PSA for vegan donors: GiveWell not ruling out animal-based aid

I don't like that its not off the table for GiveWell, but thankfully I struggle to even imagine how an animal-based human aid program could reach their cost-effectiveness bar. 

A milder ask would be for a commitment to publicize if they ever grant an animal-aid program and to allow retroactive redesignation during the relevant time period from All Grants or Unrestricted to Top Charities or some other restriction that excludes animal-aid programs.

Given that much of All Grants ends up going to top charities anyway, I don't think there's much concern about the donation being in AG or Unrestricted causing a counterfactual increase in the amount that would have been spent on an animal-aid program. 

In my view, if GiveWell starts funding a lot of livelihoods work, it should offer additional designations to allow donors to choose between lifesaving or health work versus livelihoods work, rather than having to defer to GiveWell's moral weights as between the two. So that would also likely solve the problem for donors who object to animal-aid programs anyway.

I would just make the observation that one doesn't have to buy into GiveWell's moral weights to make user of their research and recommendations. A donor can build their own moral weights, plug that into GiveWell's research, and come up with a top choice for the donation. GiveWell is in my experience pretty happy to walk larger donors through this process.

My hunch is that the animals involved in anything GiveWell would support would live vastly better lives than factory farmed animals in high-income countries. There are of course factory farms in low-income countries, but animals raised by smallholder farmers do not live in those conditions, from what I remember of a couple years in rural Kenya. 

Thanks for the post, Adam. I have eaten fully plant-based for 6 years, and have argued for GiveWell considering effects on farmed animals. However, I now think it would be better for GiveWell to keep focussing just on humans. I believe considering effects on farmed animals would tend to decrease agricultural-land-years via decreasing animal farming, and therefore decrease animal welfare due to increasing the suffering of soil animals much more than it could decrease the suffering of farmed animals (for my best guess that soil animals have negative lives). Moreover, saving human lives cheaply is the most cost-effective way of increasing animal welfare I am aware of, and GiveWell considering effects on animals would tend to decrease the funds going towards that due to alienating some of its funders.

This raises an interesting, and possibly unsolvable, point for organizations in GiveWell's shoes. 

GiveWell was created as a donor-advisory service, and to a significant degree still is (albeit most donors now entrust their money to GiveWell to disburse, so its recommendations can be executed in a timely and efficient manner). Almost by definition, GiveWell and its donors are aligned as to the primary purpose of the organization. But it's unsurprising that the donor base may have differing opinions on secondary issues like animal welfare. (By secondary, I mean that animal welfare is not the animating concern behind GiveWell's existence.)

If my memory serves, GiveWell's moral weights are mostly derived from the views of its donors, although staff views and beneficiary preferences also get some weight. I recall a sentiment that beneficiary preferences should get more weight than they do, but also a recognition that they are difficult to measure. In my estimation, there's no sound reason to defer to staff views on this issue, and I suspect beneficiary views would end up close to GiveWell's current position. As far as donor views, I speculate that they are somewhat bimodal ~ a number of donors would not really care, and a number would care a lot. So using some sort of amalgamation of donor views is likely to make very few donors happy

In a usual charity, the practical solution might be to have separate buckets for "all livelihoods work, including animal-aid" and "livelihoods excluding animal-aid" programs. But people in this community know about fungibility, and the fungibility problems on that setup would probably be significant.

Conditional on a universe in which GiveWell would recommend an animal-aid programs to livelihoods donors absent an animal-welfare adjustment, I think it may be impossible to avoid a significant problem for one subgroup of livelihoods donors.^[1] If it discloses the situation, then the donors who don't favor considering animal welfare (or apply only a minor downward adjustment) are going to preferentially fund the animal-aid program. This doesn't look much different than the two-livelihoods-bucket approach as far as funging effect. 

But suppressing the information deprives other donors of the ability to make the highest-impact choice by their own values. And the lack of transparency would be problematic for an organization whose value proposition is helping donors make effective choices with their own monies.

1. ^{^}

   For various reasons, I don't think fungibility between livelihood-focused buckets and lifesaving/health-promoting buckets at GiveWell is a major concern here.

Thank you for the post. I have a follow-up question which I hope you or Givewell can answer: Do Givewell also give cage systems to people in need, or teach them how to use cages, or both? I am asking because I have seen livelihood projects that do one of these. If my memory is not wrong, they include FAO, World Bank, the Dutch Government, and Heifer (yes, they don't just give cows, they give chickens too and teach them how to use cage systems).