Summary
Small wild fishes such as anchovies weigh around 10 grams.
Some fishing boats catch primarily small wild fish.
These boats can catch more than 3.000 tons of fish (possibly 300 million individual fish) every year.
The traditional method for killing farmed and wild fish is to let them to asphyxiate in air or on ice, which is slow and distressing.
There are effective and durable stunning equipment for wild catch fish in the market. The average costs are approximately 100.000$ per piece.
If one ignores other costs, one can simply end the asphyxia pain of 30.000 fish per 1 dollar if one assumes that this equipment will be used for 10 years.
Although convincing or forcing companies and governments to implement welfare reforms would be more preferable and cost-effective, there may be circumstances where this may not be tractable and cost-effective.
If effective animal charities which engage in welfare campaigns are already adequately funded, this could be another cause area which can also use funds effectively without the need for social change.
Introduction
Under normal circumstances, the most cost-effective farm animal welfare interventions do not pay for the welfare reform itself. Cost-effective grants typically pay the salaries and the operation costs of a small team which convince or coerce the decision makers in the food supply chain to bring about the necessary social change. This way, the overall cost of the reforms is eventually paid by the governments, companies and consumers.
However, stunning small wild catch fish cost so little that directly purchasing and distributing the necessary equipment may also pass the threshold for cost-effectiveness. In this essay, I will briefly explain the importance, neglectedness and tractability of this potential cause area and then argue that this intervention can be similarly cost-effective as some of the animal welfare campaigns. While I agree that achieving fish welfare reforms through corporate outreach and campaigning are more cost-effective in many countries and circumstances, in cases where animal advocacy is not an option or ineffective due to unfavorable socioeconomic conditions, directly paying for this particular welfare reform can also be reasonably cost-effective.
![](http://res.cloudinary.com/cea/image/upload/v1667996073/mirroredImages/tykEYESbJkqT39v64/yybi2bennlvwc7faibkg.jpg)
Importance
By the numbers, small wild fishes constitute the overwhelming majority of animals in our food system. The obvious reason for this is that these fishes are incredibly small: 10-100 grams. These are caught by the tons in the seas and oceans by large fishing boats every year. Fishing industry, like all animal farming industries, has become very centralized and “efficient” in terms of producing lots of animal products in a very small number of facilities, or in this case, fishing boats. Modern fishing boats (fishing trawlers and purse seine boats), can catch enormous amounts of fish.
![](http://res.cloudinary.com/cea/image/upload/v1667996073/mirroredImages/tykEYESbJkqT39v64/maxjqysidbz4h4r4wjat.png)
Sources: FAO Data (2016) and fishcount.org.uk, extrapolated to 2016. from: Amanda Hungerford, How Much Should We Focus on Slaughter?, Open Philanthropy Farm Animal Welfare Newsletter.
Some sources indicate that some fishing boats catch about 3000-7600 tons of small fish per year (source -page 22- “...a typical fishing boat catches on average 3350 tons of anchovies [Turkish fishers in Black Sea].”and source : “...The average catch and value for the steel hulled purse seiners is 7,600 metric tons…” [Peruvian fishers in the Pacific Ocean]). This means in a single fishing boat, more than 300 million (small) wild fish may be caught every year.
In the US, about 9 billion broilers are slaughtered each year. This means that only 30 fishing boats can be responsible for the equal number of slaughter of fish with the total number of broilers which are slaughtered in the US each year.
Unfortunately, fish also suffer a lot more during slaughter than broilers or other land animals. The traditional method for wild fish slaughter is to allow them to asphyxiate in air or on ice, which is slow and distressing. “One study found that it takes common species of fish 55-250 minutes to die via asphyxiation. Fish that aren’t asphyxiated often die by being live gutted; the same study found that live gutting can take up to 25-65 minutes to kill the fish” (from, Amanda Hungerford, “How Much Should We Focus On Slaughter”, Open Philanthropy Farm Animal Welfare Newsletter).
These numbers suggest that “each” fishing boat which catches primarily small wild fish can cause incredible animal suffering if proper stunning equipment is not installed. Here is a video from a fishing boat catching tons of anchovies in a single trip.
