On the other hand, meat offsets are a direct cost to the meat eater. They are increasing the cost of meat to that consumer.
I'd be surprised if offset costs[1] not incurred at the point of sale meaningfully impacted consumption. People respond far less to fees when they're less salient at the point of sale.[2]
...of a similar magnitude to the ones suggested by Farmkind, for example. If offset costs were, conversely, large enough to require someone to mortgage their home in order to eat a slice of pecorino, you'd see an effect
See also: Chetty, Looney & Kroft (2009, AER): Including taxes in the price shown in stores decreased consumption (n.b. for non-Americans, in the US, store sales prices do not general display the price inclusive of taxes), even though people were paying the same price!
Finkelstein (2009, QJE): When toll roads were converted from manual (i.e. you have to physically fork over money to get through a toll) to automatic (i.e. you're just charged after the fact by your license plate/a radio transponder), people increased their use of toll roads
Yeah, that's an interesting empirical question: are offsets too "cognitively distant" from the purchase to make a difference.
That said, in your Finkelstein example, We're basically talking about making an additional cost less visible, whereas in the offsetting situation, the cost is pretty visible to them, as they have to calculate it.
Is your model of how people would be offsetting one in which they are calculating it prior to picking up the product from the shelf (or at least prior to buying it at the store)? That's not really how the few people who do animal product offsets appear to be doing it: from what I've seen they generally assess their level of consumption and then purchase set-it-and-forget-it recurring offsets, to be reevaluated at most yearly. Crucially they're seeing how much they consume and then purchasing offsets.
“ However, as FarmKind’s theory of change is offsets” — it isn’t though. It’s counterfactual donations to animal charities, and the vast majority of donations they raise are not made through their offsetting tool. It’s one way to donate (and not the main one), not their theory of change
It’s probably also worth clarifying that I’m not aware of anyone that offers a way of offset consumption by causing less consumption. I am aware of ways to try offset suffering from consumption by causing less suffering from consumption :)
Carbon offsetting doesn't work even if it is a legit offset. Carbon offsetting is a scientifically flawed concept. We extract carbon in the form of fossil fuels that took millions of years to get deep in the earth, and offset it with carbon captured on the surface of the earth which will still take millions more years to return to form of the carbon we 'offsetted'. (see source for this: https://www.nature.com/articles/nclimate1804). I think that meat offsets are similarly flawed. At the end of the day you still harmed someone.
The reasons you provided don't strike me as "scientifically" flawed. If the goal is to draw down equivalent carbon to what you released, then it doesn't matter that it's a different tree or how long it was captured. It's not the form of carbon we're interested in. It's literally just removal of carbon from the atmosphere.
I do take your point, though, that we might broaden our perspective on carbon offsetting and consider more than just the tonnes of carbon taken out of the atmosphere: the replacement of ancient forest with new forest is a different but real problem.
Back to the post, though. I don't see how any of this really relates to meat offsetting? aFrom your brief reply, it sounds like you're opposed to the idea that "I can inflict harm here and reduce harm there, and call it even." have I understood you correctly?
Hi Richie, my understanding from the Nature article is that there's a limited amount of carbon we can store on the earth's surface. If we continue to extract it from under the earth's surface it at a faster rate than it can be sequestered back under the earth's surface then we can't really plan to offset it all. I'm not an expert in the field so not sure if I understand it all correctly, but the article in Nature is pretty convincing.
Yeah, regarding the meat offsets, I don't agree that it is moral to do harm and then 'offset' it by doing something good. If this were the case - then should we have human slavery? One human slave could make a whole family's life better by doing their cooking, cleaning etc. and looking after them. Should we kill healthy 18-year-old humans to harvest their organs? Organ donations from one human body can improve the lives of up to 70 other humans. If we aim to do 'net good' rather than 'good', we might even say it's immoral not to do these things.
I think it's totally fair for people to make that choice morally: to decide that its not OK to participate in the harm of one being and then reduce harm to other beings.
I would point out though that your analologies aren't exactly appropriate: you're talking about a special case where utilitarian logic tends to fall down which is "big harm to one person offset by small good to lots of people".
But in the case of, say, donating to a campaign that reduces meat consumption and then eating meat, you're saying "condemning one animal to immense suffering, and in return sparing 1 animal immense suffering". It's not comparable to "owning slave, make life slightly better for many people" and more like "own a slave and buy some other slave's freedom" which most people would say is much less obviously wrong.
In the case of killing a healthy 18-year-old to donate their organs - this isn't big harm to one person for small good to many. The person who dies could suffer the least, they lived a healthy life and are killed instantly. The people they save were likely very sick for a long time and would have had a slow and painful death. It's a relatively smaller harm to one person to prevent a worse harm for many.
I wasn't aware there are people who think it is ethical to own a slave so long as you freed another slave....
EDIT: removed claims that FarmKind's main ToC is offsets. This is not true (see Aidans comment below).
As far as I'm aware, most people feel carbon offsetting largely hasn't worked. Does that mean offsetting meat consumption is a dead-end?
