Bryan Caplan - Discrimination, Poverty, & Mental Illness

Dwarkesh Patel: Okay, today I'm speaking with my good friend Bryan Caplan, and this will actually be the second time we talked. The first time we talked was the first episode of this podcast. So yeah, I'm really excited about this.

Bryan Caplan: It's fantastic to be back here, Rakesh. Great to see how well you've been doing for yourself. And now it is my privilege to get to speak to you.

Dwarkesh Patel: Excellent. Okay. So today we're talking about your book, *Labor Econ Versus the World*, and it's a collection of your essays throughout the years and I highly recommend it. Okay. So here's my first question. What percentage of the working age population is zero or negative productivity?

Bryan Caplan: Hmm. It's a good question. So this is working age population, not the ones that are in fact currently working.

Dwarkesh Patel: Yeah. Yeah. Yeah.

Bryan Caplan: Hmm. Let's see. Wow. I'd be inclined to say probably something like three percent, although of course it's much higher if you just make someone do a job that they're not suited for. So whenever people talk about zero productivity workers, I often want to say the fact that someone has zero productivity or negative productivity at the job they're doing doesn't show that they are a zero or negative productivity person. It could just be that they are mismatched to the job. A common misunderstanding, actually. So yeah, I think it is very low for the working age population for any job at all. One of the things that labor markets do, of course, is fire you from jobs where you have really no productivity, which encourages you to search around for something where you are actually productive.

Dwarkesh Patel: So then what's the explanation for why labor force participation is like 60% or something? There's a gap of about 37% of people who could be contributing, but are not.

Bryan Caplan: Well, the biggest explanation is not working moms, rather moms, especially with young kids who don't want a job because they are busy taking care of their kids. So that is the first and foremost one: family responsibilities. There's a small share of people that are rich enough and there isn't any job that they like doing. I think that's only maybe a couple percent of the population. Then you've got a larger percentage of the population where government redistribution means that they really don't have that much of a gain from working, at least in the short run. I think often it would still in fact be better for them in the long run just to get a job, even if they don't make as much money as when receiving redistribution, because you get training and connections. So in the long run it is a better strategy. There's another chunk of people whose parents are just either really nice or suckers, depending upon how you see it, who will just take care of them. There's, of course, spouses, even when you don't have kids, or romantic partners who will take care of you, even though you are not working. And then finally, this category does overlap with the others, but there are guys who just really don't want to go and conform. Most often what they're doing is they work sporadically. They have a job, and then they are difficult, they get fired, and then they are unemployed for a while, then they find another job. One of my favorite books on poverty, *Promises I Can Keep*, talks about the problems of single moms. One of the main things it talks about is they're not happy with the fathers of their kids. The question comes up, what was the problem? They can't find jobs? And they said, no, no, he finds jobs all the time. He just can't keep them. So I think that's another part of it. If you officially say that you're still looking for work but you just don't find a job, then you'll be counted as unemployed. But if sometimes you just say, screw it, I'm not even looking, and they record that in the stats, then that will come out as saying that you aren't even trying.

Dwarkesh Patel: Yes, I'm glad you brought up poverty. So you emphasize in the book the success sequence. This is the idea that if you get married, if you have kids, you get a high school diploma and you work full time, then you're virtually guaranteed to not be in poverty. In the United States anyway.

Bryan Caplan: Oh, sorry. Yeah. In the US.

Dwarkesh Patel: So this raises a question. If these steps are so easy to stay out of poverty or to not get into poverty in the first place, and if so many people are still avoiding doing these steps, what is the explanation? Because clearly being out of poverty is something that a lot of people, you would think, would want. So are these steps harder than we think? What is the explanation for why so many people are still failing to comply with these basic steps?

Bryan Caplan: That's a great question. So when people ask me about the meaning of the success sequence, I say, imagine there's some research saying that here's a simple recipe to avoid poverty. First, be first in your class at MIT. Second of all, win a Nobel Prize. And third, marry someone who won an Academy Award. If that was the prescription, then people might reasonably say, well, hardly anybody can do that. So I'm not even going to try. It's ridiculous. It's hopeless. And then if that's what you have to do to avoid poverty, then you could reasonably say most people just couldn't possibly do it. What's interesting about the success sequence is that it does sound quite easy, right? American high school standards are very low. Basically, if you just show up and try, then you're very likely to get passed along. So you'll graduate from high school. Getting a job for working full time, even when you go and ask people who are in poverty during all but very severe recessions, normally they'll say, yeah, it's easy to get a job. It's not actually very hard. This is again, talking to actual poor people. This is not the opinion of economics professors. And then finally, waiting until marriage to have kids. That sounds really easy. It's like, well, yeah, we've got reliable contraception. It's not like you need celibacy or anything, just delay until you've got someone that you are comfortable raising the kids with. So as to why people don't do these three things, there's a few possibilities. One is that people just don't know. I'd say this is really hard not to know because you have so many people telling you. Your parents are telling you, teachers are telling you, other authorities are telling you, coaches are telling you. Furthermore, it is totally normal even in neighborhoods where people are not following this for parents to want you to follow it, which I think is a telling piece of information. There is a critique saying that this success sequence only works for people that are growing up in middle-class neighborhoods, and it is just not realistic, it is not actually a path out of poverty for other people. If that's so, why is it that parents who themselves have not followed it push it so hard? It appears to be totally normal. It is really weird for a single mom to raise a kid saying, you know what I want you to be? A single mom. It's the best. It's the only good path in life, right? That's not normal. Instead, the normal thing is to say, look at how I messed up my life. Don't do what I did. Wait, right? And then people still don't wait. There's also a story saying, look, even though people are telling you to do this, you don't have any good examples that you can see where anyone's followed it. So there isn't really any real proof. And here I say, hmm, well, probably your teachers in school follow it. So you can look at them and you can say, all right, my teacher might not be the coolest person, but they're not living in poverty, right? Bare minimum. So my teacher seems to have a path out of this. Furthermore, it's actually very normal even in poor families for there to be successful members of the family who have escaped poverty and normally do it by following the success sequence. So you have a relative who, in a family of single moms, this is the one that didn't do it, and this is the one that waited and followed the path. And you will see her at family events. People may have some resentment. Oh, she thinks she's so much better than the rest of us. But again, you can't say that they don't have any firsthand experience with it. They've got that. So I think the story that makes the most sense to me is it does require some impulse control, right? Which is not fun for people right now. So yeah, finishing high school requires that you go and sit there and be bored. I remember high school. It was really boring. So I can, especially when you've got young people that are making decisions, young people are famous for low impulse control. Same thing for working a job. A lot of what you're doing on a job is humiliating yourself. Honestly, there's a customer who's mistreating you and yelling at you, you're being rude for no good reason. To keep the job, you say, yes, sir. Terribly sorry, sir. Your pride says to say, hey, don't talk to me that way. Like, what do you know? You're the one that screwed up. You're not supposed to say that on a job. And it requires impulse control to keep your mouth shut. And then, of course, most notoriously, avoiding kids that you're not ready to have requires impulse control. And indeed, I would say that in all the work on poverty that I've read, poor sexual impulse control is actually the root of almost all the other problems, because it's one where a lapse in judgment actually leads to long-term responsibilities in a way. So you can recover from yelling at a customer, getting fired. All right, fine. You burn some bridges, but you just look around, you get another job. But you really can't easily recover, especially if you're the mom, from having a child where the dad isn't going to help you or the other parent can't help you. It does mean that for 18 years, you really are trying to, at best, juggle a lot of complicated responsibilities.