![](http://res.cloudinary.com/cea/image/upload/v1667996073/mirroredImages/tykEYESbJkqT39v64/h7keymcrhdertodesber.png)
Capture of all anchovy reported by the FAO (green indicates Peruvian anchoveta) from Wikipedia
Tractability
There are multiple stunning equipment manufacturers. I have contacted one of them after consulting with a leading humane slaughter charity (I will not disclose their name): Ace Aquatec. Ace Aquatec provides fish stunners to fish farms and fishing boats. They also claim that their fish stunners can be used for any species in any size and are very durable since they do not have moving parts (Youtube video). Stunning equipment can also be fitted on boats as well (Youtube video).
There are also other fish stunning equipment manufacturers. One of them is Blue North, which also stuns wild-catch fish. But they seem to focus on fishing cod, and they are not aiming to catch tons of fish. Ekofish group also uses stunners on boats. But they also seem to fish larger wild fish. Other stunning equipment are generally designed to be used on fish farms.
The average cost for stunning equipment in the market is 100.000$ (I am not sharing ACE Aquatec’s price since they might not like to disclose it to the public).
If the funder decides to bear the cost of the equipment, then a typical fishing company is very likely to accept to install it on its boats. Since stunning is also good for fish meat quality and makes handling of fish easier, it is in the interest of the fishers as well.
Neglectedness
Fish are generally considered as a “lower class” animal, even among people who like animals. Public attention and concern for fish welfare is unfortunately very low. Fish welfare reforms are not also very common and comprehensive, even in Europe and North America.
Concern for fish welfare is fortunately increasing, especially in the Effective Altruism community and effective animal charities. And some progress in fish welfare has been achieved. But this progress is not as extensive as in hen or broiler welfare.
One reason for the lack of comprehensive reforms was that there was not a clear cut fish welfare ask. This problem will be hopefully solved in the near future. But even future fish welfare asks will likely be focused on farmed fish which are in controlled environments (farms) and are directly in retail supply chains.
Wild-catch fish on the other hand receive less attention. Small fishes in particular are perceived as even lower value. Consumers generally believe that larger fish has better taste and quality, such as cod, sea brass, trout and salmon. People also believe that these fishes matter more than small fishes from a moral point of view as well. And since the public is not very interested in looking at numbers and cost-effectiveness, wild-catch small fish become very neglected.
Who is working on this?
There are multiple animal charities which work on fish welfare: Eurogroup for Animals, Fish Welfare Initiative, Dyrevernalliansen, Humane Slaughter Association, Global G.A.P., Compassion in World Farming, Rethink Priorities, Essere Animali, Albert Schweitzer Foundation, Environmental & Animal Society of Taiwan, Animal Protection Denmark…
These groups are conducting research and/or doing advocacy work on fish welfare. There is no charity that I am aware of which directly purchases and distributes stunning equipment.
What can a funder do?
In addition to funding animal charities where progress in fish welfare advocacy and reforms are tractable, Open Philanthropy can also initiate the founding of a charity which directly purchases and distributes stunning equipment to fishing boats which catch enormous amounts of small wild fish. There are multiple regions where this can be done: Peruvian coasts, Black Sea, Sea of Japan, Yellow Sea, and East China Sea, South African coasts etc.
Cost-effectiveness
Calculations
If we assume that a fishing boat catches 3.000 tons (3.000.000 kilos) per year, and each wild catch fish weighs only 10 grams (0.01 kilos), this means that this fishing boat catches 300.000.000 fish (300 million fish) each year.
If we assume that a fish stunner costs 100.000$ and it will be used for 10 years, this means that one can stun 3 billion fish (300 million X 10) per 100.000$.
This means that 30.000 fish can be stunned per 1$.
It is also very cheap to provide stunning for other animals. For broilers for example, broiler controlled-atmosphere stunning equipment costs about 300.000€ (source, page 35-38). As for farmed fish too, it costs just cents per fish to stun (source, pages 7-12). But these figures are still high if one considers to bear all costs for a lot of animals. For that reason, one has to look for even lower stunning costs per animal: small wild catch fish.
