Death is not the only moment a fish suffers and it’s clear that a carp sold alive and killed later may suffer more than a salmon which is slaughtered earlier in the farming process. At the same time, feed fish which are consumed by salmon are typically killed in extremely unethical ways (usually live freezing in ice slurry, asphyxiation or getting crushed by the weight of other fish) and their agony often lasts a few hours.
Because of this and the differences in farming methods, fishes may suffer more or less during their lives, making it difficult to precisely calculate the levels of suffering for the two species discussed in this post. For the purpose of this model we ignored these differences as we used an already low estimate of 1% of consumers in the hope that a very conservative approach would offset any welfare differences.
Choosing a 1% doesn’t work to mitigate this problem, i.e., if your initial estimate of relative badness was based on the number of animals killed in the process, and you notice that this relative estimate could be affected by things like painfulness of death (or by the capacity to suffer of differently sized fish), then assuming that only 1% of consumers switch is neither here nor there.
I instead read this point as saying “assume that if we persuaded 100 folks to give up carp, then 1 of those would replace their carp consumption with salmon.” So it’s talking about the replacement effect, rather than the number persuaded (the latter gives magnitude, as you say).
Yeah, this is what I meant in this point, but NunoSempere’s comment made me confused about strength of my model here and I interpret this confusion as either me getting something wrongly here or not fully getting his comment. I will think/discuss it more and update the phrasing here just to be sure it’s not misguided. Thank you for that.
Choosing a 1% doesn’t work to mitigate this problem, i.e., if your initial estimate of relative badness was based on the number of animals killed in the process, and you notice that this relative estimate could be affected by things like painfulness of death (or by the capacity to suffer of differently sized fish), then assuming that only 1% of consumers switch is neither here nor there.
I instead read this point as saying “assume that if we persuaded 100 folks to give up carp, then 1 of those would replace their carp consumption with salmon.” So it’s talking about the replacement effect, rather than the number persuaded (the latter gives magnitude, as you say).
I see, that makes some sense
Yeah, this is what I meant in this point, but NunoSempere’s comment made me confused about strength of my model here and I interpret this confusion as either me getting something wrongly here or not fully getting his comment. I will think/discuss it more and update the phrasing here just to be sure it’s not misguided. Thank you for that.