Dwarkesh Patel: Right. Right. And so I guess the most crucially, I see one of the main critiques of the success sequence, by the way, is that full-time work does almost all the work. Now, my understanding from Brad Wilcox, that's not true. There actually is a marginal effect of the other pieces besides full-time work. But even if the full-time work critique were totally correct, the question is, well, how easy is it for a single mom to do full-time work. And yeah, it's really hard. Now as to why people do it despite the fact that it's just a minor impulse control issue and then you'll have a much better life, I mean, the answer for that is like, well, hard to say. I would just, if the person were asking for advice, I would say, look, this is really easy. Do it. And they don't do it anyway. And I said, man, it was so easy and you still didn't do it. Like I told you what to do. Why didn't you listen to me? And there's the classic response of the irresponsible young person: why did you do it? I don't know. I don't know. And it's one that's been timeless, where a young person does something where it seems like it was really easy to do the wise thing. They were told what the wise thing was. They didn't do it anyway. What more can you say that points other than, gee, you've really messed up and you're no longer a child and you're now going to be paying for this for a long time. Sorry. So there's Charles Murray's story in *Coming Apart*, which is that actually the culture isn't emphasizing these values, or at least the elites in the culture are not emphasizing these values that they live by. Right. So in that way, you can say like, oh, it's the elite's fault for not emphasizing these values that these people don't know about anyways. Or I'm sure you saw Tyler's review of your book on Marginal Revolution, where he said about this. I mean, he was broadly praising your book, but he said about this point that, oh, you're underestimating the cultural factors that might get in the way of somebody achieving the success sequence. So I wonder what your reactions are to those.

Bryan Caplan: So Charles Murray, I would say, look, there's probably a little truth to this, but it's a pretty damn lame excuse. You say, yeah, well, gee, why did you go and drop out of high school and fail to get a job and then have kids before you got married? Well, the elites didn't tell me that I should do these things. Look, there's a lot of people telling you to do this stuff. Teachers are not at the highest level of elite, but relative to a poor community, teachers are elite. You got parents. You got ministers in a poor community. What is the minister telling you? Is he telling you to not do the success sequence? No, he's telling you to do the success sequence. He might also be adding that this is an evil racist society that's screwing you over, but if anything, that's a reason to go and be careful with your behavior so that this racist society has more trouble screwing you over. To say, look, I don't have a lot of second chances, so I better watch my behavior and be mindful. I think it is fair to say, look, elites could do a bit more, but again to say, well, I wanted to do all these things but the elites just didn't tell me, so blame them. It's a pretty lame excuse. It's about at the level of saying that you went and put your hand in a blender because you saw it on a cartoon. It's like, come on. I mean, if you're 10, then it's literally tell the parents, hey, well, don't let the kid watch cartoons if they're just going to copy whatever horrible thing they see Bugs Bunny doing. But if they are at the level where they know the difference between cartoons and reality, then it's like, hey, exercise responsibility there. And again, for Tyler saying I'm neglecting cultural factors, I do talk a lot about cultural factors here. I'm well aware of the fact that people tend to imitate other people. It's just the question of how good of an excuse is that. There's all my friends. So if all your friends were jumping off the Brooklyn Bridge with you, well, I don't know, maybe that's pretty foolish, is it not? Do you really even need me to tell you not to jump off the Brooklyn Bridge just because all your friends are doing it? I have some other pieces where I just say, look, nobody really is willing to actually accept these kinds of lame excuses as a rule, because it would mean that you would be going and exonerating people for horrible behavior. Someone who says, well, yeah, I did go and commit a lot of war crimes, but there are a lot of other soldiers doing war crimes too around me. It's like, well, did they kill you if you didn't participate? Uh, no. But they would have laughed at me if I didn't go along with it. Yeah, well, sometimes you need to be laughed at so you don't become a murderer. And I don't think you really need me to tell you that. It's obvious enough when you think about it. It's a very lame excuse to say, well, but the other soldiers would have laughed at me if I said, let's not burn this village down and I don't want to participate. By the way, this was also, I believe, the point of a pretty famous book, *Hitler's Willing Executioners*, which asked the question, what happened to Germans who just said, I'm not participating in the Holocaust? You might think it's a totalitarian despotism. You're going to get shot yourself or sent to the concentration camp yourself if you say, I refuse to go along with this immoral plan that you have. And what the book said is hardly any Germans were ever punished for refusing to go along. They would just say, OK, well, he's not very comfortable with this, so I guess we won't make him. It was like, that's all that it took? It's one thing to say, do you have the courage to go and die to avoid hurting a person, and it's a very different thing to say, do you have the courage to not be laughed at to avoid being a murderer? That's not that big of an ask, actually. And I will say, and I think many of our listeners will remember, are there times that you followed your conscience even though you had some peers laughing at you and you did it anyway? Is it really that hard? Come on.

