Tyler Cowen on How He Views the World

Hello everyone, my name is Andrew and welcome back to Frames of Space. In today's episode, I am thrilled to share a conversation with one of the most interesting thinkers in economics and much, much more. His name is Tyler Cowen. Now, if you're not already familiar with him, Tyler is a professor at George Mason University and the director of the Mercatus Center, which is a think tank and an organization that he's helped to shape over the years. He is deeply involved in hiring bright talent at Mercatus, and He has a knack for spotting people who can do great work on things like market-oriented policy or economic research. Now, if you've ever listened to Tyler Cowen's own podcast, Conversations with Tyler, you would know that he's a bit of a maverick in his interviewing style. For example, he famously avoids asking follow-up questions most of the time. Now, there is a method to that madness. He believes that follow-up questions often lead guests to restate what they've already said rather than covering fresh ground. This means that his conversations tend to roam across a spectacular range of topics rather than just sticking to one issue for the entire episode. And in many ways, that's Tyler's trademark approach. And trust me, this episode is no exception to the type of podcasting that Tyler is known for. Tyler is a huge generalist. He'll talk about AI, social media's effects on pessimism, the economics of inequality, cultural shifts in cinema, pretty much everything but the kitchen sink. You might also know Tyler as a self-described libertarian, sort of. He's often one of the more libertarian-leaning people in the room, even if the staunchest libertarians might not say that he's pure enough for their label. That nuance of the extent to which he is willing to identify with the label of libertarian, it really shows up in our talk where he explains how he squares his own free market views with a reality that still requires some level of things like income redistribution or government intervention. So today's episode is one episode where you might hear many parallels to a Conversations with Tyler episode. There are plenty of quick pivots, a broad range, and a willingness to see where the guest or the host might lead. So without further ado, I present to you my interview with Tyler Cowen. Hope you enjoy. Tyler Cowen, thank you so much for taking the time to come onto this podcast.

Happy to be here.

All right, so now I want you to inhabit an audience member that doesn't know who you are or what you do. Explain to the audience What is the Mercatus Center and what is the work they're designed to achieve?

The Mercatus Center is a research center at George Mason University. It has been there since 1980. Uh, the single biggest thing we do is simply support graduate and undergraduate students who want to study economics and go out and have a broader impact in the world. But we also have elements that are a policy shop. And we bring in people who have good ideas on economic policy, typically in market-oriented directions. And we support them and market their work. And we put out podcasts. And I've been direct faculty director of the Mercatus Center since 1997.

What is it like being the faculty director?

Well, there's a lot of responsibility. Fundraising is a significant part of the job, but also making sure that Mercatus sees in advance which are important ideas and has talented people to work on them. That's extremely hard. It's much, much easier to fail at that than to succeed. And those are some of my responsibilities.

What is it like? So like you're essentially in charge of hiring at a major organization. How does that feel?

Well, there are several of us who have major roles. We also have an executive director right now, that is Ben Clotzy, and most of the hiring would be done by Ben and/or our COO, or some of the hiring is just done by teams. So we have about 150 employees, and only at a high level would I be involved in their hiring, but nonetheless, indirectly, I'm responsible for all the hiring There's a lot of delegation. A lot of my main roles are external or public rather than making sure, well, if we need a new receptionist, that I'm there to interview the person. That's something I don't do.

Yeah, but one thing I find very interesting is when you talk about how the Mercator Center is for, I would maybe like, bargaining undergraduate and graduate students. Do you think there's any kind of demarcation that you want to draw between the Mercatus Center and what a lot of academia is usually like?

We are much more concerned with applying economic ideas to the policy world than a lot of academia would be. Not all of it. Uh, we're more interested in podcasts and blogging than much of academia would be. Again, I'm not saying it's an absolute distinction. Plenty of academia does that too, but we do it systematically as part of our mission. And that's a difference.

Yeah. Then there's one interesting train of this where you mentioned on one of your past podcasts that like sometimes you feel like there are people that are trapped in academia. Like, would you phrase it like there are certain people that are in academia, but they would thrive better if they were in a setting that afforded them more creative freedom?

Uh, I agree with that. I mean, to be clear, plenty of them would just fail, right? And end up with some kind of service job. But you can have tenure, maybe you live somewhere you want to live, or your spouse has a job there, you're used to an academic lifestyle, it's not that easy to leave. But I think, I don't know, 1-2% of those people could really do something pretty phenomenal if they just press the eject button and got on with their lives. And I know it would be a less secure existence, and it wouldn't be clear every day what they do in the morning to organize what's going to happen next. But like Shruti Rajagopalan, who works at Mercatus, she was tenured teaching in New York City and gave that job up. And she's, you know, 100x more influential now because she did that. That would be one example.

Yeah. But like, I guess if we were to see academia as like a bit more rigid, a bit more structured, I guess then one, like, thing that is partly an outflow of that would be, like, a specific argument that I heard that I want your take on. What do you think of the criticism that academia is largely out of touch with the rest of the world, that it's essentially like its own bubble filled with, like, its own community that doesn't really talk that much to people that don't speak its language?

Well, academia is a very large thing. And big parts of it I don't know much about. So if I look at George Mason, I learned recently that one quarter of our students major in engineering. That's high, that's a lot higher than I thought it would be, and I actually think that's great, that's great news. I don't know what those people are doing, but I would be very reluctant to say they're not interacting with the real world. At some point they will be very clearly in the real world, mostly working on real-world problems, And I think that's a sign of the system correcting itself. Now, if you go to the humanities, I think on average the performance is much worse. Economics is one of the better fields, but I do think you have far too many economists— they simply write for the approbation of their peers, for letters of recommendation for their tenure case, and they're not writing for broader impact as they should be doing.

