As OP continues researching moral weights, OP's marginal cost-effectiveness estimates for FAW and GHW may eventually differ by several orders of magnitude. If this happens, would OP substantially update their allocations between FAW and GHW?
We’re unsure conceptually whether we should be trying to equalize marginal returns between FAW and GHW or whether we should continue with our current approach of worldview diversification. If we end up feeling confident that we should be equalizing marginal returns and there are large differences (we’re uncertain about both pieces right now), I expect that we’d adjust our allocation strategy. But this wouldn’t happen immediately; we think it’s important to give program staff notice well in advance of any pending allocation changes.
Your comment seems to imply that OP's moral weights are 1-2 orders of magnitude lower than Rethink's.
I’m wary of sharing precise numbers now, because we’re highly uncertain about all three of the parameters I listed and I don’t want people to over-update on our views. But the 2 orders of magnitude are coming from a combination of the three parameters I listed and not just moral weights. We may share more information on our views and methodology later, but I can’t commit to a particular date or any specifics on what we’ll publish.
I unfortunately won’t have time to engage with further responses for now, but whenever we publish research relevant to these topics, we’ll be sure to cross-post it on the Forum!
We think these discussions are valuable, and I hope we’ll be able to contribute more of our own takes down the line. But we’re working on a lot of other research we hope to publish, and I can’t say with certainty when we’ll share more on this topic.
Several of the grants we’ve made to Rethink Priorities funded research related to moral weights; we’ve also conducted our own research on the topic. We may fund additional moral weights work next year, but we aren’t certain. In general, it's very hard to guarantee we'll fund a particular topic in a future year, since our funding always depends on which opportunities we find and how they compare to each other — and there's a lot we don't know about future opportunities.
I unfortunately won’t have time to engage with further responses for now, but whenever we publish research relevant to these topics, we’ll be sure to cross-post it on the Forum!
(Hi, I'm Emily, I lead GHW grantmaking at Open Phil.)
Thank you for writing this critique, and giving us the chance to read your draft and respond ahead of time. This type of feedback is very valuable for us, and I’m really glad you wrote it.
We agree that we haven’t shared much information about our thinking on this question. I’ll try to give some more context below, though I also want to be upfront that we have a lot more work to do in this area.
For the rest of this comment, I’ll use “FAW” to refer to farm animal welfare and “GHW” to refer to all the other (human-centered) work in our Global Health and Wellbeing portfolio.
To date, we haven’t focused on making direct comparisons between GHW and FAW. Instead, we’ve focused on trying to equalize marginal returns within each area and do something more like worldview diversification to determine allocations across GHW, FAW, and Open Philanthropy’s other grantmaking. In other words, each of GHW and FAW has its own rough “bar” that an opportunity must clear to be funded. While our frameworks allow for direct comparisons, we have not stress-tested consistency for that use case. We’re also unsure conceptually whether we should be trying to equalize marginal returns between FAW and GHW or whether we should continue with our current approach. We’re planning to think more about this question next year.
One reason why we are moving more slowly is that our current estimates of the gap between marginal animal and human funding opportunities is very different from the one in your post – within one order of magnitude, not three. And given the high uncertainty around our estimates here, we think one order of magnitude is well within the “margin of error” .
Comparing animal- and human-centered interventions involves many hard-to-estimate parameters. We think the most important ones are:
Moral weights
Welfare range (i.e. should we treat welfare as symmetrical around a neutral point, or negative experiences as being worse than positive experiences are good?)
The difference between the number of humans and chickens, respectively, affected by a marginal intervention in each area
There is not a lot of existing research on these three points. While we are excited to support work from places like Rethink Priorities, this is a very nascent field and we think there is still a lot to learn.
To ground your 1,000x claim, our understanding is it implies a tradeoff of 0.85 chicken years moving from pre-reform to post-reform farming conditions vs one year of human life.
A few more details on our estimate and where we differ:
We think that the marginal FAW funding opportunity is ~1/5th as cost-effective as the average from Saulius’ analysis.
The estimate Holden shared in “Worldview Diversification” (over 200 hens spared from confinement per dollar spent) came from a quick calculation based on the initial success of several US cage-free campaigns. That was in 2016. Seven years later, we’ve covered many of the strongest opportunities in this space, and we think that current marginal opportunities are considerably weaker.
