background knowledge and experience in animal advocacy but I could only spend about two weeks preparing this essay. Although I am very confident about the main arguments of this essay which argues that ACE is not fulfilling its evaluation authority role in a substantial, clear manner, and ACE’s evaluations do not reflect the massive differences of tractability between different animal advocacy programs, I am not entirely confident of my understanding of ACE’s evaluation processes and reasoning. So one should not immediately conclude its judgment of ACE based on this critique alone.
I want to say that although this is an essay criticizing ACE, I want the best for ACE and its team members. I am a previous donor to ACE and I always follow their content. I have no intention to discredit their work, their decisions or the work of their recommended charities and grantees. My aim is to present a better and clearer mechanism for ACE to fulfill its mission. I am also aware that the ACE’s task is hard, very hard. Unlike GiveWell, ACE is tasked to evaluate advocacy groups, which makes things much more complicated. What is even more difficult is to compare the cost-effectiveness of different programs in almost totally different fields and/or countries. The problem of “comparing apples and oranges” indeed poses a great burden. I am also aware that ACE has been the target of vicious criticism in the past. This is not one of them and I hope it will not be perceived as such. I will also try to provide as much context as possible below in order to show the difficulties which ACE had faced and is still facing, and how those difficulties are also the reasons for the problems that I am criticizing.
1.ACE’s current style of reasoning is somewhat opaque
1.1. A general description of ACE’s evaluation approach
Following Effective Altruism principles, ACE uses cause prioritization. Accordingly, ACE focuses on farmed and wild animals. There is no mismatch between these areas and ACE’s charity evaluations and grants. There are no top or standout charities in ACE’s recommended charity list which do not work in these areas . Same goes with ACE’s Movement Grants: all groups which have received these grants have at least some programs on farmed or wild animals. It is clear that ACE uses this type of prioritization very strictly. The only exception might be Nonhuman Animal Project which was on ACE’s standout charity list between 2015 and 2019.
ACE has four evaluation criteria: programs (i), room for more funding (ii), cost-effectiveness (iii), leadership and culture (iv).
The most decisive criteria which I want to focus on are programs and cost-effectiveness. For programs, ACE states: “A charity that performs well on this criterion has programs that we expect are highly effective in reducing the suffering of animals. Effective programs: i) work in high-priority cause areas or countries, ii) work toward high-priority outcomes, iii) target high-priority animal groups, and/or iv) pursue interventions that are expected to be highly effective.”
For cost-effectiveness, ACE states: “A charity’s recent cost effectiveness provides an insight into how well it has made use of its available resources and is a useful component in understanding how cost effective future donations to the charity might be. Given the limitations of strictly quantitative approaches, we have moved away from using a fully quantitative model and have instead transitioned to a more qualitative approach that analyzes the resources used and the outputs achieved for each intervention type.”
As for high-priority outcomes, which are mentioned in the programs section, ACE provides a menu of outcomes that “aim to capture the full spectrum of work done by animal advocates to try and produce change for animals.” In this menu of outcomes, there are six main categories: decreased consumption of animal products, direct help, improvement of welfare standards, increased availability of animal-free products, increased prevalence of anti-speciest values and stronger animal advocacy movement. In this menu of outcomes, ACE does not comment on which outcomes or interventions which could work toward those outcomes are more important.
In almost all of its comprehensive reviews, ACE states that: “We categorize the work of animal advocacy charities by their outcomes, broadly distinguishing whether interventions focus on individual or institutional change. Individual-focused interventions often involve decreasing the consumption of animal products, increasing the prevalence of anti-speciesist values, or providing direct help to animals. Institutional change involves improving animal welfare standards, increasing the availability of animal-free products, or strengthening the animal advocacy movement.”
ACE also states: “Currently, we find the arguments for an institution-focused approach more compelling than individual-focused approaches.”
ACE concludes that: “..., when considering charities to evaluate, we prioritize those that work to improve welfare standards, increase the availability of animal-free products, or strengthen the animal advocacy movement. We give lower priority to charities that focus on decreasing the consumption of animal products, increasing the prevalence of anti-speciesist values, or providing direct help to animals.”
As for where would programs be more effective ACE states that: “We prioritize charities in countries with relatively large animal agricultural industries, few other charities engaged in similar work, and in which animal advocacy is likely to be feasible and have a lasting impact.”.
As for which groups of animals should be prioritized, ACE states: “We prioritize programs targeting specific groups of animals that are affected in large numbers and receive relatively little attention in animal advocacy.”
Finally, ACE states that: “Each charity’s long-term impact is plausibly what matters most. The potential number of animals affected increases over time due to population growth and an accumulation of generations. Thus, we would expect that the long-term impacts of an action would likely affect more animals than the short-term impacts of the same action.”. This point explains why ACE strongly prioritizes programs which aim for a stronger animal advocacy movement.
1.2. ACE’ use of some expressions are somewhat different than their ordinary meaning
1.2.1. “relative to other charities”
In order to select its top and standout charities, ACE has to compare the effectiveness of different animal charities. As a result, some are judged to be more effective than others and therefore earn top or standout status, while others who are judged to be less effective do not earn top or standout status. And when ACE does not qualify a charity as a top or standout charity, ACE can and does justify its decision by claiming the relative ineffectiveness or the relative low outcomes of that charity.
For example, on the overview page of ACE’s evaluation of Compassion in World Farming (CIWF) International, ACE comments that : “Overall, Compassion International’s legal advocacy and corporate outreach work—while potentially impactful—seem to have achieved outcomes that are below average relative to other charities we have reviewed this year.” . Or on the overview page of ACE’s evaluation of Environment & Animal Society of Taiwan (EAST), ACE comments that: “Overall, EAST’s work seems slightly less cost effective than the average charity under review.”
Normally, when an average donor or an animal advocate reads these statements, one can expect that these charities work are less cost effective or have low outcomes relative to all top and standout charities reviewed in those years. A normal interpretation would conclude that all charities which ACE judged to be top or standout charities have higher cost effectiveness and outcomes than Compassion in World Farming International or EAST, thus have relatively more impact per dollar spent.
But this is not the case. While ACE uses the expression “relative to other charities” in a short form in its overview pages, it adds an important context to that expression in the comprehensive report pages which I believe changes the ordinary meaning of the expression.
In those comprehensive report pages, ACE uses the same expression “relative to other charities” again without any additional qualifications at the beginning. But when one goes on reading the comprehensive report, one sees a different meaning.
ACE states that: “This year, we have used an approach in which we more qualitatively analyze a charity’s costs and outcomes. In particular, we have focused on the cost-effectiveness of the charity’s specific implementation of each of its programs in comparison to similar programs conducted by other charities we are reviewing this year.” (italics are mine).
ACE also states that: “After accounting for all of their key results and expenditures, we think the cost effectiveness of EAST’s work in their corporate engagement and industry engagement programs seem slightly lower than the average cost effectiveness of other, similar programs working toward improving welfare standards we evaluated this year.” (italics are mine).