A comparison
I assume that the welfare gains of the fish stunner (the elimination of asphyxiation suffering of a small wild fish), is equivalent to the welfare gains of a cage-free system or higher welfare broiler breed for a chicken during a half-day.
Saulius Šimčikas estimated that corporate campaigns affect 9 to 120 years of chicken life per dollar spent (post). The mean is 41 years. This means that 1$ spent in this potential cause area (purchasing and distributing fish stunners), can be similarly cost-effective as interventions which affect 41 chicken-years per 1$.
30.000 (number of fish asphyxiation)/ 2 (half-day equivalence with chicken suffering) /365 (days in a year) = 41.6.
Given these numbers and assumptions, this cause area can be similarly cost-effective with past hen and broiler welfare campaigns.
One can even lower the equivalence and still reach a high impact estimation. Let’s assume that the welfare gains of the fish stunner (the elimination of asphyxiation suffering of a small wild fish), is equivalent to the welfare gains of a higher welfare broiler breed for a quarter-day. For broiler campaigns, Šimčikas estimated that the mean chicken-years affected is 15, for the fish stunner it would be 20.8.
The argument
Although I believe that this potential cause area can be cost-effective, I am not suggesting that the most cost-effective way to achieve animal welfare is to directly bear the costs. Funding animal charities which create social change via advocacy would be even more cost-effective, especially when they start to work on fish welfare. These groups can convince or force governments, corporations and eventually consumers to bear the total costs for reforms.
But I do think there are also limitations. Firstly, even in developed countries, the public can be less concerned about fish welfare, small wild-catch fish in particular, than other animal welfare. This may hinder the pressure effect of animal charities which try to convince corporations and governments that this is something important not just for them, but also for their customers and voters. Although one can be optimistic by looking at previous surveys, people can act very differently than their answers on surveys, and decision makers know this.
Secondly, animal welfare reforms were mostly successful in countries which are relatively rich. This is expected since prosperity allows corporations and their customers to bear the cost of welfare reforms, and wealthier, developed countries are more open to widening their moral circle to animals since they have already solved most of the “human welfare” problems in the first place.
I am not pessimistic about the first limitation in relatively rich countries where there are also established and powerful animal advocacy groups. I think it is very likely that we will see major progress in fish welfare in developed countries, Europe in particular, in the near future. But I am somewhat pessimistic about the second limitation. It can be really hard to push for “wild-catch small fish slaughter welfare reforms” in regions which I mentioned above. Although the costs per animal is extremely low, overall costs (about 100.000$) may deter many fishing companies from adopting this reform. While this cost may be reasonable for a European fishing company (or a boat owner), it may be ridiculous for a Peruvian or a Turkish or a Chinese fishing company (or a boat owner), with good reason since they and their customers may not be as rich as the European companies and customers.
A funder should obviously fund firstly the animal charities which have the opportunities to achieve (chicken or fish) welfare reforms through advocacy. But there is of course a limit, even a very effective charity cannot use indefinite amounts of money in the same level of efficiency of achieving welfare reforms. Each new hire or new advertisement airing or undercover investigation will have a lower marginal utility. And finally, a charity cannot indefinitely achieve welfare reforms. At some point, there will not be any more reasonable fish welfare reform to be implemented.
If one assumes that all (existing and potential) effective animal charities, which aim for primarily chicken and fish animal welfare reforms, are well funded, I believe that Open Philanthropy can also consider funding this potential cause area as well.
