Reducing animal suffering is one of the main focus areas of the Effective Altruism movement. Currently, the dominant approach to reducing animal suffering among EAs seems to be veg*n (vegetarian/vegan) advocacy. But there's a big problem with veg*n advocacy: most people who switch to veg*n diets later switch back. Animal Charity Evaluators writes:
Overall, the evidence suggests that the vast majority of people who adopt dietary changes like vegetarianism, veganism, and other forms of animal-product-limiting eventually return at least some distance towards their former dietary habits. An intervention that changes an individual’s diet usually has only a temporary effect.
Based on the data, ACE estimates that the average new veg*n lasts about 7 years, and they consider this an important factor affecting the cost-effectiveness of interventions to promote veg*nism. But I'd submit that veg*n recidivism is actually even more important than ACE realizes.
EDIT: I highly recommend this comment to read along with the rest of this post.
Shifting the net growth rate of the veg*n movement
If a quantity is growing exponentially, the rate of growth is much more important in the long run than the quantity's current value. A veg*n movement that constitutes 1% of the population and reliably doubles in size every 2 years will quickly surpass a veg*n movement that constitutes 10% of the population and reliably doubles in size every 20 years.
Exponential growth seems like a good way to think about the growth of the veg*n movement because the growth rate of the veg*n movement is related in an important way to the movement's current size. A larger movement means more word of mouth and more resources to plow back in to movement growth, creating a positive feedback loop. And indeed, Google searches for the term "vegan" seem to be on a rough exponential trend, having doubled in the past 4-5 years.
Exponential growth gets harder and harder to sustain the more you grow, and I think it'd be wishful thinking to say that the current exponential trend will continue forever. The animal rights movement has been around for decades and the vast majority of the population is still omnivorous. But the exponential model is still a powerful way to think about veg*n movement growth.
Based on the exponential model, it makes sense to pay less attention to the current size of the veg*n movement and more attention to the net rate of growth in the veg*n movement. And there are two factors affecting that rate of growth: the rate at which people come in, and the rate at which people leave.
The rate at which people come in roughly corresponds to the quality of veg*n messaging. Throwing your resources behind a message that converts at a higher rate means a higher rate of growth regardless of how many resources you have. Fortunately for veg*ns, organizations like The Humane League systematically test different messaging strategies to see which work best. (But note that more aggressive veg*n marketing strategies may backfire in the long run if they turn some people off of veg*nism and make it so exponential growth tops off sooner.)
On the other hand, many fewer people seem to be working on the other factor that effects the net growth rate of the veg*n movement: the rate at which people leave. And there are independent reasons to believe this number is more important. Sam Altman, president of Y Combinator, writes on Twitter:
a much better metric than growth rate for early-stage startup success is how much users love the product. but it's much harder to measure.
they eventually converge, but it often takes years.
in the mean time, lazy investors use growth rate as a proxy for product love, but often get tricked by growth hackers.
the best investors dig in and figure out the answer to the first-order question.
To someone like Sam, the current veg*n movement might look like a startup that has a bunch of talented growth hackers but an uninspired product. Most users abandon it. In fact, it's even true that the majority of self-described "vegetarians" eat meat.
Why is veg*n recidivism so high, and what can be done to decrease it?
Why do most vegetarians go back to eating meat? According to data collected by Hal Herzog the major reasons are, in order of frequency, declining health, hassles and social stigma, irresistible urges and shifts in moral thinking. Interestingly only 2/77 reported shifts in moral thinking made them go back to eating meat. This jives with what I’ve read about former vegans. From the small sample of ex-vegans interviews I’ve read, most seem to experience cravings or health problems and then rationalize their choices with changes in moral ideology.
From here. Animal Charity Evaluators writes:
It was also found that half of all veg*ns replaced meat with more vegetables, where legumes and grains would be a better choice nutritionally.
So it seems that the veg*n movement is not doing a very good job of helping veg*ns get all the nutrients they need in the absence of animal products.
To delve in to this question further, it might be interesting to try to understand whether veg*n recidivism is higher in areas that have more veg*n food options (likely confounded with the number of veg*ns living in an area and providing social support).
How might the veg*n movement work to reduce recidivism? These blog posts on the ethics of eating oysters and mussels seems like a good step. Educating veg*ns on how to get nutrients like creatine and choline that are mostly found in animal products could be valuable. There might also be value in copying elements of traditional, and therefore proven, vegetarian and pescetarian diets.
Why has the veg*n movement neglected recidivism thus far? Note that many veg*ns insist that maintaining a veg*n diet doesn't require much ongoing effort. I think the evidence shows pretty definitively that there's a large segment of the population that finds it difficult. So perhaps veg*ns are self-selected to find veg*nism easy.