In conclusion, when ACE is saying that X charity is less cost-effective or achieved lower outcomes “relative to other charities”, what ACE is meaning is that X charity is less cost-effective or achieved lower outcomes “relative to other charities which are engaged in similar programs”.
Those are two very different things. First, unqualified expression (“relative to other charities”) would normally be understood as the said charity is less effective than all top and standout charities, while the second, true version of the expression (relative to other charities which are engaged in similar programs), would be understood as the said charity is less effective than other charities which carry out similar programs only. And one has to read through the comprehensive reports to understand the difference.
How ACE compares charities which engage in different programs is another topic which I will explain in detail in the next section (1.3. There are some important gaps in ACE’s evaluation reasoning).
1.2.2. “cost-effective”
As one can expect, ACE uses the expression “cost-effective” a lot. Here are some examples:
1- “Compassion USA does not have plans for any significant expansions, so their room for more funding is limited. While we do not have concerns about the cost effectiveness of their program to encourage individuals to reduce their consumption of animal products, we are less confident about the cost effectiveness of their corporate outreach and public engagement programs.”
2- “We believe that xiaobuVEGAN’s work is particularly effective due to the neglectedness of farmed animal advocacy in China. As a growing charity planning to expand their programs and invest in staff training, we think that xiaobuVEGAN has room for more funding, and we do not have concerns about the cost effectiveness of their programs.”
3- “SVB’s work to increase the availability of animal-free products and their work to decrease consumption of animal products seem particularly cost effective.”
4- “We think that Vegetarianos Hoy’s programs that improve welfare standards, decrease consumption of animal products, and increase the availability of alternatives to animal products are highly effective.”
5- “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”
For an ordinary donor or animal advocate, “cost-effective” means more or less something like, the degree to which something is effective or productive in achieving desired outcomes in relation to its costs. In this context, for an ordinary animal charity donor or an animal advocate, a cost-effective intervention in animal advocacy would reduce or remove a high amount of animal suffering for a low cost.
However ACE uses the expression “cost-effective” somewhat differently than its ordinary meaning. Firstly, at least for some interventions, “the effect” or “the outcome” of the intervention is not necessarily the reduction of animal suffering, but some intermediate steps which might lead to that outcome. For example, when ACE claims that a program which involves corporate outreach to restaurants in order to increase the prevalence of plant based options, is cost-effective, the “effect” is not measured by the amount of animal suffering removed from the food supply chains, but rather the number of restaurants which add a vegan option to their menus. Even when ACE states that “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”, ACE does not mean that the effect of the intervention is the reduction of animal product consumption, it is more like reaching out to and participation of more people in these programs.
You can look at the ACE’s cost-effectiveness estimates spreadsheet for public view.
Secondly, when ACE is commenting on the cost-effectiveness of a program, for example: “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”, it is making these comments relative to each intervention type.
ACE states: “A charity’s recent cost effectiveness provides an insight into how well it has made use of its available resources and is a useful component in understanding how cost effective future donations to the charity might be. Given the limitations of strictly quantitative approaches, we have moved away from using a fully quantitative model and have instead transitioned to a more qualitative approach that analyzes the resources used and the outputs achieved for each intervention type.” and states: “The variation in cost effectiveness between charities was more dependent on which interventions the charity used rather than how they were implemented. This suggests that, rather than modeling the cost effectiveness of each charity, it would be more useful to model the average cost effectiveness of each intervention and incorporate that into our discussion of effectiveness in our Programs criterion.” (italics are mine).
For example, when ACE states “While we do not have concerns about the cost effectiveness of their [CIWF’s] program to encourage individuals to reduce their consumption of animal products, we are less confident about the cost effectiveness of their corporate outreach and public engagement programs.”, it means that CIWF’s corporate outreach program is relatively less cost-effective than other corporate programs carried out by other charities, it does not mean that this corporate outreach program is relatively less cost-effective than all other programs, or CIWF’s individual outreach program. It is entirely possible (in my view, very probable) that CIWF’s corporate outreach program is much more cost-effective than its individual outreach program in terms of reducing animal suffering per dollar spent. And at this point, I think an average donor or animal advocate would be unable to understand this important difference since ACE is not providing this important context.
Another example is this statement: “We believe that pledge programs are highly effective in decreasing the consumption of animal products.”. An ordinary person would understand from this claim that each vegan pledge program would convince a considerable number of people to become vegan, vegetarian or flexitarian and decrease their consumption of animal products for a significant amount of time. However, that is not how ACE is using this expression. In ACE’s understanding, a highly effective program which decreases the consumption of animal products, is one which is relatively more effective compared to other programs which aim to decrease the consumption of animal products, such as leafleting or humane education.
1.3. There are some important gaps in ACE’s evaluation reasoning
You might be asking yourself by now how ACE compares the cost-effectiveness of different programs (for example the cost-effectiveness difference between an animal welfare corporate outreach program and restaurant outreach program which aims to popularize vegan options) and the effectiveness of charities which engage in different programs (for example the cost-effectiveness difference between Compassion in World Farming and xiabuVEGAN).
Although I tried my best to answer this question, I could not find an answer on ACE’s website.
We have only a partial explanation which was cited above: “We categorize the work of animal advocacy charities by their outcomes, broadly distinguishing whether interventions focus on individual or institutional change. Individual-focused interventions often involve decreasing the consumption of animal products, increasing the prevalence of anti-speciesist values, or providing direct help to animals. Institutional change involves improving animal welfare standards, increasing the availability of animal-free products, or strengthening the animal advocacy movement.”
However, ACE does not comment how they are prioritizing between programs. Firstly, ACE does not indicate how they weigh the interventions which they give high priority. For example, it is not clear how much more weight does ACE put on the effectiveness of a corporate animal welfare campaign over, say, a program which aims to strengthen the animal advocacy movement, both of which are prioritized highly by ACE. Secondly, ACE does not comment on the degree of difference between high and low priority works. For example, it is not clear how much more weight does ACE put on the effectiveness of a corporate animal welfare campaign over, say, a vegan pledge program.
ACE also does not comment on how important it is for a charity to be operating in a country where there are lots of farmed animals and animal advocacy is neglected. For example it is not clear how much more weight does ACE put on an animal charity operating in China, over say, an effective yet European animal charity.
These questions are important and decisive since ACE does make those comparisons and act on them eventually by awarding top and standout status to some charities, while not awarding these to the rest of charities which it is reviewing.
It is not also clear whether there are some categories or quotas by which ACE selects its recommended charities on the basis of the effectiveness within the framework of that category. For example, is there a quota for charities which operate in regions where animal advocacy is neglected? Or is there a quota for charities which focus on research? Or is there a quota for charities which focus on animal welfare reforms or vegan advocacy?
From an outsider viewpoint, for the reasons explained above, how ACE is making these comparisons and reaching its decisions are somewhat opaque and can benefit from clearer reasoning.
Here are some suggestions which can fix these problems:
ACE can consider adding more context to its statements without expecting donors or advocates to read the comprehensive reviews in detail, and can consider using expressions in their ordinary meanings.