I argue no. If we look at the 2 main reasons why carbon offsetting has failed to reach it's potential, we can see that meat offsetting does not have these problems. As a result, meat offsetting may be an effective strategy to reduce animal suffering.
Caveat before we begin
Here I am discussing a very narrow question, which is whether the failure of carbon offsets tells us that meat offsets will not work. There are wider debates to be had around meat offsetting, such as whether it is morally permissible. I’m not commenting on that here, and I’ve seen good arguments on both sides.
This should also not be taken as an endorsement of FarmKind’s “Forget Veganuary” campaign they ran earlier this year. Evaluating that campaign involves a lot more than just offsetting, and I have nuanced thoughts about it that I won’t go into here. However, as FarmKind does have a tool that allows people to offset their consumption, this post can be taken as a defence of the idea behind that.
1. Most carbon offsetting companies are basically scams
This scam risk does not apply to meat offsetting, at least in the case of FarmKind. they send offsetting donations go to 3rd party evaluated organisations that I am very confident do reduce demand for meat or reduce animal suffering or both. See them here. Similarly, FarmKind themselves are not conducting the evaluations. They are relying on independent evaluators (e.g. Animal Charity Evaluators) who have a good track record of estimating impact as accurately as feasible.
So long as you donate to effective animal charities to offset meat consumption, there is no scam.
2. Carbon offsetting reduces the incentive to lower one’s carbon footprint and may actually drive people to increase their emissions (aka “moral licensing”)
First, the evidence for moral licensing is weaker than it sounds. A 2025 review of 115 experiments found it only showed up when people were being watched. When nobody was watching, the evidence pointed to no effect at all. Moral licensing is by no means a “psychological fact” that many people seem to believe.
Second, many offsetting programmes don’t increase the cost to consumers, which I think is crucial for them to work. It's intuitive to me that for offsetting to change behaviour, consumer must actually have to pay their own money for offsetting. For example, some carbon offsetting programmes from airline companies did not charge consumers the extra money to offset their emissions, or didn’t charge the full cost. Instead, the airline bore some/all of that cost. It’s obvious why that wouldn’t get consumers to reduce emissions, because in a very real way you haven’t imposed a cost on them.
Meat offsets probably do reduce meat consumption
On the other hand, meat offsets are a direct cost to the meat eater. They are increasing the cost of meat to that consumer. And people eat less meat when it gets more expensive. Whilst I haven’t seen a direct study on meat offsets, I’d argue there is stronger indirect evidence that it would work. This means that meat offsets could have a double benefit: reducing consumption of meat at the individual level and at the systemic level through donations.
Summary
Carbon offsetting failed because most carbon offsetting companies were scams. Orgs that meat offsets go to are checked by independent evaluators, so they are very unlikely to be scams.
The evidence that carbon offsets or any offsets cause people to increase consumption is weak; however meat offsets increase the price of meat to consumers, which we know reduces meat consumption. We should expect that meat offsets will reduce meat consumption.
Note: This post was crossposted from Planned Obsolescence by the Forum team, with the author's permission. The author may not see or respond to comments on this post.
Subtitle: If acceleration compounds, the naive extrapolation implies superintelligence within a year
This post discusses a public METR report on AI R&D acceler...
TL;DR:
* When we started Rethink Priorities (RP) in 2018, our mission was, simply put, to close the knowledge gaps that stopped people from improving the world as much as they could. That hasn’t changed much, but the landscape we’re doing it in is shifting.
* In 2027, we plan to grow our core GHD, Animal Welfare, and Worldview Investigations teams to take on emerging priorities across causes, including those that the forthcoming transformative AI is creating.
...
This post is part of the Rethink Priorities Cross-Cause Fund (CCF) series on how to allocate funding across causes when it is uncertain what matters most.
...
I'd be surprised if offset costs[1] not incurred at the point of sale meaningfully impacted consumption. People respond far less to fees when they're less salient at the point of sale.[2]
...of a similar magnitude to the ones suggested by Farmkind, for example. If offset costs were, conversely, large enough to require someone to mortgage their home in order to eat a slice of pecorino, you'd see an effect
See also:
Chetty, Looney & Kroft (2009, AER): Including taxes in the price shown in stores decreased consumption (n.b. for non-Americans, in the US, store sales prices do not general display the price inclusive of taxes), even though people were paying the same price!
Finkelstein (2009, QJE): When toll roads were converted from manual (i.e. you have to physically fork over money to get through a toll) to automatic (i.e. you're just charged after the fact by your license plate/a radio transponder), people increased their use of toll roads
Yeah, that's an interesting empirical question: are offsets too "cognitively distant" from the purchase to make a difference.
That said, in your Finkelstein example, We're basically talking about making an additional cost less visible, whereas in the offsetting situation, the cost is pretty visible to them, as they have to calculate it.
Is your model of how people would be offsetting one in which they are calculating it prior to picking up the product from the shelf (or at least prior to buying it at the store)? That's not really how the few people who do animal product offsets appear to be doing it: from what I've seen they generally assess their level of consumption and then purchase set-it-and-forget-it recurring offsets, to be reevaluated at most yearly. Crucially they're seeing how much they consume and then purchasing offsets.