How I Learned to Love Shrimp

Seth Green on why reducing meat consumption is hard and what actually works

• James Özden

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 1:11:57

This episode, I spoke with Seth Ariel Green, a research scientist at the Humane and Sustainable Food Lab at Stanford university. He recently published a meta-analysis called “Meaningfully reducing consumption of meat and animal products is an unsolved problem” (EA Forum summary here) where he reviewed over 30 papers and hundreds of interventions on the topic. Seth also writes about the science of meat reduction on his Substack, called Regression to the Meat, which I highly recommend checking out for some accessible and fun to read writing about meat reduction.

We talk about why Seth is more sceptical than most about plant-based defaults, what actually works when it comes to changing people’s food choices, why some research in this space is misleading and new interventions to shape diets and food choices that he is excited about. 

Chapters:

(00:00:00) Cold intro

(00:00:53) Introduction to Seth and his work

(00:05:38) What are defaults and why is Seth sceptical

(00:19:55) The best paper on defaults - what does it mean for advocates?

(00:28:50) What does the research on meat reduction say?

(00:34:25) Is 5 percentage points a small or big change in meat consumption?

(00:43:20) What actually works in reducing meat consumption?

(00:50:18) Potential interventions that Seth is excited about

Resources:

With thanks to Tom Felbar (Ambedo Media) for amazing video and audio editing!

If you enjoy the show, please leave a rating and review us - it means a lot to us!

The Myth Of One-Weird-Trick

Seth

I think this whole like what I'm going to describe as this one weird trick will change your life and doctors hate it. This whole paradigm, no, I'm just I'm not for it.

James

Yeah, I I think, yeah, probably I agree. Yeah, it's easy to for people who want the world to change in a certain way, when they see these things promising big results for small changes, we want to believe them because it's like this could really help animals or really improve the world. But sadly, I think reality is often this is highly overstated.

Seth

And of course, for defaults in particular, they're promising because they're a way of bypassing a difficult conversation. My point of view is that, like, I'm sorry, that conversation is difficult. We still gotta have it.

James

Today, I spoke with Seth Ariel Green, who is a research scientist at the Humane and Sustainable Food Lab at Stanford University. He recently published a meta-analysis, which is functionally a review or a summary of many different academic papers on the topic, which was called Meaningfully Reducing Consumption of Meat and Animal Products is an Unsolved Problem, where he and his colleagues reviewed over 30 papers and over 100 different interventions on the topic of meat reduction. Seth also writes about the science of meat reduction on a Substack called Regression to the Meat, which is actually maybe my favorite Substack game of all time. And I highly recommend checking it out for some accessible and very fun to read writing about meat reduction. And today, Seth and I spoke about why he's a bit more pessimistic than most about plant-based defaults, what actually works when it comes to changing people's food choices and diets, why some research in this space can be quite misleading, and possible new interventions to shape diets and food choices that he's excited about researching. He's also sorry if his meta-analysis paper is a bit of a downer, but hopefully this conversation also shows some reasons to be helpful too. And I learned a lot in this episode, so I hope you do too. Hey Seth, welcome to the podcast. How are you doing? I'm good. Thanks, Dames. How are you doing? Good, thank you. Good. We like to start everyone off with the same question, which is what's something you changed your mind on recently related to the animal advocacy movement and why?

Seth

I'm gonna talk about something that I'm thinking about, although I'm not sure I've changed my mind yet. Yeah, basically, if you talk to some people in the movement, they're gonna say, don't worry about culture, don't worry about beliefs, worry about technology. Anything we can do to accelerate the growth of plant-based proteins and like true analogs is where we should be focusing our efforts. Sometimes I feel this way. And then recently I read an interview with Wayne Seung in The New Yorker about his own journey to a seminary school. And I thought he made some really interesting points about the role that religion plays in like providing infrastructure and broader meaning to a movement, and in particular thinking about our own movement in light of other successful movements and unsuccessful movements. I have been mulling over this question about what whether I should be worried about culture or not, broadly speaking. What do you think?

James

Well, it's a big question. So wait, so your reflection is more you know, I I guess this has come in many forms. Maybe the simplistic version was if we just change institutions or develop technologies that are cheap, tasty, healthy, plant-based foods, and everyone will just switch on masks, but then I guess it's a Wayne's point, and I guess your reflection of this stuff is necessary but not sufficient, and we need to do a bunch of work to take up some attitudes to help adoption.

Seth

Well, it's funny because that could be that could be one possible read of it, but another is that the it's just kind of a cop-out to be hoping for some technological thing. We should actually be like doing the hard work in the meantime. And do we think that adoption of this stuff, even if it does get Deus X Machinata into our lives, is going to be instantaneous? I don't think so. Like if we look at you know backlash in Texas and Florida and stuff, should we be thinking in advance about what's gonna how can we make this amenable to people? So there could be a like some intermediate answers here. But I think broadly lately, my answer is that I've been thinking more about culture and less about technology.

James

Interesting. Well, we'll dive into all that. And then maybe uh as some context for listeners, I guess the backlash you're referring to is the cultivated meat bands happening in I think now seven US states. And yeah, I guess this is very timely for me because I'm currently in the in the Bay Area in San Francisco, where there's technology optimism is you know widespread and running amok. And already there's been a few conversations that sounds something like we're just gonna have artificial intelligence explosion leading to cultivated meatings for me cheap, uh, dot dot dot, end of all animal suffering. And I I agree with you that there's a slightly simplistic uh narrative, and there's other things that have to be done, uh including what Wayne and you talk about in terms of.

Seth

That's actually not that simplistic, though. Like if like uh when uh Dwarkesh Patel interviewed Louis Bollard, his very first question is, okay, so what are we doing? If we're gonna invent perfect substitutes in 10 years, what are we doing in the meantime? And it's like, wow, man, there's a lot of assumptions built into that. Uh that I owe I don't even know where to begin evaluating them. Yeah, exactly.

What are defaults and why is Seth sceptical

James

Um but that's probably not gonna be a block of the discussion today. I guess today we're gonna be talking about your work at Sanford at the Humane and Sustainable Food Lab and all the interesting work you've been doing there, and maybe more broadly in terms of analyzing the literature and can creating new research on to what extent people actually change their diets towards more plant-based foods and less animal-based foods. And I think the first thing we I think we can talk about there is defaults. So maybe do you want to um, in a nutshell, explain what a default is and the research you've done on the space and what you guys have found? Yeah.