Cost-effectiveness can be lower and higher
First uncertainty is about the projected costs. The cost calculation made above does not cover the cost of staff who need to manage the process of purchase, communications, and supervision. One cannot just select a fishing boat in the sea and click “upgrade stunners” as one can in a video game. This cost calculation also does not cover the energy cost of stunning equipment. I assumed that staff costs would be negligible since there will not be a major effort necessary to bring together consenting parties (funders who want to buy the equipment, manufacturers who want to sell their equipment and fishers who want to upgrade their boats for almost free). I also assumed that energy costs would also be acceptable for fishers since stunning improves fish meat quality and makes handling of fish easier (and they get new equipment for free). If those assumptions are wrong, then the cost-effectiveness may be lower. On the other hand, one can also negotiate prices. Some fishers may agree to install stunners if the funder covers some of the costs, not all of it. One can also negotiate the price with the manufacturer. If cost per stunner decreases, then the cost-effectiveness may be higher.
Second uncertainty is about the effectiveness of the stunners. Fish stunners are not widely used yet, so the evidence of their effectiveness in alleviating the suffering of fish may change in the future. Also, small fish are usually caught in big fishing nets and brought near the boat. In this process, a significant amount of small fish can be crushed before they are even pumped on the boat. This can be solved by loosening the nets and providing fish enough space before they are pumped on the boat and the stunning equipment. But fishers can also ignore this. If the effectiveness of the stunners is lower than I assumed, cost-effectiveness may be lower.
Third uncertainty is about the moral status of small fish. One can argue that fish, and small fish like anchovies matter or suffer less than I assumed. If my assumptions are wrong, then cost-effectiveness may be lower. But one can also argue that I underestimated the value of small fish suffering. If that is true, cost-effectiveness may be higher.
Final uncertainty is about indirect effects:
a) This intervention may be counterproductive by making fishers, companies, governments reliant on philanthropists. If fishermen, companies and governments would wait for the philanthropists to pay for the costs, rather than implementing the reforms themselves in the near future, this would waste lots of money. If that is the case, the cost-effectiveness may be much lower. But I do not think this is the case in many regions.
b) This intervention can also motivate governments to make fish stunning compulsory for all fishers. Fishers which have their stunners installed can also lobby the government as well. This can happen if a funder can install stunners on the majority of fishing boats. Although this would cost a lot, the number of large fishing boats is not that big in many countries. One can also negotiate with the government in order to pay for the stunners in exchange for legislation (in Peru for example). If that can be achieved, the cost-effectiveness may be much higher.
c) This intervention has no moral advocacy involved. So if one believes that other interventions which do involve moral advocacy have significant future impacts, one may discount the relative cost-effectiveness of this cause area. My view on this is that awareness raising, movement building activities are mostly instrumental (if they are designed as such) to achieve animal welfare campaign objectives. For that reason, I do not think that one needs to double-count the effect of moral advocacy.
d) The foundation of this potential cause area may also draw other donors to effective giving for animal welfare. For many people, including myself, “the life you can save meme” which basically claims that you can save a person’s life by donating 2.500$ (current amount is higher) to Against Malaria Foundation was very impressive. Animal Charity Evaluators tried to create a similar meme in the past. But it did not work since firstly, it was very weird for most people (there were negative numbers within the ranges of effects per donation, which meant that the donation could be harmful to animals (?)). And secondly, the models used were problematic and eventually they were dropped. Currently Animal Charity Evaluators provides a list of top and standout charities without giving an estimated effect per dollar. That is reasonable since charities use funds to create (social) change in the food system. One can never say that if one writes a check to a charity, that charity will create a definite social change in the foreseeable future. No animal advocate thinks that they will immediately and/or definitely get a welfare commitment from a corporation after they receive a grant. These uncertainties about tractability of social change can put off some donors.
This cause area, on the other hand, has a very clear message for donors: for 1$ you can end the suffering of 30.000 suffocating fish. This can motivate some people (other than Open Philanthropy) to donate more and perhaps more importantly, this can draw people to effective donating as the life you can save meme did (fish you can save?).
Although Open Philanthropy’s farm animal welfare grants seem very focused on tractable and cost-effective cause areas, there are so many projects in animal advocacy that do not seem that tractable and cost-effective. These are funded nevertheless by other funders because they impress funders and/or funders cannot see more cost-effective funding opportunities. Creating a clear, cost-effective funding opportunity can improve the cost-effectiveness of other grants made by other organizations and donors if some are drawn to this potential cause area.