Veg*n recidivists leave the veg*n movement, leading to them being underrepresented in the movement's decisionmaking. The leadership of the veg*n movement is thus made up entirely of non-recidivists.
Conclusion
As you might have guessed, veg*nism advocacy isn't my current guess for the best cause and I don't restrict my diet. (I won't get in to details on this because I'd prefer to keep the discussion focused.) But I know I appreciate people providing informed, thoughtful outside criticism of the causes I support, and I guessed many veg*n Effective Altruists feel the same way, so that's what I tried to do for them. Even though I don't think veg*n advocacy is the best cause, being veg*n seems admirable from my perspective and I thank veg*ns for their commitment to sentient life.
I think a major part of why people lapse completely is that there exists a widespread belief that veg*nism is a hard and fast label, and falling off the wagon is irreversible. Many ex-vegans I know now eat completely omni diets that are simply at odds with their moral beliefs. I think that promoting the idea that 80/20 veganism is okay will result in a lot less animal product consumption overall. This article explains the way this black-and-white thinking negatively affects peoples' beliefs around their diet and personal identity.
Religious communities have a lot to teach us here. Picture the following recruitment strategy:
Eventually, this kind of community could function like a local atheist spiritual community of some kind, and be a platform for spreading other effective altruism ideas.
In most parts of the world, veganism is not the norm, which I believe is a big part of why people find it so challenging long-term. Focusing on universities strengthens the perception that veganism is merely part of a normal period of experimentation. Once there are people who live in communities where more than 50% of people are veg*n, I suspect it will be much easier to both retain those people and convert new people.
i agree with the 80/20 idea, and i wonder we apparently 98% of our movement keeps stuck in black and white thinking. for instance in this case, we are apparently thinking that the question to what level of consumption the vegans are sliding back is irrelevant, but it's not, of course. for one thing, moderate use still saves a lot of animals, for another, every reducer helps make full time veganism more easy, by increasing demand (and thus the supply), the social acceptability, the convenience with which to eat vegan etc... a society where 60% of people are 75% vegan is probably a lot better and closer to a vegan society than one where 10% of the people are full time vegan...
also, something jonathan saffran foer told me: you know the example of the guy that eats meat again because once he was at an airport and it was the only thing he could it? he fell back entirely. i think that has to do with the fact that we see veganism too much as an identity, and as something binary. when we fail, we might as well give up
i know the second point is perhaps in contradiction with the first, but maybe they both have some merit somewhere...
If 'veg*n' isn't a word that everyone uses yet, I think there should be a different one. It's hard to not read it as a misspelling of 'vegan.'
It seems kind of weird that things recede after seven years. Like, if it were a willpower thing I'd expect it to be shorter. I'm not just going to decide after a few years that my eating habits are too hard to keep up.
Did you see anything about people saying why they started eating animal stuff again?
"Like, if it were a willpower thing I'd expect it to be shorter. I'm not just going to decide after a few years that my eating habits are too hard to keep up."
This seems reasonable, but the reasons given were health issues followed by hassles, and only then uncontrollable cravings. Health issues would probably take a while to develop and then a while more to respond to by switching back.
Re. hassles, I can imagine someone switching to a vegan diet when it is relatively more convenient and then switching back when it stops being so because your environment changes, e.g you switch when you're near stores that stock a good selection of vegan products, and then stop when you move to a different area.
I use a harm-minimisation strategy when I'm in food deserts or traveling, but I suspect I'm very rare. For many vegans I know, not eating any animal products at all is part of their identity, so being forced to consume them because of a lack of availability would be incredibly emotionally stressful for them and create a lot of cognitive dissonance.
Seven years is an average; the distribution seems pretty skewed.
Yes, see the text right under the "Why is veg*n recidivism so high" heading.
I suspect that veganism is also done at the same time as organic/local/raw and other things that drain willpower and might be unhealthy and expensive and more socially awkward. You would need to tease out the different types of veganism to see the respective growth and recidivism rates.
An average of seven years is consistent with the hypothesis that the problem is vitamin B12 deficiency. Our bodies store enough B12 that it takes anywhere from months to decades for symptoms of a severely deficient diet to become clear.
This is a pretty poor explanation for people stopping eating vegan. You could apply the same argument if the average time someone remained vegan were anywhere from 2 months to an entire lifetime.
Interesting point. Should there be a greater focus on advertising and promoting how to eat a healthy vegan diet? While I'm not sure how effective this would be, I'd like to see more people pressuring government health departments to advertise the health benefits of being vegan, as my brief reading on the topic indicates that, for the government at least, it would save them money in the long run as the public health burden would be lower.
I think that would be a good idea. Would it make sense to have Facebook ads and Google Video ads to promote NutritionFacts.org? We might want to link to a page that distills the information on the videos into a single video or webpage, as the site has a lot of videos!
