Garett Jones — Immigration, national IQ, & less democracy

Garett Jones 0:00 Go ahead and run your experiments in Iceland. Let's run that for 50 years and see what happens. It's weird how everybody's obsessed with it running the experiment in America, right? They'll balance the long run budget on the, kind of, the backs of the poor and the middle class. Anything that lowers the innovation in the world's most innovative countries has negative costs for the entire planet in the long run. But that's something you'd only see over the course of 20, 30, 50 years, and libertarians and open border advocates are very rarely interested in that kind of timeframe. It's worth thinking through why it is that the successful so-called monarchies aren't really monarchies, right? They're really oligarchies. We should presume that the average skill level of voters, the average traits that we're bringing from our ancestors are having an effect on our current productivity for good or ill.

Dwarkesh Patel 0:41 Okay. Today, I have the pleasure of speaking with Garrett Jones, who is an economist at George Mason University. He's most recently the author of The Cultural Transplant: How Migrants Make the Economies They Move to a Lot Like the Ones They Left but he's also the author of 10% Less Democracy and Hive Mind. We'll get into all three of those books. Garrett, welcome to the podcast.

Garett Jones 1:06 Glad to be here. Thanks for having me.

Dwarkesh Patel 1:09 Um, first question. Is, isn't The Cultural Transplant still a continuation of your argument against democracy? Because, uh, isn't one of the reasons we care about the values of migrants the fact that we live in a democracy? So, should we view this book as part of your critique against democracy rather than against migration specifically?

Garett Jones 1:28 Um, well, I do think that, uh, governments and productivity are shaped by the citizens in a nation in, in almost any event. Um, I think that even, as we've seen recently in China, even in a very strong authoritarian dictatorship, which some would call totalitarian, even there, the government has to listen to the masses. So, the government can only get so far away from the masses on average, even in, uh, an autocracy.

Dwarkesh Patel 1:56 If you had to split apart the contribution though, um, the, the impact of, of migrants on, let's say, the culture versus the impact that migrants have on a country by voting in their political system, um, uh, h- how, how would you split that apart? Is, is the m- is mainly the impact we s- the cultural impact we see for migration due to the ability of migrants to vote or because they're just influencing the culture just by being there?

Garett Jones 2:20 I'll cheat a little bit because we don't get to run experiments on this, so I just have to kind of guess, uh, make an informed guess. I, I'm gonna call it 50/50. Um, so the way people, uh, the way citizens influence a country through formal democracy is important, uh, but citizens end up placing some kind of limits on the government anyway. And the people in a country are the, they're the folks who are gonna work in the firms and be able to either establish or not establish those complicated networks of exchange that are crucial to high productivity.

Dwarkesh Patel 2:52 I wanna linger on hive mind a little bit before we talk about The Cultural Transplant. Um, if you had to guess, do, do the benefits of national IQ come from having a right tail of elites that is smarter or is it from not having that strong of a left tail of people who are, you know, lower productivity, more li- likely to commit crimes and things like that?

Garett Jones 3:12 Uh-

Dwarkesh Patel 3:12 In other words ... Yeah, go ahead.

Garett Jones 3:14 Yeah, yeah. I, I think, uh, the upper tail is gonna matter more than the lower tail, um, in, uh, in the normal range of variation. Uh, and I think part of that is because, uh, nations, at least moderately prosperous nations, have found tools for basically reducing the influence of the least informed voters and for, uh, basically being able to keep productivity up even when there are folks who are sort of disrupting the whole process. Um, you know, the, the, the risks of crime from the lower end is basically like a probabilistic risk. It's not like it's, it's not like some, uh, zero to one switch or anything. So, we're talking about something probabilistic, and I think that, uh, it's the, uh, the median versus the elite is the, is the contrast that I find more interesting. Um, uh, median voter theorem, you know, normal, the way we often think about democracy says that the median should be, matter more for determining productivity and for shaping institutions. Um, and I tend to think that that's more important in democracies for sure. So, when we look at countries, if you just look at a scatter plot, just look at the raw data of a scatter plot, if you look at the few countries that are exceptions to the rule where the mean is the br- mean IQ is the best predictor of productivity compared to elite IQ, um, the exceptions are non-democracies and South Africa. So you see a few, uh, places in the Gulf where there are large migrant communities who are exceptionally well-educated, exceptionally cognitively talented, um, and that's associated with high productivity. Those are a couple of Gulf states. It's probably Qatar, the UAE, might be Bahrain in there. I'm not sure. Um, and then you've got South Africa. Those are the, those are the countries where the average test scores, it doesn't have to be IQ, could be just PISA, TIMSS type stuff, um, those are the exceptions to the rule that the average IQ, the mean IQ is the best predictor of national productivity.