ACE can consider providing more reasoning about its comparisons between different programs and charities which engage in different programs and operate in types of countries.
ACE can consider creating categories for different charities which engage in different programs and evaluate their status in each different category. For example: “ charities which engage in programs which aim to improve animal welfare”, “ charities which engage in research programs”, “charities which engage in programs which aim to spread veganism”, “charities which operate in regions where there are lots of farmed animals and animal advocacy is neglected”, “charities which focus on movement building” etc.
In ACE’s defense, it is very difficult to compare the effectiveness of different programs. “Comparing apples and oranges” is indeed a problem for an evaluation organization. However, I think it is fair to say that one should at least say clearly that it is weighing apples when it is evaluating apples and not oranges, and vice versa. I think it is also reasonable to expect from an evaluation organization an explanation of how it is comparing apples and oranges since it is actually comparing apples and oranges. Finally, even if ACE is making these comparisons separately in each category, I think it is better for ACE to state those categories (best apples in the apple category, best oranges in the orange category), and make its recommendations in those separate categories. I can imagine that this may sacrifice some simplicity since there will be multiple lists rather than the current single list, but I think a coherent and clear complexity is preferable to an incoherent and obscure simplicity.
2. ACE is omitting from making substantial claims about the most effective ways to help animals.
2.1. The elephant in the room for funding
On ACE’s website’s opening page it reads: “Have you ever wondered how you can have the greatest impact helping animals? So we have. Animal Charity Evaluators conducts research to answer that very question. Explore our website to learn how you can achieve the greatest good with your efforts to help animals”
This is a perfect opening page for an evaluation organization. The problem is that ACE is not answering that very question. My point here is not that ACE’s evaluations are defective, or its answers are bad (I will make this point later on, but not in this part). It is more fundamental: ACE is omitting from evaluating the effectiveness of different interventions for farmed animals, thus it is not even giving a substantial answer or answers to this very question.
To be fair, ACE makes three points. First, as mentioned above, we should prioritize farmed and wild animals since their numbers are higher and these fields are neglected. Second, as mentioned above, ACE states that it finds the arguments for an institution-focused approach more compelling than individual-focused approaches. Third, as mentioned above, we should consider our long term impacts more carefully, since they will be the most important ones.
But from a practical point of view, these points are either too obvious, too general or too abstract. They do not help an animal advocate who wants to decide which interventions should be prioritized in its organization or a donor who wants to know which charity is impacting the most with additional donations.
In order to give practical guidance to these persons, ACE has to answer this obvious question: Which intervention(s) are the most effective to help animals?
There are many interventions in farmed animal advocacy: corporate outreach for animal welfare reforms, campaigns for animal welfare reforms, vegan pledge programs, protesting for the end of factory farming, restaurant outreach for more vegan options, outreach for meatless monday programs in schools, campaigns for legal change, humane education programs, developing and promoting alternative proteins, building a community of protesters, media campaigns for raising awareness, building a community of alternative protein entrepreneurs etc.
ACE seems to accept that almost all of them can be similarly effective, and do not argue for much prioritization between these programs. For example, ACE does not comment whether animal welfare corporate outreach is more or less effective than restaurant outreach for more vegan options. Nor does it comment whether meatless monday outreach is more or less effective than corporate campaigns for animal welfare.
While ACE does argue for some prioritization of institutional interventions over individual interventions, even this claim is made very softly. There are many recommended charities and movement grantees which focus primarily on individual interventions (mostly vegan outreach). Some examples are standout charities like xiabuVEGAN and Dharma for Animals, and ACE Movement Grantees like Viral Vegans and Gyvi Gali.
So at the end of the day, ACE does steer donors and advocates to farmed and wild animal welfare cause areas, but does not direct them to particular effective interventions. Imagine that GiveWell defends the view that one should donate to charities which operate in Africa but does not comment on which type of interventions most effectively improve the wellbeing of the global poor. Imagine that on GiveWell’s website there are no comparisons of effectiveness between bednet distributions, food aid, malaria medicine distributions, school renovations, deworming, cash transfers, vitamin A supplements, playpumps, vaccine programs and many more. And there are charities in its top charity list which carry out almost all of these programs. Imagine that GiveWell says it prioritizes health charities over education charities but in its top charity list there are some education charities as well. Unfortunately, this hypothetical scenario resembles the current evaluation approach of ACE.
Now let’s look at what GiveWell is doing. First and foremost, GiveWell focuses a lot on programs: “The program an organization chooses to implement is a major factor in the overall impact of its work and thus its performance along our criteria. Because its choice of program(s) has such a strong effect on the amount of good an organization can accomplish, we typically focus on identifying promising programs before we start looking for organizations that are implementing them.” (italics, underlines and bolds are mine). Currently, GiveWell is prioritizing only four types of interventions: medicine to prevent malaria, nets to prevent malaria, supplements to prevent vitamin A deficiency, and cash incentives for routine childhood vaccines. One should note that, in the global health & development field there are hundreds of interventions. But GiveWell prioritizes only four. It has even removed deworming and direct cash transfers, since those are deemed less effective compared to other interventions. I think GiveWell’s prioritization approach, which puts forward a few interventions out of hundreds of interventions which are deemed to be the most effective, by looking at the empirical evidence and theoretical reasons, is a good example of how an evaluation organization should function.
In ACE’s defense, ACE does not have the same level of evidence available to GiveWell, in its own evaluations. ACE states that: “There is little available evidence to support the effectiveness of any given intervention” and that “There is currently no empirical evidence that reviews the effectiveness of movement building in animal advocacy” and that “Due to the lack of research on the extent that animal advocacy research results are used by the movement to prioritize and implement their work, our confidence in the short-term effects of this intervention is low” and that “Currently, there is no peer-reviewed research available about influencing the availability of animal-free products through institutional outreach.” and that: “We believe that increasing the availability of animal-free foods will decrease the consumption of animal products. However, we note that the body of empirical research—especially measuring causal effects—is limited to a few field experiments conducted at universities.” While there are many studies which calculate the effectiveness of different global health and development interventions, the same volume of literature does not exist in animal advocacy cause area.
In response to this, I have three replies. Firstly, there is evidence of tractability for at least some interventions, for animal welfare campaigns in particular. National statistics show a steep reduction in the share and the number of chickens in caged systems after the success of animal welfare corporate outreach and campaigns. While I will make this point in detail in the last part, I have to point out that I disagree with ACE’s claim that “there is little available evidence to support the effectiveness of any given intervention” (italics are mine). There might be little evidence for some interventions (for example restaurant outreach for vegan options), but that does not mean there is little evidence for all possible interventions.
Secondly, if there is not enough evidence for making substantial evaluations, then I think ACE should make this its first priority to conduct research to fill this gap. As a matter of fact, ACE claims that it is already doing that: “Have you ever wondered how you can have the greatest impact helping animals? So we have Animal Charity Evaluators conducts research to answer that very question.” (italics are mine). However when one looks at ACE’s research page, one cannot find research which answers that question. Although ACE has some research briefs on some interventions (but not all of them), there is no research which compares the effectiveness of different interventions. ACE needs to conduct research which can be instrumental to compare different programs, but currently there is no such completed research by ACE. This is also evident since ACE itself admits that there is not enough research, conducted by others or ACE. One should also note that ACE provides many useful resources such as interview series and roundtable series with leading animal advocates, and research briefs. But those are not instrumental for solving the main problem.