Seth

So first I want to start with something that you just said, which is we were talking about changing diets. I'm gonna come back to this exact phrase a lot. Something that uh Jennifer Chan at the Benner Food Foundation said we're distinguishing changing diets from changing a single meal. Interesting. Okay, so a default is a setup to some choice you're making. Let's say it's a meal where one option is more or less pre-selected for you. So if you don't like make any active choices or do anything, that's the option you get. Simple example is go to Chipotle, you're you're filling out the or Chipotle's online ordering, you're filling out what you want in your taco tofu or something is pre-selected for you. So if you just skip that question, you get tofu. That's a very minimal type of default. You could have it be more extreme. For instance, you might have a long message that says, Chipotle is proud to serve you tofu, comma, a healthy and sustainable blah, blah, blah. And if you don't want tofu, please write a three-paragraph essay about why you would like another thing. That's like on the other end, you can make your defaults as like onerous as possible. But the basic idea is the way of setting up the choice so that if you're not paying attention, you just go ahead, you just choose a thing. And it's meant to be like cheap, easy to scale, and it can theoretically operate without coming to your conscious awareness. I have not conducted any default studies myself, although I am working on one now with some folks at Greener by default and another researcher at a different school. I'm really excited about that. But the context here is that a lot of people in the animal welfare movement are very excited about defaults as a behavioral change strategy. Um I'm less excited than they are. And the reason is that last year I wrote a meta-analysis of all interventions that intended to reduce consumption of meat and animal products. This kind of gets into like what I do, what we do at our lab. And this meta-analysis looked only at studies that met certain, in my opinion, low, but actually in practice quite stringent quality bars. And those are it had to have at least 25 subjects in treatment and control. So you have to have 50 people or more. It had to be a randomized control trial, no quasi-experimental stuff. And back to your point about changes in diet versus changes in meals, it had to measure meat consumption at least a single day after the treatment was administered to people. So that is a good example is if you had done a default study and it was like, you know, the conference is offering tofu, and then you could either choose tofu or not, and then that's the entire outcome measurement. We wouldn't have included that because that tells you about changing a meal. We would have included it if it said, Oh, by the way, what'd you eat for dinner the next day? Or something like that. Anything in the category of like do changes last. And when we did this, we found that every single default study published by December 2023 didn't actually qualify. So the first yeah, so it's kind of a like condemning beginning already. Yeah, yeah, this is this is tough. This is tough stuff. Like we were I didn't really know that everybody loved defaults in advance of starting this literature. I kind of just waded into it blind. But I think this result surprises people. But the basic problem, the most fundamental problem, is this exact thing, which is that most of these studies just do their one and done. They put the default in, and that's the end of the story.

James

Can you explain why that doesn't work then? Because uh, you know, in a way, I don't expect a default, let's say on Monday to affect their diet on Friday, but I do expect that the default happens every day of the week. It has a similar impact every day of the week. So, like, does it actually matter? That's not measuring diet change, like change in meals five days away or six months away.

Seth

And complex question you're asking, because ultimately that is a way of saying, what do you believe is happening outside of the scope of the study? And then the reader's beliefs about like, oh, I okay, I don't think it's affecting you five days later, and then I'll I'll make that additional assumption. But that's not actually in the data itself. So that's like a actually a hard question. You know, the data do not speak for themselves. What do we think they mean? Always that's a hard question. To your point about what if the default is five days a week, that's interesting. And this is the idea, this is I think the promise of default that you can just if they if they work one day a week, then you can just scale them up to five. Do we really think that's the same thing though? Like it's much more plausible to me that if you get a default once a week, you don't even notice it. You just say, okay, I don't know, I got I got tofu today. But if like every single day you start noticing that your lunch does not have meat or that you have to make a choice to get meat, I think you're a lot more likely to say, okay, I need to like make a decision on this menu. So I do not think that the scaled up version of the default that we have in our mind really is the same thing because the more present it becomes to people, I think the more likely you are to have a backlash. Because then it stops being like, oh, this one conference didn't want me to have meat. It starts being like, oh man, my school has like an agenda against meat. And that's where students start complaining, and that's where dining halls start getting mad.

James

Yeah. Yeah, this sounds kind of similar to, I guess, a conversation. But Katie Cantrell from Green and Body Default also came on the podcast uh maybe some months ago. And yeah, she said that there's almost this tension where the more, let's say, strong or obvious the defaults, uh the way the more powerful is in terms of reducing consumption, but also the more likely it is people will almost rebel against it and find ways to get around it. So it's actually quite a fine line. And so I guess your point is kind of saying, sure, it works one day of the week because it's so small and it not noticeable. But when you do it five days, people start thinking maybe I should just go to a different place to buy food, or maybe I should I know that if I ask twice, I can get the meal I actually want, and people start utilizing those other options. Is that right? Yes, that's exactly right.

Seth

And this goes to my I'll I got a lot of things that I think about defaults, but the only one that I think that's important is basically goes back to your earlier earlier point, James, about okay, well, we have to look at this little data and make a model of the world, and that's how we interpret what we see. My model of the world just does not include a lot of big effects from small changes in general, like in regards to food and regards to, you know, whether having a small loan, if you're a person like a subsistence farmer, does getting a small loan make change your life? Like it looked like that for some pilot studies, and then it doesn't look like that once you scale that up. When we talk about defaults, the more pre- exactly what Katie said. And I want to stop here and say that like I'm kind of you know establishing myself, I guess, as a critic of defaults, but the people doing this work at GBD and other organizations, they're smart. They've thought about this and they've seen it for themselves. Like I know that they're thinking about this question. So I'm actually not sure there is that much daylight between us. I think I'm just I'm gonna be candid here. I'm not in a position where I'm fundraising on behalf of a default-centric organization. So I can just say this. I actually don't really know how much daylight there is between us. But my basic thought is that the more you default people in towards eating plant-based, the more it starts to look like Neatless Monday, and we know that people don't like that.

James

Yeah. Interesting. And and yeah, I think this is something you alluded to just before, which is part of it is um there's almost a mindset question, which is some people start out as inherently relatively skeptical, thinking most things don't change people or their behavior of their diets by huge amounts, so we should assume most things are going to be quite small. And some people maybe now should be more optimistic. I think Lewis in podcast we had with him said, you know, his default for most interventions is they don't work. And when they work, this is a pleasant surprise. But I think for most people, this is not the case. But it seems like you're more in the case of uh things don't really work until proven otherwise, or you expect things to be small changes just because it's hard to change the world.

Seth

Yeah. And if we really want to get spicy with this, I would say that a lot of the research in food in particular that shows that small changes can have big effects is almost certainly fraudulent. Well, that is a uh bulk plan. Do you want you wanna say more about that? Yeah, yeah. There's uh the if you look at a lot of these papers, they all say, and uh and diners go through their lives and make like 215 unconscious choices a day or something like that. And the the site is always Wansink and Sobol 2007. This is a book called like mindless eating, and the lead author of it, Brian Wansig, uh was uncordially dismissed from his position at Cornell because of what we now call p-hacking, which is a way of like slicing the data a million ways until you get what you are looking for. I think cherry picking is the right way to think about this. So when I look at this stuff, I see like, okay, not only is like the whole like the foundational text that everyone is still citing, one of its co-authors is now, in my opinion, fairly disgraced, or at least not welcome in the academic community anymore. It is a red flag for me that people are still citing this as the foundational text of like why mindless eating is such a powerful determinant of our behavior. I feel kind of sad when I think about that.

James

Yeah, yeah. There's actually some there's some other uh academics that had similar fates, right? That have done a bunch of these stuff on on nudges. Is that right? I think I'm thinking of probably Francesca Gino. Yeah, yeah, exactly.

Seth

Some people have said this about Dan Arielli, although he's he retains his position. And if you're listening to this, Dan, I'm sorry if you didn't do anything. I think this whole, like, what I'm gonna describe as this one weird trick will change your life and doctors hate it. This whole paradigm, no. I'm just I'm not for it.

James

Yeah, I I think, yeah, probably I agree. It's like yeah, it's easy to for people who want the world to change in a certain way, when they see these things promising big results for small changes, we we we want to believe them because it's like this could really help animals or really improve the world. But sadly, I think reality is often this is highly overstated.

Seth

And of course, for defaults in particular, they're promising because they're a way of bypassing a difficult conversation. My point of view is that, like, I'm sorry, that conversation is difficult. We still gotta have it. Hmm.

