November 16, 2017
One of the greatest problems in the world today may be the suffering of animals in the meat industry. In 2015, roughly 9.2 billion animals were raised and slaughtered in factory farms in the US alone, in conditions likely to cause extreme suffering.
This problem seems to be incredibly neglected. Many experts now believe that animals have conscious experiences, and are capable of experiencing pain. We tend to give much more weight to the suffering of humans than to the suffering of animals - this is potentially a form of “speciesism”, valuing animals much less than they deserve.
There are things we can do to help solve this problem. Three main types of intervention look promising: persuading people to change their diets, lobbying for better welfare standards, and developing alternatives to animal products. However, the evidence in this area is not as strong as it is for global health interventions.
This profile sets out why you might want to work on improving animal welfare - and why you might not. This area looks particularly high-impact if you think animals’ capacity to suffer is similar to that of humans, and the treatment of animals is unlikely to improve naturally as humanity makes progress.
How can you tell where your resources will do the most good?
- Heuristics: We can use rules of thumb to focus our attention. In particular, we might look for important problems that are being neglected by others, or interventions for which there is lots of evidence, and a possibility to gather more (value of information).
- Quantitative estimates: We might look at studies which estimate the cost-effectiveness of interventions, based on empirical data from randomised control trials (RCTs), althought there aren't RCTs for many of the most promising interventions in animal welfare.
- Historical evidence: For interventions, like corporate campaigning, where it isn't possible to run effective RCTs, we might try to assess whether the intervention caused the company to change, and then use figures on the scale of the change to estimate the impact of the intervention.
Each year, tens of billions of animals are raised for meat and slaughtered on factory farms. This is many times more than the total number of humans alive today (~7 billion). A report on the conditions of modern factory farming by the Humane Society of the United States details how dire these conditions can be. Chickens raised for meat are kept in densely populated sheds with large amounts of waste accumulating. Egg-laying hens are packed together in small cages, while male chicks are ground up alive or gassed. Some dairy cows are kept inside all year round and more than half of them are separated from their calves within 14 hours of birth. Breeder pigs are kept in stalls for years where they cannot even turn around.
It is also becoming clear that factory farming is harming the environment. A report by the Food and Agriculture Organization estimates that animal agriculture accounts for 14.5 percent of all human-induced emissions.
The problem of animal suffering looks even larger in scale when we also consider wild animals, which vastly exceed factory farmed animals in number. For example, it’s estimated that between 0.97 and 2.74 trillion wild fish are caught and slaughtered each year for human consumption. Many more wild animals suffer in the natural environment.
Especially given the scale of the problem, animal welfare seems incredibly neglected. Around 97% of philanthropic funding in the US goes towards helping humans. The remaining 3% is split between the environment and animals. Even within the funding spent on animal welfare, only 1% goes towards farmed animals, although over 99% of domesticated animals are farmed animals.  In total, an estimated $10-40 million is spent each year on reducing animal suffering in factory farms, roughly a tenth of a cent per animal. Work to improve the welfare of wild animals is even more neglected.
There is a small but growing base of evidence on animal welfare interventions, suggesting promising ways to make progress on this problem, and that some approaches could be cost-effective.
Campaigns to try to get large companies to reduce their impact on animal suffering are one of the most promising types of intervention. Corporate campaigns to date have resulted in cage-free pledges from around 100 companies, sparing about 60 million hens annually from confinement. Lewis Bollard (Program Manager for animal welfare at the Open Philanthropy Project, and fund manager for the animal welfare Effective Altruism Fund) conservatively estimates that these campaigns will spare about 38 hens a year from cage confinement per dollar spent - and may save up to 250 hens a year from confinement (depending on how the money spent on these campaigns is counted).