Dwarkesh Patel 5:14 Hmm. Uh, interesting. Um, uh, d- does that imply the fact that the, um, at least in certain contexts, the elite IQ matters more than the left tail, does that imply that we should want a greater deviation of IQ in a country?

Garett Jones 5:30 Um-

Dwarkesh Patel 5:30 That you can just push a button and increase that deviation. Would that be good?

Garett Jones 5:33 No, no, I don't think so. Uh, oh, if you could just increase the deviation, um, holding the mean constant-

Dwarkesh Patel 5:39 Yeah.

Garett Jones 5:39 ... yeah, I think so in the normal range of variation. Yeah, yeah, yeah. So ...

Dwarkesh Patel 5:43 Mm-hmm. Is it ... so-

Garett Jones 5:44 Um, and I think that it has more effects, it, you know, it's people at the top who are, um, tend to be coming up with the, the big breakthroughs, the big scientific breakthroughs, the big intellectual breakthroughs that end up spilling over to the whole world, basically, the, the positive externalities of innovation. This is a very almost Pollyanna-ish, uh, Paul Romer new endogenous, new, uh, new growth theory thing, right? Which is-... the innovations of the elite just swamp, uh, the negatives of the low-skilled among us.

Dwarkesh Patel 6:14 Can we just apply this line of reasoning to low-skilled immigration as well then? That maybe the average goes down, the average IQ of your country goes down if the, if you just let in, you know, millions of low-skilled immigratio- uh, immigrants, and maybe there's some cultural effects of that too. But, you know, you're also going to... The- that, the elite IQ will still be preserved, and more elites will come in through the borders along with the low-skilled migrants. So then, you know, since we're caring about the d- deviation anyways, uh, m- more immigration might increase the deviation, uh, and then, you know, the, we just, uh, eh, that's a good thing?

Garett Jones 6:47 So notice what you did there, is you d- you did something that didn't just, uh, increase the variant. You simultaneously increased the variant and lowered the mean-

Dwarkesh Patel 6:55 Yeah, yeah.

Garett Jones 6:55 ... and median, right? And so I think that, uh, hurting the mean and median is actually a big cost, especially democracies. And so that is very likely to swamp, uh, the benefits of, um, the small, the small probability of getting higher elite folks in as part of a low-skilled immigration policy. So pulling down the mean or the median is... That, that's a, that's, that swamps, that swamps the benefits of increasing variance there. Yeah.

Dwarkesh Patel 7:26 Yes. But if you get rid of their, migrants' ability to vote, and I guess you can't do th... But let- assume you could do that.

Garett Jones 7:31 Yeah.

Dwarkesh Patel 7:31 What exactly is, like, uh, h- what is the exact mechanism by which the, the, the cultural values or the lower median is impacting the elite's ability to produce these valuable externalities? You know, like, there's a standard comparative advantage story that, you know, they'll, they'll do the housework and the cooking for the elites and make them-

Garett Jones 7:51 Yeah.

Dwarkesh Patel 7:51 ... even more productive.

Garett Jones 7:52 Yeah, well-

Dwarkesh Patel 7:53 Yeah.