Dwarkesh Patel: But so I guess putting aside the question of moral blame, to the extent that the argument you're making there when you talk about the willingness of, let's say, Germans in Nazi Germany, you're making the argument like, oh, human nature is very perceptive to what your peers think, and they do want to disappoint the people you're working with and so on, within your community. So that implies that causally speaking, the community matters a lot or the culture matters a lot. So like, what could be, I think, Robin Hanson reacting to Tyler Cowen's review of your book, he said, I guess he was kind of amusing. So I don't know if he's necessarily endorsing this, but he said, you know, maybe this kind of justifies cultural imperialism, where a culture that has better values when it comes to success sequence adjacent stuff, they could kind of impose them on cultures that have worse values here. So do you think anything can be done to improve the culture when it seems dysfunctional in these ways? Does greatly cutting back on redistribution or making it more conditional upon good behavior count as something we can do?

Bryan Caplan: I think that's a really obvious thing. It's to just say, look, it was pretty easy for you to go and avoid these problems. And so we're going to treat you differently because your problems were avoidable and you didn't avoid them. So that could just be lower amounts of payments or cut people off sooner, or to say, look, you're not eligible. These are all possibilities. So there's that. In terms of other things to be done, you could actually go and teach the success sequence in schools. I don't know that it's going to be all that persuasive because it really is just repeating pretty much what you're already telling kids anyway. I do have some thought that if you just ramp up the level of preaching by a factor of 10 or 50, then maybe it would work. There is a lot of research saying that various things don't work. And then the question is, well, they don't work because they have zero effect or they don't work because the dosage is too low. And honestly, I often think the problem is really the low dosage. It's just the question of, is this an organization that actually has the steel to give a very high dosage? For example, foreign language education, the kind of foreign language education that goes on in American high schools essentially has zero effectiveness. And yet people do learn foreign languages. How do they learn foreign languages? You basically learn it when you take the intensity of a normal high school program and you ramp it up by a factor of 10 or 50, and then people learn. Now, given how much time schools are already spending on foreign language, which is typically in, say, a place like California, three years out of high school, you don't really have the time to ramp it up by a factor of 10 or 50. It would just absorb the entire program. And that's where you say the amount of effort that would be required to go and actually get fluency in Spanish is so high that we're better off not trying. For something like this, where people really are messing up their lives at a pretty early age, then maybe, and especially if you're not currently giving them three years of advice, you might only be giving them a few hours per year when you add it all up. Maybe if you did go and multiply that by a lot, then it would actually have the effectiveness you're looking for. It's the kind of thing I say it's worth just trying and seeing whether it works if we just go and make a much bigger effort here. And also just consider A/B testing where some people just get the lecture and some people you go and actually meet people who messed up their lives. I honestly don't think it would be at all a bad idea to go and recruit homeless people to go and talk to kindergartners. You say, especially if you're in a poor area, you see these guys just sitting on the side of the road. Well, let's bring them in and talk to them and find out what's going on. Ask them what they think they did wrong with their lives, if anything. I think people would just be so squeamish. You don't want to go and traumatize five-year-olds by having them meet a homeless person and have them talk about his crack problem or whatever. But I would say, why not? If you just convince one kid in the class not to become homeless from that experience, or again, if one isn't enough, do it 10 times. Do it twice a year until you finish high school. It seems like it's at least worth a try. It couldn't cost much.

Dwarkesh Patel: Yeah, yeah. Going slightly off topic, does the fact that foreign remote workers are paid many multiples less than Americans who are also remote workers, does that suggest that foreigners are lower productivity than you think? Maybe there's some cultural barriers that are preventing them from being as valuable as Americans are. Maybe there's some other factor, but in any case, why is the fact that programmers in India who are working remotely are paid much less than programmers in America who are working remotely? Does that suggest that actually the people in India are lower productivity than you think?

Bryan Caplan: It definitely suggests it. However, we have tests to show the suggestion is wrong. Because we know what happens when you are a remote programmer in India. We also know what happens when you move that remote worker to the US or another first world country. And then we see the wages skyrocket. So it seems like the most you could say is there might be some interaction between being foreign and being remote. At least we know that being foreign and being present in a first world economy in no way predicts lower pay, and probably also no difference in job performance really. So now as to why that is, the one thing is that we've had remote work take such a huge increase during COVID that right now we perhaps we're just not in equilibrium yet. So right now, it may be that it's taking time for first world companies to say, wait, we know we could get better workers that are as good for all this money in the third world. So maybe that's what's going on. But I have to think that there is a reason why remote work, first of all, was rare before. And second of all, why most firms are eager to return to in-person work. So I think that there are some serious problems with it that are maybe a little hard to articulate. My best guess here is just that while it may be not that bad to go and switch to remote work when you already have a well-functioning team that you have assembled for in-person work, it's probably a lot harder to get a well-functioning team that has never been in person. So essentially you might think of in-person work as being an investment in being able to cooperate well, whether or not you're in person or remote in the future. And if you've never been anything other than remote, then maybe the team building doesn't work so well. But again, the other possibility is that we're just not in equilibrium yet. The economic argument comes down to if you really can get someone just as good for a third of the price from another country for remote work, then why don't you do it? Right now, of course, there is the answer of lower productivity, which again, I say is plausible until you actually see what happens and move that same worker into the first world. And you say, nope, turns out he was just fine. And the problem was probably remote work, or maybe there's something about remote work and being in another country as well that might make it happen. By the way, I do have a friend who is actually an American living in Spain, and he applies for remote jobs and it doesn't work for him either. So it's not just something about being Spanish. It seems to be that a US company isn't interested because he is in Spain, even though he is an American. And that is a problem for him. Now, again, you might say, why does he just claim to be in Michigan? The problem is he has been in Spain for a long time and therefore his whole CV is Spain oriented. So it looks like he is not well integrated into the US labor market at this point. But yeah, it is a very good question.

Dwarkesh Patel: A potential explanation occurs to me. I don't know if this makes any sense. But could you apply the signaling explanation here? So just like college, going through the immigration process signals that you have high conscientiousness and high conformity. And it gives you more information about their background as far as crime and other things are concerned, and even hiring within America would give, right? So maybe it's like a signaling story, like going through immigration, just the same way college is.

