I am a generalist quantitative researcher. I am open to volunteering and paid work. I welcome suggestions for posts. You can give me feedback here (anonymously or not).
Hello Stien. Thanks for the context. I assume you are sufficiently concerned about ALI's governance to be confident they should not be recommended (for now), and that therefore there is no need to complete their evaluation. However, have you considered publishing an incomplete evaluation with some information about your concerns about ALI's governance?
You estimate the work targeting layers of Sinergia Animal in 2024-mid-2026 helped 0.32 layers per $, and that of Çiftlik Hayvanlarını Koruma Derneği (ÇHKD) in 2024-2025 helped 2.6 layers per $. These estimates are 2.91 % (= 0.32/11) and 23.6 % (= 2.6/11) of your estimate of 11 layers helped per $ for similar work from The Humane League (THL) in 2024. Have you considered not recommending Sinergia Animal or ÇHKD given this, and that the interventions are very similar?
Here are the estimates for the cost-effectiveness in suffering-adjusted days (SADs) averted per $ of the charities you assessed and recommended this year:
Çiftlik Hayvanlarını Koruma Derneği (ÇHKD): 30.7 (0.7 to 196).
Greener by Default’s work in 2025-mid-2026: 2.14 (0.33 to 13.0).
Sinergia Animal’s work targeting layers in 2024-mid-2026: 5.73 (0.01 to 106).
The ranges above do not account for uncertainty in pain intensities and moral weights, which I think are the 2 largest sources of uncertainty.
You are assuming excruciating pain is 45.6 (= 11.17/0.245) times as intense as hurtful pain. This implies 1 min of “severe burning in large areas of the body, dismemberment, or extreme torture” (excruciating pain) is as bad as 45.6 min of “awareness of Pain is likely to be present most of the time” (hurtful pain). I believe 1 min of excruciating pain is way worse.
Excruciating pain in humans would “mark the threshold of Pain under which many people choose to take their lives rather than endure the Pain”. I doubt people taking their lives to avoid 1 min of excruciating pain would do so to avoid 45.6 min of hurtful pain. In this pain, “Individuals can still conduct routine activities that are important in the short-term (e.g. eating, foraging) and perform cognitively demanding tasks, but an impairment in their ability or motivation to do so is likely to be observed”.
I made a similar comment in 2025, and this year privately. I know you have considered them. However, I do not understand why you believe the intensity of excruciating pain is so low relative to what I consider plausible.
Here is a quick calculation illustrating that the suffering random diets in the US causes to poultry birds can be offset donating much less than the 185 $/person-year suggested by FarmKind. There were 4.72 poultry birds per person in the US in 2024. ACE estimated THL's cage-free accountability work in 2024 helped 11 layers per $. Say each layer lives for 1.34 years (= (60 + 80)/2*7/365.25), which is the mean between the lower and upper bound of the range of "60 to 80 weeks" assumed by the Welfare Footprint Institute (WFI). Then THL's cage-free accountability work in 2024 improved 14.7 layer-years per $ (= 11*1.34). I calculatethat moving 1.67 layer-years (= (0 - (-2.27))/(-0.333 - (-1.69))) from conventional cages to cage-free aviaries increases the welfare of chickens as much as removing 1 broiler-year in a conventional scenario. So I infer THL's cage-free accountability work in 2024 increased the welfare of chickens as cost-effectively as removing 8.80 broiler-years per $ (= 14.7/1.67). Suppose the conditions of poultry birds in the US are as bad as those of broilers in a conventional scenario (I think they are better due to layers in conventional cages having higher welfare than broilers in a conventional scenario). Random people in the US could offset the suffering their diets cause to poultry birds donating 0.536 $/person-year (= 4.72/8.80) to interventions increasing the welfare of chickens as cost-effectively as THL's cage-free accountability work in 2024, which is 0.290 % (= 0.536/185) of the offset suggested by FarmKind.
Is it accurate that the lowest numbers you arrived at come from donating entirely to shrimp slaughter?
Yes. "I estimate donating 0.279 $/person-year (= 0.106*2.63) to SWP is enough to offset the suffering the food consumption of random people in the US causes to farmed animals". This is "assuming SWP is as cost-effective as HSI has been", and HSI advocates for electrically stunning shrimps.