Thirdly, while I do not think this is an optimal solution, if ACE sincerely believes that “there is little available evidence to support the effectiveness of any given intervention” and the situation is so futile that one simply cannot make substantial claims about the effectiveness of different interventions, I think it is best for ACE to restructure and rebrand as a hits based “fund” rather than an evaluation organization. I am suggesting this not because I disagree with ACE’s recommendations (I have my disagreements, but those are not relevant at this point), but because ACE is seeming to imply that evaluation is not possible, at least not in a substantial and practical way. If ACE holds the view that it cannot meet the challenge of evaluating the effectiveness of animal advocacy programs, then it should adopt a different role (like a fund).
Currently, ACE on one hand claims to be taking on the challenge of evaluating the effectiveness of different animal advocacy interventions, on the other hand however it omits this challenge by declining to answer the fundamental question that an evaluation organization has to answer: Which intervention(s) are the most effective to help animals?
2.2. Animal Charity Evaluators is acting more like a fund than an evaluation organization.
In its current and past recommendations ACE has a number of charities that do not have a track record of directly impacting animal welfare. These are recommended because of their potential future impact. One can categorize these into two groups. First group consists of charities which engage in research which might impact animal welfare if they reveal important truths and charities which try new (promising) interventions such as alternative protein development or legal reform which might impact animal welfare if they succeed. Some examples are Nonhuman Rights Project, Good Food Institute (at least in 2016), Faunalytics, Wild Animal Welfare Initiative.
Second group consists of charities which operate in regions where animal advocacy is neglected. These are also prioritized for their potential future impact. Some examples are, xiabuVEGAN, Good Food Fund, Dharma Voices of Animals in 2021. These two groups of organizations which are awarded top or standout charity status constitute the majority of ACE’s recommended charities.
ACE’s Movement Grants program functions almost exclusively like a fund. ACE states that:
“Movement Grants is our grant program aimed at building and strengthening the global animal advocacy movement. We are interested in funding groups working on various approaches to animal advocacy, especially those that are underfunded, target large numbers of animals, and are in regions with a relatively small animal advocacy movement. We provide Movement Grants for three main reasons:
We believe that a broad, pluralistic animal advocacy movement is more likely to be resilient—and hence more impactful—than a narrow, monistic animal advocacy movement.
There is little available evidence to support the effectiveness of any given intervention, and we think that animal advocacy is more likely to be successful by continuing to fund a wide range of interventions.
Many countries do not yet have a well-established animal advocacy movement—by funding projects in those countries, we can build up the movement in neglected regions and address animal suffering as a global problem.” (italics are mine)
To be clear, I have no objections against the existence of funds, fund management and grant making. However, ACE’s primary role is evaluation, not fund management or grant making. These are secondary to its role. In fact, these are so secondary that they are not even necessary. An evaluation organization can perfectly fulfill its role of evaluating interventions and charities without engaging in any fund management or grant making at all.
In ACE’s defense, when ACE was founded in 2015, Effective Altruism animal welfare cause area needed both an evaluation organization and a fund. So I think it is understandable that ACE assumed both roles at the beginning.
However, currently there are animal welfare funds other than ACE: Open Philanthropy Farm Animal Welfare Program and Effective Altruism Animal Welfare Funds. So it is not a necessity for ACE to assume this role, next to its evaluation role.
There are two problems with ACE’s role as a fund.
Firstly, ACE mixes these roles. This would not be a huge problem if ACE had one top charity list evaluated by their track record à la GiveWell and a seperate hit-based giving list. But this is not the case. Some of ACE’s top and standout charities are recommended because of their potential future impact, some of them are recommended because they have good track records and engage in tractable programs. One cannot use the same criteria for evaluation and hits based giving. As a result, it is not clear how ACE is making reliable evaluation when executing both charity evaluation and fund management all at once. And it makes the “comparing apples and oranges” problem even worse: “%80 probability of having X kilos of apple pies in 2-3 years vs. %10 probability of having Y liters of orange juice in 20-30 years”. As explained above, ACE is currently unable to answer fundamental questions about the effectiveness of existing programs. Adding the potential future impacts to these calculations renders ACE’s task almost insurmountable.
Secondly, ACE assumes its fund manager role at the expense of its evaluator role. Like all organizations ACE has limited staff. In fact, it has only five research staff. Fund management is hard work. It involves research, evaluation (taking a hits-based approach), grant making, and follow up assessment. So currently ACE is using its limited and precious resources (human and financial), for its fund management work, rather than its primary role: evaluation. This could be fine if ACE was perfectly executing its evaluation role and was not mixing both roles as pointed out above. However, as explained above, ACE has shortcomings in its evaluation methods and desperately needs more empirical and/or theoretical research for answering fundamental questions about effectiveness of different programs. Under these circumstances, it is not an optimal use of ACE’s resources to focus on “building and strengthening the global animal advocacy movement”.
I do not mean to diminish ACE’s position, on the contrary, I think ACE could have a much stronger position and higher impact in Effective Altruism animal welfare cause area if it could double down on its primary evaluator role. A good example is the relationship between GiveWell, Open Philanthropy, and Effective Altruism Funds. GiveWell is an evaluation organization, while Open Philanthropy and Effective Altruism Funds are grant making organizations. Open Philanthropy and Effective Altruism Funds, benefits greatly from GiveWell’s work as an evaluator. Both of those funds channel a significant portion of their grants to GiveWell’s top charities and use some of their resources for low probability-high impact opportunities. If GiveWell did not exist or did not prioritize effective charities in its current form, these funds would be in a very difficult position. They would had to assume evaluator role next to their funder role with limited expertise and experience .This is a good example of effective cooperation and division of labor: GiveWell is better positioned (has more experience and expertise) to make better judgments about the most cost-effective solid funding opportunities, while Open Philanthropy in particular, has a better understanding and experience of hits-based giving.
I think a similar cooperation and division of labor could be much more effective than the current situation where we have three funds instead of two but do not have an evaluator which makes substantial prioritization.
2.3. A framework for a new approach
So how should ACE evaluate? Frankly, I do not know for sure. I do not have the competence or the experience to comment on this question authoritatively. For that reason, I am fairly uncertain about the validity of my suggestions in this part and I am sure that someone with more competence and experience (most probably ACE staff) would give a much better answer to this question. But since this essay is not an open thread, I will put forward my suggestions, for the purpose of showing at least in a crude way, how things could be different and more coherent if ACE adopts a different approach to evaluation. I will take GiveWell’s approach as a model and argue that ACE can take a similar approach.