Yeah. So one of the things you mentioned about the Mercatus Center and something it does differently is it does a lot more podcasts. Do you see podcasts in general, like especially podcasts from academics, as a way to bridge the gap between, say, like humanities academia and like the real world?

I don't think they need to be from academics. Maybe on average academics are worse at doing podcasts because they, are not forced to get to the point by their other training. But I think if you have good content and a clearly defined niche, you can get considerable mindshare from important and influential people, uh, by doing podcasts. And there's still room for more. So, uh, if you do something Joe Rogan-like, you can make a lot of money. That's not mainly how people can succeed. But if you want an actual audience, I would say try doing a podcast. Just make it good and don't worry whether or not you're an academic.

Yeah, make it good and make it consistent, and that's all that matters.

And have it, you know, it should have a clear image. So if your podcast is, well, you know, I focus on the Dobro guitar, like, that's an obscure thing, right? But in fact, there are people out there who play the Dobro. They probably don't already have their own podcast. And if you're thinking of doing that, you should do it.

How do you see your podcast in that case?

Mine is a funny mix between being quite general and pretty well known and being quite niche. It's hard to characterize. I cover far more topics than would be wise for almost any podcast, but part of the niche of mine is in a way like me showing off that I can cover so many topics. And that's a niche thing, but we have a big audience and, you know, we don't make any money or have ads or, you know, sell it in any way. But my plan is just to keep on going. I think I have one of the world's best and smartest audiences.

I guess then if your podcast covers so many topics, how do you decide which areas you want to study very deeply and which areas you just want to remain like a generalist in?

Well, it's really pretty selfish and arbitrary on my part. So if I want to learn an area, I'll like think of a guest, or if a new book takes my fancy, I think, well, that I like that book because it was interesting to me, I should read more in the area, and then I'll just do it. So there's a lot of people who are probably very good guests, but I don't do them. You know, I'm not paid. And I just don't want to go through that topic. No, no slight intended to them at all. So if I ever turn you down for my podcast, any listener out there, that's probably the reason why. Not that I think like you're not good enough. Yeah.

Okay, so one, one—

He's a guy with some YouTube channel. He travels the world. It seemed very interesting, but I felt I'd have to spend like 20 hours listening to him on YouTube. And he is very good, but it just didn't feel to me like that, that's where I'm at right now.

But theoretically, if you found some like Russian historian that talked about like some like event or some conflict that you thought was really interesting, you, would you be willing to spend like 20 hours studying for that?

Oh, more than 20 hours. You know, I might read 40 or 50 books on it. So, uh, yes, it's quite arbitrary and selfish, basically.

Yeah. Do you ever enjoy, like, hosting a podcast where you're talking about one topic and then you suddenly, like, abruptly pivot to, like, a completely different topic? And then, like, do you enjoy asking, like, very abrupt questions that feel unrelated to, like, whatever topic you had just been discussing?

Well, people say I do that all the time, but to me they feel very related. So I don't feel that that's what I'm doing. I would admit, like, how other people describe me is more accurate than how I would describe myself. But it's all about like, how does the person think? That's what I'm trying to get at. And I think I'm along the same lines of questioning.

Do you, do you enjoy asking follow-up questions or do you think follow-up questions in an interview are overrated?

They're overrated. Most people are not good at answering follow-ups. They repeat what they said already in an inferior way. There are exceptions to that. But it's hard to get into an actual back-and-forth dialogue. Now, I just recorded one with Scott Sumner, and I did have some follow-up questions, and I felt he was very good at not just repeating what he said the first time. So I'm glad I did that. But I would say at least three-quarters of guests, I think follow-up questions are mostly a mistake.

Well, I guess just to add my own editorializing into the ring, I guess my take is that my favorite episodes of yours are the ones where you do have follow-up questions. Like, I really loved your conversation with Jonathan Haidt, because you did ask quite a bit of follow-up questions and I think the guest answered those questions pretty well.

Well, I think that's consistent with my view though. If only some people are good at follow-up questions, those people on average are in some way better guests, right? So if the better guests are your favorite episodes, it may or may not be causal that you like them because I have follow-up questions. They're also just better guests. It doesn't mean you necessarily think I should do that with everyone.

Yeah, but like, it does mean that like, there is a selection effect where the guests where you tend to ask follow-up questions, those tend to be better episodes because the selection effect dictates that like in general, those guests are better on average in the first place.

And on average, I have more respect for them, whether or not they're better. I would say that's correlated, but someone could dispute that. And if I have more respect for them, that's another way it might make the episode better.

Yeah, I guess then one question I have as a result is like, when you're podcast hosting different guests, how do you determine whether a guest's arguments are worth pushing back on? And like, how do you determine when it's not worth like asking those sorts of follow-up questions and just moving on?

Well, any guest I've heard already on YouTube or their other podcasts, almost everyone has those. So I make a judgment But even apart from that, somehow I feel intuitively in the first minute or two a sense of whether they can do it, even if I had to approach the whole thing blind. Uh, so you can just tell by how engaged people are in having a sequence of claims that are new rather than repetitive, even on the first round of an answer, you get a sense of how a second round might go.

So like sometimes that's a judgment call that you make like during the episode itself and not beforehand.

Absolutely. I mean, I have put in work beforehand, but even if I hadn't, I feel like I would have a reasonable sense of how it would go.