James

And maybe sticking on defaults for a little bit. So we kind of alluded to a couple of problems, which is well, A, the quality of the research isn't particularly good, so it's hard to have a really rigorous idea of where that effects are. You know, and there's related things, which is maybe people that their behavior changes uh after the default gets introduced, so this rather this when equilibrium is actually much less powerful in changing their food choices compared to uh the first day. I guess what other problems do you expect expect there to be? And like, I guess I just have a question on like the leakage problem. One is when leakage being, you know, people just go somewhere else. I can imagine in university dining halls, this is a problem because let's say there's you know, there's five dining halls they can go from, only one's doing this default strategy, so they can just go somewhere else, or they can ask a couple of times. But then there's other cases where, you know, in I think in New York, a bunch of hospitals do this kind of double default. There, I think it's not so easy to you know, for better or for worse, the patient can't can't go anywhere, right? They're they're kind of stuck there. So I can imagine leakage there isn't as much of a problem. Like, do you broadly agree? Do you think is it like relatively context-specific how this stuff works?

Seth

Absolutely. I think context-specific is this like perennially under underrated explanation. And also, I would say hospitals are a great place for default. I expect that to scale up pretty well because I think that's a way to catch someone at a point where they're having a moment of reflection about their life and their lifestyle. That's a great place to say, like, maybe you want to have some tofu. Maybe this is a good, like, a day to start eating tofu. And also I think NYC hospitals have put a lot of work into making the plant-based options delicious. Like there's some real nice stuff from gosh, Michael Grunwald has an article about this in Canary Media from last year or 2024 rather, in which people say, like, oh, it's pretty good. And that that's a positive update, right? So that's uh and whether the food or not, the food is good or not, it matters a lot. I guess I'm less when I've like actually worked with universities and restaurants and stuff and done a little bit of this looking at this reason at this work, I'm not so sure that they make the plant-based food good. I think they make it in many cases for the vegans. So everything else gets a lot of attention and the like the plant-based option is kind of an afterthought. So if you default people into that and you cause them to have a negative update about the quality of plant-based food, well, that's not good. Yeah. So I don't really know how this is gonna work. I know that the NYC hospital seems to have done a really good job with it.

James

Yeah. Yeah, I I agree. I think it's always tough when interventions ultimately are out of the control of people who want them to happen. And then it's it's hard to make sure the quality is up to our bar at least.

Seth

That's what it means to scale up. It means that someone else who doesn't care the way you care is actually in charge now.

James

Yeah, that's tough. To have interventions that are kind of where the quality is kind of hard to get wrong, where it's gonna be so standardizable. Yeah, it feels like there aren't that many things that meet that bar, right?

Seth

Yeah. One would be have everybody watch 10 minutes of Dominion. That's uh I mean, if we can make that happen, it's a really hard, you know, it's hard, obviously, but that's really standardized. So this is a this is like a different way to approach the whole question of persuasion.

James

Yeah, makes sense. Some of the points you had on defaults were some of the studies they also ignore what animal or what meat is eaten. Can you talk more about how that might actually influence the final results of these studies?

The best paper on defaults - what does it mean for advocates?

Seth

Specifically with the interspecies substitution question. There often is just not a lot of good data on that question. Probably the best paper on defaults to date is by Joel Ginn and Greg Sparkman, 2024, which is like I think called like can we default a vegan or something like that?

James

And just for reference, we'll we'll link all these below. I'm sure you'll reference a bunch of studies. People can find these in the in the show notes.

Seth

Yeah, I think I got the title wrong, but uh whatever. Yeah, yeah. They do defaults at a station at three separate universities. So already great intervention. Got a lot of different things happening. But you know, I love this multi-site thing. At one of those schools at RPI, the staff did not play ball. They said, like, oh, we're they said to the students, like, ps, there's meat. Don't worry. Like this, like researchers like watched these interactions happen. So now we're only looking at two schools because the third one they just basically said, like, sorry, it didn't work. Although that's interesting too, because that's a way of saying defaults where they work is a different question than how effective are defaults.

James

Yeah. Then you almost made major just seem like two out of three will also have this problem, and then that reduces everything. But yeah, yeah, I think that's that is a very important consideration.

Seth

It's just a hard problem. It's hard to get people to do stuff. Anyway, so at the schools where it actually worked, which were Tulane and Lehigh, they know did they have a station where you have a plant-based option by default and the meat basically is hidden so that you can like request it, but if you don't request it, you get like the plant-based option. Yeah. They notice a lot of leakage on days where this happens, people, many people go to other stations and we don't know what they did there. So you could make a lot of you have to make some assumptions about how much meat they ate, but we know that like they didn't eat the plant based thing at this station. But did they go away and eat twice as many burgers as they would have because they were annoyed? I mean, probably not, but that could be pretty bad. Or maybe these are like the most hardcore carnivores, so they're the most likely to infect. make sure that they get meat so like there's no actual change in that ground population. Or the problem is we don't really know enough about in this particular study or really any of these studies about what the alternatives are in the plant-based station. If it's like plant-based meatballs versus like a quarter pound of uh meat-based meatballs and someone doesn't like that, they doesn't like the way it's presented, so they go somewhere else and they have like, you know, double cheeseburger with bacon, maybe that's twice as much cheat meat. I don't know. Yeah, yeah. And we have we don't know anything about like the interspecies stuff in those contexts.

James

Yeah, because it it's always yeah I guess there's almost two questions like that they look at meals rather than actually quantity of meat. And then we don't even care about quantity of meat. We actually care about the number of animals. So yeah if if they actually went from beef to chicken or fish, then it could be far worse. And all these studies seem like on the whole don't actually try to account for different products.

Seth

Not as far as I know. There might be some that like make some efforts at this but I don't think there are any randomized control trials that do this. And I just want to say like I'm not blaming the researchers for this. We have been talking to schools about this. It's this incredible process to get data on how many chickens were eaten today. That's like actually a complex problem.

James

Yeah. Yeah no I I can imagine it's actually a total I mean I'm I'm amazed that universities almost collaborate and want to do this stuff at all. I agree.

Seth

I often wonder like what like I talk about that amount I kind of want to ask like can we like let down our guard right now and I just want to know like why are you talking to me? I'm annoying and I'm asking you to do this annoying stuff. What is in it for you? It's actually kind of like a complex question I think because they probably do have goals but it might be oh gosh I could talk about this for a long time.

James

Have you asked them at all?

Seth

Like what are the saying I did actually ask at one point I said like oh what motivates you and we dig we got like uh some stuff about the sustainability goals and we got some stuff about you know trying to connect towards plant-based students or students who keep to a plant-based diet. A lot of inclusivity concerns have come up um I think inclusivity inclusivity is underrated always as like a thing that actually motivates dining services.

James

Yeah like you said I'm I'm grateful to them for doing this at all I think the this research is still far better than hypothetical experiments of you know people just take things in a box. So I think the authority is far better than that.

Seth

I think running a university dining services is hard. You just gotta like you're putting out a lot of fires you're dealing with a lot of student complaints and then we come in and we ask for this additional thing and then as we're digging we realize that the seafood item for the day is called like seafood chef's special and we know that there are seven units of that consumed we don't know anything else and that's because they're doing things to meet their needs. The the the chef's special the seafood special is probably like oh man we got so much seafood we just have a big pot of seafood we need to like bring it in. How many animals like someone knows how many animals that is but that's like might be deep in the supply chain. This is actually it's like a hard problem. And why would they though? They don't care.

James

Yeah yeah yeah this is very interesting. Is there anything else on defaults do you think that people miss or do you think that's most of the key considerations why and I guess I think your overall take on this blog post which we'll definitely share was you you think oh like when when defaults scale up that actually they'll affect meat consumption by around one to two percentage points compared to now which is obviously much smaller than you know I've seen as high as 80% in some cases. So I guess that's like your overarching thesis. But yeah is anything else that we missed.

Seth

You know it's funny though James I'm actually not so sure those are different because if 10 let's think about this for a second. Let's if uh if 5% of people order plant based at a at a university as the rescue meat which is like you know there's 4% of people are vegetarian and you have an 80% increase in that, you've got about one to two percentage points increase.

