Vasco's post isn't a clear example of naive utilitarianism in this sense because he isn't recommending any action that is clearly highly conventionally immoral and norm-breaking. The only action he is recommending, if any, is not donating to global health charities. 

While it would be unfair to compare Vasco's possible recommendation to a recommendation for "conventionally highly immoral and norm-breaking actions like stealing and murder," conceding that exclusion doesn't really help me that much. There's still a wide scope of action in which one could act on a principle what looks something like young children's lives as a class are likely net-negative and so we should generally avoid saving those lives unless it would be rights-violating to do so. ^[1]

For example, I could vote against costless consumer-safety laws because they are lifesaving, or vote against hospitals having to provide unfunded lifesaving charity care to those who are unable to pay for that care (EMTALA in the US). There are other reasons one might conceivably vote against such legislation (e.g., heavy libertarian ideology, or awareness of the rather significant costs of EMTALA).  

But if I'm going to apply the reasoning of Vasco's post when evaluating lifesaving global health charities, I'm going to need a good reason not to apply it to strangers drowning in a pond or votes on legislation. And I'm not thrilled with the candidate arguments that I can come up with to confine the post's logic to the narrow grounds of charitable donations (even though I accept the reasons for not extending them to murder, theft, etc.). Without a convincing way to cabin the post to charitable donations, I think most of us are not willing to consider biting the bullet and (e.g.) voting to repeal lifesaving legislation even though having more people die of exploding toasters and emergency-room refusals would presumably lead to fewer factory-farmed animals.

1. ^{^}

   I went with rights-violating with a nod to this post (the top result for naive utilitarianism on Google), and because I wanted to minimize strong effect of partiality commonly associated with conventional morality.

I think most people (I would say over 99 percent outside of EA) would consider avoiding saving the lives of the poorest and most disadvantaged humans because they might make animals suffer, while rich countries continue to cause most animal suffering a "conventionally highly immoral and norm-breaking action.", thus meeting your second criteria.

I also think most people (although perhaps not over 95 percent) would consider this pursuing utilitarianism through ruthless means.

To perhaps labour the point, the idea saving the lives of really poor people is a good thing, is a pretty strong conventional norm, even if most people don't take that seriously enough to do much about it.

Stopping doing something is still an action.

So for me at least this meets your criteria for naive utilitarianism.

Thanks for the great clarifications, David! Strongly upvoted.

My actual reason to disagree is that I place much lower weight on animals than you, and I would axiomatically reject any moral weight on animals that implied saving kids from dying was net negative. I cannot give a tight philosophical defence of that view, but I am more committed to it than I am to giving tight philosophical defences of views. I suspect that if GiveWell were to publish a transparent argument as to why they ignore those effects, it would look similar to my argument - short and unsatisfactory to you. (Note; I work at GiveWell but this is my own view.)

I upvoted this comment for honesty, but this passage reads to me like committing to a conclusion ("saving kids from dying cannot be net negative") and then working its way backward to reject the premise ("animals matter morally", "saving kids from dying causes more (animal) suffering than it creates (human) welfare") that leads to a contradictory conclusion. That seems like textbook motivated reasoning to me? It doesn't seem like a good way of doing moral reasoning. I think it would be better to either reject the premise or to argue that the desired conclusion can follow from the premise after all.

Personally I think it's very much not obvious whether the meat eating problem is genuine. But given that the goodness of a very large part of the EA project so far hinges on it not being real, and given that it's far from obvious whether it's real, I think it would be useful to make progress on that question. So I'm glad that @Vasco Grilo🔸 and others are trying to make progress on it and a little discouraged to see some pushback (from several commenters) that doesn't really engage with Vasco's arguments/calculations.

(It does seem like, as @Ben Millwood🔸 has commented, any harm caused to animals by donating to global health charities is much smaller than the harm of not giving to animal charities. So maybe a better and more palatable framing for the meat eating problem is not, "Is giving to global health charities net negative/positive?" but "Is giving to global health charities more/less cost-effective than giving to animal charities?")

Hi Karthik,

Your comment inspired me to write my own quick take, which is here. Quoting the first paragraph as a preview:

I feel pretty disappointed by some of the comments (e.g. this one) on Vasco Grilo's recent post arguing that some of GiveWell's grants are net harmful because of the meat eating problem. Reflecting on that disappointment, I want to articulate a moral principle I hold, which I'll call non-dogmatism. Non-dogmatism is essentially a weak form of scope sensitivity.

