going vegan is better than the first-order math suggests

TL;DR: People underestimate the value of vegan diets based on naive first-order calculations. Veganism is an unusually good signal against existing dietary norms, which can massively boost impact. In my (possibly unrepresentative) case, signaling effects have already easily 25x'd my lifetime impact.

People have argued that going vegan is an inefficient way of reducing farmed animal suffering (e.g. 1↗, 2↗). They point out that the lifetime individual consumption changes from one person going vegan reduce animal suffering (through reduction in demand) around the same amount as a $1000 donation to the most effective animal welfare charities. If they would incur more than $1000 in costs from switching / being sad / staring sadly at other people eating animal products, they should offset rather than boycott.

If we assume this number is correct and ignore some other classic arguments¹, I will argue there is another compelling reason to go vegan: Going vegan is an unusually valuable signal against bad dietary norms.

Going vegan is costly

Costly signaling is more↗ effective↗. If your friend posts on their Instagram story about the unending suffering of giraffes with necks too short to eat, you probably will not pay much attention. However, if your friend quits their lucrative dream job to volunteer planting shorter trees for short giraffes, you've gotten a bunch of evidence that they think this is very important. You will give the giraffe problem a second look and think about why your friend is sacrificing to work on it.

Going vegan is costly. People like eating animal products and not eating them limits your options significantly. Vegans are viewed more negatively than most out-groups↗ and going vegan will, in most social environments, damage your social status. Many omnivores are also uninterested in dating vegans↗.

As such, going vegan is a strong signal that you feel strongly about reducing animal suffering and animal rights² (and also a great signal that the person is willing to sacrifice their personal happiness for something they think is important).

Veganism is uncommon

Most people are not vegan↗. While people generally know about veganism, media representations tend to be negative↗ and it's easy to have wildly wrong/misjudged beliefs about the outgroup↗. For many of your friends, you will be their only vegan friend.

As a back-of-the-envelope calculation here, in a graph of 300 million people where each person knows 100 random other people, about 60% of people will know no members of a group of 1.5 million (estimates approximate veganism at ~0.5% of the population). As such, a randomly chosen vegan here would add another 60 people who now directly know a vegan. This is likely an underestimate (of the impact of a random vegan); since we know existing vegans cluster together in a group, "wasting" connections on each other and on the same non-vegans.

In many ways, knowing one vegan is way more than knowing zero. The ideas and beliefs are more tangible. The visualization of veganism might default to the (hopefully positive) image of the friend, rather than a vaguer conception with negative valence. As a somewhat analogous piece of evidence, we see discussing unrelated concepts with members of the opposite party↗ reduces short-term polarization.

Meat-eating is a normative practice

People eat meat for many reasons besides taste and health, including as signals of wealth or masculinity, or due to tradition. The presence of animal products is assumed based on the communities' default behavior (and changing defaults can have outsized impact↗).

Norms here can shift. Diets have and continue to change rapidly around the world↗, from social trends, changes in taste, and increased access to different foods. Additionally, preliminary research shows that signaling can be effective↗ in this domain.

Even in terms of retention, lack of social support is a common reason for exiting veganism and a barrier to re-entry↗. Just intuitively, making a leap to veganism is much less scary if you know someone who has already done it rather than being forced to break the norm yourself.

Moreover, conversion to veganism is only one goal here. Knowing an ethical vegan requires confrontation with the reality of farmed animal suffering and with the somewhat foreign idea of caring about animal welfare. Non-vegans that are sympathetic to animal welfare considerations will be more likely to switch to plant or lab-grown meat alternatives as they emerge. They will be more open to "weird" ideas for reducing animal suffering↗. They will be less hostile to other new vegans in the future or will create more positive media representations.

In addition, if value lock-in↗ is possible in the near future, shifting societal attitudes towards animal welfare could be significantly more valuable than shifting current dietary practices. Depending on your belief in lock-in, the outcomes of transformative AI futures, and in the value of future animals, this might be the most important part of veganism.

Veganism is a frequent signal

Left unprompted, I don't talk much about animal welfare. This doesn't matter much, because meals are embedded deeply into social interaction and me not eating animal products comes up extremely often. And even if I try to avoid the topic, people will (at least outside of the Bay) ask why and how and what I can and can't eat, so I end up talking about veganism more than I (selfishly) would like.