Garett Jones 7:53 ... taking all the institutions as given, which is what a lot of open borders optimists do, they take institutions as given, they take cultural norms as given, um, all that micro stuff works out just fine. I'm totally, I'm totally on board with all that sort of Adam Smith division of labor, blah, blah, blah. Um, but institutions are downstream of culture. And, uh, cultural norms will be changing, partly because of what I call spaghetti theory, right? We meet in the middle. When new folks come to a country, there's some kind of convergence, some part where people meet in the middle, um, between the, the values, uh, that were previously existing and the values that have shown up, uh, that migrants have brought with them. So, you know, like I, I call it spaghetti theory because, um, when Italians moved to America, that got Americans eating more spaghetti, right? And if you just did a simple assimilation analysis, you'd say, "Wow, everybody in America eats the same now. Like, they eat burgers and spaghetti. So look, the Italians assimilated." But migrants assimilate us. Um, uh, Native Americans certainly changed in response to the movement of Europeans. Um, English-Americans certainly changed in response to the migration of German and Irish-Americans. So this meeting in the middle is something that happens all the time, and not just through democratic channels, just through the sort of soft contact of cultural norms that sociologists and social psychologists would understand.

Dwarkesh Patel 9:15 Um, now, I'm sure you saw the book that was released, I think, in 2020 c- titled, uh, Wretched Refuse, uh, wh- where they showed a slight positive relationship between, uh, immigration-

Garett Jones 9:25 (clears throat)

Dwarkesh Patel 9:25 ... and, you know, pro-market, uh, laws, and I guess the idea behind that is there's selection effects in terms of who would come to a country like America in the first place, but also-

Garett Jones 9:33 But they never ran the statistical analysis that would be most useful, I think. They said that, uh... So this is Powell and Nowrasteh.

Dwarkesh Patel 9:40 Yeah.

Garett Jones 9:40 They ran a statistical analysis that said, and they said, "In all of the statistical analyses we've ever run, we've never found negative relationship between low-skilled migration, any measure of it, and changes in economic freedom." And, um, I actually borrowed another one of Powell's datasets, and I thought, "Well, how would I check this theory out, the idea that changes in migration have an effect on economic freedom?" And I just used the normal economist tool. I thought about, "How do economists check to see if changes in money, changes in the money supply change the price level?" That's what we call the quantity theory, right?

Dwarkesh Patel 10:14 Mm-hmm.

Garett Jones 10:15 The way you do that is on the x-axis you, you show the change in the money supply, and on the y-axis you show the change in prices, right? This is Milton Friedman's idea, money is always in-

Dwarkesh Patel 10:24 Yeah, yeah, yeah.

Garett Jones 10:24 ... everywhere a- ... Yeah. Inflation's always and everywhere monetary phenomenon. So that's what I did. Uh, I did this with a, with a, um, a student. Uh, we co-authored a paper doing this, and the very first statistical analysis we ran, we looked at migrants who came from countries that were substantially more, uh, corrupt than the country's average, and we looked at the, the different, the relationship between chan- an increase in migrants from corrupt countries and subsequent changes in economic freedom. Every single statistical analysis we found had a negative relationship.

Dwarkesh Patel 10:58 How-

Garett Jones 10:58 We ran the simplest-

Dwarkesh Patel 10:59 How-

Garett Jones 10:59 ... estimate you could run, right? Change on change. Change in one thing predicts change in another. They somehow never got around to running that very simple statistical analysis. Ch- one change predicts another change.

Dwarkesh Patel 11:11 Hmm. But, but what about-

Garett Jones 11:12 We found negative relationships every time, sometimes statistically significant, sometimes not, always negative. Somehow they never found that. I just don't know how.

Dwarkesh Patel 11:21 (laughs) But what about the anecdotal evidence that in the US, for example, the, in the periods of the greatest expansion of the welfare state or of governed power during the New Deal or Great Society, the levels of foreign-born people were at, like, historical lows? Uh, i- is that just a coincidence, or w- what do you think gives, uh...

Garett Jones 11:38 I'm not really interested in, uh, migration per se, right? My story is never that, like, migration per se does this bad thing, migrants are bad. That's never my story, right?