I decided to spin off a quick take rather than replying here, because I think it would be interesting to have a discussion about non-dogmatism in a context that's somewhat separated from this particular context, but I wanted to mention the quick take as a reply to your comment, since it's relevant.

Thanks, Karthik.

I'm tempted to point out that increasing the population may not increase meat consumption much via effects on prices

The prices would have to increase by an unreasoble amount for this to change my conclusions. For a random person globally, and in China, India and Nigeria in 2022 to cause as much suffering to poultry birds and farmed aquatic animals as the person's happiness, the animal suffering would have to be 6.45 % (= 1/15.5), 2.89 % (= 1/34.6), 19.3 % (= 1/5.17) and 43.3 % (= 1/2.31) of the one I calculated. In addition, I have assumed no growth in the consumption of animals per capita, whereas I expect this to increase as real GDP per capita increases.

meat consumption among the extreme poor is much lower than on average

I acknowledged the people helped by GiveWell and AIM would cause less harm to animals than random people, but I do think this resolves the meat-eater problem.

The harms would be smaller for a random person helped by such GiveWell’s grants or Ambitious Impact’s organisations. I assume they have an income below that of a random person in the respective country, the supply per capita of meat excluding aquatic animals roughly increases with the logarithm of the real GDP per capita, and I guess so do the number of poultry birds, farmed aquatic animals excluding shrimp, and shrimp per capita. Yet, self-reported life satisfaction also roughly increases with the logarithm of the real GDP per capita. So I believe the harms to farmed animals per person increase roughly linearly with self-reported life satisfaction, at least across countries. As a result, it is unclear to me whether the harms to farmed animals as a fraction of the human benefits would be higher or lower for a random person than for a random person helped by such GiveWell’s grants or Ambitious Impact’s organisations.

factory farm consumption is likely much smaller for beneficiaries in remote areas not reached by large scale animal agriculture

Great point. This is the kind of consideration supporters of extending human lives would ideally investigate. Note it is also harder for life-saving interventions to reach remote areas.

My actual reason to disagree is that I place much lower weight on animals than you, and I would axiomatically reject any moral weight on animals that implied saving kids from dying was net negative.

Nitpick. I think valuing animal welfare as highly as I do implies saving lives in many countries is harmful nearterm, but the overall effect may well be beneficial or harmful (not just harmful).

AIM is a more interesting case to consider because they are clearly more cause-agnostic than GiveWell and so can't (or wouldn't want to) make the same claim. However, that makes for a very simple hedging/offsetting defense. Given uncertainty about moral weights and risk aversion over the amount of value created, AIM should optimally fund both GHD and AW work.

I do not see how AIM being beneficial overall justifies them starting organisations which may well be causing lots of harm nearterm.

It is admirable that you acknowledge that you are not using “reason and evidence to do the most good” and that you presumably accept that you have no leg to stand on when trying to persuade nativists who assign zero weight to people who live in other countries to give more to those who live abroad.

Like Ian Turner I ended up disagreeing and not downvoting (I appreciate the work Vasco puts into his posts).

The shortest answer is that I find the "Meat Eater Problem" repugnant and indicitative of defective moral reasoning that, if applied at scale, would lead to great moral harm.^[1]

I don't want to write a super long comment, but my overall feelings on the matter have not changed since this topic came up on the Forum. In fact, I'd say that one of the leading reasons I consider myself drastically less 'EA' since the last ~6 months have gone by is the seeming embrace of the "Meat-Eater Problem" inbuilt into both the EA Community and its core ideas, or at least the more 'naïve utilitarian' end of things. To me, Vasco's bottom line result isn't an argument that we should prevent children dying of malnutrition or suffering with malaria because of these second-order effects.

Instead, naïve hedonistic utilitarians should be asking themselves: If the rule you followed brought you to this, of what use was the rule?

1. ^{^}

   I also agree factory farming is terrible. I just want to find pareto solutions that reduce needless animal suffering and increase human flourishing.

I did not vote on the post, but I considered downvoting it on the grounds that Vasco has made versions of this argument many times across the Forum, and I think repetition is bad for discourse. If I had a nickel for every time I've read "I Fermi-estimate that corporate campaigns for chicken welfare are 1000x GiveWell", I could offset the harms caused by several meat eaters /j

I strong downvoted because the title is unnecessarily provocative and in my opinion gives a misleading impression. I would rather not have this kind of thing on my forum feed

I disagree-voted, rather than downvoted, but I could understand downvoting on the basis that, like so many animal welfare pieces on the forum, this one smuggles in a ton of unstated and highly controversial beliefs about the value of animals. I feel like pieces like this should come with a disclaimer at the top saying, “warning: this essay rests on highly unconventional beliefs about the value of animal welfare”.