Bryan Caplan: Yeah, the problem is the immigration system, I just don't think is really very meritocratic at all. It's quite arbitrary. I mean, remember, we even have the diversity lottery. So if you see that someone got into the US by winning the immigration lottery, you're inclined to say, well, if he's just a lottery winner, he just got lucky. So I should think that he's not that good, right? That's not actually the normal reaction. So I don't think, and that's a case where there's an enormous amount of luck involved. And then again, for other stuff, it's just so obvious that it's mostly about family reunification, really. So it's like, look, this person was lucky enough to have relatives. I mean, there is some level of commitment to go and fill out paperwork and so on. But yeah, as to how much that makes you enthusiastic about hiring the person. And then on top of all of this, we actually have a whole separate body of evidence on illegal immigrants, where employers definitely seem to be more reluctant to hire illegal immigrants, but almost all of that appears to be just the problem of compliance and getting caught, right? And we know this because we actually have the experiment of when you regularize illegal immigrants, then very quickly they get a big raise. So again, if you're an employer and you really thought that it's only the legal immigrants that have signaled something good about themselves, then when you see that there's an illegal immigrant who was regularized, then you should hold it against them and say, well, since he didn't go through the normal process, then I have to think that he's of low quality compared to other immigrants. This does not seem to be the normal reaction that employers have. The normal reaction is, great, now I can hire the guy I wanted to, plus I don't have to worry about getting fined or shut down or raided or anything else.

Dwarkesh Patel: Yeah, that makes sense. What does the education system look like in equilibrium given there's no government cuts? Because if there's credential inflation, isn't it just like an escalating cycle? So eventually, I mean, it wouldn't make sense eventually to get to the point where you need a PhD to be a janitor. But what does the equilibrium look like?

Bryan Caplan: Yeah, great question. So I would say actually that if you keep the level of funding constant, then I would think that we've reached equilibrium already. There are periods when the share of Americans with college degrees was actually quite flat. So basically during the late eighties, early nineties, I believe the share of, like, 25 to 29 year old Americans who would finish bachelor's degrees was quite flat for about 15 years, which then led many people to say, okay, we've reached equilibrium and now that's over. Then afterwards, it resumed its upward increase, and it went from like 20, 25%. Now we're getting more like 35% of people in that age bracket have finished four-year degrees. As to what were the shocks to the system?

Bryan Caplan: Yeah, that's an interesting question, and I don't have more than speculation here. Part of what's going on, of course, is that much of the rise is driven by women. So there is a growing disparity between male four-year college graduation and female four-year college graduation. And so I think a lot of what's going on is that this is a period when it's becoming more and more normative for women to go to college and to get their four-year degrees. And there's, as I said, people are conformist. So once, if you are a young woman, you see this is what almost all the other young women are doing, then you're very inclined to go and do that.

So in terms of sheer affordability, this is overall a period when college was becoming less affordable. It's easy to exaggerate that because there's also so much financial aid. And then furthermore, of course, people are getting richer on top of that. So that is complicated, but nevertheless, at least I would say that it's probably not driven by changes in affordability too much.

Let's see. I mean, another story is just the rising inequality is another reason to go and get more education. So when the gap between high school graduates and college graduates goes up, this is a reason for more people to go. So there's probably that too. So you can get changes in just the structure of the economy, higher return to quality, that kind of thing also matters.

But yeah, I would say the idea that there's just an actual tendency for education to rise no matter what, I think that's wrong. There's this old line attributed to Herb Stein that anything that can't go on forever won't go on forever. Very reasonable point. As to how far it can go though, oh yeah, there is something else that I don't think we have really good data on, but I'm very strongly suspicious of it, which is that schools are constantly slightly lowering their standards over time in order to make it easier to finish.

So that's something where it's very clear they lowered their standards between '65 and today, right? The curriculum was just much more intellectually demanding back then. It is harder to say with great confidence the lower standards between 2000 and 2022, but that is definitely my suspicion. It's one where if you are a college professor, especially in a lower rank school, you'll appreciate what the pressure is. Every year you have a class of students, most of them don't want to be there and really are not academically inclined. And then the question is, how many of them am I going to fail out?

Now, if you have any pride in your work at all, you'll start by failing the people who never showed up and didn't even do the exam. So it's like, okay, well, I can fire them with good conscience. Plus, I don't even know what they look like, so it doesn't hurt my feelings, and I don't feel bad for them when I flunk them. But then there's the question of, what about the people that showed up and tried, but got a zero? What do I do about them? Most professors' pride still says, no, I can't pass that person. It's sad, but I can't. But when you do it for 30 years in a row, probably there's a general tendency each time to go and fail maybe 0.2 fewer people per year, and over time this happens.

And then there's also a culture of what is an acceptable share to fail. And then as the standards fall, you get weaker students there. And then the process continues to amplify where standards keep going down. So I do think that there's been a lowering of standards, which means that more people want to go to college because they can actually have a prayer of getting through. And then when you do that, that does spur further credential inflation because you've got too many people with low quality with the degrees in hand. And so you need an additional degree piled on top to get much confidence in quality.

Dwarkesh Patel: Right. So then why don't more employers care about the hardness of a curriculum for the college that they're interviewing students from, in comparison to like what level of degree they have? So they'll care like, does he have a master's or a bachelor's? They'll care less about like, did he go to UChicago or Harvard? I mean, I've heard that UChicago is much harder, right? But employers don't seem to care that much. Like, that's an even stronger signal of conscientiousness that he went to UChicago. As long as the incoming class had a similar GPA and SAT score, shouldn't employers care about that?

Bryan Caplan: Good question. So here's what we know about the actual payoffs of different kinds of college degrees. The single best predictor of payoff is your major. And the majors that are thought by most people to be harder generally have much better payoffs. That's right. So step one is employers are doing that. They are much more eager to hire STEM majors. Econ majors, for your listeners who are trying to figure out what they want to do, earn a lot more than business majors. You might think they're the same, but the world treats econ majors almost as well as CS or electrical engineers. When I discovered it, I was kind of amazed. I'm like, what a deal. I always tell my students, economics is the highest paid of all the easy majors. And when they start getting flustered and say "easy," it's like, look, this is not CS. We're not getting vitamin D deprivation from going and doing your work for 72 hours straight underground. Come on. So it's harder than business, but it's not hard. Absolutely, just put in a little bit of effort and you'll get through, as long as you can do pretty basic advanced math. If you can take a derivative of an easy function, you can get through econ. Yes.