Yes. So on a— I would consider this an unrelated note, but maybe you could find some kind of tether to this.

No, it's highly related. I know this already.

All right. On a highly related note, Tyler Cowen, could you explain why you are a libertarian?

Well, most libertarians would say I'm not one, just to start off the bat. Uh, but I look at it this way. Let's say I go to a public event with a lot of well-known people. Say there's 100 of us there. Odds are I'm one of the 3 most libertarian people in that room. So if someone says I'm a libertarian, I don't run away from it. They have pretty good reasons for thinking it, but the people who are like true capital L libertarians would mostly say I'm not, I'm not close enough to it for them. So I don't outright just say I'm a libertarian, but on economics, I'm much more free market oriented than most of my peers in the economics profession. In general, I think it's wrong to send people to jail for things they do to themselves. That makes me pretty libertarian on social issues. On foreign policy, I'm more of a hawk than many libertarians. I'm not sure how much that rules me out. I guess I think of it as the more libertarian point of view, but I certainly think that's contentious, right?

Yeah. Then I guess something that I would wonder is like, to what extent do you think you're like— so obviously you would consider yourself to be a libertarian, so maybe that's sufficient. Do you think that your libertarianism is a product of like moral conviction or utilitarianism? Like one way I would phrase it would be like, do you think libertarianism is like inherently correct from like a fundamental philosophical point of view? Or are you a libertarian because this specific political context suggests that libertarianism would be the most like utility maximizing ideology?

I once wrote in my book Stubborn Attachments that I'm a two-thirds utilitarian, And that was tongue-in-cheek. I would say on economic issues, I'm almost completely consequentialist or even utilitarian. On a lot of social issues, I would grant some kind of strictly enforced mental morality by law might be better. I just think it's wrong to coerce people. So on that, I would say it's mostly not utilitarian. Then on foreign policy, Maybe I'm a Hobbesian that I think we need to ensure that civilization continues in the world. That will involve some rights violations.

Wait, what rights violations?

Well, U.S. foreign policy causes a lot of rights violations. Some of them are wrong and unjustified, but some of them are necessary. And I just think we need to do it, that otherwise the world will collapse into some state of affairs you know, say pre-1945.

What rights violations from the US's foreign policy establishment do you think are justified?

I don't think there's a good abstract answer to that question. I would just say when you fight any war, typically you kill a fair number of innocent people. Not saying all those wars are justified. Vietnam War was not, for instance.

But the Iraq War was not justified.

So sorry.

The Iraq War was also not justified.

That's more complicated, and there's several Iraq Wars, right? A Korean War, I would say, was justified. So it's going to depend, uh, but it's also taxation at home, and that's coercive as well. And in earlier times, you know, it may or may not have required a draft, and that's clearly coercive. So there's some heavy-duty coercion going on there. It's one reason libertarians would say I'm not a libertarian.

Yeah, but like you would— so like, but you would say that you are a libertarian. So I guess I want to flesh out—

I wouldn't insist on it. Again, I wouldn't deny it, but I don't insist on claiming that word. I'll just say I'm more libertarian than most of my peers, and other people can call me what they want.

Yeah, so like you're not opposed to it, but it's not like a hill that you would die under, like calling yourself a libertarian.

Correct. And I think on foreign policy, my prescriptions lead to a more libertarian world than what the libertarians claim they want, which is strict non-interventionism. People are going to disagree about that.

Yeah. So then I guess one thing that I'm very interested in fleshing out is your, your, how your ideology relates to economics. So like, would you consider economic efficiency to be one of the most important goals for people, like for society to achieve right now? And if so, why?

Again, there's a lot of big concepts behind that question. I want to know what's the policy area, what's the country you have in mind? But think of economic efficiency as roughly equivalent to growing wealth. We have a number of places in the world today that are not growing their wealth, like the United Kingdom. And they're pretty miserable and they're failing. So I'm a big believer in having GDP growth. Absolutely. At all costs? No. But most of the time it's going to come together with other good things.

Yeah, but not more liberty. Not at all. Like, not at all costs. Like, like, you would acknowledge that there are some times where there is a trade-off between efficiency and fairness.

Let's say if the UK were to execute all of its alcoholics that people would stop drinking and they would grow more rapidly. I would not favor that. That would be an example.

Yeah. Would you favor, like, Amazon's exploitation of its laborers in terms of, like, constantly keeping their mind on the clock and trying to maximize output as much as possible, even if it leads to, like, cruel working conditions?

There's a lot of premises behind that question. I've never been inside an Amazon plant. My impression from a distance is that those are better jobs than what the people would get otherwise. So I tend to be for them. I'm open to more evidence, but that would be my immediate response.

Okay. Do you think that economic efficiency is, like, sufficiently prized among economists? Like, do you think that economists prize economic efficiency to the extent that it is worthwhile to do so in society?

Most economists will say they do, but then they'll go away and take their actual views from NPR and The New York Times. So in that sense, I'll say no. But if you just polled them, I think you'd get pretty reasonable answers. So it depends on how you're going to look at this.

So do you think that economists are engaging in preference falsification in terms of what they say they would prize compared to what economic policies they would prize?

No, I think they're mostly sincere. Some of them might be saying it cynically to go along with the crowd, but I think it's more likely a conformism that people grow into believing.

Could you elaborate on the whole thing about conformism that they grow into believing?