2.3.1. Stage I: Deciding which programs are effective
First and foremost, as explained above, ACE should start research projects which aim to identify programs which are evidence based and cost effective. This is crucially important since as GiveWell indicates: “organization’s choice of which program to implement is one of the most important factors affecting the impact it can have” (italics, underlines and bolds are mine. I considered adding capital letters too, but decided that that was too much).
ACE can conduct shallow reviews of a larger number of programs and then conduct more intensive reviews of the programs that seem promising. ACE can also partner with other research organizations like Faunalitics or Rethink Priorities. ACE can also ask leading charities which conduct certain programs to make their case, and later critically review their evidence. In fact, ACE seems to have a good template for this process on this page. But the outcomes of this process seem to be lacking.
While making cost effectiveness calculations, ACE should qualify only direct effects on animals as “effect”. It should not count intermediary steps like more vegan options in restaurants, or movement building or number of participants in a pledge program as “effect”. If the effects of an intervention is not tractable enough to be able to show tangible effects on animals, then it should not be qualified as cost effective, at least for the time being. But it could qualify as cost effective if more evidence becomes available in the future.
After these processes, ACE should be able to publish write-ups of these program investigations like GiveWell. Here are some examples: cash transfers, vitamin A supplementation, Aspirin for cardiovascular disease prevention. clean cookstoves. GiveWell seems to have reviewed 84 programs, it is currently prioritizing only 4 programs.
At the end, ACE can also make judgments on the effectiveness of different interventions and can finally make some prioritizations. I think this is the most important stage, and it can fix ACE’s main shortcomings.
2.3.2. Stage II: Determining charities which have a track record of engaging in effective programs, executing them relatively cost-effectively than other charities, have room for funding, and are transparent. (Top charities)
After establishing the effectiveness of programs, selecting top charities is relatively easier.
ACE would select its candidates and evaluate them on the basis of their track record of engaging in effective programs, executing them relatively cost-effectively than other charities, having room for funding, and transparency. These are more or less similar to GiveWell’s criteria for evaluating charities.
The major difference between ACE’s current approach and this alternative is that this approach is insistent on the track record of engaging in effective programs which are supposedly decided by the research process explained above. Currently ACE does not apply this form of prioritization.
Another minor difference is that I removed ACE’s “leadership and culture” from this evaluation criteria since the impact of this criterion is redundant in the track record (past success) of the organization. I think this criteria is useful and necessary if one is taking the hits based giving approach like a fund (which explains why ACE use it since it is assuming that role as well), but it is not necessary for an (effectiveness) evaluation organization (which explains why GiveWell does not use this criteria). Let me be clear, I think ACE’s intention of promoting healthy organization culture practices in the animal advocacy movement is valuable and laudable. I also believe that diversity, equity, inclusion, and protection from harassment and discrimination are also very important. However, I think ACE is assuming more roles than necessary, again. I think these important issues can be addressed and assessed by another organization which has the resources and the expertise (for example one organization which focuses on racial inequalities is Encompass). I think since ACE’s capacity is already overstretched, it is not able to make the best of this issue. In order to have a real insight about the leadership and the culture of the organization, one needs to conduct in depth research, such as participant observations or open-ended interviews with multiple members of the organization. Currently, ACE is mainly conducting an online survey among the members of the organizations. And as ACE points out “...surveying staff and volunteers could lead to inaccuracies due to selection bias, and that surveys may not reflect employees’ true opinions as they are aware that their responses could influence ACE’s evaluation of their employer”. Therefore it may not be a good criterion to take into consideration while evaluating the current effectiveness of the organization, while it may be a good data point for a hit based giving decision.
Another difference of this approach is the scope of candidates for evaluation. ACE should not evaluate charities which carry out research, like Faunalitcs or Wild Animal Initiative. ACE should not also evaluate charities which do not engage in tractable effective programs or in promising programs which will prove its effectiveness in the future. This is also in line with GiveWell’s model. GiveWell does not consider research organizations or organizations which engage potentially impactful programs which yet lack tractability, as candidates for its top charities. Providing funding for these organizations may also be impactful and even more impactful, from a hit-based giving perspective. But these decisions should be considered and taken by “funds” like Open Philanthropy and Effective Altruism Animal Welfare Fund or simply by individual donors. As pointed above, ACE should not mix evaluation with hit-based fund management.
2.3.3. Stage III: Determining charities which do not have a long track record but are engaging in effective programs (Incubation Grants)
ACE can also consider evaluating charities which engage in effective programs but which currently do not have a long track record, and/or are not currently very cost-effective but can overcome these if given the possibility of scaling. Examples of these kinds of charities may be organizations which engage in effective programs in regions where animal advocacy is neglected. ACE should evaluate these charities in a different category since these are operating in a different context.
For example, let’s assume that The Humane League is a top charity which engages in animal welfare campaigns which is considered by ACE as an effective program. The Humane League achieves many commitments from corporations and affects millions of animals with a budget of millions dollars. Now let’s imagine that another organization from Chile (say Vegetarianos Hoy), which engages in the same effective program (animal welfare campaigns), but due to lack of scale and unfavorable socioeconomic conditions it does not have the same level of cost effectiveness as The Humane League, yet. I think ACE can evaluate Vegetarianos Hoy and other charities which engage in tractable, effective programs in different regions, in a separate category and recommend those who are relatively cost-effective in this category. These organizations should at least have the potential to be top charities of ACE if they are provided enough funding for scaling. This is in line with GiveWell’s now inactive Incubation Grants. GiveWell, currently does not use this name to better reflect that these grants are not exclusively used to incubate early-stage organizations.
One may object to this “stage III” since it may seem contradictory to one of the points that I made earlier: separation of evaluation and fund management roles. In response, I think there is a significant difference between evaluating and funding organizations which engage in intractable interventions, and evaluating and funding organizations which are engaging in the same interventions that already worked elsewhere but just need some funding to scale them in another region. While the former is a form of hit based giving, the latter seems more like a safer bet.
One may also object to this framework by pointing out that ACE cannot function like GiveWell because it is evaluating something entirely different: advocacy. While I agree that ACE cannot function exactly like GiveWell and advocacy evaluations involve larger intervals of uncertainty, I think the main reasons which makes ACE look so different from GiveWell at the moment, have less to do with the things it is evaluating, but more to do with its design of evaluation. As explained above, I think the main issues that separate ACE from GiveWell are firstly, ACE’s reluctance of making claims about the effectiveness of different interventions, i.e. prioritization, and secondly, ACE’s role of fund management which goes hand in hand with its evaluator role. If ACE could make those prioritizations and give up its fund management role, it could very much function like GiveWell, with perhaps larger uncertainty intervals but nevertheless similarly. Sure, in its models ACE will have to make lots of assumptions and grapple with uncertainties, but one should also note that GiveWell also makes lots of assumptions and operates under uncertainty.
Until now, my criticism and suggestions were mainly formal, procedural or structural. In the next part I will make a substantive argument about ACE’s preferences between different interventions.