James

Oh no no I've seen 80% reduction in meat meal order I I I feel like in in some of the New York City hospitals I feel like I heard that that was maybe I'm making this up in the hospitals in particular I as I said I think that's a really good situation for them.

Seth

My last point about defaults and I'm really sorry to be such a downer on this is I'm not totally sure what a default at a restaurant looks like. So like I the Chipotle ordering online one is a good example. You could I can I can like picture this but I okay I walk into Pizza Hut and what's a default at a Pizza Hut? Would it just be like one pizza is said like order our veggie pizza small letters meet options available upon request Pizza Hut's not gonna do this. That's just not that's not how Pizza Hut works like so if it's ever yeah you can like make one option bigger you can put a little sign next to it that says this is the Ceph special but that's not a default.

James

Yeah I feel like there's some things you can do but probably won't have the impact well as big of an impact as we would like for example maybe all the cheese is plant based by default or I know in coffee shops sometimes the the milk is oat milk by default but of course that these are important sources of animal suffering but not the largest source of animal suffering. So it's still one getting at the the big cructs of chicken for example.

Seth

Exactly and if we want to talk about a places where defaults probably are going to work I think oat milk is really good and I'm I love that blue bottle coffee is making that its default and likewise I like that Starbucks has eliminated the upcharge for plant-based milk that a lot of its or most of its stores I think everything except the ones that like operate in targets because they're doing a whole different thing. These things are great. I just want people to not be disappointed when we try to scale these things up to universities or dining halls or anything and we find that in general meat eating is a stubborn, deeply seated behavior.

James

Yeah yeah and I think I guess anyway to take away from for me for this isn't you know we shouldn't do this it's we should do this but we shouldn't only count on this and we should make sure we're pursuing a range of other interventions that are focused on this issue to make sure we kind of try to cover all the bases or hit different audiences and different contexts and all that kind of stuff. Is that how you view it as well?

Seth

Yeah and there's really no reason that the default has to be the only thing you do it can be done in simultaneously with other things. You could make a whole like you could start your dining hall with a big thing about why meat is bad for the environment and then have plant-based defaults. So people are already primed to think about it.

What does the research on meat reduction say?

James

Why not? I mean coming back to this meta-analysis you mentioned earlier on which you know none of these default studies made into but maybe can you just say more about the meta-analysis more broadly in terms of what did you guys find in terms of likely effect sizes and what intentions were actually considered in that pool?

Seth

To recap, we had these quality-based criteria about sample size, study design and outcome measurement this paired P-A-R-E-D paired the literature down from like a thousand studies to about 35 papers or 35 yeah 35 papers and like 107 interventions. So right away that's the headline finding is that like oh there's just not that many papers that met this bar. I think there would be more today. Today would probably be like 50 or 60 if we had done it if we had like kept it up. The fun part was figuring out what are the different categories that we could group these things into and how big are the effect sizes per group so the reason that it's fun to think about the categories is because like you can slice and dice 35 papers a lot of ways. Like we wanted to have a big group that was just like psychology studies, but come on, like what is psychology? Of course all things are like psychological if I like make you watch Dominion that's a psychological intervention. But like we were talking like ooh social psychology which is like norms trying to get people to think about whether meat is cool or not. And this other stuff which is like response inhibition training which is getting people to if they don't want to eat meat getting them to like train training them to like look away and do something else.

James

It's very funny to I'm not gonna I people tried this like is anyone gonna do this?

Seth

Yeah well um I'm not summarizing it like the best right now but the response inhibition training R IT is a real thing. Actually it's funny I just put in a grant that was trying to basically like well we haven't gotten the grant yet so never mind but let's say theoretically someone might do something like where you you might put on like AR goggles or something and have them if you like look at meat in my flash the word like no in front of you. Why not?

James

No way that's so funny. I I'm I'm like how like why would people do this? Like how would this get like what settings would this happen? I guess I don't understand. Okay so there are people out there who say they want to reduce their meat so you might be like all right time to put a shut up here are our goggles yeah yeah I I guess fair enough yeah if people say this it's like if I guess for me it sounds kind of funny is it sounds very it's like you're trying to stop an addiction sounds like kind of very intense in that way but I guess maybe in some ways that is actually the case. What if you got shocked with a cattle prod every time you ate meat why not?

Seth

That's what it sounds like anyway okay so this is one we got response inhibition trading is like broadly speaking the categories we looked at the 35 papers this is one possible breakdown there where others are first you can try to persuade people that's good talk to them you can give them environmental reasons animal welfare reasons or health reasons to stop eating meat and then the other category is this automatic stuff which I'm gonna call nudges broadly because that includes defaults although we didn't look at it at any in this particular study. Others would be like putting the vegetarian option at eye level at a cafeteria so you just look at it and it's big and it's like not the it's just the the obvious thing, the salient thing. Yeah. The other study in this category, I love this it was uh trying to get people to eat less fish sauce at a university in Thailand. So the reason they were interested in this is because of sodium content but they were effectively reducing the amount of meat that goes in and to your concerns James fish sauce tends to be drawn from like very small fishes. So in fact a couple grams of fishes could mean like how many sardines is that like a lot right these things are tiny. The intervention is they put a hole in the center of the spoon where you would serve yourself fish sauce from as you were bringing it over to your pad tie.

James

That's so funny.

Seth

Great intervention. Great idea so but that's like that's choice architecture that's like actually kind of like the getting shocked with a cattle prod thing because they're just making it more annoying to get the animal product.

James

As much as you want. Yeah I guess yeah there's been some studies on just making sound so silly making the serving spoon smaller. I guess it's kind of like that it's making it harder to well it's kinda different uh making it harder to get what you want to yeah yeah my colleagues worked on that serving spoon study we're talking about like a small it's like a that's by An H Boschke and others at Stanford it's published in BMC something BMJ, sorry.

is 5 percentage points a small or big change in meat consumption?

Seth

It does seem silly, but on the other hand, if there were some magical thing that was really clever that made people eat less meat, you wouldn't be reading about it in an academic paper. We would already know. Like in some ways we're just testing the obvious things to see like okay well what would that what will happen? And I love this Thai intervention because it's so creative. I would not have thought of like putting a hole in the spoon. Anyway sorry so we've got intervention categories are persuasion, animal welfare, environment health got this nudge stuff. And we've got as I said earlier psychology. Almost all the psychology studies all but one are trying to make vegetarian options seem like a reasonable popular or normal choice. They're norms studies. And the other one is indeed response inhibition treatment by Camp and uh Lawrence. Okay so then you have these categories and you what you do is you take each study's results and you turn it into amount of change in terms of standard deviations. So that's a little bit of an unintuitive concept but basically you take the spread of like imagine that you were doing an intervention to get people to score better on the SAT. James, did you take the SAT?

James

Uh I I did a long time ago a long time ago do you apply to American universities? Where are you from? Uh well I grew up in Dubai I was considering going to the US but I did not in the end you grew up in Dubai man that's a okay well we'll get back to that.