Persuading individuals to go vegan/vegetarian may also be a promising approach. The expected value of going vegetarian seems to be significant. This case is strengthened by the fact that each individual going vegetarian may make it more likely that others go vegetarian. Distributing leaflets is cheap, so leaflet distribution could be cost-effective even if only one in every few hundred people decided to go vegetarian as a result. There have been some preliminary studies looking at the impact of leafleting, but the results are not yet particularly conclusive. ACE’s leafleting study found a small proportion of people who received a leaflet about reducing animal product consumption did report stopping eating meat, relative to no change in a group who received a ‘control’ leaflet. However, the sample was too small for these results to be statistically significant.
Overall, RCTs and cost-effectiveness analyses suggest that there are opportunities to have a large impact on animal welfare. The evidence from corporate campaigns seems particularly promising. However, the evidence base here is still relatively small, which suggests we should be less confident in these estimates of direct cost-effectiveness. This is because we might think that most interventions are not very cost-effective, so we should be skeptical of weak evidence of high impact. However, it also suggests that it may be worthwhile to invest more resources into evaluating the impact of animal welfare interventions. This could help us to make better decisions in future - focusing our efforts elsewhere if these interventions do not seem to be effective, and scaling them up if the results are more promising.
To summarise, we believe that it’s possible to have a large impact in animal welfare because:
- Scale: Tens of billions of animals suffer and are slaughtered in factory farms annually. Orders of magnitude more animals suffer in the wild.
- Neglected: Less than 3% of all philanthropic funding goes towards helping animals, and only a tiny proportion of that funding actually going towards the animals who suffer the most.
- Cost-effective: Corporate campaigns have had some large successes in persuading big companies to change their practices, and rough calculations suggest that these campaigns could be extremely cost-effective.
- High value of information: Doing more research into this issue could help us to decide how many resources we should devote to the problem.
Some people object to the idea of reducing factory farming by claiming that eating meat is natural - or that we need to eat meat in order to be healthy. There are a few important points to note in response to these concerns.
First, just because something is “natural” does not necessarily mean that it is good. For example, it is natural for children to go unvaccinated, with a large proportion dying at a young age. But this state of affairs seems clearly wrong.
Second, even if some amount of predation were natural and necessary, factory farming is not particularly natural. Hens are not naturally kept in tiny cages indoors, and cows are not naturally separated from their calves. At best, wanting to do things that are “natural” could justify personally hunting animals and eating them - but not buying factory farmed, prepared meat from the supermarket.
Moreover, many of the unnatural conditions in factory farms are avoidable. We have the resources to raise animals in a more humane way, alleviating their suffering.
Many people defend the mass production of meat saying that we need meat in order to be healthy. However, it’s far from clear that humans today need to eat animals or animal products, with some evidence suggesting that vegetarians and vegans are in fact more healthy than meat-eaters. 
There are still risks that a vegan diet can lead to deficiencies in certain micronutrients, such as B12 and Omega 3. It’s relatively easy to find vegan foods that contain these nutrients (such as dark green vegetables and fortified cereals/drinks), and/or to take supplements, but this does require a bit of thought and effort.
Animal welfare seems to be a promising cause area. But there are also a number of reasons why you might be unconvinced by this analysis, or why you might think that a different cause area is likely to hold even greater opportunities to do good.
You might think the evidence for the effectiveness of interventions in this space is not strong enough
The evidence base here is still relatively weak, especially when compared to global health interventions. It is normally clear exactly where and how more money directed towards global poverty can improve and save lives. By contrast, much less research has been done on animal welfare interventions.
However, even if there is not enough evidence to say that the intervention is definitely high impact, there seems to be a significant probability that it is very high impact, which could make the expected value high. Some sorts of work, particularly additional research, may also help us to gain more evidence. This is valuable because it helps us make better funding decisions in the future.
Strength of evidence aside, you might choose to prioritise human-centred cause areas over animals simply because you think that improving human lives is higher priority. Deciding how to prioritise animal welfare relative to the problems faced by humans depends on a number of complex issues:
1. The significance of animal suffering relative to human suffering
Though it seems likely that animals have the capacity to suffer and feel pain, it seems reasonable to be more confident of human consciousness than of animal consciousness. We have direct evidence of the former, but we still know very little about which physical or functional characteristics are needed to create conscious experience. Depending on how much more confident we think we should be of human consciousness, this might suggest assigning considerably less moral weight to animals than humans, and therefore prioritising interventions in human welfare.