Vasco’s analysis implies a much wider welfare range than the one we use.
We’re not confident in our current assumptions, but this is a complicated question, and there is more work we need to do ourselves to get to an answer we believe in enough to act on. We also need to think through consistency with our human-focused interventions.
We don’t use Rethink’s moral weights.
Our current moral weights, based in part on Luke Muehlhauser’s past work, are lower. We may update them in the future; if we do, we’ll consider work from many sources, including the arguments made in this post.
Thanks again for the critique; we wish more people would do this kind of thing!
Hi Ariel,
As OP continues researching moral weights, OP's marginal cost-effectiveness estimates for FAW and GHW may eventually differ by several orders of magnitude. If this happens, would OP substantially update their allocations between FAW and GHW?
We’re unsure conceptually whether we should be trying to equalize marginal returns between FAW and GHW or whether we should continue with our current approach of worldview diversification. If we end up feeling confident that we should be equalizing marginal returns and there are large differences (we’re uncertain about both pieces right now), I expect that we’d adjust our allocation strategy. But this wouldn’t happen immediately; we think it’s important to give program staff notice well in advance of any pending allocation changes.
Your comment seems to imply that OP's moral weights are 1-2 orders of magnitude lower than Rethink's.
I’m wary of sharing precise numbers now, because we’re highly uncertain about all three of the parameters I listed and I don’t want people to over-update on our views. But the 2 orders of magnitude are coming from a combination of the three parameters I listed and not just moral weights. We may share more information on our views and methodology later, but I can’t commit to a particular date or any specifics on what we’ll publish.
I unfortunately won’t have time to engage with further responses for now, but whenever we publish research relevant to these topics, we’ll be sure to cross-post it on the Forum!
We think these discussions are valuable, and I hope we’ll be able to contribute more of our own takes down the line. But we’re working on a lot of other research we hope to publish, and I can’t say with certainty when we’ll share more on this topic.
Thank you again for the critique!
Several of the grants we’ve made to Rethink Priorities funded research related to moral weights; we’ve also conducted our own research on the topic. We may fund additional moral weights work next year, but we aren’t certain. In general, it's very hard to guarantee we'll fund a particular topic in a future year, since our funding always depends on which opportunities we find and how they compare to each other — and there's a lot we don't know about future opportunities.
I unfortunately won’t have time to engage with further responses for now, but whenever we publish research relevant to these topics, we’ll be sure to cross-post it on the Forum!
(Hi, I'm Emily, I lead GHW grantmaking at Open Phil.)
Thank you for writing this critique, and giving us the chance to read your draft and respond ahead of time. This type of feedback is very valuable for us, and I’m really glad you wrote it.
We agree that we haven’t shared much information about our thinking on this question. I’ll try to give some more context below, though I also want to be upfront that we have a lot more work to do in this area.
For the rest of this comment, I’ll use “FAW” to refer to farm animal welfare and “GHW” to refer to all the other (human-centered) work in our Global Health and Wellbeing portfolio.
To date, we haven’t focused on making direct comparisons between GHW and FAW. Instead, we’ve focused on trying to equalize marginal returns within each area and do something more like worldview diversification to determine allocations across GHW, FAW, and Open Philanthropy’s other grantmaking. In other words, each of GHW and FAW has its own rough “bar” that an opportunity must clear to be funded. While our frameworks allow for direct comparisons, we have not stress-tested consistency for that use case. We’re also unsure conceptually whether we should be trying to equalize marginal returns between FAW and GHW or whether we should continue with our current approach. We’re planning to think more about this question next year.
One reason why we are moving more slowly is that our current estimates of the gap between marginal animal and human funding opportunities is very different from the one in your post – within one order of magnitude, not three. And given the high uncertainty around our estimates here, we think one order of magnitude is well within the “margin of error” .
Comparing animal- and human-centered interventions involves many hard-to-estimate parameters. We think the most important ones are:
There is not a lot of existing research on these three points. While we are excited to support work from places like Rethink Priorities, this is a very nascent field and we think there is still a lot to learn.
To ground your 1,000x claim, our understanding is it implies a tradeoff of 0.85 chicken years moving from pre-reform to post-reform farming conditions vs one year of human life.
A few more details on our estimate and where we differ:
Thanks again for the critique; we wish more people would do this kind of thing!
Best,
Emily