Well, they're surrounded by other graduate students, faculty. That's the audiences they need to please. Academia in economics, it's not left-wing, but it's quite Democratic Party oriented. So those are the views, either explicitly or implicitly, that you're encouraged to have. You keep on running into evidence for those views, and over time you grow to hold them, and maybe you, you held them to begin with. It's just a very common story. I think there's some amount of cynicism, but I sometimes like to say sincerity is the most underrated motive, and I think that's true here as well, even with politicians. I think a great number of them are pretty sincere.

Yeah, it reminds— so this is, this is gonna sound hackneyed, but maybe it's actually completely relevant. Um, it reminds me of this one quote from the video game Metal Gear Solid 2 where like they're talking about the prevalence of like echo chambers online where like there are like cardinal truths that neither clash nor mesh. So when you're trying to like, like build up your own worldview, you can just like only look at the truths that like sort of seem to validate you. And like, as a result, when you retreat into like those echo chambers, like nobody is invalidated, but nobody's right.

Um, but mainstream Democratic Party-leaning academia in economics, it's hit by it pretty strongly.

Yeah. Um, so I was, I was gonna say on an unrelated note, but I'm sure that this is actually completely unrelated even if I don't realize it. Um, would you be opposed to, like, within the US's current economic system, would you be opposed to income redistribution on purely utilitarian grounds?

I think there's a utilitarian case for income redistribution, but I think the strongest case for income redistribution is that you need some of it to keep your democracy up and running. So the force of that second case and the kind of redistribution you need will typically outweigh the force of the first case. I would agree that both are valid cases, but the second one is typically what's going to really matter.

All right. Elaborate on, like, your characterization of the second argument then.

Well, most voters, including Republicans, favor income redistribution. Social safety net. And I think we're very far from a world where that is not true. And it's true in all the wealthy countries, all the democratic wealthy countries. So you have to take that somewhat as a constraint. I don't think it's going to change. So the kind of oppression you would need to blunt those desires— you could say to people, well, we should redistribute much less it would be better to invest all that in science or taking care of immigrants. It could even be true from a utilitarian point of view that that would be better, but you're not going to persuade your citizens of it. So you need to go along with a version of a social safety net that will allow enough people to agree so your semi-free society keeps on progressing and holding free elections. And really, a lot of countries have succeeded in doing that. I think it's quite impressive.

Are you saying that you would be more like sympathetic to the argument for income redistribution on political grounds rather than moral grounds?

They both point in the same direction, and the political one is stronger, is how I would put it. The moral grounds argument, it depends what else you're doing with the money. So if you would take that money and send it to very poor people in Africa or use it to support their immigration, like, the moral argument then goes away. But if it's, well, have a safety net or, you know, give subsidies to agribusiness, well, it's clear the moral argument then points in favor of the safety net.

Do you think there are any important negative externalities that come from large levels of income inequality within a society?

Any— that's a dangerous word to use. I'll say this, there's a book coming out in 2025 by Lane Kenworthy where he goes through all the evidence and he shows it's very difficult to show that income inequality causally has bad effects. And Lane is on the left. It's a very good book. So when you say any, I don't know, like it's probably going to be something.. But I think those effects are significantly overrated. And if anything, income inequality might be a symptom of some bad features of your society, like too much crony capitalism. Take Russia as a super obvious example. But it's not like the causal reason why your median voter has high blood pressure, say.

Yeah. All right. Can you explain the concept of the hedonic treadmill to our audiences?

Well, the way I think of it is that some people, as they acquire more wealth or a better situation, they come to expect more and they don't end up all that much happier. I'm not saying I agree with this claim, but that's how I understand the hedonic treadmill argument. I think it's pretty clear that people with some qualifications, but they have more wealth, typically they're better off. Is my view. I think there's less of a treadmill than is commonly believed.

But maybe there is a very sharp law of diminishing returns when it comes to the net benefit, but like from the iPhone 7 to the iPhone 11, where tons and tons of money was invested in trying to improve like the iPhone just a little bit from generation to generation. But the actual net utility that it grants people or the net increased amount of joy that it gives people is not actually worth the enormous amount of money that Apple puts into making new iPhones every few years?

Well, when it comes to well-being, I think there's a cardinal element and an ordinal element, and I'm pretty sure the cardinal element maxes out. Wait, sorry.

Could you explain?

At some sum of money. You can debate the level.

Explain what a cardinal element and what an ordinal element is.

Roughly, the cardinal element is your happiness. Do I get more joy if I'm a billionaire and I can buy a better ice cream cone? I think in fact the billionaire cannot buy a better ice cream cone than what I can buy. But that said, there's other aspects of your well-being that simply come from the desire to be able to do certain things. So if a multi-billionaire gives away wealth, to help solve some problem in Africa. I don't know that that makes the person happier, uh, but it's a preference the person had, and it's part of their well-being. Maybe it's a kind of satisfaction. People who have children, they may or may not be happier in the sense of being relaxed, feeling at ease, but they might have more joy. They might also have more frustration. They're probably more stressed. They have less ability to do what they want. I think quite a few of them are somewhat less happy, but they're more satisfied that they have one kind of life rather than another. So happiness is complicated. The fact that one part of happiness, you know, maxes out at some income level doesn't mean that all of human well-being does. So that's why I don't fully buy into this hedonic treadmill thing. It's not all hedonic. Is one way to put it.

Would you accept the premise that, like, within first world countries like the US and in the world at large, like, general people's general quality of life and, like, material living conditions have improved dramatically over the past few decades?

Uh, probably. Again, I want to know the country. When you say few decades, you know, I want to know the years. But in general, yes, living standards have improved. But at a slower rate than what they used to do, say, you know, 1948 through 1973, for instance.

