Owen Cotton-Barratt of the Global Priorities Project wrote an article on comparing human and animal interventions. His major conclusions include:

1. Indirect long-term effects dominate considerations.
2. Changing behavior of far-future humans matters more than alleviating immediate animal suffering.
3. Helping humans has better flow-through effects than helping non-human animals.

The analysis effectively concludes that helping humans is more important than helping non-human animals but I believe it misses a few important considerations.

(These are fairly quick thoughts about which I have a lot of uncertainty; I’m publishing them here for the sake of making the conversation public.)

What flow-through effects occur?

This article does not address what flow-through effects exist with global poverty interventions or how substantial we should expect them to be. (I looked through GPP’s materials to see if they have written on this elsewhere but I couldn’t find anything.) This is a common problem–a lot of people claim that flow-through effects are an important part of the benefits of global poverty interventions, but say little about what effects actually occur.

People often bring up economic growth as a positive flow-through effect of global poverty interventions. While it’s true that effective interventions probably promote economic growth, the estimates of how much growth they produce are non-robust. GiveWell’s analyses of its top charities barely discuss economic flow-through effects, which suggests that they are either small, too difficult to detect, or both. (Right now GiveDirectly is conducting an RCT to test the macroeconomic effects of its programs; at the time of this writing, GiveWell believes that the effects are non-negative, but does not have strong evidence. (I’m not associated with GiveWell; I’m inferring this from reading GiveWell’s public writings.))

It’s not even clear that flow-through effects are net positive. It’s conceivable that most global poverty interventions do indirect harm by suppressing local economic production. There is not strong evidence that this does happen, but it’s possible, and most people are too quick to assume that flow-through effects are net beneficial. Beyond that, it’s possible that economic growth actually hurts the long-term future, perhaps by increasing existential risk or improving humans’ ability to create lots of suffering (e.g. by increasing the scale of factory farming).

How strong are flow-through effects?

It seems reasonably likely that the best anti-factory farming interventions are 10 times better than the best GiveWell top charity, and possibly more like 100 times better (see e.g. ACE’s estimates for online ads and leafleting). Even if GiveWell top charities have positive flow-through effects, it’s implausible that these effects could be 10 times larger than the direct impact. Perhaps one could argue that these charities have a huge positive influence on the far future, but the mechanism by which this could occur is unclear; and interventions to change humans’ behavior toward non-human animals could have similarly large impacts on the far future. In fact, I expect that spreading concern for animals matters a lot more for the far future than preventing malaria or intestinal worms.

Conclusion

I’m not entirely sure that this contradicts anything in the original article. Owen writes

A push towards vegetarianism is one sort of value-shifting intervention. It is possible that this or another such intervention could be more effective than direct improvements to human welfare, but in order to assess this we need to to model how changing societal values today will influence the behaviour of future generations. This should be a target for further research.

It does seem true that changing human behavior matters a lot more than changing the behavior of non-human animals, since humans have a lot more power. (For that matter, some humans’ behavior matters a lot more than others’. Getting extremely wealthy people to donate money has a much greater effect than getting ordinary people to donate.) But I don’t think we can say with any confidence something like “global poverty interventions have more beneficial long-term effects than animal welfare interventions, therefore we should prioritize global poverty.”