Personally I find repugnant the idea that we should let little kids suffer from malaria because of the chance they might grow up to eat chicken. The same logic could be used to justify school shootings.

Thanks, Nick!

All 4 of GiveWell's top charities and almost most Ambitious Impact charities operate in Sub-Saharan Africa

Great point. GiveWell has made 956 M$ of grants impacting people in Nigeria, 87.7 % (= 956*10^6/(1.09*10^9)) of the amount granted impacting people in China, India and Nigeria (the 3 countries I analysed). Ambitious Impact has incubated 5 organisations in Nigeria significantly decreasing human mortality, 62.5 % (= 5/8) of the ones in China (0), India (3) and Nigeria (5).

My revised estimated "harm" would be something in the realms of... (see below for reasoning)

(Chickens 0.841 x 0.375) [0.375 = 0.25*0 + 0.5*0.5 + 0.25*1]   +  (Shrimp 1.08 x 0.1 ) +  (Fish  0.178 x 0.5) = .512 DALYs harm caused

Which is less the the DALYs  averted from a life saved of around 0.9 

For harms to animals 56.9 % (= 0.512/0.9) as large as the benefits to humans nearterm, there is still room for saving lives to be harmful overall due to future growth of the consumption per capita of animal-based foods throughout the lifetime of the people who are saved. I think the nearterm harms to animals per person are larger for people living in urban areas, and urban population as a fraction of the total has been increasing in the coutries I analysed (note the vertical axis below starts at 15 %, not 0).

I relied on aquaculture production, but supply (production plus net imports) is what matters for the harms to animals, and it looks like Nigeria imports lots of fish. From Ogunji and Wuertz (2023), "Nigeria is far from self-sufficiency in its fish supply, with a deficit of about 2.5 million tons being imported, but national initiatives envision an expansion in the future to increase the supply of the population with high-quality protein and the prevention of malnutrition". If the 2.5 Mt refers to net imports of farmed fish, which I think tends to be cheaper, my harms to farmed aquatic animals excluding shrimp in Nigeria in 2022 would become 9.62 (= 1 + 2.5*10^6/(290*10^3)) times as large. This would increase the harms to farmed animals as a fraction of the direct benefits to humans in Nigeria in 2022 to 1.41 (= (0.841*0.375 + 0.178*0.5*9.62 + 1.08*0.1)/0.908), which is already higher than 1, thus suggesting decreasing mortality is harmful nearterm. The situation may well be worse due to future growth, as hinted by what I quoted from Ogunji and Wuertz (2023). Below is Nigeria's aquaculture production. It may well not matter much, if the vast majority of aquaculture supply is imported, but it illustrates there may be large jumps over the next few decades. From 2000 to 2021, aquaculture supply per capita increased from 0.204 (= 25.7*10^6/(126*10^6)) to 1.26 kg (= 276*10^6/(219*10^6)), becoming 6.18 (= 1.26/0.204) times as large. If aquaculture supply (production plus net imports) per capita grows as much as that over the next few decades, the harms to animals can easily become way larger than the direct benefits to humans (as opposed to just 1.41 times as large), even without accounting for growth in the supply of poultry birds and farmed shrimp.

Obviously even if I'm right, given the uncertainty here this far from precludes net harm under your framework.

Agreed. Even neglecting longer term effects, I illustrated above just accounting for net imports of farmed fish could make the harms to farmed animals larger than the direct benefits to humans nearterm. As another example, using RP's 95th percentile welfare range of shrimp of 1.15, instead of the median of 0.031, and neglecting net imports of farmed fish as I did in the post, the harms to farmed animals as a fraction of the direct benefits to humans nearterm would increase to 4.86 (= (0.841*0.375 + 0.178*0.5 + 1.08*0.1*1.15/0.031)/0.908), which is higher than 1.

In principle, I strongly endorse maximising expected impartial welfare regardless of the probability of harm (in practice, I believe decreasing the probability of harm is often a good heuristic to increase expected impartial welfare). However, I would say the case for supporting animal welfare over global health and development is especially strong for people who care intrinsically about minimising the probability of causing lots of harm.