Most farmkind donors probably don't agree that shrimp suffering can substitute for chicken or pig suffering, or if so, expect it to convert at a rate much less favorable to shrimps than you or ACE.
I assumed sentience-adjusted welfare ranges proportional to "individual number of neurons"^0.5, which implies that of shrimps is 0.1 % of that of humans. This ratio is 8.0 % in Bob Fischer's book about comparing welfare across species, which is what Rethink Priorities (RP) uses as the default value in their Donor Compass tool. ACE also uses the value in Bob's book or something very similar. So I would have obtained a lower offset using ACE's or RP's preferred value.
Maybe if they deeply engaged with your reasoning around shrimps they would agree, but with more modest engagement I expect them to land closer to FarmKind's model.
I expect them to land on a much smaller offset than FarmKind's as long as they endorse offsetting the suffering of all farmed animals by decreasing the suffering of chickens. ACE estimated THL's cage-free accountability work in 2024 decreased 88 SADs per $, and I calculated random people in the US increase the suffering of farmed animals by 138 SADs per person-year based on ACE's 2025 meal replacement model. In this case, random people in the US can offset the impact of their diets on farmed animals by donating 1.57 $/person-year (= 138/88) to interventions as cost-effective as THL's cage-free accountability work in 2024, which is 0.565 % (= 1.57/278) of the offset suggested by FarmKind.
@Aidan Kankyoku, here is a quick calculation illustrating that the suffering random diets in the US cause to poultry birds can be offset donating much less than the 185 $/person-year suggested by FarmKind. I am tagging you because this was not in my initial reply to your comment.
The cost of offsetting the suffering a diet causes to farmed animals depends a lot on the question being asked. How much money is needed to help each type of farmed animal affected by the diet about as much as the diet harms it? Or how much money is needed to decrease the suffering of farmed animals as much as the diet increases it, regardless of which animals benefit? The cost also depends on the cost-effectiveness estimates used, and on how much they are adjusted for conservatism.
FarmKind’s Compassion Calculator answers the 1st question. It recommends donating 278 $/person-year to offset the suffering random Western adult omnivores cause to farmed animals. It models conservatively on purpose, as it wants donors to be confident that their donations are enough.
For the 2nd question, I estimate 2.04 $/person-year for random people in the United States (US) based on Animal Charity Evaluators’ (ACE’s) results, and 0.279 $/person-year based on my own numbers for the suffering caused to farmed animals, and the past cost-effectiveness of the Shrimp Welfare Project’s (SWP’s) Humane Slaughter Initiative (HSI).
Changing one modelling choice at a time in FarmKind’s spreadsheet shows how much each choice affects the cost. For example, offsetting the suffering of birds raised for meat with cage-free corporate campaigns changes the cost from 278 to 109 $/person-year, and barely requires comparing welfare across species, as almost all birds raised for meat are chickens.
Which estimate is more useful depends on what a donor wants an offset to do, and on how much margin they want for uncertainty.
Hello Stien. Thanks for the context. I assume you are sufficiently concerned about ALI's governance to be confident they should not be recommended (for now), and that therefore there is no need to complete their evaluation. However, have you considered publishing an incomplete evaluation with some information about your concerns about ALI's governance?
You evaluated and not recommended the Aquatic Life Institute (ALI), Crustacean Compassion (CC), and Dansk Vegetarisk Forening (DVF), but only published the evaluations of CC and DVF. Could you say why you did not publish the evaluation of ALI?
You estimate the work targeting layers of Sinergia Animal in 2024-mid-2026 helped 0.32 layers per $, and that of Çiftlik Hayvanlarını Koruma Derneği (ÇHKD) in 2024-2025 helped 2.6 layers per $. These estimates are 2.91 % (= 0.32/11) and 23.6 % (= 2.6/11) of your estimate of 11 layers helped per $ for similar work from The Humane League (THL) in 2024. Have you considered not recommending Sinergia Animal or ÇHKD given this, and that the interventions are very similar?
Here are the estimates for the cost-effectiveness in suffering-adjusted days (SADs) averted per $ of the charities you assessed and recommended this year:
The ranges above do not account for uncertainty in pain intensities and moral weights, which I think are the 2 largest sources of uncertainty.