You might also think that there are other reasons that, for example, make humans living in poverty worse than animals being kept in a cage. You might believe that humans’ greater cognitive complexity means that their capacity to suffer is greater or more significant. Perhaps freedom and dignity are more important for humans than they are for animals, for example. Or perhaps you think that more complex forms of consciousness - the ability to reason and reflect, for example - are more important than pleasure and pain, and that only humans are capable of experiencing these.
2. The indirect effects of poverty interventions versus animal interventions
Human societies are capable of development in a way that animal societies are not, and so we might think that the indirect effects of human-focused interventions will be greater. However, improving attitudes towards animals might increase our circle of empathy generally, which could itself have positive indirect effects.
However, improving the lives of humans could also have negative indirect effects for animals, since people generally eat more meat as they get richer. This is sometimes known as the meat-eater problem.
3. How likely you think it is that the problems faced by each will get solved anyway
Relatedly, it’s also worth considering how likely the problems faced by humans and animals are to be solved without us intervening. There may be good reason to expect that humans will be naturally motivated to improve their lives and the lives of others. It’s less clear whether humans will be naturally inclined to improve the lives of animals.
This might be a point in favour of prioritising them in our altruistic efforts. Even if we think that humanity’s ability to empathise with animals is likely to increase naturally, it’s not clear this will happen fast enough to outweigh our increasing ability to ‘accidentally’ harm animals as a result of pursuing other goals. We probably harm many more animals today through the practice of factory farming than our ancestors did hundreds of years ago, despite the fact that our empathy for animals appears to have increased - simply because it’s easier for us to harm animals at scale, and to participate in harming animals (e.g. by purchasing meat) without having to cause the harm ourselves directly.
The long-run future could be incredibly morally important. We could have billions or even trillions of descendants. This means that even a low probability of improving the lives of future generations, or ensuring that humanity survives into the future at all, could be significant. (We make this case in more detail in our profile on the long-run future.)
In the case of animals, focusing more on the long run future might mean focusing more on changing attitudes and reducing speciesism, and less on short-term farming practices.
However, the extent to which you might choose to focus your altruistic efforts / donations on the long-run future, and what that means in practice, depends on some judgement calls:
1. How much do future lives matter?
Since future lives don’t exist yet (by definition), some people think we have less moral reason to help them than those who do exist. On “person-affecting” views, an action can only be good or bad if it is good or bad for someone - and so actions that will benefit future generations of people or animals cannot be either good or bad, since there is no-one specific they are good or bad for.
If you don’t think future lives matter, then you’ll want to focus on alleviating the most immediate forms of suffering. It’s also worth noting that person-affecting views might also affect how important you think certain kinds of animal advocacy interventions are - since animals live relatively short lives, the impact of many of these interventions will be on animals that don’t yet exist, giving less moral weight to these views.
However, we do think there are some convincing reasons to object to person-affecting views (not least that they lead to some counter-intuitive conclusions), and that future beings do matter morally, which we discuss in more detail in our profile on the long-run future.
2. Is there anything we can do to improve the lives of future sentient beings with any certainty?
There is an empirical question of how easy it is for our actions today to have any impact on people and animals living in the long-run future. You might think that future and presently-existing living beings are equally important, but that it’s much easier for us to do things that help those currently alive, and that we can therefore have a greater impact by focusing on them.