All right. How about material living conditions for the US since the turn of the century?

You mean since 2000?

Yes.

That's quite complicated. Overall, they've improved, but it's been disappointing, I would say.

Okay, so then I guess the question is like, do you think that all of the technological innovation that Apple did to go from like the iPhone 7 to the iPhone 11, do you think that that actually led to any kind of substantial improvements in happiness compared to if that money was funneled to help deal with like poverty?

I don't think I know the differences between those two iPhones. Late iPhone is much better than early iPhone. Uh, you have to ask, like, what's this— the institutional system you have in mind that turns Apple resources into poverty alleviation? What I see is that the success of the iPhone led to a lot of copying and a lot of cheaper smartphones, and those actually have done a lot to fight poverty. So I would look at it that way.

You think that they did more to fight poverty than if, like, the government were to increase taxation on those corporations to fund, like, a greater child tax credit?

Probably, yes. You look at the impact of smartphones in Africa, Philippines, I think it's been pretty phenomenal.

All right, so then would you reject the premise that there is a sharp law of diminishing returns when it comes to the US economy's current allocation of resources and its like, like slavish— what I would characterize as like a slavish devotion to like marginal technological improvements on leisurely things?

I think those marginal improvements are part of a broader competitive environment that spreads around the whole world and makes many billions of people much better off. So they don't upset me at all. I say bring it on.

Yeah. So you're broadly— would you say that you're broadly supportive of them?

Absolutely.

Yeah. On a completely related note, what are some ideological positions that, like, you've changed your mind on in recent years?

Well, what's recent years? What's ideological? You know, at first I thought Bitcoin was a bubble and a fraud. I don't think that anymore. Is that an ideological view? I'm repeatedly distressed at how pessimistic people are willing to become, and that's surprising me. So I keep on changing my mind. I think, well, they're not going to get that pessimistic, but I just see so much more negativism than I would have expected in the 1970s, and this is disturbing. I think it's a real threat. To our polity, to our civilization.

Do you think— what would you— are you familiar with a certain strain of argument that, like, society's increased pessimism is a direct result of technological innovation to improve things like social media algorithms?

There's a lot of meta studies that look at the correlation between individual happiness and social media usage. And those correlations are very, very weak. I think they're stronger for girls in the age range of something like 12 to 14. Then they're not so weak. But to blame societal pessimism on social media usage, I don't see that supported by the data.

Okay. What is your explanation for societal pessimism then?

I don't know. What I observe in broader history is you have very large social mood swings, such as before World War I, that do not seem to have obvious causes. So if today's does not have an obvious cause, of course that's a puzzle. But in a way, it's the kind of puzzle we should have been expecting. Yeah.

So this is something that you mentioned to Jonathan Haidt when you were talking to him. I guess then, like, you, you don't find it believable that social media incentivizes like, like social— you don't think that social media or like news consumption like inherently incentivizes negativity bias?

I don't know. I see a lot of people my age or older who watch a great deal of cable TV and they become negative from that. So no, I'm not convinced it's social media. I don't see that in the data. I think it reveals that maybe if you're an intellectual and you read other intellectuals on Twitter, you see they're worse or stupider or more biased than you thought. I think that definitely happens, you know, the Martin Gury effect.

But maybe the reason why you see those—

TV is still super important.

Okay, so you think TV might play a larger role in negativity bias than social media?

Well, I'm reluctant to say TV is causal, but as a transmission belt, it could just be social moods shift through contagion effects. The origins of the shift, you know, remain obscure, and then it gets channeled through all the different things we have— television, social media, whatever. Uh, that's maybe the best model I have at the moment.

Yeah. I guess that—

I don't think it's a very good model, to be clear.

Yeah, but like theoretically, if I were to like show you studies that suggest that like social media or like online news was correlated with increased negativity in some way, would that like change your mind?

Well, I've read a lot of those studies. What I see is a bunch of poorly done studies that find big correlations, and then when you do an actual randomized control trial, there's still an effect, but it's quite small.

So you think—

and they believe the randomized control trials more than the big bad correlative studies. So I think it's, you know, a small factor in pessimism and negativity, but nothing like what John Haidt says it is.

So do you think that academics are engaging in cloudspotting when they try to say that things like online news or social media are responsible for increased negativity and pessimism?

Well, I've seen a number of specialists review Jon's book. One review I think was in Nature, another was in the Daily Beast, and they were quite negative and scathing, saying that Jon got this wrong. So the specialists in academia I've seen often getting this right. It's the sort of casual drive-by academics who turn on NPR and read The New York Times who are more likely to be getting it wrong.

I guess when it comes to like the way that you view the world, how do you guard against confirmation bias?

Well, how does anyone, right? You try to know a diverse set of people, people who are smarter than you, people who know more than you. You try to travel and so on. Uh, does that guard against it? Does that cement it in? 'Cause you say, oh, I'm wonderful, I'm doing all these things, I can't have confirmation bias. I don't know, you do it, you can, right?

Yeah. I guess then the question is like, do you ever want, like, do you ever like internally try to recalibrate the amount of moral clarity that you deliver arguments in? Like, do you ever wonder if like something you said you delivered with an unearned amount of confidence, for example?

I think the level of confidence people have when they speak stems mainly from their temperament and not from their actual epistemic level of certainty. And it's good when listening to people or reading them to keep that point in mind.

Yeah, but maybe like a lot of people who consume political content often mistake temperament for how much like epistemic certainty that person has when they're delivering those arguments.

Sure, I try to teach them not to do that, but it's hard.