So anyway, the major is the single biggest one. Now, there is a big literature on the return to selectivity of colleges. Definitely very subtle differences like Harvard or Chicago. It's very hard to pick up much there. But again, I think you could just say, yeah, well, employers don't distinguish because hardly anyone goes to those schools. They are really close, in fact, in terms of their quality. They are going to distinguish between Harvard and, say, Cal State Northridge, something like that.

There is a puzzle about why don't employers seem to care more than they do about the selectivity of college. It seems like part of it gets picked up in the major because harder colleges, people are more likely to do harder majors. And then another part of what's going on is people are looking at your overall background. So when we say that selectivity doesn't matter much, this is adjusting for a lot of other facts about you. So if we didn't, then we would see that we do actually have a noticeable gain from the selectivity of college.

So I guess my main answer would be that they already are using a lot of information to distinguish it. And then really it comes down to, for some further fine grain differences, why don't they pay more attention to those? And I think the easy answer is the fine grain differences just don't come up that often. And then on top of all of it, so much of worker quality really does have to be observed. So it's enough to get you in the door, but there's enough uncertainty remaining where you can say, well, look, it's the difference between having a 78% chance the person works out and a 76% chance the person works out. That's almost a rounding error. So if I have a little bit of other information on the person that would normally be 76, that can put me over the top and get me to give that person a chance. So I think that what's going on is actually quite a bit more reasonable than you might think if you don't know a lot of the details.

Dwarkesh Patel: Right. Yeah. Okay. So another point you make in the book is that market discrimination is often caused actually by government interference and then it will solve itself in equilibrium. But if you just look at college discrimination of Asians, right? The government is not requiring Harvard to discriminate against Asians.

Bryan Caplan: Yes.

Dwarkesh Patel: And in fact, it lowers the signal that Harvard is able to send if they do aggressive affirmative action. And even actual companies that are in the market proper, like Google or something, they'll do affirmative action too, right? I mean, you could say a little bit of that is required by discrimination law, but probably not to the magnitude they're doing it. So what is the explanation for why that hasn't been solved out by just competition?

Bryan Caplan: Right. So there's multiple things going on. So for colleges, the key thing is that they're nonprofits. So I have this running argument with my colleague Alex Tabarrok. There is a whole economics of discrimination based upon the idea that, for example, if it was really true that women earn 70% as much as equally qualified men, that any moron has a get-rich-quick scheme: just fire all the men at your firm, replace them with equally qualified women, and save a massive amount on your labor costs. So that's sort of the classic economic argument. And I think it's basically sound. But the key premise is precisely that you're dealing with organizations where there is someone that is willing to do something that sounds ugly in order to get rich. And so I'm always telling them, look, this does not apply to nonprofits like academic departments or universities. So the first thing that I would say is that nonprofits, at minimum, the pressure on them to not discriminate is much weaker than for for-profits. So I would distinguish those two quite strongly, actually. And I'd say actually nonprofits, you should definitely expect them to do more discrimination. And the question just is what kind of discrimination are they inclined to do, right?

So this is one where in the old days when there was a lot of anti-Semitism at top schools, then the top schools would be engaged in discrimination against Jews, right? Well, how could they afford to do that? Easy. They're sitting on a mountain of money that isn't theirs, right? They're not going to go bankrupt if they have a tight Jewish quota. It's not like there's any skin off the nose of the admissions department, and therefore they can engage in whatever discriminatory preferences they may have precisely because they are a non-profit.

Dwarkesh Patel: So then what explains the discrimination at Google, for example?

Bryan Caplan: Yes, so I was just about to get to that. Now in the case of for-profit businesses, again, there is a certain kind of superstar company like Google where they are so rich that at least you might say, huh, they could probably discriminate for 100 years too and still survive. So there are a few companies like that where they have just been so incredibly successful that now they're sitting on a mountain of money. And then they're not up against the wall saying, oh my God, if we don't do everything great, then we're going out of business.

Now, it's important to remember, there are very few companies like that. They're famous. Also worth pointing out, the companies that think they're in that position forever, it's not like Harvard. There are a whole bunch of companies that were in the same position as Google that are now bankrupt. So even there, I think the people at Google do have some sense that there are limits to this.

Now, in terms of what the actual role of discrimination law is, I think that in the case of colleges, they are definitely going further than the law requires because they're true believers there. And the true believers are especially concentrated in the admissions departments. So that's where I'd say the main thing is the nonprofits, although even there they do have the threat to hang over the heads of other people in the administration, which is, well, we don't want to be perceived as doing clearly less than other comparable schools. That is an important thing if you want to avoid lawsuits, is to be obviously the group or the organization that cares less than everybody else and does less. So for nonprofits, I think there is a bit of that.

In the case of for-profit companies, I think actually that discrimination law is a major part of the reason why they actually are supportive of affirmative action, because they don't want to be less supportive than any comparable company, at least in an obvious way, for fear that then they will paint a target on their own backs. So I have a couple of pieces that I wrote on this. I think they're going to be in a different book, if I remember correctly. Because actually, this is the first in a series of eight books of my best blog posts on EconLog from 2005 to 2022.

Dwarkesh Patel: Oh, cool.

Bryan Caplan: Right. But anyway, what I say in another piece is that just imagine what would happen if you were a well-known firm and you just officially announced, "We strongly oppose any kind of witch hunt about discrimination. This firm is very fair, and we scrutinize all accusations of racism and sexism very carefully to make sure that you are not making false accusations against an innocent person." Now imagine that you very loudly say this. Right now, of course, there's protests and boycotts and so on that you might get. But on top of all this, if you ever get sued, that is going to be discovered and they are going to introduce it and say, "This is a firm where they actively punish people for trying to insist upon their rights." And you say, "Hey, we were in compliance with the letter of the law." Well, how is a jury likely to see it when they say, "This is what a normal firm looks like, where you have a bunch of propaganda about how we are doing everything in our power to help oppressed groups. And then there's another firm, and then you, and then we show, and what do we have at this firm? This firm is where you say, 'At our firm, the battle against racism and sexism has long since been won. We consider it an extremely minor problem. And our presumption is that accusations will be false. We'll listen, but we are firmly convinced in the fairness of our own system. And so we are going to carefully scrutinize and discipline people who level false accusations.'" But I think if you were that firm, you really would be painting a giant target on your back. It would not be safe.