Another idea could be to translate the videos or make transcriptions and translate those.
Agree with that. Far too many vegans ignore B12.
The difference between a dominant approach and a measurable approach could explain why ACE recommends vegan advocacy groups, since the impact of investigations that groups like DxE are harder to measure, even if they're more effective. There's also a lack of charities directly trying to reduce animal suffering in general.
Two ideas:
(1) Direct Action Everywhere (a group of which I am a part, full disclosure) promotes community building as a way to address this. It seems like social support can address many of the reasons for recidivism both by providing moral support and providing a group of friends who can help address issues you face. On the other hand, social ostracism for your diet seems likely to be a challenge to remaining veg*n. I'd like to see diet on this, though I suspect self-reports wouldn't be that trustworthy on this subject.
What do you think of this as (part of) a solution?
(2) I believe the Humane Research Council's study found that people who were vegan for animal-related reasons were less susceptible to recidivism. Perhaps keeping the focus on an animal-centered message would help? Again, I'm not an unbiased observer since this is the approach I take. Curious for others' thoughts.
I don't think I have special insight in to either of those questions. My purpose was to point out the likely existence of low-hanging fruit in spreading info on how to eat a healthy vegan diet. (BTW, I didn't mention this in the post, but I suspect the experience of "meat cravings" may be your body telling you that you're not getting enough of particular vital nutrients.)
I'm not sure there is enough data here about veganism as opposed to vegetarianism to draw many conclusions. It would also be interesting to see how many vegans for ethical reasons have reverted back and compare the two growth rates of ethical vs dietary vegans.
It may be better to look at the amount of meat being bought year to year per capita to see overall trends. 50 million people eating vegetarian for one day a week will have a much larger impact then 100,000 more vegans.
Wouldn't veg*n recidivism specifically be more important for evaluating veg*n outreach programs like leafleting?
Maybe, or we could compare areas with tv/radio/internet advertising/ leafletting and the resulting change in non vegan products bought in that area, which could take in a much larger data sample.
Both would be important to see what works, and maybe reaching out to companies that sell vegan products to see if they have market data could help.
Thanks for the article. I'm surprised you didn't mention the Faunalytics (HRC) study that underpins the ACE analysis. Our analysis touches on many of the points you mention above. For you and your readers, here is a link to the full study: https://faunalytics.org/feature-article/study-of-current-and-former-vegetarians-and-vegans/
Additionally, we have more analysis coming from the results including length of adherence to veg*n diets as well as qualitative findings on motivations and barriers.
Thanks very much for commenting. I'm having a bit of trouble accessing the attachments on the page you link to, e.g. my browser tells me this page has a redirect loop.
If you want to highlight important things from the study that I got wrong or left out, I'd be happy to link to your comment in my post.
The attachment links for the Faunalytics study are now fixed. Thanks again for point out the issue.
We cover a lot of ground in those reports, and much of it corroborates what you say in this post. I'm not sure where the assumption of exponential growth comes from, however. Our research has shown that the proportion of vegans/vegetarians has not increased substantially in the past decade. There may be exponential growth with Google searches of the term "vegan," but that doesn't mean the vegan population is growing.
Rather, the reduction in farmed animal slaughter and the rise of vegan food alternatives is being driven by "part-time" abstainers. That's where we see the exponential growth. With that said, I certainly agree the growth rate is more important than the absolute number of vegans. The challenge is that research on low-incidence populations like vegans/vegetarians is hard, especially when tracking changes over time.
In addition to helping people maintain a nutritious and plentiful diet while being veg*n, animal advocate also need to work on making veganism less of a purist identity and to make it more inclusive. Following are the recommendations copied from our report:
Improve Vegetarian/Vegan Retention
Thank you! I linked to your comment near the top of my post.
So this data suggests the people who stick with veganism tend to be people motivated by multiple factors?
A quick defense of my exponential growth model: Changes in the size of the vegan movement are a result of a tension between tendencies for it to grow and tendencies for it to shrink. If you're correct that the size of the movement has been basically flat, that suggests those tendencies are currently canceling each other out. But that doesn't mean this way of thinking is totally useless. (It could be significantly misleading if there's a subset of the population that's naturally much more inclined to veganism though.)
It's a useful contribution, thanks, we need more like it! (Though I'm more in your camp. In fact I think recidivism is a big concern across EA as a whole. I've struggled with it myself as a veg*n of varying varieties. However - I'd guess like you - I prefer to avoid public picking apart of my personal choices. When I tried optimising my allocation of willpower recently, veg*nism turned out on reflection to consume a lot of this, though my impression is that I find it unusually hard and unpleasant. :S )