3. ACE is currently underrating the effectiveness of programs which aim for animal welfare reforms.
As explained above, ACE is not prioritizing heavily between existing interventions to help farmed animals. I think this is a major error not just from a formal point of view, but from a substantive point of view as well. One set of interventions is significantly more tractable than the rest of the interventions: interventions which aim to improve animal welfare standards by corporate outreach and campaigns. I will call them “animal welfare campaigns” in short. I will also focus on cage-free campaigns in particular in order to keep it short, and comment on broiler and fish welfare campaigns later on.
3.1. Massive differences in tractability
Animal welfare campaigns have a short history: almost two decades. The first major cage-free commitment was made by Whole Foods in 2004. But the real momentum began after 2015 (only 7 years ago) when a number of animal charities focused on this intervention and Open Philanthropy provided major grants to organizations which carry out these programs. This is a good example of why prioritization and picking effective programs are crucial in doing the most good for animals.
These campaigns have been very successful to convince or to coerce companies to adopt cage-free standards in their egg supply chain. Although there was some uncertainty about whether corporations would actually stick to their commitments, most of them did implement the reforms. The change can also be seen in government statistics.
Source: Report from The Humane League Labs
Cage-free campaigns are also very successful internationally. You can see hundreds of commitments in Chicken Watch. European Union statistics also show a steep reduction in caged eggs (source).
The welfare gains by cage-free standards are also real and significant. Layer hens in cage-free systems can step on soil, walk, turn around, flap their wings, sleep on high perches, make dust baths, lay eggs in nesting boxes. All those things are very important for their welfare and they are unable to do these things in cage systems. One can look at the Welfare Footprint Project’s website for more information.
The good thing about animal welfare reforms is that one can track the effect of the campaigns on the animals. Another important thing is that animal welfare reforms have a “lock in” effect. Corporations and consumers are forced to buy and consume cage-free rather than caged eggs since the latter option is removed from the menu of options.
Other interventions in the farmed animal advocacy lack this kind of tractability.
Interventions to spread veganism for example, aim to convince people to become vegan through education seminars, pledge programs or leafleting etc. Although more people becoming vegan is a good thing, the question we should ask ourselves is whether these interventions work effectively to bring about the change we want to see. These interventions are intractable since one cannot tract participants and check whether they have become vegan after these interventions. After participating in these events and reading materials, participants can simply not become vegan or reduce their animal product consumption. Although ACE points out that “Some empirical studies suggest that self-monitoring—which is part of taking a veg*n pledge—reduces meat consumption, at least in the short run. Other studies that measure the impact of veg*n pledges suggest that some participants adopt a more plant-based diet for several months after engaging with a pledge.”, as ACE again says: “Previous studies mostly rely on self-reported consumption data, which can be subject to misreporting and biases.” I think the most important data (which is missing in ACE’s reports and reviews) that one should take into account is that after all these years and efforts, the number of vegans is not increasing. One can look at YouGov data for the UK and Gallup data for the US. By the way, in these surveys participants are asked about their self-identity and diet preference, but many people do eat meat despite their answers in surveys (probably due to social desirability bias), thus one can suspect that the real numbers can be even lower.
Same issues with tractability also apply to institutional outreach to influence the availability of animal free products. A restaurant can put some vegan options in its menu as a result of the outreach, but customers may simply not choose them and we may never know whether this intervention reduced the number of animals used and in particular how much it reduced animal product consumption. ACE’s research brief only makes a weak claim that this type of intervention can succeed in making vegan options available, but it doesn’t even claim that the availability of animal free options would lead to lower animal product consumption which is the main objective and counts as the real “effect”. So it is fair to say that this intervention is not tractable at all. Another weakness of this intervention in my view is that one cannot say that it is very neglected. Market forces already play their role here. If there is a sufficient demand in the market for the availability of vegan or plant based options in a restaurant or a market, it is fair to expect that they would do this themselves without the outreach. At least some of the restaurants would do this proportionate to market demand, and the rest would follow if the first movers make decent profit. On the other hand, if there is not enough demand for these goods, then restaurants would eventually drop them from their menus. I can understand that there might be some cases where outreach may accelerate or improve this process but one should not expect a radical change from the existing supply and demand outcomes, since this intervention neither supplies a new good to the market (as alternative protein intervention does) or nor change the demand (as vegan outreach does).
Same issues with tractability also apply to “movement building”. ACE, states that it “Each charity’s long-term impact is plausibly what matters most”, and argues that long term impacts should be given high priority while evaluating the effectiveness of animal charities. This can also explain the foundation of ACE’s Movement Grants whose aim is “building and strengthening the global animal advocacy movement”.
I have no objection to the idea that long term impacts and movement building are important. But I find it odd to isolate and decouple movement building impacts of a charity from its programs (which may be animal welfare campaigns, vegan advocacy or increasing the availability of plant-based options, etc.) which actually aim to impact animals. Isn’t that the point of movement building? And if let’s say a charity builds a bigger movement, the begging questions are: “What is that movement for? What will these more numerous or better organized advocates do to help animals?”
Well, probably they will primarily continue their existing programs since movement building activities would also promote these programs and emphasize their importance. One cannot build a movement without promoting and emphasizing the importance of the very thing that movement is doing. For example, if an animal charity’s main programs are about vegan advocacy or increasing the prevalence of plant-based options, the “movement” that this charity builds would almost surely focus on these programs in the future as well and/or the movement in question can be mobilized for these programs. So even if one agrees that movement building is important, one has to see that it functions as a “multiplier”, multiplying the impact of the organizations’ main program(s) which directly affect animals. As a consequence, if a charity carries out programs which have low tractability (and probably are ineffective), one cannot conclude that due to “movement building” those programs will have much higher impact in the future.
As for alternative protein development and promotion, I think one can make the theoretical argument that if one can develop price and taste competitive plant-based products, consumers may substitute their animal product consumption with these new goods. One can also make the case that factory farming is inherently inefficient since it is first feeding the animal to consume it later, which wastes resources, while plant based food production may use all its raw materials to produce its final product without or with little waste. Since this intervention is novel (do not have a track record of success or failure) and has a decent set of theoretical reasons to back up its case, I agree that funding these interventions from a hits based giving point of view makes sense. However, empirical evidence is not yet clear. Although plant-based foods’ growth in the market is observable, we do not have clear evidence that these goods are substituting animal products, which is our final aim. It is also unclear how much and which type of animal products are being substituted. If alternative proteins only substitutes dairy and beef, then they would hardly remove much animal suffering from the food system, since those animals are large and provide so much animal products. In order for this intervention to be tractable and impactful, one needs to show not only that these goods are in many restaurants and retailers, or that their market share is growing, or that they received a lot of investment, but also that these products substitute significant quantities of animal products which are mostly responsible for high animal suffering (mainly chicken, eggs and fish). As far as I know, we do not have that evidence yet. In fact, there is a study which shows that substitution effect is limited.