Seth

Anyway SAT is a standardized test that American high schoolers have to take for smart and a hundred points per section is a standard deviation like that's how that's how they're just scored. And if you did a tutoring intervention that raised their points by 100 points they would you would get a standardized mean difference of one because it the number is equal to the amount of standard deviations that someone changed. This is a little less intuitive in terms of animal welfare stuff because you have to know what like the every standard deviation is different. But we're talking the actual outcomes are in terms of like percent of a standard deviation. And little drum roll the answers we found were that every result was tiny. Like 0.07, 0.09, 0.21 for the choice architecture stuff, the nudge stuff, but that was just two studies and like I think the overall answer was 0.07.0 yes 0.07 so in SAT terms this would be like about seven points and in animal welfare terms this isn't like strictly true but broadly speaking you can think of this as like one or two or maybe three percentage points reductions. Wow so it it is very small. It's small it's small effects I mean a lot of these interventions are doing pretty small things. So this is back to the earlier question about like do small things have big effects? Like probably not to give some context here like I'll be talking about two studies in particular. So I mentioned the uh study about it's a sweet uh about placing the vegetarian option at eye level it's uh this is Anderson and Nilander called nudge the lunch it's a study from 2021 there's three options at this university cafeteria it's in Sweden you can get meat fish or veggies right and they just randomly alter the ordering of those things on a big billboard menu when you walk into the cafeteria. On days when the vegetarian option was eye level meat consumption goes down by five percentage points. Great and then you like read the study closely this is what it it means to do meta-analysis and then you see on those days that fish consumption went up by four percent so four percentage points sorry so what's actually happening is that meat gets reduced by about one percent this translated to an effect size of like 0.1 something it was but even though it was like a small effect but there wasn't that much variation in how much people order meat so it's actually like a big change because it's like all right well it is the only thing that changes whether they eat meat or not. That's like one example of a, you know, it looks like a pretty good effect and if you cared about the and well if you cared about global warming you would say like nice five percentage points. But if you cared about animals and if you care about small animals like you and I do, not so good.

James

Hmm I mean that is like an extremely important reason like why you should almost read the small print or why I'm glad people like you are reading the small print of these studies because you know you you can read the abstract and read the title you go wow five percentage points that's pretty good for a small thing and then yeah you read a bit further down and you're like oh actually that's not so good.

Seth

So yeah I think it was a very important lesson in the I don't know man like I said it's not so good. Am I like am I gonna down myself here?

James

Like one percentage point if you do it like let's say we got that to all Swedish universities, it's a lot of people and yeah I mean it totally if you scale this and everywhere does this and globally goes down by one percentage point it's like wow that could be like several billion animals or something. It's like well that sounds great.

Seth

If we're talking about one percent of but what what's one percentage of all shrimp that are farmed in a year. A huge number right yeah yeah exactly billions probably anyway but a couple billion yeah yeah and so the final thing I'll say about this study also is that you've also pinpointed the like the actual hard part about this meta-analysis is that unfortunately you cannot skim these papers and because things like not everyone thinks meat is the same thing. You know this is very like when I talk to I talked to a person who grew up reasonably religious and Jewish and she said like oh well I I I don't eat meat but I eat fish. It's like what? Like but like in the Torah there's like a definition of what it means to eat meat meat and it does not include fish. And so you're just what you think the word meat means is like this very coded thing. So uh there's no like consensus opinion on this in the literature which means you actually have to read the papers to see what they're doing. And Anis Hakazetti you you Oh yeah sure it's a we can go let's talk about as many studies as you want by uh Andy Jalil and Josh Tassoff and one other person called a low cost climate intervention and then some other words. But it's about they at a small school they have an economics introduction lecture replaced by a lecture about why meat is bad mostly in terms of environmental reasons and uh a little bit of health stuff. And the cool thing about their study is that it was at a place where students' dining purchases are actually tracked over the course of their whole time there. You can like say we can basically see who eats what and that led to a 5.4% reduction for percentage points reduction in total meat consumed. And what I like about this study is that it had this long-term outcome measure. So where that lasted for like three years. And also that even though the it was mainly about the environmental consequences, there was still a reduction in chicken and in fish.

James

Sounds very cool. Yeah I think maybe we were watching this because this one almost seemed too good to be true is you know they watched an hour long thing about the environmental impacts of animal farming and then three years later still there's a meaningful impact uh reduction in their meat consumption. So yeah this one was surprising to me I guess what was it saying for you?

Seth

So hard to say on the one hand like I I can I'm a human being and like other humans I can tell stories. Like I want to tell a story here about how like you get people when they're 18 and it's by the way it's not a video it's like the the professor giving you this like impassioned lesson about the thing which is cool. Really cool intervention idea. And I can believe that it would like if it caused like two people to go vegan out of all the classes you would actually observe a 5.4% reduction more or less because they that's like a go down from all to nothing. And college students made choices like this. Something gets us to be vegan those of us who are vegan but on the other hand talk about things that are hard to scale like this particular professor gave this particular lecture and he did I think he did a really good job probably I don't know man like if you scale it up to University of Michigan or whatever and you say like all right faculty you've got to give this lecture some of them are going to really sandbag that so scaling is going to be sounds like the exact same as some dining holes won't prepare particularly good plant-based options.

James

In this case maybe some of the lecturers won't give particularly compelling speeches about why actually matters.

Seth

Yeah I mean I'm gonna maybe I'll make my middle name scaling up is hard one word just stuff's like tricky what can I say?

James

Yeah and maybe one minor thing we should probably mention for people who are less in the weeds of statistics I'd say um when you said it went down by five percentage points can you explain what you mean by percentage points rather than percentage?

Seth

I'm talking about like out of a hundred percentage points like if if uh meat was consumed at like half of all meals it went down to like 45% of all meals. So yeah the actual percentage points is like the fraction out of a hundred.

James

Yeah because otherwise there's a weird way you can do that she went down by 10% but it's only five percentage points. So yeah I think it's it's useful to clarify how those things are different.

Seth

That's why I I like to think about things in terms of percentage points because we, you know, if you read like medical literature you get this all this like relative risk or odds ratio and stuff which is all like relative to the baseline incidents cancer rate went up by a factor of three but it's like but only one out of a million people got that cancer in the first place. So still talking about yeah like I would talk about what's the actual risk.

James

Cool. Okay so we spoke about okay D volts. Yeah you know a few percentage points pretty good we we should do it maybe not as big as some people might hope. The mental analysis findings broadly similar looking at these two different categories of it was I guess psychology, choice architecture and Rami the talking to people animal welfare environmentalist and health persuasion tactics. Cool persuasion nice and maybe maybe diving into some of those categories or you know if people are listening to this and thinking okay boy you know that's kind of shitting on my parade and you know I I thought all these things worked so well and they're asking what does work? What does work What would you say in terms of maybe things in those categories or things you haven't covered yet? What what things do you think are the most promising to change people's food choices and diets?

Seth

If I were talking to young people, I would talk about the environment. I think the Jaleliel, that's basically the Jalelial et al. study and a couple others that have informed my worldview on this question. And I think that there is a way to frame this argument in a way that gets it actually to not just be cows. If I were talking to older people, this has to be done the right way, but I would talk about health. And the reason it has to be done the right way is again this problem that people might sub into smaller animals because they've been told that I like our health authorities, for instance, for a long time. And I one supplementary finding from our paper I didn't highlight is that we also looked at studies that tried to reduce just red and processed meat consumption and only reported that instead of all meat. And those are like on average more effective. I think they had an effect size of like a fifth of a standard deviation, which is not huge, but you know, that's good. But it would I would put a big asterisk next to those studies until we know.

James

Yeah. So must so the broad claim is it's actually much easier to reduce red meat consumption compared to other types of meat because maybe people it's much more saying that that's like a a bad thing, something compared to chicken or fish.