3. How concerned should we be about the possibility of extinction, and is there anything we can do about it?
If you believe that there is a serious threat of human extinction in the next century (or less), and that there are actions we can take now to reduce that threat, then it could be more important to focus on reducing those threats than on improving the welfare of either humans or animals in the more short-term. Suppose there was a >50% chance of an asteroid hitting the earth in the next 50 years that would wipe out all life. Suppose further we were confident that there were things we could do to detect and prevent the asteroid impact, but that doing so would take a great deal of resources. In this situation, we would probably be willing to prioritise putting resources towards this over putting resources into reducing factory farming right now - precisely because it’s little use improving the lives of animals if the whole planet might soon be wiped out.
Of course, this depends a great deal on how likely you think the asteroid (or other existential threat) is, and how confident you are that any attempts to thwart it can make a difference. Short of being 100% certain of the threat and exactly how to prevent it, we wouldn’t want to divert all the world’s resources away from more immediate problems - but we would likely seriously reprioritise. Even if the threat is relatively small, it seems like we should be investing some resources into anticipating and preventing a potential global catastrophe, given how much is at stake. We think this kind of work may currently be too neglected given its importance, because it is so abstract. This perspective and the reasons to take it seriously are outlined in more detail in our profile on the long-run future.
- The problem of animal suffering is huge in scale: each year, at least tens of billions of animals live in dire conditions and are slaughtered in factory farms.
- The current state of evidence suggests we should put a high probability on animals being conscious and able to suffer.
- The problem of animal suffering is incredibly neglected relative to its scale, with less than 3% of total philanthropic funding going towards animal welfare, and less than 1% of that funding going towards factory farming, the practice that accounts for more than 99% of animals used and then killed by humans.
- There is also good reason to think that the problem is relatively tractable: campaigning of various forms has been shown in the past to be effective at changing individual and institutional behaviour as well as regulations around the treatment of animals, and innovation in the food industry seems a very promising way to make it easier for people to avoid animal products.
- However, these interventions are not as well-proven as some focusing on human welfare, such as global health interventions.
- Whether you believe this to be the most important cause depends on: how important you think it is to have strong evidence of effectiveness; how confident you are that animals are conscious; whether you think our treatment of animals will likely improve anyway, and how important you think it is to focus on the long-run future of civilisation over more immediate problems.
This is part of a series of articles setting out the key ideas in effective altruism. Click "next" to keep reading.
If you'd prefer to get a pdf of these resources in your inbox, sign up below. (Once we've sent you some introductory reading, you'll receive roughly one article per month, no spam.)
“These farm animals—sentient, complex, and capable of feeling pain and frustration, joy and excitement—are viewed by industrialized agriculture as commodities and suffer myriad assaults to their physical, mental, and emotional well-being, typically denied the ability to engage in their species-specific behavioral needs. Despite the routine abuses they endure, no federal law protects animals from cruelty on the farm, and the majority of states exempt customary agricultural practices—no matter how abusive—from the scope of their animal cruelty statutes. The treatment of farm animals and the conditions in which they are raised, transported, and slaughter within industrialized agriculture are incompatible with providing adequate levels of welfare.” Humane Society of the United States, “An HSUS Report: The Welfare of Animals in the Meat, Egg, and Dairy Industries” ↩︎
In 2012, a group of neuroscientists signed the Cambridge Declaration on Consciousness, asserting that “The weight of evidence indicates that humans are not unique in possessing the neurological substrates that generate consciousness. Non-human animals, including all mammals and birds, and many other creatures, including octopuses, also possess these neurological substrates.” ↩︎