Finally, there is one other intervention which some animal charities (which are also in ACE’s recommendation list) engage in: Meatless Monday campaigns. Meatless Mondays campaigns usually target public institutions like schools and universities. The good thing about Meatless Monday campaigns is that it doesn’t just provide an option which can be ignored by the consumers, it removes the option of animal products for lunch meals on mondays. One can therefore make the inference that at least for those meals, animal consumption is reduced. That is how ACE seems to calculate the effectiveness of Meatless Mondays: “We estimate that switching from a typical diet to a vegetarian diet for one year spares approximately 30 land animals. We treat one meatless meal as the equivalent of between ⅓ and ½ of a vegetarian day”.
However there are a lot of moving parts here. Firstly, the institution can simply distribute meals with animal products to other meals of the week. Second, the students can compensate for not eating meat for that meal, with eating more meat at home. Third, it is not clear which type of meat is eliminated, the effect could be much lower if it is just minced beef. Fourth, if the institution or the participants compensate the eliminated meat with eggs, chicken or fish, it could have a negative effect. Fifth, students or workers can bring their own meal or choose to have lunch elsewhere. Finally, while there is an expectation that Meatless Mondays campaigns will convince people to decrease their animal product consumption further and have an impact beyond the cafeteria, given the constant rise of animal consumption and stagnant vegan-vegetarian share of population, this expectation is not supported by evidence. One can also read the comments to this forum post, from which I developed most of these points.
In conclusion, animal welfare reforms massively outperforms other interventions in terms of tractability.
3.2. ACE’s current position with regards to animal welfare reforms and the implications of this critique
As explained above, ACE is not making substantial claims about the relative effectiveness of different interventions. While ACE is stating that it highly prioritizes institutional focused interventions, it does not comment about how much it highly prioritizes the institutional focused interventions relative to individual interventions and it does not also comment how it prioritizes between institutional interventions. Finally, one can interpret that in ACE’s view many charities engage in movement building if they engage with other advocates or organizations even if they do not have a seperate movement building program, so in ACE’s view, an animal charity which has a individual focused intervention can also have high scores since it also engages in movement building which is considered as institutional focused intervention by ACE. If you look at the comprehensive reports of recommended charities, a significant number of charities are considered to be movement building, even if they do not have a seperate program.
Furthermore, although ACE has many charities in its recommendations which engage in animal welfare campaigns, ACE expresses some concerns about animal welfare reforms in the footnotes of its comprehensive reviews: “On average, our team considers advocating for welfare improvements to be a positive and promising approach. However, there are different viewpoints within ACE’s research team on the effect of advocating for animal welfare standards on the spread of anti-speciesist values. There are concerns that arguing for welfare improvements may lead to complacency related to animal welfare and give the public an inconsistent message—e.g., see Wrenn (2012). In addition, there are concerns with the alliance between nonprofit organizations and the companies that are directly responsible for animal exploitation, as explored in Baur and Schmitz (2012).” (italics are mine).
I think this comment in the footnotes should be explained in more detail by ACE. First of all, the cited article, Wrenn (2012) argues that “...abolitionism, as developed by Francione, is the only morally consistent approach for taking the interests of nonhuman animals seriously. Further, it is suggested that the newness of the abolitionist movement and the mainstream nonhuman animal welfare movement’s dismissal of abolitionism has thus far prevented any substantial abolitionist success.” (italics are mine). Given the negative outlook of Francione towards animal welfare reforms, one could expect that ACE might be discounting animal welfare reforms a lot, even if on average it still sees it as “a positive and promising approach”. Because, according to Francione, “Animal welfare reforms are not [even] ‘baby steps’. They are big steps in a backward direction. The problem is use, not treatment. The goal is to abolish animal use, not to regulate treatment. The means to the goal? Go vegan and educate others about veganism.” (italics are mine). Another issue I would like to point out is that while ACE voices concern about animal welfare reforms following a critique à la Francione, it does not mention the same critiques for other interventions. This is interesting because according to Francione almost all of the interventions which ACE is evaluating and approving as effective, do not meet the “moral baseline” of Francione. I think if ACE is discounting animal welfare reforms for Francionist arguments, then the same arguments should lead to the discount of other interventions like Meatless Mondays, vegan pledges, diet change advocacy which aims for “reduction” of animal products, any institutional outreach, or the use of graphic images which show animal suffering in factory farming. Those who are familiar with Francione’s literature know perfectly well that Francione is not just against animal welfare reforms, he is pretty much against most of the interventions mentioned above, including vegan pledges.
Second issue that I am somewhat concerned about in this detail is that Francione’s viewpoints are almost antithetical to current effective animal advocacy approaches. Of course, an organization should be open to diverse viewpoints, and we should never try to suppress them. However, if these diverse viewpoints lead to dysfunctionality, where members do not even agree about the fundamentals of their task, and mainly operate by uneasy concessions and/or opaque reasoning in order to reach a joint resolution, this may not be an optimal framework for both of the viewpoints. If this is the case (which may not be, I am just speculating about a probability), it might be better to present these different viewpoints in separate dissenting votes, rather than in joint decisions.
While from my understanding ACE is underrating animal welfare campaigns, due to the reasons explained above, animal welfare campaigns should be rated and prioritized much highly (in absolute and relative terms). The implications of this approach would probably result in animal charities which primarily engage in animal welfare campaigns occupy all of top and standout charity statuses, ranked according to the number of animals they affect by the commitments and the implementation of the commitments they achieved, the scope of the welfare gains of the reforms, moral weight of the animals relative to their sentience or intelligence (differences say between fish, chickens or shrimps), and their operation costs. This approach would probably result in leaving out animal charities which primarily engage in programs which are not tractable.
As a matter of fact, one can say that Open Philanthropy and Effective Altruism Welfare Funds are already prioritizing animal welfare campaigns, even if they are funds which are functioning on a hits based giving approach. When you look at their grant reports, one can see that the overwhelming majority of their grants go to animal welfare interventions. They do also fund novel interventions which are not yet tractable such as alternative protein development and promotion, and wild animal welfare research (which, as explained above, makes sense from a hits based giving perspective). However, apart from very few exceptions, neither Open Philanthropy nor Effective Altruism Welfare Funds, seem to provide grants to organizations which engage in restaurant outreach for more plant-based options, vegan pledge programs, vegan outreach, or meatless monday campaigns, even from a hits based giving perspective. ACE on the other hand, seem to be in equal distance from each type of programs and seems to be engaging in hits based giving more freely than Open Philanthropy and Effective Altruism Welfare Funds, by attributing top and standout status and providing grants to a wider range of charities, even if it is proclaiming to be evaluator organization, not a hits based fund.
3.3. Concerns and replies
Is animal welfare campaigns all that we should be doing according to this approach?
Yes and no.
First, an unapologetic yes.
If animal welfare campaigns are relatively more tractable and impactful than other interventions according to the existing evidence and reasons, then of course we should be prioritizing animal welfare campaigns over other interventions. Not doing that would result in more animal suffering. Given that we have so much work to do in this field, I think not prioritizing according to evidence and reason has a high opportunity cost.
Prioritization is the key to doing the most good and not acting on evidence and reason on this issue would contradict with the basic principles of effective altruism. I think animal advocates should invest more of their efforts and resources into animal welfare campaigns rather than other interventions. Donors should also channel more of their resources into these projects.