Dwarkesh Patel 11:48 Mm-hmm.

Garett Jones 11:49 As you know, right?

Dwarkesh Patel 11:50 Yeah, yeah.

Garett Jones 11:51 So my story is that migrants bring, uh, cultural values from their old country to their new country, and that sometimes those cultural norms are better than what you've got, and sometimes they're worse than what you got, and sometimes it's just up for

Garett Jones 12:02 debate.

Dwarkesh Patel 12:03 So if you had to guess, what percentage of the world has cultural values that are equivalent to or better than the average of America's?

Garett Jones 12:11 Oh, equivalent to or better than?

Dwarkesh Patel 12:13 Yeah.

Garett Jones 12:14 Uh...I mean, just off the top of my head, maybe 20%? I don't know. 30%? I'll just-

Dwarkesh Patel 12:19 Yeah.

Garett Jones 12:19 ... throw something out there like that, yeah.

Dwarkesh Patel 12:21 So, I mean, like-

Garett Jones 12:21 Per country averages, right? Yeah, yeah.

Dwarkesh Patel 12:24 Yeah, yeah. Um, currently we probably don't have, uh, w- w- well, it would probably be hard for like 20% of the rest of the world to get into the US. Um, w- w- would you support some pro- uh, possib- policy that would make it easy for people from those countries specifically to get to the US? Just, uh, have radical l- immigration liberalization from those places?

Garett Jones 12:44 Um, that's really not my comparative advantage to have opinions about that. But like, substantial increases of people who pass multiple tests. Like, let's take the low-hanging fruit and then move down from there, right? So, people from, uh, countries, uh, that ha- um, on average have, say higher savings rates, um, higher, uh, education levels, higher S... what I call SAT Deep Root Scores, and, um, countries that are, say, half a standard deviation above the US level in all three, right?

Dwarkesh Patel 13:19 Why do they have to be higher? Why not just equivalent? Like, uh-

Garett Jones 13:22 I mean-

Dwarkesh Patel 13:22 ... you, you got all the gains from, like, trade and plus it can't be, you know, equivalent so it's, there's no tr- trade off.

Garett Jones 13:27 No. Part of the reason is because the entire world depends on US innovation. So, we should make America as good as possible, not just slightly better than it is. So, very few firms would find that their optimal hiring policy would be, hire anyone who's better than your current stock of employees. Would you agree with that?

Dwarkesh Patel 13:42 Yeah, but you have to pay them a salary if you're just, uh, if s- somebody just comes to the US, you don't have to like pay them a salary, right? So, if somebody is better, th- if somebody's producing more value for a firm than the salary you would pay them, I think like, that probably makes sense.

Garett Jones 13:53 Is, is a firm's job to maximize its profits or to just make a little bit more than it's making right now?

Dwarkesh Patel 13:58 Maximize profits? But-

Garett Jones 14:00 Yeah, there you go.

Dwarkesh Patel 14:00 Okay.

Garett Jones 14:00 So you pa- you find the best people you can.

Dwarkesh Patel 14:03 No-

Garett Jones 14:03 You know, uh, Forbes teams that are hiring don't just say, "We want to hire people who are better than what we got." They say, "Let's get the best people we can get. Why not get the best?"

Dwarkesh Patel 14:11 Yeah, they can get-

Garett Jones 14:11 That was Jim- Jimmy Carter's, that was Jimmy Carter's, uh, biography, Why Not the Best?

Dwarkesh Patel 14:16 But you can do that along with getting people who are, you know, in e- on expected, uh, terms but y- as good as the existing Americans. That gives you all the

Narrator 14:24 microphone.

Garett Jones 14:25 Why would you... I don't get why, why you would want this. This seems like crazy, right? What are you talking about?

Dwarkesh Patel 14:29 I don't see how... But I- I- I'm not sure-

Garett Jones 14:31 Why not the best?

Dwarkesh Patel 14:31 ... what's the trade off there?

Garett Jones 14:32 Huh?