Conclusion
The most cost-effective way to improve farm animal welfare seems to be animal welfare campaigns for hen, broilers and fish (including constructive corporate outreach) in regions where those campaigns are tractable. Providing financial support for those groups would obviously be more cost-effective since these groups can convince or force governments, corporations and eventually consumers to bear the total costs for reform.
However, in cases where this approach is not tractable and cost-effective, this potential cause area also provides a cost-effective funding opportunity. One of the constraints of funders is cost-effective funding opportunities. This potential cause area can reasonably provide funding opportunities which can make use of millions of dollars while saving billions of fish from a horrific death.
Thanks, great post, and far better and more specific than mine. After looking a bit into the issue, and pending and pending seeing arguments or evidence to the contrary, I think wild fish slaughter is the single [non-longtermism-motivated, object level] cause most deserving of marginal resources.
If there was a charity (or subset of HSA) working on this, It'd be my default go-to for individual giving.
A few things bits from my research I think complement this post well
Hi there!
Thank you very much. It is nice to see someone else thinking almost the same things, it's like seeing someone murmuring the same song you also play in your head.
Thanks for the complements as well.
[1] Fish suffering (wild fish in particular) looks so big that it makes me question if we are missing something. One thing I am trying to find out is the duration of asphyxia in small wild fish. The only "sources" I can find is some videos on YouTube.
[2] Yeah, I saw that one too. I assume that the main bottleneck in the past was the lack of effective stunning technology. I think this is still an issue, since there are so few stunner manufacturers.
[3] Very interesting idea. Substituting wild-fish fish feed with plant-based fish feed would also decrease the demand for wild-fish catch. The difficulty is that, since most wild-fish fishing is not driven by demand, but rather constrained by the supply of fish in the seas and the fishing quotas, one has to dramatically decrease the fish demand in order to create a situation where fishers would rather stay at the docks even if they have not filled their quotas and there are plenty of fish in the seas. That may be hard to achieve.
Hi Enginar,
This is a very interesting post, thank you for writing it! It fits very well into the idea of megaprojects for animals, or ways of using additional funding, which may become increasingly relevant as more funding comes into EA.
I've been working on a report on improving wild caught fish welfare at Animal Ask and the report should be relatively soon. I certainly share your thoughts that smaller fish should be the priority.
I have some (not necessarily decisive) worries about how this fits into the long-term strategy of animal advocacy. As you say, it is not exactly moral advocacy (though we should acknowledge that sometimes changing attitudes follows as a result of welfare improvements). I think moral advocacy is important not just for the current animal welfare ask, but for enabling future asks, so this seems less desirable from the perspective.
Hi there!
Thank you for the comment and your work on fish welfare.
I agree, moral advocacy is also important and may have long term impacts. My point is that "not all" moral advocacy interventions or funding opportunities may be as cost-effective as directly purchasing a wild fish stunner. The main question is: does every animal advocacy group create as much (long term) impact as relieving the suffering of 30.000 fish per 1$?
Also, given that effective animal advocacy groups are limited in number and their ability to use funds to create impact are also limited due to their capacity and social conditions, this intervention may also be a very reasonable addition to the avenues of helping animals.
I think same logic applies (even strongly) to other cause areas. One can say that advocacy for effective international aid and effective development reforms might have even more impact in the short and long term. But given that the possibilities are limited and tractability is low, Give Well's funding opportunities are very impressive even if they do not have as much long term impact as a broad political or social change.
Thanks, this is an interesting post! (I was going back reading some posts on aquatic animal welfare and came across this).
I think the crux for me is:
I think this paragraph is hiding a lot of the work for me. I'd be interested in reading a follow up (even in a quick BOTEC-y form) for:
I guess I don't know enough to assess whether the heuristic of comparing "stunning v. asphyxiation" to "half-day of cage-free / higher welfare breed v. baseline" is sensible.
Also curious where this list of examples is generated from?
Have you considered brushing this up and sending this as an idea for Charity Entrepreneurship to consider incubating as a charity (or consider applying yourself if you think you would be a good fit)?