Seth

Yes, perhaps that was true up until uh a couple weeks ago when we got the new food pyramid, though. Like I don't also this other question of like, do you have temporal validity? Do you think your findings hold into the future? Of course, none of us knows the answer to that question. Not even in theory. So those are two things I'm I would be like cautiously optimistic about. I do think that choice architecture works in some settings. And if I had my druthers, I would have a lot of people watch Dominion. Uh, there's a study that came out after our paper was published by Peter Landry and Lisa Kramer called How the Sausage is Made, in which they get University of Toronto undergrads to talk to watch like either 11 minutes or 16 minutes of a Dominion, and then they ask them what they want to eat, and they also ask them like a week later, and they give them a like a chance of getting like a 50% chance of having their meal compensated. So it's like a pretty real outcome. And that one reduces pork consumption by a lot and reduces meat consumption overall by quite a bit. So you know it's how big is quite a lot in overall compared to some other things we we've spoken about. Gosh, I forget the pork consumption number, but I think it's like 13% percentage points reduction in meat overall a week later. Pretty good. It's like the largest, most credible effect size that I remember in any of the papers I've ever looked at. If you would talk to Peter, he would say that actually undergrads at the Business School of Toronto are not even the ideal sample here. Like we really want to talk to like, I don't know, UC Dublin or whatever they call it, UCD, or like Hamster College in America in America, like the you know, the liberal leaning amenable students.

James

Yeah. Yeah, or like the environmental or biology department, who I can imagine are slightly more open-minded towards animal issues.

Seth

And the challenge is that when you turn Dominion on, you if you're anything like me, you're gonna be very inclined to turn it off very quickly if you have the choice. A study is a good way of getting people to like, you know, as Stanley Milgram taught the world, it's really easy to get people to do stuff in a study. They like want to please the experimenter. So that's I'm talking about this famous like shocking experiment from the 70s. So I think that's a good avenue forward. I don't know how we would turn that into like a big campaign, though.

James

Yeah, it's hard to sit people down and and make them watch a documentary. But I guess in a way, this is you know, some thing that I guess we've been doing for a while is there are a bunch of college groups in the US and other countries that do host things like film screenings, and it seems like this is actually a good thing or, you know, and they should keep doing this and yeah, and it can be like surprisingly powerful.

Seth

On average, I'm not so keen on documentaries. The the Dominion study really studies uh stands out in terms of its effect sizes. My colleague and principal investigator Maya Mather did a separate study on I think game changers or not game different side uh game changers. What the hell? One of them found pretty small effects that didn't really last.

James

Yeah, I I've heard this that I think Louis Politon, his his substack some time ago, had a good piece in documentaries, and I think his conclusion is there's a wide range of effectiveness, and you know, yes, there's lots of ones that don't work, but there's a small number, I think domain seems to be in this list, they're just like highly persuasive, and we should focus on getting these out into the world and maybe playing these as screenings rather than some of the random ones.

Seth

I mean, I certainly agree with that. And when you take this out of its particular context and think about it for a minute, some movies are better than others. Everybody knows this.

James

It makes total sense. It's not exactly rocket science. Yeah. Yeah. That's cool. Okay, so documentaries is is uh well, okay, yeah. One documentary. Yes, maybe one definitely Dominion, maybe a couple others. I can imagine some Earthlings, which is quite Dominion-esque. Maybe that's up there, maybe it's old and dated now. Uh Earthlings I found very challenging.

Seth

I I started I got to the point where they were gassing the dogs and I could not continue. God. Yeah, that's intense. Yeah.

James

Okay, so we have small, okay, high-quality documentaries that say spies no persuasive. What else?

Seth

I know you're gonna want me to talk about what else works. I knew this was gonna happen. And I'm gonna say, like, I don't know. And the reason that I'm a researcher is that I love saying I don't know, even though I don't like when people look at me so sad about when I say it. You're not you don't look sad right now, but like people don't listen.

James

Well, no, I I laugh because it's like there's so many funny researcher tropes. You know, it's like more research is needed, I don't know, it's context dependent, it's like, god damn it, we just want clear answers. I think we've covered all three of those. Thank you.

Potential interventions that Seth is excited about

Seth

Are there any like more research is needed? I was gonna say that, but I didn't want it, it's just too annoying, so I didn't say it. But like for sure, I think that I think that I mean, I'll tell you what I'm thinking about now in terms of what I want to test. I want to know, can we get people to think about the small body problem as having moral implications?

James

Yeah. And by that you mean people going from large animals like cows to chickens and fish would actually kill more animals overall.

Seth

Yep, that's exactly the problem. Or the small animal replacement problem as it's sometimes been called. As far as I know, there's been no studies where we ask people, like, hey, did you know a cow is a lot bigger than a chicken? Do you know how much bigger? I think they're gonna, I think, first of all, I want to ask people this. I want to know, I want to know, like, how big do you think a cow is? No, you idiot. It's a thousand pounds, whatever. Like I think that most people are very disconnected from our agriculture and also have no idea like what's happened. Like, you know, chickens now grow to their slaughter weight in like 40 days. That's an update of people. The fact that they can't barely stand up is an update. I want to know if there's some way to get people to think like, okay, here are the size differences. Here's how many more animals you want to eat. You'd have to kill to get a pound of meat if it comes from a small animal. Do you think this matters? And I've been thinking about this study for like a year, and I haven't actually gotten anywhere with it because I can't figure out how to ask this last part. Do you think it matters? Like if you ask people this in a survey, they're gonna be like, no. End of discussion. Like that's not meaningful.

James

That's funny. I I thought you were gonna say you can't ask that because the leading question, and people be like, yes, of course. Well, that too, yes. I don't know what they're gonna say.

Seth

The point is that I don't think the answer is meaningful. I think I have to actually really get to people's hearts here.

James

Yeah. Could you not almost tie that with a one of this like you get a 50% chance of getting the meal you want, and one of the options is is is a beach. Chicken or a beach.

Seth

We could do that. We've called it, it's called it like an incentive compatible outcome measure. I suspect that you're gonna get no effect on people's behavior just by telling them this thing. Well, a friend of mine who I will always be grateful for actually being honest with me about this, said, I think that both of those animals are in the food chain, so why would I care whether I whether two of them need to die or one or whatever.

James

I mean, it would be interesting to test. I I agree. I think I don't think I've seen anything like this, and this is a perennial problem for animal advocates. So I mean, personally, I I'd be very interested to this easy MS BS. I'll leave the details of making it work. We're thinking about that now.

Seth

That's kind of my like when I need to I need to like stop procrastinating and just do a study. I mean, it's funny about research. Like, there's so many ways I can writing a substack where I talk about research or talking to you, this is a lot of fun. This is like the part where we get to say, like, oh, what's what's true, what's false? But like the actual thing where you stare at a piece of paper and come up with a research idea, it's very hard. I don't really know if I'm good at it. Think like, oh I'm okay, I'm okay at it. I shouldn't understand. But like the this is the hard part, right? Especially not just like I have an idea, it's like, okay, you have a research topic, the small body problem. How do you translate that into a question? What's your question? Questions are always hard. Other questions? So another thing you another thing you did a study on?

James

Uh additional veggie options. You want to say about that? Yeah, absolutely.

Seth

This is a study I worked on with Jessica Hope, Jacob Peacock, and Maya Mather. It's an online survey environment. It's called the paper, it's called what is it called? It's something about giving fresh veggies the boot or something, or it's got a long title. And we might change it as we submit it to a journal. But it's about we uh we ask we ask people to think that they're like in a consumer type environment where they are selecting pens, t-shirts, and tacos. We randomize the order between those three things, and the actual treatment is when you're filling out your taco. How many plant-based options do you get? So it's it looks exactly like the Chipotle menu. And in one option, you get the number of plant-based options that are actually at Chipotle, which is chips and veggies and guacamole, and card and uh what's it called? Sofritas. Sofritas is like a carnitas analog made out of uh soy. Soy probably yeah, yeah. Yeah. The other treatment is we take away the sofritas, and then the third treatment is we add to sofritas and chips and guac or plants and veggies and guac something called like chickenitas, which is meant to be a chicken-based analog to but a fake thing. But don't really but we do we like label it as vegetarian, but we don't talk about it in in other contexts, and we just see what people choose. Again, this is hypothetical, they're just like going through the survey, and we did this in a pretty well-powered sample of like a few thousand people, and we found changes of this will shock you, one to two percentage points per additional arms. Like 1.1 percentage points for like like no options to one option is like one, or like one or one and a half percentage points of more veggie options, and then adding the second option again, it's like 2.4 percentage points change. More people order veggie options.