Dwarkesh Patel 14:33 No. No, I'm not saying you don't get the best. But I- I'm saying once you've gotten the best, what is the harm in getting the people who have equivalent SAT scores and, and the rest of the things you mentioned?

Garett Jones 14:43 I think part of the reason would be, you'd want to find out... I mean, if you really want to do something super hardcore, you'd have to find out what's best for the planet as a whole. What's the trade-off-

Dwarkesh Patel 14:50 Mm-hmm.

Garett Jones 14:50 ... between, um, having the very best, uh, most innovative, talented, frugal people in America doing innovating that has benefits for the whole world, versus having an America that's like 40% better, but where the median's a little bit, the median of skill is a little bit lower, right? Uh, because the median's shaping the productivity of the whole team, right? This is what you, you know-

Dwarkesh Patel 15:11 Yeah.

Garett Jones 15:11 ... what it means when you believe in externalities, right?

Dwarkesh Patel 15:14 But if you have somebody who's equivalent, by definition, they're not moving the median down.

Garett Jones 15:19 That's... You're, you're totally right about that.

Dwarkesh Patel 15:22 Yeah.

Garett Jones 15:22 But like, why wouldn't I want the best thing possible, right?

Dwarkesh Patel 15:26 Okay.

Garett Jones 15:26 I'm still trying to figure out why you wouldn't want the best thing possible. You're trying to-

Dwarkesh Patel 15:29 I just don't... I'm not-

Garett Jones 15:29 ... come up with a story for why I don't want the best thing possible.

Dwarkesh Patel 15:30 ... disagreeing with you.

Garett Jones 15:30 I'm like, "Why not?"

Dwarkesh Patel 15:32 I'm not disagreeing with you. I'm just-

Garett Jones 15:33 Yeah, yeah.

Dwarkesh Patel 15:33 ... I'm a little bit confused about why that, uh, that precludes you from also getting the second best thing possible at the same time. You- 'cause you're not limited to just the best, right?

Garett Jones 15:43 Well, eh, because the second best is going to have a negative externality on the first best. Everything's externalities. This is my worldview, right? Everything's externalities. You bring in the second best-

Dwarkesh Patel 15:54 But it seems like-

Garett Jones 15:54 ... you're like, you're not... that person's gonna make things on average a little worse for the first best person.

Dwarkesh Patel 16:00 But it seems like you were explaining earlier that the negative externalities are coming from people, from countries with, uh, low SAT scores. And by the way, SAT, you can explain what that means just for the audience who's not familiar with how you're using that term.

Garett Jones 16:11 Oh, yeah.

Dwarkesh Patel 16:12 Um-

Garett Jones 16:12 So, uh, there, there's three prominent, uh, measures in what's known as the Deep Roots literature, and, uh, that are widely used. Uh, two are S and A, that's state history and agricultural history. That's how many thousands of years your ancestors have had experience living under organized states or living under settled agriculture. And then, the T score is the tech history score. I use the measure from 1500. It's basically, what fraction of the world's technology were your ancestors using in 1500 before, uh, Columbus and his expansive conquest ended up upending the entire world, um, the world map? So S, A, and T are all predictors of modern prosperity, but especially when you adjust for migration.

Dwarkesh Patel 16:54 Gotcha. We can come back to this later, but one of the interesting things I think from the book was you have this chapter on China, and the Chinese people as a sort of unstoppable force for free market capitalism.

Garett Jones 17:06 Mm-hmm.

Dwarkesh Patel 17:06 Um, and it's interesting, as you mention in the book, that China is the poorest majority Chinese country. Um, what do you think explains why China is the poorest, um, majority Chinese country? Maybe, are there like non-linear dynamics here where, uh, if you go from 90... 40 to 90% Chinese, there's positive effects, but if you go from 98 to 95% Chinese, there's too much

Narrator 17:26 No.