So, right. And this again is how there has been an expansion of discrimination law over time. So I mean, if you say like sexual harassment law, at the federal level, as far as I know, there is no law against sexual harassment. Rather, this is just a judicial extension of the original '64 Discrimination Act. But it started with saying, all right, well, look, obviously it's illegal to discriminate. That's what the law says. Well, what if you're just treated badly? Well, hmm, well yeah, I guess that could be discrimination too, right? Yes, well, is it discrimination because of your sex? Yes. Well, what if women get paid the same amount, they get promoted the same amount, but there's a bunch of calendars with women in bikinis around the firm? Could that be discrimination? And this is the origin of the law of the hostile workplace.

Now, sometime in the '70s, there was one of my very favorite all-time court cases where there was a bisexual employer who was accused of sexual harassment. And he said, "It's not discrimination. I harass men and women alike. Totally. I am totally within the letter of the law." And if ever a person that was totally legally in the right was that guy, that guy, unquestionably, he had the law on his side, but he lost anyway. So the court basically said, look, we're not going to let bisexuals get away with sexual harassment when other people can't do it. No, we're not going to allow that. And so it continued and then expanded to what we have now, where people can get sued because some employees told jokes or graffitied a bathroom. So that is at the level where we're at.

Now, again, this doesn't mean that most firms have much to worry about, but it's hanging over your head. And it's a reason, at least for prominent firms, to avoid doing anything, because prominent firms are like piggy banks. They're full of money and people are just waiting there. "Hey, Elon Musk has a lot of money. What if we could go and sue him? He could pay. Let's go after him." If it's just some guy's paint store, then probably not much is going to happen to you. Although even there, you're just a little bit worried.

Now, the other thing to remember is that a lawsuit is not just a loss of money. It's also enormously stressful. So don't underestimate just the worry of that. And just think about what it is like if you've been sued once, how do you feel about getting sued twice? I think almost anyone who's been sued once, they like, "That was a terrible experience for me," and they will be trying to think of ways to avoid having it happen again. One thing, of course, is to be super cautious, but there are a lot of ways of being super cautious. There's also the one where, well, I profile people who might sue me and I don't hire them. Something else you can do.

Dwarkesh Patel: Okay. So I want to ask you about mental illness. So you're aware of Hanania's book where he explains that the unitary actor model can't explain American foreign policy. It's not one rational actor. It's these diverse interest groups that are within the establishment.

Bryan Caplan: I think I see where this is going.

Dwarkesh Patel: Right. Right. So yeah, I was just about to say, can't you treat individual personalities this way? So, you know, there's different ways to think about this. Maybe you can think of modules within the brain that are trying to entice you this way or that, and they temporarily could get ahold of you. Maybe, I don't know, personality parts, you know, ego, whatever. There's different ways you can talk about this. But could you apply a similar model to individuals where you say, just as you wouldn't say the average American is responsible for the mistakes of American foreign policy, the individual is not responsible for the weird things that parts of their personality makes them do, right?

Bryan Caplan: Right. So I guess my first reaction to this is just introspection and say, look, I don't experience my mind as being like a bunch of interest groups that are pushing and shoving.

Dwarkesh Patel: But you're not mentally ill, right?

Bryan Caplan: Yes. Well, so here's the thing. There's two versions of it. One is just to say that everybody's like this, and then the mentally ill basically have worse interest groups. And then the other one is to say that normal people are unitary, and then the mentally ill are outliers who are not unitary. So these are two different stories.

The main thing that I would say is there is actually a whole literature on hyperbolic discounting or self-control problems in general that tries to think of the mentally ill as being people that just fail to go and do the best long-run plan and instead focus on short-run plans. What I say is, this is actually just a deeply wrong story about what's going on with at least the severely mentally ill.

Here's the thing. The difference between people that get diagnosed as severely mentally ill and people who just have some moderate problems is precisely that the severely mentally ill are stubbornly convinced that they want to do the objectionable stuff they want to do. So basically, if you're someone that says, "Gee, I have these violent urges, I need to go to a psychiatrist, I'm trying to control my violent urges, what can I do?" All you'll get classified as is a moderately mentally ill person.

But on the other hand, the person who is a full serial killer, every second of their day they're stalking prey. That person seems very unitary, more unitary than most people actually, and yet we call that person severely mentally ill. And that's the person who probably, if they're ever caught, will be found not guilty by reason of insanity. Why? Because they were so organized and determined and focused.

Dwarkesh Patel: Wait, but I would say somebody like Ted Bundy, maybe this isn't how the legal system works. I know Ted Bundy, actually, the story, at least I watched a documentary and I read the Wikipedia article. Or I mean, I watched a fictionalization and then read the Wikipedia. It was, I believe it was the Zac Efron version. Anyway, I think that the fictionalization was quite accurate. So I don't think he's going to be a good example for what you want.

Bryan Caplan: So let's go with it. Okay. Well, so some other serial killer who actually is unitary. I actually don't know the story that well, but I think we have more sympathy for the person who has fits of bad behavior but is generally a nice person. So we do understand, oh, this is a person who is not a unitary actor and therefore we can sympathize with the part of him that's not this way. We sympathize more with them than the person who is wholly mentally ill but is just pursuing a bad aim throughout, right?

So I think that our understanding, but I say there's still something weird about that because you say, okay, so we got a regular person and we're not worried about them, or we think they're a unitary actor. Then you got the severely mentally ill, we think they're unitary. And then we come down to the person with fits, and we say they're the people where our regular story doesn't fit. I mean, that might be right, right? But it's weird. It's like, hmm, so the people that we normally think of as severely mentally ill also seem to be very unitary. So our story is not that they are fighting against uncontrollable urges. Rather, they identify with the horrible urges. There's really nothing to them other than this person who stalks other people to torture and kill them. And this is the main thing they do for fun. It's like, gee, really? I mean, I have read more than my fair share of biographies of serial killers. So, yeah, it's the best story about them. You know, there's a little bit of, "Oh, gee, why am I this way? Oh, well, time to go and murder and torture another person."

In the other case, that's one where I will say that I'm just more confused as to what's going on. Again, there are some cases where at least I am very strongly inclined to say that it still seems to be very strategically designed to avoid social consequences. So someone who is not a homeless-level alcoholic where they drink so much that no one even wants to have anything to do with them, but just someone who is periodically drunk to the point where their family is very distressed about it. Again, it's easy to say, "The devil has a hold of me." But I was like, well, does it though? Or is it more that it's a convenient excuse for you to do some of what you obviously have a very strong desire to do? I'm not saying that is absolutely every case, but still.