James

I'm surprised that the the no to one option was actually so small. Wait, and Shorty can't meet.

Seth

Sorry, no plant-based. That's my bad. The no option had veggies and guac. Got it. Okay. So it's like this like uh meat analog thing is actually what we're testing.

James

Got it. Okay, so it seems like, well, yeah, I guess we don't know actually what the impact is from growing from zero to one in terms of overall meat sales, but you can probably imagine having no plant-based options is not great for variations going to one, it it's uh one is going to one is good. But going to two and three seems is pretty minor in terms of how much it actually impacts things.

Seth

Yeah. Or maybe it's not if we imagine that the study was actually taking place at Chipotle in person, and maybe the chicken needs is incredibly delicious and it was cheaper and something like that. You can like you can keep adding this. This is a very simple test of just introduce one or two more novel protein options to see if people choose them. There is indeed like the effects are pretty weird. Like in the group with uh uh three plant-based options, more people order steak than in any other group. Why? I don't know. There's always just random noise in studies. Or maybe there's some reason that I can't think of. But we do notice that as you introduce more and more plant-based meat options, the percentage of people who choose veggies in guac goes down. So like the total number of people who choose plant-based options goes up less than the change in each new option. So like, you know, the the chicken, the chickenitas might get like three percentage, three percent of people to choose it, but yeah, some of those otherwise would have chosen veggies.

James

Yeah, yeah. So you're kind of the overall share goes up a little bit, but it's kind of being divided amongst all the available options. Yeah. I think that's kind of like.

Seth

I mean, these studies, I mean these if we take this at face value, which you know you might not do, you might say, like, oh, hypothetical studies, whatever, I want to see this like in Chipotle. If you do take it at face value, you'd say, this is good because it gives vegetarians more options. Good, right? On the other hand, what we were looking for in this study was an effect size big enough that we could literally take it to Chipotle and say, hey, there's unmet demand for this chicken eat this thing. Like, can you can you invent this and sell it?

James

Which could have worked. I mean, there is some unmet demand, but I guess it's probably not going to meet their bar of introduce introducing this.

Seth

And I was just saying, we like we thought that a reasonable starting point for like there's unmet demand would be like five percentage points change.

James

Yeah, because I guess for them, if they're cannibalizing their veggies and goak, it's like, I don't know, how good is that for them? Because then maybe then they don't have good economies of scale in producing that product, and it's like I don't know.

Seth

Or just every new thing at a restaurant like Chipotle is logistical overhead.

unknown

Yeah.

Seth

If Chipotle fully automated, if it became a robot restaurant, which I it's funny because the founder of Chipotle at one point wanted to do this. There's like a I think it was Chipotle, he like had a little side venture.

James

Yeah, well, I think I think Sweet Green is gonna do this, is my understanding as well.

Seth

Any of these like fast casual places with a lot of venture capital money, they're definitely trying. They're definitely somewhat like then you can add a lot of options. So maybe at that point, actually, there's some idea that that kind of automation will like lower the bar. But you know, I'm not like cheering for us to get rid of human labor here. Yeah, yeah, yeah.

James

That's for another podcast.

Seth

Yeah. I'm not saying work is good, but that's what's complicated.

James

Couple last questions. One, uh, so I guess one thing you mentioned that you were particular potentially excited about that kind of hasn't been tested much, which is this tell people directly about the their impact on food choices on different sizes of animals and in number of individuals. That's one. Is there anything else that's in this category of interesting but not yet tested that you're kind of keen to explore?

Seth

Yeah, I feel like I'm it's funny. Academia sometimes has a real like defensive posture where you don't share what you're working on because you don't want someone else to scoop you. But I have never had that happen to me. I guess my field is too obscure, so I'm just gonna go ahead with what I'm working on. We're working on a study where we make the default steadily more onerous. So, like, how much effort do you need to put in to opt out? And we're gonna measure meat, you know, because meat consumption and also backlash. Like, the more onerous it gets theoretically, the more people are gonna say, like, you know, you can ask them, I agree with this approach, or like I felt like this study was trying to force me to do something I didn't want to do. You can ask like directly, what did you think? So I'm interested. I'm curious about that. I think there might be like some sweet spot where defaults actually are no longer automatic but are combined with arguments, for instance, about the environment and inclusivity. That I think that might work pretty well. Interesting.

James

Yeah, because I think yeah, this all came up in the conversation with Katie. It's uh yeah, it's trying to move from just changing the immediate food choice they make to somewhat uh having some education or maybe some long-term behavior change, yeah, via this kind of like information stimulus. Yeah, I think it'd be very cool to see to what extent that helps or it annoys people.

Seth

Who knows? I don't know. I think that there's a lot of opportunity for just testing this stuff and a lot of um but back to our your first question about culture versus technology, like, you know, there's always gonna be vegan subplots on TV, for instance. I think there was one in and just like that, the sequel to Sex in the City, where there was like, you know, a hunky vegan gardener or something. I heard this third hand, and I didn't watch it, I'm sorry. But like uh if I were if I could wave a magic wand, I would put this kind of stuff into oh my god, I don't even know what people watch anymore. Like, is is modern family still on? Whatever is the modern equivalent of modern family, Bob's Burgers or something. No, Bob's Burgers actually did do a vegan episode. Anyway, and I would just try to make it like a really good plot line, and not in which the vegan is annoying, but in which the vegan is just like a normal person doing it.

James

Yeah, yeah. Yeah, yeah. Yeah, I think yeah, I think some of this is happening in the in the climate world, both in terms of trying to how to subtly nudge people towards low carbon behaviors in a way that they're not the freak ordering lettuce or I'm not flying and therefore I'm not coming on this holiday. You guys are all evil. So, how do you kind of normalize and kind of celebrate these things a bit more?

Seth

So I like the idea of normalization. I mean, but on the other hand, we've got a lot of reason to think that that won't work because like we can think of lots of vegan celebrities who, by dint of their veganism, are perceived as weird. Mobile.

James

I don't think Billie Eilish is seen as weird as I think it is kind of maybe how hard they lean into it. And yeah, Billie Eilish is fairly outspoken, I would say, as well.

Seth

Oh, good for her. I actually didn't know that.

James

Yeah. Is there anything else that you guys are working on uh at the Humane and Stable Food Lab?

Seth

There's lots of lots of projects in the mix. Those are the ones that I'm involved in. I'm also doing a default thing, and I'm thinking about the last one I'm excited about, and this one's very nascent, is back to our earlier question about culture change at universities. So you go in and you talk to dining services and you ask, what are you why are you talking to me? Like, what are you what do you think? Like, am I what is in it for you? I actually like if you reframe that question as like, tell me about your needs and tell me about how plant-based diets are consistent with them, it stops being an annoying question and starts being like, how can we help each other? So I've been thinking lately about I am hoping to run a study where I talk to young people who are working on this stuff at the university level.

James

Like That's cool. So people who who are working in the catering and dining services section.