Garett Jones 17:26 I think it's just... I think just communism is dumb and has terrible like sometimes decades long effects on institutional quality that I don't cr- really quite understand. So, I'd say North Korea, if we had good data on North Korea, North Korea would be an even a bigger sort of Deep Roots outlier than China is, right? Like don't, don't have a communist dictatorship in your country seems to be pretty, a robust lesson for, uh, national prosperity. China's still stuck with a sort of crummy version of that mistake still. North Korea, of course, is stuck with an even worse version. So, I think that's... I... My hunch is that that's, you know, the overwhelming issue there. Um, it's, it's something that it's, it's sort of a... China's stuck in an in... Currently China's stuck in an institutional cul-de-sac and they just don't quite know how to get out of it, and it's, uh, bad for a lot of the people who live there on average. If the other side had won the Chinese Civil War, things would probably be a lot, lot better off in China today.

Dwarkesh Patel 18:22 Yeah, um, uh, but what- what does that suggest about the deep roots literature if the three biggest countries in the world, China, India, and America, um, it- it underpredicts their performance? Or sorry, in the case of China and India, it, uh, it- it overpredicts their performance, and in the case of America, it underpredicts. Does that suggest that maybe the ... H- how reliable is this if, like, the three biggest countries in the world are not, uh, adequately accounted for?

Garett Jones 18:45 Uh, well, you know, communism is a really big mistake. I- I think that's totally accounted for right there. Um, I think India's underperformance isn't that huge. Um, the U.S. is a miracle along many ways. Um, it's ... We should draw our lessons from the typical country and I think, uh, population-weighted estimate, I don't think that basically one-third of the knowledge about the wealth of nations comes from the current GDP per capita of China, India, and the U.S., right? I think much less than one-third of the story of the wealth of nations comes from those three. And, uh, again in- in all three cases though, if you look at the economic trajectories of all three of those people, or all three of those countries, uh, they're all, uh ... China and India are growing faster than you'd expect. And also I want to point out, this is the most important point actually, um, when we look at, uh ... When Caplin made this claim, right? Bryan Caplin has made this claim, right?

Dwarkesh Patel 19:37 Yep. Yep.

Garett Jones 19:37 That the SAT ... That the ancestry scores, the deep roots scores don't predict, um, the prosperity of, uh, the- the low performance of India and China. He only checked the S and the A in the SAT scores.

Dwarkesh Patel 19:53 Okay.

Garett Jones 19:53 Which letter did he not predict? Which letter did-

Dwarkesh Patel 19:55 The T.

Garett Jones 19:55 ... he never test out?

Dwarkesh Patel 19:57 Yeah, yeah, yeah.

Garett Jones 19:57 He never tested the T. What do you think happens when you test the T?

Dwarkesh Patel 20:01 Does it predict, uh, China and India and America?

Garett Jones 20:03 Uh, they start, they ... T goes back to being statistically significant again.

Dwarkesh Patel 20:07 Uh-huh.

Garett Jones 20:08 So with T, which we've always known is the best of the deep roots scores, somehow Caplin never managed to measure that one. Just as Powell and Norostei never managed to run the simplest test, change in, uh, migrant corruption versus change in economic institutions. Somehow, like, the simplest tests just never get run.

Dwarkesh Patel 20:26 Okay. And then w- what is the impact if you include T?

Garett Jones 20:29 I mean, if you- if you look at T, then, um, then, uh, contrary to what Caplin says, uh, the deep roots, that deep roots measure is sig- statistically significant.

Dwarkesh Patel 20:38 Okay. Um-

Garett Jones 20:39 Yeah.

Dwarkesh Patel 20:40 ... interesting.

Garett Jones 20:40 The puzzle goes away.

Dwarkesh Patel 20:41 (laughs) Interesting. Um-

Garett Jones 20:44 Yeah, so somehow these guys just never seem to run, like, the simple things, the transparent things. I don't know why.

Dwarkesh Patel 20:49 The, um, the, um-

Garett Jones 20:50 Weird, huh?

Dwarkesh Patel 20:51 The, uh, the- the one you mentioned from, uh, w- was it Norostei, the name of the guy who wrote the Wretched Refuse? The-

Garett Jones 20:58 Yeah, yeah.