I guess the way that I think about mental illness is we have this very large category that has emerged in modern society. It is such a large share of humanity that winds up getting diagnosed. And a lot of what I wind up doing is saying, well, we can cut out a lot of this. A lot of this really doesn't make sense to think of it as a disease. And then we can remove half for some reasons, we can remove another 30% for other reasons, another 10% for some other reasons. And at the end, we're left with some cases where I'll just say, I don't know, maybe it's just hard to say.

So, you know, my son the other day was asking me, like, what would make you most convinced that a person really did have what we think of as a mental illness? And I said, honestly, it would come down to I have known the person really well for a really long time. And they always seem very honest and reliable. And then they say, "I just feel something strange happening to me inside of my head." Right. And he said, "Isn't it weird that it's so heavily based upon firsthand anecdotes?" And I said, well, the problem is that there isn't really any data for a stranger that would convince me they're super reliable. I really do have to have a lot of firsthand experience of the person to take their word for something like that. So that's really where I would come down.

So Thomas Sowell, who inspired a lot of this, he often spoke as if it was just a conceptual truth. And I don't think that's true. I think it's more of one where it's just, what is the story about what's going on in someone else's head that makes the most sense? So when someone is doing something very strange, does it make the most sense to say that person has an illness that makes them do that, or does it make more sense to say that person has just very different goals than I do, but it doesn't mean that they are sick in any meaningful, in anything other than a metaphorical sense?

Now, seeing an example that I thought about quite a bit is like, what if I had a kid who wanted to become a nun? So like, I want to go and marry Jesus and be celibate and have no grandkids. And honestly, if you know anything about the value put on grandkids, that would be one of the worst things for me to hear. And I would just be like, "No, what are you talking about? This is just what I want to do." Right? And yeah, it would be tempting to say, "No, no, you're sick. We have to go and get you a doctor to explain to you how this is all based upon a delusion, that you have this invisible spouse that is going to be your husband but he isn't really real, and you're giving up your actual genetic future for this thing that has imposed itself upon you." But it's not really what's going on. The person is taking seriously a bunch of doctrines a lot of other people believe and trying to live them consistently. And yeah, I object because I think that they're mistaken, right? And of course, I especially object because it's my genes, not just my genes, but also, look, I've got plans for being a grandpa. I'm like, you're cutting into those plans. And I was planning on playing with my grandkids, not going and visiting you at a convent. That's not the life I want for myself, but it's not the life I want for you. So therefore you're sick, right? Very tempting, but just seems like a lie.

Dwarkesh Patel: Yeah. Yeah. All right. Some rapid fire questions. So first is, why is it that multinationals are just so much more effective than companies from developing countries? You would think that they're actually familiar with the culture, they know the problems that the country faces, so they would be much more effective businesses. Why are companies that are founded in America that are foreign to the places where they're working, why are they just so much more effective?

Bryan Caplan: Right. So first of all, just to let people know, this is not just our opinion. There is a very impressive body of research saying this very thing. So most associated with the research, Nicholas Bloom and John Van Reenen. But in any case, I'd say that a lot of the answer is that the culture is the problem. The culture is the problem. So if you have a nepotistic culture, to say, "Well, we need to come up with a hiring practice that works really good in our nepotistic culture." It's like, no, no, no. The whole problem is the nepotistic culture. You got to switch over to a different cultural norm because nepotism is a bad way to run a business. Meritocracy is the way to go. And this is what multinational companies bring in. So that's one of the most obvious ones.

But then similarly, just having a lackadaisical attitude towards punctuality. This is another cultural problem that exists in some countries to a much greater degree than others. Every country has some late people, but some countries, lateness is almost a virtue. I was actually talking to a guy who worked at, I believe it was Toyota in Brazil. And so Brazilians, he told me, they all have a terrible problem with punctuality. And what did Toyota Brazil say?

Bryan Caplan: Like, if you want to work at Toyota Brazil, punctuality, we insist on this. You will be fired, zero tolerance for lateness. This is our system. So they actually, to a much greater extent than in Japan, openly declared war on the cultural norm that it doesn't matter exactly when you show up.

I think that's probably a lot of the answer. The main thing that multinationals do is find the universal truths of business, and then they try to teach them to places where these norms go against the culture, to bring them into the world of modern business, raising productivity and giving them the opportunities that people in other countries have.

Now, to be fair to Toyota Brazil, if you want to work in some inefficiently managed Brazilian firm where they respect Brazilian cultural norms, you can. But Toyota paid so much better that it was considered a highly desirable job. So it's not just that some outside firm comes in and says, "Your culture sucks, you have to do it our way." Rather, there's an outside firm that comes in and says, "If you're willing to bend on the culture, we will give you a lot more money than you usually get in this country." And it turns out a lot of people in those countries are happy to take that deal.

Dwarkesh Patel: So if cultural norms explain so much of the difference between the productivity of firms across different countries, then does that make you less bullish about open borders? Because then you're importing people who have cultural norms that are just making companies ineffective from the places where they come from, right?

Bryan Caplan: Right. Well, I'm not worried, because basically companies in first world countries are very similar to multinationals in terms of what they do. They don't just say, "Whatever people in this area or whatever applicants think is okay is okay." Instead, they have this list of cross-valid rules for the right way to run a business, and then they say, "Look, if you want to be here, you're going to have to do them."

You could go and run your own business where you don't have to show up on time and see how that works for you. But if you want to work at our store, you have to do this. What businesses in the first world do is take workers who on their own would not follow these rules. Just because you are more open to punctuality than people in Brazil doesn't mean you're all that thrilled about it. A lot of what employers do is take people that are not really well suited for a team initially and weld them into a team. This is what coaches for athletic teams do: you take a disorganized rabble and whip them into shape. This is the same thing that businesses do routinely.

You have some managers at the top who know how things need to be, and then you have a bunch of flawed human material that shows up. What are you going to do with it? You could just say, "Oh, I can't work with this material, it's too hard, I give up." But businesses are not run by those people. Instead, the normal thing is you have a manager who says, "First, I know what needs to be done. Second, I need to figure out a way to get people to do it, even though it's not their strong desire anyway." If you read any book on management, it will give you a whole pile of advice on what to do.