You mentioned this effect might be balanced out by reduced quality of life increases for poorer Nigerians compared to richer ones, but I don't think it would come close.

@NickLaing^[1], you adjusted down my estimate of the harms to animals, but not my estimate of the direct benefits to humans. The real GDP per capita in Nigeria in 2022 was 5.55 k 2021-$/year, or 13.8 2017-$/d (= 5.55*10^3*0.905/365.25), which is 6.42 (= 13.8/2.15) times the consumption of someone living at the extreme poverty line of 2.15 2017-$/d. So, assuming welfare is proportional to the logarithm of the daily consumption in 2017-$/d, the welfare of someone at the extreme poverty line would be 29.2 % (= ln(2.15)/ln(13.8)) of that of a random person in Nigeria in 2022. As a result, the harms to animals as a fraction of the direct benefits to humans would increase from your estimate of 56.9 % to 1.95 (= 0.569/0.292), which is again higher than 1.

1. ^{^}

   I am tagging you because I forgot to discuss this in my initial comment.

Thanks for the comment, and welcome to the EA Forum! I think family planning interventions increase welfare nearterm via decreasing human population nearterm when farmed animals have negative lives. However, I do not know whether this is good or bad overall because the longer term effects on animals can be harmful or beneficial.

Nevertheless, I am not confident that saving human lives in China, India or Nigeria is harmful to animals. Even if it is so for farmed animals nearterm, it can still be beneficial overall:

• I would say at least chickens’ lives can become positive over the next few decades in some animal-friendly countries. Relatedly, I would ideally determine the welfare burden per animal per year by country, although it is unclear to me whether I am over or underestimating it. Furthermore, I guess better worsening conditions now imply a longer time until reaching positive lives, and therefore a longer time until increased consumption of farmed animals being beneficial.
• I can see saving human lives being beneficial due to decreasing the number of wild animals with negative lives, although no one really knows whether this is the case or not.
• It is unclear to me whether saving lives increases or decreases person-years. It increases these nearterm via increasing population, but may decrease them longterm, as lower child mortality is associated with lower fertility, which can lead to a smaller longterm population. Note human welfare may be decreased in this case.
• I assume improved human conditions increase the success of animal welfare interventions, for example, via greater willingness to pay for higher welfare products. In any case, I expect more targeted approaches explicitly optimising for animal welfare to be much more cost-effective.

I also think family planning interventions decrease person-years, so I can easily see them decreasing human welfare (relatedly).

Stylistically, some commenters don't seem to understand how this differs from a normal cause prioritisation exercise. Put simply, there's a difference between choosing to ignore the Drowning Child because there are even more children in the next pond over, and ignoring the drowning children entirely because they might grow up to do bad things. Most cause prioritisation is the former, this post is the latter.

As for why the latter is a problem, I agree with JWS's observation that this type of 'For The Greater Good' reasoning leads to great harm when applied at scale. This is not, or rather should not be, hypothetical for EA at this point. No amount of abstract reasoning for why this approach is 'better' is going to outweigh what seems to me to be very clear empirical evidence to the contrary, both within EA and without. 

Beyond that issue, it's pretty easy to identify any person, grant, or policy as plausibly-very-harmful if you focus only on possible negative side effects, so you end up with motivated reasoning driving the answers for what to do.

For example, in this post Vasco recommends:

In addition, I encourage people there to take uncertainty seriously, and, before significant further investigation, only support interventions which are beneficial in the nearterm accounting for effects on farmed animals.

But why stop at farmed animals? What about wild animals, especially insects? What about the long-term future? If taking Expected Total Hedonistic Utilitarianism seriously as Vasco does, I expect these effects to dominate farmed animals. My background understanding is that population increase leads to cultivation of land for farming and reduces wild animal populations and so wild animal suffering quite a bit.. So I could equivalently argue:

In addition, I encourage Vasco to take uncertainty seriously, and, before significant further investigation, only support interventions which are beneficial in the nearterm accounting for effects on wild animals.

These would then tend to be the opposite set of interventions to the prior set. It just goes round and round. I think there are roughly two reasonable approcahes here:

1. Pick something that seems like a clear good - 'save lives', 'end factory farming', 'save the world' - and try to make it happen without tying yourself into knots about side-effects.
2. Really just an extension of (1), but if you come across a side effect that worries you, add that goal as a second terminal goal and split your resources between the goals. 

