Regarding the concern about whether it’s useful to think about how to end factory farming, my intuition is that having an endgame in mind will do much to help guide us there. Even if the endgame is just more humane animal farms, I think making that more explicit will help us shape strategies today.
The project of improving farmed animal welfare is a decades-long project, and it seems highly suboptimal to not plan what outcomes we’d like to be achieving decades on down the road.
I’m sympathetic to a lot of what you say in this, including the fact that welfare reforms can and are an important part of the road to ending factory farming. It just unlikely that they will be all of that road (or even most of it).
Regarding the concern about whether we ought to even seek to end factory farming (or animal farming broadly), my views on this have been updated towards the affirmative to this based on Jeff Sebo’s arguments (EAG talk and paper). Essentially, he argues along the moral circle expansion angle: If we’re the sort of people who tolerate the human-caused suffering of factory farms, even if factory/animal farms are someday not so bad, then we’re more likely to accept other forms of exploitation that will lead to significant suffering (e.g. the exploitation of digital minds).
Last point on this: Even if the animals impacted by these reforms suffer only have as much, that’s still thousands of hours of equivalent annoyance level suffering per animals (per your spreadsheet). Though this takes nothing away from the good done by these reforms, to me this still qualifies as pretty horrible factory farming.
We should be happy by the progress we have made, but there is still a long road ahead.
All in all I’m now thinking that switching from battery cage to cage free averts ~60% of suffering (per your figures), and switching from conventional to BCC-approved broilers averts 30-40% (your figures plus a downward estimate for the breeds growth rate concern I mention in the other comment).
Thanks for this Saulius. This is a slightly positive update for me that both cage free and broiler reforms are more impactful than I thought.
One concern I have with the Welfare Footprint study (caveat: I have no experience in animal welfare science or with the BCC). The Welfare Footprint study people say (bolding added by me):
We analyzed the following scenarios, for which data on broiler welfare was available: (1) a baseline scenario represented by the use of conventional fast-growing breeds (e.g., Aviagen Ross 308, 708, Cobb 500) reaching a slaughter weight of 2.5 Kg at 42 days and (2) a reformed scenario, represented by the use of a slower-growing strain (ADG: 45-46 g/day), reaching the same slaughter weight in 56 days. This is a growth rate consistent with typical figures achieved by various of the breeds approved under the BCC, also referred to as medium- or intermediate-growing broilers, also falling within the acceptability of other welfare certification schemes.
My concern is with the claim that reaching 2.5kg in 56 days necessarily is the growth rate experienced by BCC chickens, or whether the final BCC birds are actually faster growing.
Given the watering down concerns Farm Forward discussed, and that I wasn’t sure how up to date WF’s study was, I looked into the different breeds accepted by the BCC (you can see the latest breeds GAP approved here) and how long they take reach WF’s cited slaughter weight of 2.5kg.
Of the 11 breeds GAP approves, 2 of them (Aviagen Ranger Classic and Aviagen Ranger) reach 2.5kg below significantly below the 56 days benchmark WF uses (they reach 2.5kg at 50 and 51 days, respectively). 5 to 6 days may not seem like much, but remember we’re only talking about a difference between 14 days (difference between reaching 2.5kg at 42 vs 56 days) that accounts for half the suffering these animals experience. The other breeds were either around 56 days to reach 2.5kg, or in the case of the Hubbard Redbro significantly above it.
Worst case scenario then, if you just use average weight gain as a simple welfare proxy then I estimate these BCC approved breeds to be half as less bad off as Saulius estimates.
I find this concerning because 1) companies will likely congregate to the fastest growing breeds still available, and 2) it possibly illustrates the watering down concern.
Nope, I think that is mostly (though not 100%) correct. My impression is that OpenPhil in particular is both more opportunity- and operationally-constrained than it is by funding. I do think though that they (and other funders) ought to do more active grant-making to try to identify non-CWR opportunities to fund (though they could very well already be doing this).
I also agree with your point that few if any other approaches could absorb significant amounts of money currently (though I also expect that there’s many orgs you could talk with trying more novel approaches who would disagree with us here, so perhaps I’m just not sufficiently aware of them).
My point is more that many of the EA funders seem to have found a local optimum with CWRs, and if we put more efforts into exploring we would find other approaches that also look very promising. What I’d like to see is more work from EAA and funders to incubate and help build new approaches. I realize that that can be a difficult role for these organizations to play though.