A recent report from Luke Muehlhauser at the Open Philanthropy Project on “Consciousness and Moral Patienthood” reviews in great detail different ways of assessing the consciousness of different creatures. The report concludes that we still know very little about what features are necessary, sufficient, or indicative of consciousness, and takes a more skeptical view of many of the studies of animal consciousness (pointing out, for example, that there is likely a selection bias in those researchers who choose to study animal consciousness.) However, the report still ultimately suggests assigning relatively high probabilities to the likelihood of consciousness in various animals - 90% for chimpanzees, 80% for cows and chickens, and 70% for rainbow trout - based on how similar these creatures are to humans in various relevant ways. Though these probabilities might seem surprisingly low, they still seem high enough to justify taking the possibility of animal suffering very seriously, especially given the scale of the problem. ↩︎
U.S. EPA, Ag 101: Dairy Production, 27 June 2012 ↩︎
USDA APHIS NAHMS, Dairy 2014: Dairy Cattle Management Practices in the United States, February 2016 (page 7). ↩︎
“Previous studies suggest that fish experience pain and fear and that, for commercially-caught fish, the severity and duration are likely to be high. This study seeks to assess the number of such animals… Fisheries capture statistics (tonnage by species) published by the FAO were used, along with estimates of mean weights for different species, to estimate the global number of fish caught annually.” Estimating the Number of Fish Caught in Global Fishing Each Year. ↩︎
Source: Charity Navigator’s Giving Statistics, 2015 https://www.charitynavigator.org/index.cfm/bay/content.view/cpid/42 ↩︎
Source: Animal Charity Evaluators, “Why farmed animals?” https://animalcharityevaluators.org/donation-advice/why-farmed-animals/ ↩︎
“Animal welfare is an area that receives a lot of attention and funding, but most of this is focused on pets and animals used for lab testing; farm animals receive less attention. This may be because people don’t want to think about where their food comes from, because they find it disgusting or don’t want to feel as if they should change their diet. Certain events have created short-lived increases in public interest, such as Mad Cow disease or the horsemeat scandal. However, this interest wanes, because people generally trust the federal government to address public health issues.” Notes from a conversation between the Open Philanthropy Project and Adam Sheingate. ↩︎
The Open Philanthropy Project’s report on the Treatment of Animals in Industrial Agriculture cites the following main sources of funding in this space, on which this estimate is based: The Humane Society of the United States has an annual budget of $1 million/year for Farm Animal Protection (though resources from other parts of the organisation may also be directed towards farm animals, potentially raising the total up to ~$10 million); six smaller advocacy organisations have annual budgets in the range of $500,000 to $2 million, and Farm Sanctuary, which has a budget of ~$9 million per year. ↩︎
There are also psychological reasons to expect we might neglect this problem. We tend to be most motivated to help when we can empathise with a specific living being, and when we feel personally responsible for alleviating their suffering. But these motivating factors are absent for factory farming. The problem is somewhat hidden, affects many anonymous beings, and is something no single person feels much responsibility for.
In addition, we may be biased against taking the suffering of animals seriously. This bias is sometimes called “speciesism”. There is experimental evidence that people think of animals as having far fewer mental capacities than they in fact do. This bias is perhaps because of the widespread acceptance of meat consumption and farming practices. For someone who eats meat, acknowledging the harm this causes is likely to be emotionally difficult. This means that they may have a strong incentive to turn a blind eye and find ways to convince themselves that the problem is not so big after all. ↩︎
“Counting just the ~$2.5 million spent on corporate cage-free campaigning over the last few years, and conservatively assuming that the campaigns only accelerated pledges by five years, these campaigns will spare about 250 hens a year of cage confinement per dollar spent. And even if you add the $1.5 million disbursement for the first year of our three grants, and the ~$12.5 million (at most) spent both on Prop 2 in 2008 and all egg undercover investigations ever done in the U.S., these campaigns will still spare about 38 hens a year of cage confinement per dollar spent. In my view, the assumption that these campaigns only accelerated pledges by five years is very conservative. It seems equally likely that these companies would never have dropped battery cages, or would have merely transitioned to “enriched” cages. For instance, as recently as March 2015, a coalition backed by McDonald’s, General Mills, and other major food companies issued a report which largely endorsed “enriched” cages as an alternative to cage-free systems.” (From OPP’s profile on The Humane League and corporate campaigns.) ↩︎