Yeah.

You can't step out of your temperament. So like me, if you speak directly, and I think usually I do, people will think you're more certain than you really are. But that doesn't mean I should go off and cultivate some complicated roundabout style to confuse people so that then they think I'm more uncertain.

Yeah. Or maybe something that you could do, or maybe something that you've been trying to do, would be to deliver those arguments but say, I believe, I think, rather than saying it with like, this thing is true.

I think that's phony though. When people do that, I trust them slightly less and the audience tunes it out. So no, I don't want to do that. One thing I try to do is present in a lot of different media so people can cross-compare if they want. There's written, there's spoken, I'm on podcasts, there's my podcast, there's other podcasts. Which I do a lot of. There's Twitter, what I write for Bloomberg, and it's somewhat all in different voices and tones. And, uh, you know, maybe that's useful for some people. They still have to sort it out though.

Yeah, they still have to sort it out and basically try to like, they, they still have to manually disentangle like how confident you are in your temperament from like how much epistemic certainty you have when you are delivering those positions.

Yeah, I would say it's on Marginal Revolution when I'm most free to do thinking out loud, and that would be the closest to the real me.

Yeah. Um, so when people give you that sort of like feedback about what they think about, uh, like your takes, do you have any advice to people about how to deal with harsh feedback? Like, is there a right balance between taking it too seriously or not seriously enough?

Well, most people should just get used to it. It's a sign you're having impact. And you're going to encounter it every day if you're successful. So you should take the content seriously, but don't let it depress you.

Yeah, maybe find—

Every day I get a lot of harsh feedback.

Yeah, find a way to disentangle like the negative things they said about your work from like how you should feel about yourself on a philosophical level.

That's impossible though, because my work is a big part of me. I think you just have to compartmentalize even more than that. Like, if they don't like my work, that's fine, but they're in essence not liking a part of me. I just have to deal with that. Yeah, but we'll like that part of me.

Yeah, so obviously maybe there are like some better ways to deal with that and some worse ways to deal with that. So like, are there like times where you give like a political hot take, and then like the negative feedback that you receive, you've tried to find ways to disentangle like the worthwhile feedback from the not worthwhile feedback. Maybe like also try to disentangle the substance of their arguments from the tone that they use when they're talking about you.

Yeah, one thing I like about blogging is you return to the same topics over time, and on Marginal Revolution there's a search function which works well, and anyone can use that and see how my thoughts have evolved on something. So I feel that's another way that I'm trying to give people entry into, like, the real view. Now, a lot of people don't want to do that. I mean, that's up to them. I think most of our audiences often, they don't really want epistemic uncertainty or a person's real view. They want a weapon for building their own social alliances and coalitions. And they will use me and indeed other people toward that end. I don't know what to do about that. But the EMR search engine, to me, it's a great tool of seeing how in fact I do change my mind.

Yeah. And Bitcoin, maybe it's a nice way to see you on in a more like stream of consciousness manner.

Yeah. But again, most people don't use it, but I can't make them do it.

Yeah.

I'm just saying, oh, I think at the beginning of every blog post. You know, they'll just dismiss it.

Yeah. Um, what are your most left-wing political positions?

I don't know anymore what is a left-wing position. So I used to think my pro-free speech views were quite left-wing, but it seems now they're right-wing.

Well, they were right-wing, and now they're like Neither left. I'd say like now they're like neither left-wing nor right-wing.

Maybe like my free trade views used to be right-wing. They're now somewhat more left-wing.

Oh, because of the tariffs.

Because of, you know, Trump being a protectionist, people on the right have become much more protectionist. I have not much, maybe a little. Like I would admit there are more cases where the national security argument matters than I used to think. But I'm still basically a free trader, unlike, say, Oren Cass. And, uh, that doesn't put me on the right anymore, but it used to.

Yeah. So I guess within this specific political environment, would you say that you have any strong, like, opinions that might be coded as left-wing?

Well, I have a very negative opinion of what happened on January 6th, say. That would put me as left-wing, I guess. Again, classically, it would make me a conservative, but today it's more of a left-wing view.

Unfortunately so, yeah. Do you think political commentators tend to be more anxious and more neurotic than most people?

Most people is a big category, but I meet a lot of them and they seem pretty neurotic to me. I haven't met the whole cross-section of everyone else. I think they're, they're often quite neurotic. And, uh, at least formerly, the left wing was much more neurotic, but that may be shifting. I'm seeing the right wing adopt more of the memes of the left in some funny way, like rallying against the plutocracy, all the outrages that are going on supposedly. Uh, so I think it's changing.

Yeah.

Okay.

So what are your theories for why, like, people in politics tend to be more neurotic?

Well, you're underpaid, right? You have to be somewhat intelligent, probably pretty intelligent. So you could, at least in theory, make more doing something else. So you need some personal motivation. It just so happens a lot of our personal motivations are negative. And so these commentators end up being neurotic. If you're just happy, healthy, you know, maybe you'll start an ice cream business and accumulate $20 million in net wealth. Like, isn't that just a better thing to do?

Yeah. Um, to what extent do you think movies and films have affected your political worldview?

You don't know all of what goes into your view. I've watched a very large number of movies and films. I would say that I think on net they have made me more anti-authoritarian, but I would offer that answer with a lot of caveats that you don't really know what affects you always.

Yeah. Do you like, are there any specific movies that like you watched when you were growing up that were formative to the way that you view the world, at least consciously, or is it just hard to say?