Personal Results

I have been vegan for ~6 years. In that time: my friends have donated ~$25,000 to animal welfare charities. I feel very confident they would not have gone vegan or donated to effective charities if I had not been vegan. Ignoring any other positive impacts (like future donations or dietary impacts), the total measurable signaling impact of my veganism so-far is around 25x what my lifetime personal dietary impact will be. The real impact will likely be much higher, since I expect them to continue donating to effective animal charities in the future, they also have positive spillover signaling effects, and I'm deliberately ignoring non-measurable impacts.³

Some details:

1. Four of my friends have gone vegan because of me⁴. Two of them also became EAs and do/will donate to animal charities (and I don't think they would have if I was not vegan).

2. I would describe most of my non-vegan friends as sympathetic to animal welfare concerns, whereas most were apathetic or unaware of the ideas when I first met them.

3. Essentially all of my acquaintances have tried vegan alternatives to animal products. My impression is this ended with my friend having a more positive view of the vegan alternative (except for vegan cheese and vegan shrimp, which are hated).

   Relatedly, the inconvenience of accommodating my dietary restrictions has led my non-vegan friends to eat vegan meals pretty frequently⁵.

4. Some people have told me they switched to buying "more ethical" animal products. I have not put any time into determining how valuable this is or how conscientious they were, so I do not value this at all, but it probably (?) doesn't increase animal suffering

Caveats That Seem Plausible

I'm already in a community which cares about animal welfare:

• This does make signaling less relevant to you. Still, most people have social connections beyond a single community. Also, being vegan within a vegan-friendly community is significantly easier, so consider cooperating on the collective action problem, rather than free-riding the signaling from other members.⁶

My numbers are wildly high and most vegans have much lower signaling impact

• This is probably true, but I think it's easy to underestimate signaling impacts because people don't communicate (or even know) how much their beliefs are impacted by their friends
• Even if most vegans had little signaling impact, the expected value seems likely to still be pretty high from the few vegans with lots of impact

Diminishing Returns to Signaling:

• Signaling effects will start having significant diminishing returns after large-scale adoption of veganism, but we are not close to that number.

Advocating for veganism is an ineffective strategy (e.g. Matt Ball here↗)

• I am narrowly trying to pin down the real impact of going vegan, not arguing that advocating for dietary changes should be our primary plan for improving animal welfare

This is still much less important than welfarist interventions which improve the lives of factory farmed animals:

• Welfare interventions are extremely important, shockingly neglected, and effective. Consider donating to charities like this one↗.⁷

We should consume unique noteworthy diets to be more memetic:

• It is useful to have a single (or a few diets) to coordinate around, rather than a hodge-podge of different ones.
• Sincerity is especially important here. I don't advocate polluting the intellectual commons by intentionally consuming (and arguing for) a diet that is not the one you think is best

Strict veganism is too difficult. We would do better by coordinating around an easier diet to maintain and signaling that it's okay to be imperfect

• I think people interested in this should check out Ozy Brennan's article on Ameliatarianism↗ and BB's on eating beef and dairy↗ rather than trying something more standard like vegetarianism
• As always, making the diet easier trades off with costliness of the signal and value of switching. Personally, I've found it to easy to advocate for eating beef and milk over eggs and chicken while being vegan myself

Veganism could be negative signaling:

• If most vegans are chill and fun to be around, increasing the number that people meet should reduce negative polarization. This is mostly an argument to be more positive, unthreatening, and pleasant to interact with.
• Maybe just bringing up animal suffering is so uncomfortable for most people that existing as a vegan is negative signaling↗. This seems pretty cursed if true, but doesn't match my experience

But what about net effects on insects and wild animals/what about how farmed animals live net positive lives/etc.:

• I have assumed that factory farming is net negative for the purposes of this post, so please feel free to ignore the post if you believe veganism is net bad.
• In terms of wild animal/insect suffering, I imagine that people taking these causes seriously is usually downstream of them caring about animal welfare. As such, signaling from veganism may still be net positive in that regard, even if wild animal claims are right↗, but I do think people should take the actions they think are otherwise optimal

Some forms of meat-reduction can increase animal suffering

• Look here for an example of this claim↗. I agree that promoting diets that can offset consumed animal products with others (like increasing egg consumption through vegetarianism) could lead to net-negative outcomes, but I think specifically welfare-motivated veganism is more clearly motivated and not likely to cause this.

Other counterarguments:

• This is one list of ways in which reduced animal consumption could be more neutral↗