I'm a big fan of Dale Carnegie's *How to Win Friends and Influence People*. A lot of what he says is, for example, give everyone a good reputation to live up to. Don't tell people, "You are a terrible late person, I hate you." That's not very motivating. Much better to say, "Notice that you were on time, excellent, great." And then if someone is a little late, say something more along the lines of, "Gee, normally you're on time and I can count on you, and it was really hard for me to run the business today." Obviously there are some people that don't respond to being nice, and people like that generally get fired after a while.

By the way, you're probably familiar with the book by Barbara Ehrenreich, *Nickel and Dimed*?

Dwarkesh Patel: No.

Bryan Caplan: So she went undercover as a famous journalist, did a bunch of low-skill jobs, and then wrote about her experience and how terrible it was and how hard it is to be poor in America. That's the usual write-up of the book. I've read it. I think it's a fantastic book because there's a little bit of propaganda of that kind, but most of it she just describes very neutrally what happened to her. And what happened to her is quite different from what her summary of what happened actually was.

When I read the book, I say, here's a worker who has a very negative attitude, and employers don't appreciate it and try to get her to get a better attitude and do her job. There's a lot of that going on. Furthermore, the other thing you really get out of the book is that people managing very low-skilled workers are probably a lot more emotionally abusive and disrespectful than people managing higher-skilled workers. But when you read the book, you realize why, for two reasons.

One is that with higher-skilled workers, they have a lot more pride in their work. You don't need to be mean to them to get them to do their job well; it's more internally motivated. Whereas with someone who's washing dishes, it isn't like they wake up every day and say, "My calling is to wash dishes." You really do need to be harder on someone to get them to do that job well.

But probably an even bigger factor is: who manages a small town diner? Is it someone who has great social skills and just knows the right way to make everyone feel like an important special person? If you were that good with people, you'd be managing something much better than a small town diner. Most people who manage a small town diner are probably someone who was a good worker in that diner and got promoted, or maybe they started their own business and ran it. So a lot of why it can be unpleasant to be a low-skilled worker is that you're being managed by people whose main experience is being a low-skilled worker. If they had top social skills, they'd be managing a higher status team. That's something else you got out of the book.

Dwarkesh Patel: Right, right. Okay. Final question. What do you think of Tyler's idea in his book *Average is Over*, that basically the labor market is bifurcating into people who can work with computers and people who are getting substituted by computers, and that's the dominant theme of the labor market?

Bryan Caplan: I'd say that the last 20 years of empirical research say that's wrong. Here's my reading of empirical research. From about 1980 to about 2000, there was a fall in the real wages of very low-skilled workers and a corresponding rise in that of very high-skilled workers. But from around 2000, that fall stabilized or moderately reversed. What we've actually seen since then is some fall in mid-skill wages rather than low-skilled. The facts don't really fit that story.

Then during COVID, there have been so many changes in labor markets, but one of the main ones is suddenly there is a great desire to attract those good workers. As to what's going on, I think there's a lot going on. One is the hangover of some of the most enormous redistribution in American history and world history. Even when it's cut off, there are a lot of people who previously were working because they had debts to pay, and then they paid off all their debt during COVID, and now they are in a really good position. Maybe their parents say it's okay for you to live at home without paying rent because of COVID.

There are some people who had enough money to retire early and they're scared of COVID. And then on top of all this, there is still this enormous overhang of fiscal stimulus, which was accommodated very readily by the Federal Reserve. If you look at how much nominal GDP increased, from the bottom to about three months ago when I last looked, nominal GDP increased by 20%. This is an astronomical increase in demand, almost unprecedented. Normally during recessions, the government has some contra-cyclical demand policy to moderate the fall in demand, but there's still a fall. This is one where the response was so enormous that it greatly overpowered any fall in demand. Demand went up, not down.

So that's another thing going on: there's been a very large increase in demand. That means it's very easy for workers of all kinds, but especially workers that deal with other people face-to-face. These are the industries where people are most worried about COVID. A lot of people say, "Hey, can I get a job where I'm just teleworking and I don't have to worry about the disease?" And once people get started doing that, there's a lot of desire to cling to where you were.

So I think that story is basically wrong. There is a rising payoff for IT skills, but this is just one small part of a much larger world. It's a big mistake to think of this as the big thing that's happening. There are a lot of factors going on. In terms of the description of the facts, the period where low workers were doing badly was 1980 to 2000. Since then, and actually before COVID, there was a moderate recovery. And since then, it looks like a big recovery for them.

Dwarkesh Patel: Cool, cool. Awesome. This was a lot of fun, Bryan. Thanks for coming on.

Bryan Caplan: Oh, yeah. My pleasure. Thank you very much. And the book again is *Labor Econ Versus the World*, and you can get it on Amazon. This is the first of a series of eight books of the very best things that I wrote over 17 years. You can get them all for free on the blog, but then you'll have to sift through thousands of posts. So this is where I try to curate them and organize them by theme. If there's some particular topic that interests you more, you can buy the book on that and get all of what I consider to be the best things I've written on that topic.

Dwarkesh Patel: Yeah, let me add that the pricing is incredibly fair on Amazon. It's well at $12 for paperback and $9.99 for the ebook. I love the format of these essay compilations where it makes it really easy to binge read them, just a couple pages at a time. It's a great way to compartmentalize thoughts. So yeah, I highly recommend the book.

Give a shout out to the two groups that gave me the idea for this. Les Rung put out a little gift box of little books of their best essays, and I got one and I really liked it. I read a lot of stuff that I wouldn't normally have read because I had these little hard copy books. Then my friend Mike Humer decided to self-publish an intro philosophy textbook, *Knowledge, Reality, and Value*. When he told me, I asked him what the pain in the neck was, and he said it's super easy. That was the one that said, "Okay, why don't I try this and see what happens?" Then I got the idea of doing eight books. The next one will probably be out in like two months.

Bryan Caplan: Oh, cool.

Dwarkesh Patel: That one will be *Voters as Mad Scientists*? Or no, actually, it will be *How Evil Are Politicians?* Essays on demagoguery. That will be the second one.

Bryan Caplan: I see. Okay. Looking forward to it.

Dwarkesh Patel: All right. Thanks a lot. Talk to you later.