Well, my movie origins are in the 1970s, and if you know cinema of that era, it's really quite anti-authoritarian. There's a lot of movies like One Flew Over a Cuckoo's Nest, or early '80s, a lot of movies where governments form conspiracies. You have Star Wars where you root for the Rebel Alliance against Darth Vader.

Yeah.

Probably it had an effect on me. Now movies now are not like that at all, but that's what I grew up with.

Yeah.

Dosis.

I wonder if movies nowadays are more focused on like defending the quote unquote system and like doing like maybe movies today are more like that compared to how, for example, Star Wars in 1977, like George Lucas is pretty far left, and he, like, he used— like, the Empire was a stand-in for the United States during the Vietnam War for the first Star Wars movie. So maybe that affected you.

It probably did. And I think also movies of that era, uh, there's a sense in which showing the erotic is a kind of dare and an expression of freedom, and that has completely changed. It's not libertarian in the policy sense, But you could say it breeds broadly libertarian attitudes against certain kinds of moralizing, including from the public sector. So if something like gay marriage comes along, which I hadn't really thought of until it became a movement, but it was very easy for me to say, oh, well, of course I'm for this. And we did it and I was happy and I think it's worked out fine.

Yeah, that does make me wonder a little bit how the newer generation, like all the movies that they're growing up with, all the TV shows that they're growing up with, how that'll affect their political worldview in ways that are probably very hard to tangibly, tangibly pick out, but are still very large and influential, even if we can't fully put our fingers on them.

My worry is that it's too conformist more than any single view, but we'll see, right?

Yeah. Could you elaborate on that a little bit?

Well, the '70s and also the late 1960s, the anti-hero is big in cinema, and the anti-hero is taken quite seriously. Over time, the anti-hero morphs into just another establishment hero who was nominally an anti-hero, but not really. Okay, so like, not allowed to do anything bad or too defined.

Someone that's like nominally Somebody that's like nominally an anti-hero, would that be like Jack Bauer in 24? Would that count? Yes. Yeah, like he's someone that's like nominally an anti-hero, but in reality the show is very neoconservative and very like, like supporting, supporting like torture methods to the point where like, like a Supreme Court justice tried to use that to like that show to defend torture.

Um, yeah, and I see that as the trend of the last 20 years. That kind of establishment conformism.

Yeah. How much faith do you have that the next generation will handle social media well and responsibly?

I think they'll handle it much better. But social media also are on the verge of changing radically with AI. I don't think we know how they'll change, uh, but you'll have AIs that can manage your social media life Now, will people use them to only pick out the fun parts, or will people use them to just overdose on all addictions? I think more the former, but that's coming very soon. So the whole landscape is going to change. It's one of my objections to the John Haidt view is you're writing regulations for a world on the way out. It was like, you know, the antitrust cases— Microsoft earlier, AT&T— you come at the thing too late.

Yeah, but maybe imagine like there's like a, like a best case scenario and the worst case scenario where like the best case scenario would be that you are like very, very much having your finger on the pulse when it comes to how AI will change these landscapes. And like you do these, this government regulation just as quickly as technology is changing. And then like the middle case scenario would be that like you're writing regulation for a world that's on its way out. And then the worst case scenario is that you don't bother to regulate it at all. Maybe that's the case.

I think the worst-case scenario is you write in regulations that stifle changes and you end up stuck with the problems you have now.

How bullish are you on AI in that case?

Well, what do you mean by bullish? I once wrote the prediction, I think it will boost growth rates by half a percentage point a year. Everyone I know considered this either absurdly weak or absurdly high. Was I bullish or bearish? Compounded, I think I'm bullish. Year to year, it's not that big a deal, though it's still a lot of money. I think it's a hard thing to predict.

I remember talking with Dylan Matthews a few months ago, and one statistic he suggested, like, as a plausible outcome that really stuck in my mind would be the idea that, like, there could still be maybe like a 5 to 15% extinction risk from AI. Do you think that— so like, I guess, like, you would say that in the aggregate, you might be a little bit more bullish on AI. What would you say to people who would accept the premises that AI will have a large impact on society, but that, like, we should be more skeptical that AI could be used for a force for good rather than for ill?

Well, any new technology ends up used both ways. But I would just say, if we cannot put intelligence to on-net good uses, we're probably doomed anyway. Yeah, but I am more optimistic. But I think in Bayesian terms, you need to think through— if you're a pessimist, you have to be very pessimistic about a lot of other things, like humans getting more smart, or more of the world developing, or more people having access to non-AI technologies that might be destructive. So, uh, I think there's a greater chance AI saves us than kills us all.

Yeah, but maybe like the— like there's a greater chance that AI saves us than kills us all. But like the— like the risk that it kills us all is still non-negligible and we should like focus more on the threat that it might kill us all than we might focus on the hope that it would save us all.

We're not very good at collective international coordination, especially with China. They're burning more coal. They're aggressing against Taiwan. They won't even set up a nuclear hotline with us. So the AI equilibrium, it's a race to the top. Uh, you know, if you, if you're really worried, well, your, your outcome is that the Chinese AI kills you. So I would rather see that we try to get there first and do something with it better than what they will do. But you're not going to stop it.

This sounds like the nuclear arms race all over again.

Correct, with different details, uh, but the nuclear arms race, we more or less managed to stay ahead, and that was the right thing to do.

How much do you use AI for your own work at the Mercatus Center?

I use AI every day a few dozen times, you know, no matter what I'm doing. It's not— yes, Mercatus, but if I'm out there traveling, I ask GPT questions. So it's not just Mercatus and not just my work. It's my life.