Brian Tomasik makes a convincing case that an individual’s choice to go vegetarian does make a difference, at least in expectation. It seems that each individual vegetarian is unlikely to make a difference to the number of animals raised. But if they do make a difference, that difference is likely to be very large, so the expected value will be large. ↩︎
This is also relevant when it comes to the problem of wild animal suffering, which we mentioned may be many of orders of magnitude larger a problem than even factory farms. Though it’s not currently clear how tractable this problem is, the potential gains from exploring potential ways to solve this problem could be extremely high, given the scale and neglectedness of the problem. ↩︎
The Oxford Vegetarian Study looked at the health of 6,000 vegetarians and 5,000 nonvegetarian control subjects in the UK between 1980 and 1984. “Cross-sectional analyses of study data showed that vegans had lower total- and LDL-cholesterol concentrations than did meat eaters; vegetarians and fish eaters had intermediate and similar values… After adjusting for smoking, body mass index, and social class, death rates were lower in non-meat-eaters than in meat eaters for each of the mortality endpoints studied… the health of vegetarians in this study is generally good and compares favorably with that of the nonvegetarian control subjects.” ↩︎
“Research shows that plant-based diets are cost-effective, low-risk interventions that may lower body mass index, blood pressure, HbA1C, and cholesterol levels.” Nutritional Update for Physicians: Plant-Based Diets. ↩︎
“Vegans tend to be thinner, have lower serum cholesterol, and lower blood pressure, reducing their risk of heart disease. However, eliminating all animal products from the diet increases the risk of certain nutritional deficiencies. Micronutrients of special concern for the vegan include vitamins B-12 and D, calcium, and long-chain n–3 (omega-3) fatty acids. Unless vegans regularly consume foods that are fortified with these nutrients, appropriate supplements should be consumed. In some cases, iron and zinc status of vegans may also be of concern because of the limited bioavailability of these minerals.” Health effects of vegan diets, The American Journal of Clinical Nutrition. ↩︎
It’s worth noting that there are a number of complex steps and assumptions involved in reaching a conclusion about whether to prioritise human or animal welfare interventions, especially if we’re considering specific types of interventions. First, there is the question of how confident we should be that different types of animals are conscious, which turns out to be incredibly complicated. Next, there is a question of how this assessment of consciousness should be combined with other factors to assess the relative moral weight of humans vs animals - a separate issue, which might depend on things like more detailed theories of wellbeing or how “unified” the conscious experience of different beings are. Finally, even if we can decide how much relative moral weight we should assign to animals, we then need to consider the scale of the problems faced by / the suffering inflicted on different species - even if we assign humans twice the moral weight of animals, for example, we might still prioritise animal welfare interventions if we think that animal suffering is more than twice the scale of human suffering. ↩︎
We might expect lifting people out of poverty, for example, to have all kinds of knock-on benefits resulting from these people then being able to contribute more to societal progress. Since animals do not live in such organised societies, it seems less likely that there would be such additional benefits beyond just reducing their suffering - more content animals seem less able to contribute to societal progress. Owen Cotton-Barratt suggests that this might be a reason to favour interventions focused on human welfare over those focused on animal welfare. ↩︎
Though it’s worth noting that if your goal is to increase empathy broadly, there may be other / more direct ways to do this than by focusing on animal welfare worth considering. There are also some arguments against explicitly working on moral advocacy here: https://rationalaltruist.com/2013/06/13/against-moral-advocacy/ ↩︎
Especially if human progress is not accompanied by similar levels of moral progress. ↩︎
- 1.Introduction to Effective Altruism
- 2.Efficient Charity — Do Unto Others
- 3.Prospecting for Gold
- 4.Crucial Considerations and Wise Philanthropy
- 5.The Moral Value of Information
- 6.The Long-Term Future
- 7.A Proposed Adjustment to the Astronomical Waste Argument
- 8.Three Impacts of Machine Intelligence
- 9.Potential Risks from Advanced AI
- 10.What Does (and Doesn’t) AI Mean for Effective Altruism?
- 11.Biosecurity as an EA Cause Area
- 12.Animal Welfare
- 13.Effective Altruism in Government
- 14.Global Health and Development
- 15.How valuable is movement growth?