By contrast, if your genuine goal is to pick an intervention with no plausible chance of causing significant harm, and you are being honest with yourself about possible backfires, you will do nothing.

Thanks, Alex.

Why? It seems clear that you aren't GiveWell's target audience. You know that, and they know that. Unless someone gives me a reason to think that Animal Welfare advocates were expecting to be served by GiveWell, I don't see any value in them clarifying something that seems fairly obvious. 

Many people who donate to GiveWell's interventions care about animal welfare, often donating to animal welfare interventions at the same time. Some of these people may want to know about harms caused to animals nearterm due to supporting GiveWell's interventions. Some of these people may even endorse RP's median welfare ranges, although still support GiveWell's interventions due to not wanting to maximise impartial welfare. In general, people have complex preferences about their giving, so I think it is better to be transparent instead of assuming no one would care about the additional information.

In general, donating $100 to a charity with suboptimal impacts on human welfare but improved impacts on animal welfare is going to be strictly worse - for both human and animal welfare - than donating $90 to the best human welfare charity and $10 to the best animal welfare charity.

I agree. However, it would still be good to go from your 2nd allocation to one where the 10 $ still go to the best animal welfare organisation, but the 90 $ go to an intervention which is more cost-effective than the best human welfare intervention, which may be one global health and development intervention with improved impacts on animals.

Similar thoughts would seem to apply to also other possible side-effects of AMF donations; population growth impacts, impacts on animal welfare (wild or farmed), etc. In no case do I have reason to think that AMF is a particularly powerful lever to move those things, and so if I decide that any of them is the Most Important Thing then AMF would not even be on my list of candidate interventions

I agree with this prioritisation framing, and commented 4 months ago the meat eating problem is mostly a distraction in this sense. However, many people do not think there is a single most important thing, and so may be open to donating to a global health and development interventions with improved impacts on animals even if donating to animal welfare would be more cost-effective. In addition, it still seems worth analysing the meat eating problem to arrive to more accurate beliefs about the world, and because, in some hard to specify way, many value decreasing the probability of causing harm more than prioritising the most cost-effective interventions.

Equally, GiveWell or AIM's donors can offset if they are worried about this. That seems much better than GiveWell making the choice for all their donors.

GiveWell has made many other choices for all of their donors, and the ones related to how much they value saving lives (as a function of age), and increasing income influence way more money than what would be needed to offset potential negative impacts on animals.

Thanks for the comment, John. From the abstract of David Roodman's paper on The Impact of Life-Saving Interventions on Fertility (written in 2014), which is the best research I am aware of on the topic:

In places where lifetime births/woman has been converging to 2 or lower, saving one child’s life should lead parents to avert a birth they would otherwise have. The impact of mortality drops on fertility will be nearly 1:1, so population growth will hardly change. In the increasingly exceptional locales where couples appear not to limit fertility much, such as Niger and Mali, the impact of saving a life on total births will be smaller, and may come about mainly through the biological channel of lactational amenorrhea. Here, mortality-drop-fertility-drop ratios of 1:0.5 and 1:0.33 appear more plausible.

This suggests saving lives in low income countries decreases fertility, as you said, but still increases longterm population, because the drop in fertility is smaller than the drop in mortality.

In any case, if saving lives did not change longterm population, it would still increase population nearterm. Right after saving the life, the population will be counterfactually larger by 1 person (who was just saved). Consequently, saving lives would still increase person-years, and therefore the consumption of animals.

Saving lives could decrease person-years, and therefore the consumption of animals, if it sufficiently decreases longterm population. However, besides this going against David Roodman's main conclusions, it could then easily decrease human welfare, as it would then decrease the total number of human years lived. Moreover, it may decrease animal welfare too if farmed animals already had good lives by the time human population is significantly decreased.

Hi Rebecca.

Contrary to what one could conclude by applying pure utilitarianism, saving someone's life is good even if they go on to do bad things.

Would you agree with saving the life of a suicide bomber who was about to be shot to prevent a detonation which would kill lots of people? In this case, saving the life of the bomber would imply not saving the lives of lots of people. If one prefers not to save the bomber in order to minimise the number of killings, one should also be open to not saving humans in order to minimise the number of animals killed?

The discussion always seems to come up in response to saving the lives of poor people in third world countries, and I have a sense that some people don't react super viscerally to what's actually being said because the humans in question are so distant from them.

The farmed animals in question are even more morally distant, so I am more worried about people being biased towards underestimating the effects on animals.