To what extent do you think that from a utilitarian standpoint, AI will have a negative effect on hiring decisions?

It will lead to much better matching. Not everyone will like that outcome. I worry there could be too much early tracking. Overall, that's probably going to be a good thing. Uh, but it's going to discourage people who might have fought through the system and made it otherwise. So it's not an unalloyed good thing. And AI will be great at picking out people in disadvantaged environments who otherwise would not get a chance. And that will be a huge benefit. Huge.

Would you support an increased use of AI in the classroom, in the learning environment for schools?

Well, in all of my classes, I use it. And I teach the students how to use it. So it's not just I would support, I do support. Most other faculty don't do that. They're way behind the curve. What happens is students use the AI to cheat, which is the worst outcome, and we need to get our act together.

Yeah. So basically you're saying that AI is going to be used no matter what. So you might as well teach students how to use it in a responsible way.

And for instance, in one of my classes, I assigned 3 papers One of the three had to be written with AI, and they had to tell me what they did, and I graded them on the use of AI in addition to grading the paper. There's different ways you can do this, but things like that, people need to learn. I mean, what are we waiting for? We're the university, right?

Can you try to steelman the case for AI doomerism to me?

There's different versions of doomerism. To me, the most plausible version is simply that any major technological change will cause significant disruptions. They can be largely positive, but so much of our current world is so inflexible, maybe we just won't be able to hand it— handle it. Now, that's not doomerism in the sense, oh, AI kills every person and exterminates even the people hiding in the bunker in New Zealand. But that argument makes sense. If you're talking about AI rises up and manufactures the pathogen, I would just say for those claims, there is neither evidence in peer-reviewed literature nor in market prices. So right now it's pseudoscience.

That makes me kind of curious. Do you think a lot of the cases for AI skepticism are less a product of like, they're going to kill us all and more just a product of like, AI is progressing so quickly and like it's going to cause so much change to society at such a fast rate that it will be hard for people to keep up with the fast rate of change in society.

Yes, a lot of people don't want their worlds to change, and that's understandable. There's things about my world that might change, change with the internet I didn't want to change. But again, the benefits will be very great.

Yeah. So maybe to this, this is once again, this is completely related, but maybe there's some like analogy that can be drawn about how like the world is changing. People will feel uncomfortable by the increased use of AI in the same way that like people have felt uncomfortable about the fact that like Star Wars is becoming more racially diverse, but like it, that change would still be like a net positive or something like that.

But the AI changes, I think, will be quite large. And if you have to tell a parent, what should my child or maybe grandchild study to be ready for this new world? I think none of us have a really good answer for them. We don't know. And that can be true even if you expect it to be quite positive. So you could say back to the parent, the child, well, your income will be 30% higher than otherwise, and you'll live 7 years longer. That's great. But still, when they say, okay, but what should I study? You still don't have a good answer. And they're going to be more stressed. And in the meantime, they don't know what to do.

Do you think it's plausible that the reason why you might be relatively more bullish on AI is because you're part of the quote unquote elite part of society that's like, like significantly like better in terms of socioeconomic status and in terms of like education such that you can keep up with the fast rate of change that AI will create, more so than like most people will?

Well, I would say I'm much more threatened by AI as a competitor than most people. So AI is a word cell. It's already on the verge of being a better economist than I am. It certainly knows much more than I do, especially, you know, like the O1 Pro model. So AI is kicking my butt, uh, it's not obvious I'm a winner here.

Yeah, but like, even if carpenters and gardeners, they're the safe ones. Yeah, but even if you're not a winner, you would still support it on net because you think that like the benefit it will provide to society far outweighs the cost.

Yeah, but you say support, like it's always about what margin, how much support, support in which ways. We have to make all these very specific decisions and they're hard to get right. And just like supporting it better than wanting to ban it, but it doesn't get you that far.

Yeah. All right, one of my last questions. When it comes to like people, like people are always raised in like a specific context, like we exist within the context that we live in. How much do you think people, when it comes to their personalities and their behaviors, how much do you think is that is an inevitable product of the context that they were raised in? And how much do you think people can rise above the environment that they grew up, whether it be the culture or like the parents that parented them? Like, how much do you think people can rise above all of that?

Well, I don't like the word inevitable. A few years ago, I visited Liverpool, the house where Paul McCartney grew up. His family was really poor. It's striking. And he's a billionaire. And he didn't do it through education or any kind of affirmative action, but he did it. And you can find examples like that in every walk of life. It's very hard, but you see it happening all the time. So it's not inevitable.

It's not inevitable. It's just very hard.

Correct.

Yeah. All right. Uh, final question. What are some movies that you would recommend to people in order to understand how, like, the general political ideology or the general political atmosphere towards, like, that people had in the past was a lot less conformist than it might be in the present?

Well, like so many of these questions, you'll probably get a better answer from the O1 model or O1 Pro, better yet. I'm serious. So that's the sort of thing. The number of possible movies out there is very large. I've actually seen a large percentage of them. Many I have not seen since the 1970s, which is now a long time ago. I could give you a decent answer and say Ballad of Easy Rider, The Godfather, Star Wars, One Flew Over the Cuckoo's Nest. It would be a good answer, but GPT already beats me on this question. So ask it and it will give you a better answer.

Tyler Cowen, thank you so much for taking the time to come on to this podcast.

Andrew, thank you. My pleasure.

Thank you for listening. If you enjoyed this episode and would like to hear more, make sure to follow this podcast on Spotify, Apple Podcasts, or anywhere else that you get your podcasts from. Until next time, goodbye.