How do you compare human and animal suffering?

On estimates, from Lewis Bollard's recent AMA:

Our current very rough estimate is that our average $ spent on corporate campaigns and all supporting work (which is ~40% of our total animal grant-making) achieves the equivalent of ~7 animals spared a year of complete suffering. We use this a rough benchmark for BOTECs on  new grants, and my best guess is this reflects roughly the range we should hope for the last pro-animal dollar. 

Rethink Priorities' estimate "Corporate campaigns affect 9 to 120 years of chicken life per dollar spent" (on average, not on the margin).

Founders Pledge also did a direct comparison between The Humane League and the Against Malaria Foundation here. They've done their own research, and summarizing others' and theirs together: "Cost-effectiveness analyses have found that for every dollar donated, the lives of between 10 and 160 birds are affected."

Charity Entrepreneurship has some research on this, and Animal Charity Evaluators has estimates of number of animals affected by corporate campaigns for specific charities, too; you can see their reviews of recommended charities working on corporate campaigns, and this model and this spreadsheet. (I'm an intern at ACE, but only speaking for myself.)

But also see Global poverty could be more cost-effective than animal advocacy (even for non-speciesists) by Peter Hurford.

I would guess most arguments for global health and poverty over animal welfare fall under the following:

- animals are not conscious or less conscious than humans
- animals suffer less than humans

See A Debate on Animal Consciousness, Why might one value animals far less than humans? and this thread here.

The arguments for the former mostly depend on nonhuman animals not being sufficiently self-reflective in the right way, and the arguments for the latter depend on brain size/complexity or particular welfare-relevant capacities that humans have that nonhuman animals lack or have less of, e.g. preferences for the future.

But also see Rethink Priorities' research:

https://www.rethinkpriorities.org/publications#moralweight

https://www.rethinkpriorities.org/invertebrate-sentience-table

ASENT's research:

https://www.lse.ac.uk/cpnss/research/ASENT

And Luke Muehlhauser's older work:

https://www.openphilanthropy.org/2017-report-consciousness-and-moral-patienthood

https://www.lesswrong.com/posts/2jTQTxYNwo6zb3Kyp/preliminary-thoughts-on-moral-weight

we can't know how much we are improving animal suffering

There's also skepticism about the specific interventions supported to help nonhuman animals, because their welfare effects actually seem plausibly bad, or because the evidence is just much less rigorous, or because of interactions with wild animals (as discussed in this thread).

My own view is that farmed vertebrates are very likely to be conscious and brain size/complexity is relevant and probably the most important factor, and I would tentatively use something between square root and linear weighting for brain regions involved in hedonistic experience. I also think that nonhuman animals (farmed and wild vertebrates) suffer about as much in practice as some of the worst off humans in the world, because of the horrible conditions they face. I also worry about the meat-eater problem, so that helping humans may harm nonhuman animals. Overall, I currently prioritize helping farmed chickens, and doing research on how to best help them and other nonhuman animals (including prioritizing between species).

Hi! I agree with basically everything written here, in particular about their lives probably not being worth living. My sense is that this depends less on differences in intensity of experiences across species, which makes it a useful starting point for my thinking. I admittedly know less about on-the-ground conditions than activists in this area, but if their lives are void of good experiences, and include at least some subjectively bad ones, its hard to come up with a rationale for how they could have worthwhile lives. 

So, conditional on focusing on near-term problems, I think there is a very good case for prioritizing factory farming (and many EAs do!). I'm less certain about the longtermist point you make. If factory farming phases out eventually without EA effort (which seems likely to me), then your efforts aren't counterfactually ending an indefinite future of factory farming, just speeding this transition up. Preventing extinction or totalitarian lock-in really would create a counterfactual stream of goodness that's (approximately) indefinite. Though this also assumes the future is likely to be a stream of goodness, rather than badness; here's a related discussion on this point you might find useful.

We can't measure suffering of course across species. (Really, we can barely measure it among humans.) So we have to rely on extrapolation from our own experience, which in a way amounts to extrapolating from one datapoint. My intuition says that non-humans animals don't have a full consciousness by humans standards, and that their moral value is correspondingly less. I feel relatively confident in that judgement. But given scale of factory farming, how neglected the issue is among the general public, and that it intuitively it feels like at least chickens and mammals have some consciousness I still focus my support for EA causes on animal welfare ones. 

I think that if you simply make a point estimate of the scale of human and animal suffering you could end up with them being comparable, (see eg here https://forum.effectivealtruism.org/posts/TEu5SroJYRezrMuf7/how-much-physical-suffering-is-there-part-ii-animals) but you could also easily end up with human suffering or animal suffering being much larger. One major reason the EA movement works on both causes is I think a form of risk-aversion, not because it is what an expected value calculation suggests.