You are assuming excruciating pain is 45.6 (= 11.17/0.245) times as intense as hurtful pain. This implies 1 min of “severe burning in large areas of the body, dismemberment, or extreme torture” (excruciating pain) is as bad as 45.6 min of “awareness of Pain is likely to be present most of the time” (hurtful pain). I believe 1 min of excruciating pain is way worse.
Excruciating pain in humans would “mark the threshold of Pain under which many people choose to take their lives rather than endure the Pain”. I doubt people taking their lives to avoid 1 min of excruciating pain would do so to avoid 45.6 min of hurtful pain. In this pain, “Individuals can still conduct routine activities that are important in the short-term (e.g. eating, foraging) and perform cognitively demanding tasks, but an impairment in their ability or motivation to do so is likely to be observed”.
I made a similar comment in 2025, and this year privately. I know you have considered them. However, I do not understand why you believe the intensity of excruciating pain is so low relative to what I consider plausible.
Here is a quick calculation illustrating that the suffering random diets in the US causes to poultry birds can be offset donating much less than the 185 $/person-year suggested by FarmKind. There were 4.72 poultry birds per person in the US in 2024. ACE estimated THL's cage-free accountability work in 2024 helped 11 layers per $. Say each layer lives for 1.34 years (= (60 + 80)/2*7/365.25), which is the mean between the lower and upper bound of the range of "60 to 80 weeks" assumed by the Welfare Footprint Institute (WFI). Then THL's cage-free accountability work in 2024 improved 14.7 layer-years per $ (= 11*1.34). I calculate that moving 1.67 layer-years (= (0 - (-2.27))/(-0.333 - (-1.69))) from conventional cages to cage-free aviaries increases the welfare of chickens as much as removing 1 broiler-year in a conventional scenario. So I infer THL's cage-free accountability work in 2024 increased the welfare of chickens as cost-effectively as removing 8.80 broiler-years per $ (= 14.7/1.67). Suppose the conditions of poultry birds in the US are as bad as those of broilers in a conventional scenario (I think they are better due to layers in conventional cages having higher welfare than broilers in a conventional scenario). Random people in the US could offset the suffering their diets cause to poultry birds donating 0.536 $/person-year (= 4.72/8.80) to interventions increasing the welfare of chickens as cost-effectively as THL's cage-free accountability work in 2024, which is 0.290 % (= 0.536/185) of the offset suggested by FarmKind.
Hello Aidan.
Yes. "I estimate donating 0.279 $/person-year (= 0.106*2.63) to SWP is enough to offset the suffering the food consumption of random people in the US causes to farmed animals". This is "assuming SWP is as cost-effective as HSI has been", and HSI advocates for electrically stunning shrimps.
I assumed sentience-adjusted welfare ranges proportional to "individual number of neurons"^0.5, which implies that of shrimps is 0.1 % of that of humans. This ratio is 8.0 % in Bob Fischer's book about comparing welfare across species, which is what Rethink Priorities (RP) uses as the default value in their Donor Compass tool. ACE also uses the value in Bob's book or something very similar. So I would have obtained a lower offset using ACE's or RP's preferred value.
I expect them to land on a much smaller offset than FarmKind's as long as they endorse offsetting the suffering of all farmed animals by decreasing the suffering of chickens. ACE estimated THL's cage-free accountability work in 2024 decreased 88 SADs per $, and I calculated random people in the US increase the suffering of farmed animals by 138 SADs per person-year based on ACE's 2025 meal replacement model. In this case, random people in the US can offset the impact of their diets on farmed animals by donating 1.57 $/person-year (= 138/88) to interventions as cost-effective as THL's cage-free accountability work in 2024, which is 0.565 % (= 1.57/278) of the offset suggested by FarmKind.
@Aidan Kankyoku, here is a quick calculation illustrating that the suffering random diets in the US cause to poultry birds can be offset donating much less than the 185 $/person-year suggested by FarmKind. I am tagging you because this was not in my initial reply to your comment.
Readers of this post may be interested in my post How much does it cost to offset the suffering an average diet causes to farmed animals?. Here is the summary.
Hi Hazem. Nice post.
Hi architg. Nice post.