It’s good to be careful about wild animal (feed fish) effects and I recommend looking for interventions that are robustly positive taking them into account, but I think the analysis of wild fish effects here is missing too much of the picture. I don’t think this necessarily flips the conclusion again and means the intervention is robustly positive; I just think the appropriate response now is cluelessness, and we should look for something that we’re more confident is robustly positive, and/or try to better understand and weigh the effects.
The price elasticity of supply of wild-caught fish can be negative when there’s overfishing, because reducing fishing pressure (the percentage of the population caught per period) allows the population to recover enough to allow more fish to be caught in the long run. So the number of fishing deaths for feed fish could actually decrease from a shift towards salmon.
In well-managed fisheries with quotas, the elasticity should be basically 0 when the quotas are binding, so catch wouldn’t change. It can be positive if there’s underfishing and either there’s no quota or catch is below the quota, so feed fish catch would increase with a shift towards salmon.
Wild-caught feed fish populations might decrease in fisheries from increasing demand (if there’s no quota that’s binding). It’s not clear whether that’s good or bad. Also, of course if the population decreases, so will deaths, just not necessarily due to fishing (although fishing deaths may decrease, too, if overfishing). On the other hand, the populations of their prey may increase in response. Even anchovies eat crustaceans. Is this a good trade?
Thanks for the comment Michael. It’s really informative.
Do you have any good sources that describe your points in more depth? These would be good for us to follow.
In regards to robustness and cluelessness—I agree. This is probably the strongest update based on the new considerations highlighted in the blog post. Similar conclusion was formed, especially among senior campaigners, although we used different phrases for that during discussions.
I think there are few studies that directly estimate the supply curve or its price elasticity, but you can instead often infer the sign of the price elasticity of supply from common model assumptions + stock/fishery data, which are more widely available.
It’s worth mentioning that wild animal effects can also make us clueless or pessimistic about interventions that reduce the production of terrestrial animal products, e.g. diet change and alternative proteins, I think mostly because of land use effects on wild animal populations.
(Edited for clarity.)
It’s good to be careful about wild animal (feed fish) effects and I recommend looking for interventions that are robustly positive taking them into account, but I think the analysis of wild fish effects here is missing too much of the picture. I don’t think this necessarily flips the conclusion again and means the intervention is robustly positive; I just think the appropriate response now is cluelessness, and we should look for something that we’re more confident is robustly positive, and/or try to better understand and weigh the effects.
Copying from this comment and flipping things to reflect increasing demand for feed fish from a shift towards salmon: https://forum.effectivealtruism.org/posts/zu7D6DKMcB5Jwq5Ey/updates-to-faunalytics-animal-product-impact-scales?commentId=NGgFEfH7LB8o5uZSm
The price elasticity of supply of wild-caught fish can be negative when there’s overfishing, because reducing fishing pressure (the percentage of the population caught per period) allows the population to recover enough to allow more fish to be caught in the long run. So the number of fishing deaths for feed fish could actually decrease from a shift towards salmon.
In well-managed fisheries with quotas, the elasticity should be basically 0 when the quotas are binding, so catch wouldn’t change. It can be positive if there’s underfishing and either there’s no quota or catch is below the quota, so feed fish catch would increase with a shift towards salmon.
Wild-caught feed fish populations might decrease in fisheries from increasing demand (if there’s no quota that’s binding). It’s not clear whether that’s good or bad. Also, of course if the population decreases, so will deaths, just not necessarily due to fishing (although fishing deaths may decrease, too, if overfishing). On the other hand, the populations of their prey may increase in response. Even anchovies eat crustaceans. Is this a good trade?
Thanks for the comment Michael. It’s really informative.
Do you have any good sources that describe your points in more depth? These would be good for us to follow.
In regards to robustness and cluelessness—I agree. This is probably the strongest update based on the new considerations highlighted in the blog post. Similar conclusion was formed, especially among senior campaigners, although we used different phrases for that during discussions.
These issues are discussed in https://reducing-suffering.org/#fishing (largely from a suffering-focused/negative utilitarian perspective), especially:
https://reducing-suffering.org/should-fishing-opponents-be-happy-about-overfishing/
https://reducing-suffering.org/wild-caught-fishing-affects-wild-animal-suffering/
https://reducing-suffering.org/marine-trophic-level-contains-total-suffering/
On what Peruvian anchoveta eat, see https://en.wikipedia.org/wiki/Peruvian_anchoveta#Description , and Peruvian anchoveta account for 20-40% of fishmeal produced from whole wild-caught fish depending on the year, if I recall correctly.
For more on the kinds of models used by fishery experts:
https://en.wikipedia.org/wiki/Maximum_sustainable_yield
https://haddonm.github.io/URMQMF/
For some direct estimates of price elasticity of supply for anchovies that came out negative, from Vietnam:
https://munin.uit.no/bitstream/handle/10037/6370/article.pdf;jsessionid=4E475668FEFAD00890D8766F0595DD81?sequence=1 / https://www.ccsenet.org/journal/index.php/jsd/article/view/27124 (they don’t directly report elasticities, but you can derive these from their models or check that supply is a decreasing function of price)
https://folk.ntnu.no/skonhoft/EDE%20Middlmen%200519.pdf , based on http://www.rimf.org.vn/ctkhcn/chitiet/TAYNAMBO
I think there are few studies that directly estimate the supply curve or its price elasticity, but you can instead often infer the sign of the price elasticity of supply from common model assumptions + stock/fishery data, which are more widely available.
It’s worth mentioning that wild animal effects can also make us clueless or pessimistic about interventions that reduce the production of terrestrial animal products, e.g. diet change and alternative proteins, I think mostly because of land use effects on wild animal populations.