Seth

Well, first I want to talk to advocates who have tried to get more plant-based options, and then I want to talk to their counterparts at dining services to get a sense of like how did we, how do we like cross these, how do we create cultural connections between us and what worked, what didn't work, what kind of requests are amenable. And you know, there's like this garbage can model of organizations where organizations are big and complex, and some of those some people have some problem and other people have other solution, and if they like link, then things happen. Yeah. And so like I guess I want to know. I want to know I want to study that up close.

James

That's cool. That's exciting. So the three projects are yeah, there's there's this, you know, testing people's reactions to small animal replacements and seeing if that's at all sympathy inducing. Yeah. One is the last one you just said, which is dining services and student advocates.

Seth

And then Rami the other one is um altering the onerousness of default.

James

Oh, yes, that's it. That's a cool one.

Seth

Yeah. It's very exciting.

James

And I'm I'm sure I guess we'll we'll see these out of the way. Yeah, I hope so.

Seth

Geez, I mean, I'm so I like I get in my own way sometimes when I write these papers. That's the truth. It's like really hard to think about like how to operationalize these things. Like, what's our dependent variable? What's our outcome? What are we looking at? Like, you can think about that question forever, but at some point you gotta just write the paper.

Closing questions

James

Nice. Well, there's many advocates who work on this stuff, so I think it's very useful to have people like yourselves and the others at your lab doing this kind of deep and important research. So thank you. Thank you for all that you do. It's kind of exciting. And then we always finish off with a few of the same closing questions. So, Seth, what's one bit of news you're excited about or grateful to hear recently?

Seth

I am excited that Norway has announced a policy to phase out the most popular breeds of fast-growing chickens, which are called the Ross 308 and the COVID 500, because fun fact, a chicken is in fact an RD project worth hundreds of millions of dollars. That what we're talking about here is that the kind of chickens that like take like 40 days to grow to maximum high weight, and then many of them have serious health issues as a result of that, they topple over, they can't like support their own weight, they're too top-heavy because people's people love the breasts in particular. And I want to talk here for a second about uh chicken breeds, because it's a very interesting question. Uh I looked into this a little bit recently. Uh when we talk about uh better chicken commitment and the European chicken commitment, which are these policies intended to get people off these fast growing breeds, it's really interesting to me that it might not mean what people think it means. Like the alternative called the red bro, I don't know why it's called that. I think it takes like four I think she takes like forty eight days to grow to maximum weight. You know, I'm not an animal scientist. I actually don't know if that's like a meaningful, like how much suffering I should be thinking about this in terms of production. I think when people think like heritage breeds, they're thinking of like the kinds that used to take like many months. They're like the you know grandparents' chickens. It's not that. It's something like a little different. And I think that's good, but I'm not an expert on this question.

James

Yes, I I think the win is huge. So I think that the largest chicken producers in Norway have committed to phasing this out over a relatively quick time period, which is would be, I think, the country that is by far the leading discharge, which is so consequential because I think of like the 80 billion vertebrate land animals we kill, I think something like 70 billion is chickens. Absolutely.

Seth

What we really want from this is a demonstration that in fact you will raise chicken prices by less than the doomsayers say. So they're gonna be like, oh, it's gonna make it unaffordable. And it's like, okay, it was a couple cents a kilogram. Next argument. Yeah. Yeah.

James

We'll see. But I maybe answer your question of how good are the welfare benefits, I I would always refer people to the welfare footprint institute. We have all the best research on this. And you know, my read on the research on the bed chicken commitment is roughly reduces their suffering by about 50% relative. So the fast-growing breed has around well, the slow-growing breed has 50% less suffering compared to the fast-growing breed. Even though they do live a bit longer, but just the quality of their life is not like so much better in terms of they can actually move around and they don't have breathing problems and they don't overheat, and all this kind of like the litter dries out, loads of stuff. So I'd recommend people cool.

Seth

And so when you're talking about slow-growing breeds, you are talking about like the red bro and these like next nearest alternatives.

James

Yeah, yeah, yeah, yeah, yeah, yeah. The most basic like I think they did a study based on the next, like next most commercial, like feasible alternative.

Seth

Is there a welfare level at which you would eat chicken? James.

James

Well boy, that's a tough question. Personally, I don't think so. At least I see no reason for me to do so. I guess I'm a pretty precautionary principle kind of person, and I felt that that'd always be a question in my mind is did this have suffering at some point? And you know I think some people have spoken about being um uh a welfare terrian or a pasture terrian. I don't know exactly what the words are. If you can be sure that animals have have positive lives. To me, it's a high bar.

Seth

It's a high bar, yeah. I think I would eat I would eat eggs if I were pretty pretty happy with how the chickens were raised. I don't think I would eat chicken itself.

James

Yeah, yeah, yeah. I think you know, this defensible backyard and claims he made uh if you rescued them, et cetera, et cetera. Okay, well, what are some meat recommendations you might have for listeners? So you mentioned a bunch of papers, we'll link those. Are there things besides your blog, Regression to the Meat, which actually I didn't say this already, but I love the name. I think it's one of the best blog names of all time. We'll link that. Anything else you'd recommend for people?

Seth

First of all, I like that you've phrased this question. What are your meat recommendations? Of course, I'm gonna say carnitas and steak and whatnot. Media, media, media. Media, you said that. Media meat. I'm surprised. That's my bad. That's on me. I really like Warren Belasco's book, Food, The Key Concepts. It's like in the field of food studies. I think a lot of folks in our movement have read a lot of stuff about like activism. I want people to read about like, what does food mean to people? And when I talk about like why I'm not so keen on defaults, a part of it is because I think that food means a lot to people. And it's like not a crazy claim. I think everybody knows that. Once someone I know very well said that meat was how he knew his mom loved him. And it's like, you know, I'm not gonna just default this person into being vegan. They're gonna notice this stuff. Anyway, so I food is a really good overview of like what is what is food like mean for society and for people. I would say everything written by two authors in particular at Vox, Marina Bolatnikova and Kenny Torreya. And they write a lot about animal welfare stuff in a way I think that they get it.

James

Yeah, I agree. I I think some of their coverage is some of the best stuff for people who will also enjoy this podcast. So we'll leave this. And how can people get more involved in your work or follow you more generally?

Seth

I have no social media presence to speak of besides my Substack, so that's regression to the meat.substack.com and our seminar, my the lab I work at, the Humane and Sustainable Food Lab at Stanford, we have a seminar series that is public and we welcome more participants. And we'll put a I think James will put a link to that.

James

That's right. It's exciting. Yeah, I think it may have been to one of them. I I do recommend it. I think there is no better way to get like I don't know, reading a research paper is something or reading a summary, but just being able to hear someone who did the research talk about it in a bit more depth and you can ask some questions, uh, for me feels a lot more fruitful, totally.

Seth

And just and by the way, um, the recap group uh run on a Brian Research and Phonalytics is another group in this group that there, you know, you can just show up. Like a lot of this stuff is happening. And my final comment on that is that on a spectrum of like how much we drink the Haterade, I'm an extremist. Like other people at the lab will give you you'll like hear what they say and you'll feel good.

James

This is good to know that this isn't maybe we should put this earlier on in the episode so people can have it. Yeah, yeah. This is not normal. My behavior is abnormal. I mean, I I think I think it's good to balance optimism and and some pessimism. I feel like we've had lots of optimistic people in the podcast. So it's nice to have some some people call it realism, some people call it pessimism.

Seth

I don't know if I'm pessimistic. I think like we I really do.

James

Skeptical.

Seth

Skeptical I would go with. I really do believe that we can convince the world to stop doing factory farming. But I think we can't get there without doing the hard work of talking to people about what food means to them. Yeah.

James

That's a great place to end it. Well, thanks so much, Seth. It's been great to have you.

Seth

Thank you, man.