Second, a qualified no.
Firstly, of course, everyone is free to disagree and do what they think is right. My entire point rests on the assumption that the argument made above is supported by evidence and reason, and therefore is convincing for the reader. If one thinks the opposite, then of course one should not be following this advice.
Second, prioritization of animal welfare campaigns is constrained by room for funding. And if animal welfare campaigns are supported to the point of having no more room for funding (and staff), one may look for the second most effective option. However, given that there is a lot of room for funding and there is a lot to be done, this prospect is not likely.
Third, another likely prospect may be to work on and fund promising and novel interventions such as alternative protein development and research for finding new or better ways of helping animals (with a focus on wild animal welfare) or other novel projects supported with reasons. As I mentioned above, these novel interventions may also be deemed effective (even more effective) when more evidence becomes available in the near future.
But these exceptions does not mean that we should not be prioritizing tractable interventions over intractable ones.
I think this answer is similar to the answer which can be given to the question “are bed nets all there is?” First, an unapologetic yes: one should be donating to the most effective funding opportunity available. Second, a qualified no. First, you can do whatever you want with your money, no one can tell you what to do. Second, there could be instances where the Against Malaria Foundation does not have more room for funding and one should look for the second best opportunity and finally, there can be hits based opportunities which can also be effective like funding research for gene drive in mosquitos. But these exceptions should not lead us to the conclusion that we should not be prioritizing between different programs and put bed net distribution programs in the same category with other intractable programs.
Shouldn’t we aim higher?
Yes, but only if we can achieve higher.
Some advocates are not very passionate about animal welfare campaigns because these reforms are incremental as they do not end animal exploitation or lead directly to animal liberation.
While I share this frustration myself and find the desire for more progress for animals laudable, one has to see that aiming higher does not immediately result in achieving higher. There are many aims such as “ending factory farming”, “banning slaughterhouses”, “cutting the global production and consumption of animal products by %50 by 2040”. I would be happy to see these changes, but when deciding about what to prioritize we should not be just judging options by our preference for the end goals, we should also look for the tractability of these interventions. And unless we have evidence that these are in fact tractable, it may not be an effective strategy to always aim high, since that would have an opportunity cost of not carrying out projects which are tractable.
I think this answer is similar to the answer which can be given to the question “shouldn’t we aim higher for improving the welfare of the global poor?”. Yes, but only if we can achieve higher. Some advocates may be less passionate about health interventions in Africa because these reforms are incremental, they do not end the poverty of the global poor. They may be willing to aim for more such as “comprehensive economic development”, “large scale institutional reforms”, “recognition of colonialism and reparations”, “overthrow of global capitalism”. Although there might be some probability to achieve these goals, and everyone has the right to pursue whatever project they like, it is hard to say that the effective strategy is to always aim high, since that would have an opportunity cost of not carrying out projects which are tractable.
Finally, I think these concerns are underestimating how high they are in fact aiming, and overestimating our capacities to reach these goals. Global animal product consumption is at an all time high and increasing (over 330 million metric tons). “The global farmed animal advocacy movement”, only has about 200 million dollars in support and consists of (at most) thousands of people where it has the most power, and only hundreds or even tens of advocates in the rest of the world. We should also not ignore the unfavorable socioeconomic conditions worldwide which limit the public support for animal rights.
I think this concern also underestimates the progress animal welfare campaigns have achieved. The welfare of hundreds of millions of animals improved thanks to these campaigns and there is still much to do: cage-free campaigns in Asia, Africa, Latin America, broiler campaigns worldwide, and new fish and wild animal welfare campaigns which have the potential to impact billions of animals.
Shouldn’t we aim for a broader and pluralistic animal advocacy movement?
While I think diversity and pluralism are central values, I do not think we should necessarily value diversity and pluralism of programs. I think ACE and the Effective Altruism community (or the part of the community which is engaged in animal welfare cause area) should affirm more prioritization rather than more diversity or pluralism of programs if we are to become more effective and help more animals.
The need for prioritization is evident where ACE highly prioritizes farmed and wild animals. But if we follow the pluralism of programs logic to its conclusions we should not be stopping at farmed and wild animals, we should also include programs which aim to help companion animals or lab animals or captive animals etc. If one agrees that we need prioritization, not pluralism, in the case of selecting animal groups, one should also agree that we also need prioritization, not pluralism, in the case of selecting programs.
Finally, neither this essay nor ACE possess the power to radically change the advocacy landscape. A lot of advocacy groups would continue their existing programs regardless of what ACE or my essay claims.
What if this critique is not correct in its substantive claims?
The main critique of this essay is structural rather than substantive. And even if you did not find the substantive claims in the third part of this essay convincing, it should not devalue the claims made in the first and second part of the essay. I am more concerned with ACE’s omission from making substantive claims, whatever those claims will be. I am ready to update my views if more evidence and reason is presented. I think animal advocates, effective altruist animal advocates in particular, need to have more discussions about the effectiveness of different interventions. And ACE has a distinctive role to enrich and validate the arguments in these debates, by using its resources and expertise.
Conclusion
In conclusion, I would like to underline that this essay is aimed to be a constructive criticism, not a destructive one. I would also like to express clearly that although I believe ACE has some defects, those defects are not related to a lack of hard work, good intentions, talent or honesty. My main criticism is more about what ACE is not doing, not about what ACE is doing. As explained above, these issues can be solved if ACE is willing to act more decisively and assume its primary role rather than doing many things all at once.
A final note in defense of ACE: animal advocacy is very controversial. If ACE or anyone else makes substantive claims about the effectiveness of a certain program which even implicitly implies that other programs are less effective or even ineffective, many advocates would take these claims personally. In comparison, if one makes a substantive claim about the effectiveness of a certain program in global health and development cause area, which even explicitly argues that other programs are less effective, most people would be more like “oh, I liked that [other] program, but okay, I will consider my donations accordingly”. In animal advocacy on the other hand, many advocates have a strong emotional and moral attachment to their prefered intervention and even make this part of their identity. You can find a lot of people who would put “vegan” in their bio in social media or in this forum, but you would be unlikely to find a lot of people who would present themselves as “dewormer”, “ethical bed net donor”. I am not saying that this is a bad thing, in fact, it is a good thing that people take these issues so seriously that they make it part of their identity. But my point is rather that the effectiveness debates occur in a much different way in animal advocacy. And I think it is understandable when an organization is somewhat cautious about making bold claims which is very likely to cause a lot of reaction which might lead to divisions and conflict within the animal advocacy movement.
However, I think ACE should take on this challenge and deal with it, and leave those concerns around movement unity to other actors in the movement to resolve (another point about division of labor). Effective Altruism animal welfare cause area needs effectiveness discussions more than ever. We will need to prioritize even more in the near future when fish welfare asks and wild welfare asks mature. Since there will be so many programs, we cannot afford to not prioritize between different programs. I believe ACE can do much more good if it can be more instrumental to help and inform those prioritizations in the animal welfare cause area.