Dwarkesh Patel 20:58 That was-

Garett Jones 20:59 Powell and Norostei, yeah. Yeah.

Dwarkesh Patel 21:00 That ... You said you did the regression on institutional corruption, uh, and from the countries they come from? Is that ... Is that right?

Garett Jones 21:06 I, and so yeah, the- the measure they use, I just took, I took Powell's-

Dwarkesh Patel 21:09 Yeah.

Garett Jones 21:09 ... dataset from another study.

Dwarkesh Patel 21:11 Yeah.

Garett Jones 21:11 And it was the percentage of ... It was basically, um, the percentage of your nation's population, the percentage increase in your nation's population from relatively poor or corrupt countries. They had multiple measures.

Dwarkesh Patel 21:24 Uh-huh.

Garett Jones 21:25 So-

Dwarkesh Patel 21:25 And- and what is on the y-axis there?

Garett Jones 21:27 Y-axis is change in economic freedom, that's my preferred one.

Dwarkesh Patel 21:30 Gotcha.

Garett Jones 21:31 There's also a change in corruption one, which is a noisier indicator. Um, you get much clearer results with change in economic freedom, so.

Dwarkesh Patel 21:38 Gotcha, gotcha.

Dwarkesh Patel 21:39 Um, now does the idea it's getting harder to find stuff and great stagnation, does that imply we should be less worried about impinging on the innovation engine in these, uh, countries that people might want to migrate to? Because worse comes to worst, it's not like there are a whole bunch of great new theories that were going to come out anyways.

Garett Jones 21:58 Uh, no. I think that, I think that it's always good to have great things, um, and new ideas. Yes, new ideas are getting harder to find, but, um, that, but that the awesome ideas that we're still getting are still worth so much, right? If we're still increasing lifespan a month a year, uh, for every year of research we're doing, like, that just seems great, right? A decade, that adds a year to life, so.

Dwarkesh Patel 22:22 Mm-hmm.

Garett Jones 22:22 Just to use a rough, uh, ballpark measure there.

Dwarkesh Patel 22:25 But, so we have a lot of these countries where a lot of innovation is happening. So, let's say we kept, uh, one or two of them as, you know, immigrate- uh, s- havens from any potential, uh, downsides from radical changes. You know, we already have this in the case of Japan or South Korea. There's not that much migration there.

Garett Jones 22:41 Mm-hmm.

Dwarkesh Patel 22:41 What is, uh, w- what is the harm in then using the other ones to decrease global poverty by immigration or something like that?

Garett Jones 22:48 Well, um, it's obviously better to create a couple of innovation powerhouses, um, rather than none, right? So obviously, that's- that's ight. But instead, I would prefer to have, um, open borders for Iceland. If the open borders advocates are right and open borders will have not- noticeable effect on institutional quality, then it's great to move, have our open borders experiment run in a country that's lightly populated, has a lot of open land, and, um, has good institutional quality. And Iceland fits the bill perfectly for that. So, we could preserve the institutional innovation skill, uh, the institutional quality of the- the, what I call the I7. Uh, that's, you know, China, Japan, South Korea, the U.S., Germany, UK, France. And choose any country out of the hundred, out of the couple of dozen countries that have good institutional quality. Just pick one of the others that aren't one of those seven. Pick one that's not an innovation powerhouse and turn that into your open borders, uh, country. Um, you could, uh, if you wanted to get basically Singapore levels of population density in Iceland, that'd be about 300 million people, I think. I think that- that's about what the numbers end up looking like, something like that.

Dwarkesh Patel 24:00 But the entire-

Garett Jones 24:01 So you could put-

Dwarkesh Patel 24:01 But- but, uh, the value of open borders comes from the fact that you're g- coming to a country with high conglomerations of talent and capital and other things, which is, uh, n- not true of Iceland, right? So, the, i- i- isn't the entire (laughs) place an open border?

Garett Jones 24:13 No, no. In fact, the whole point of open borders is that there's institutional quality and there's some exogenous institutions that make that place more productive than other places.

Dwarkesh Patel 24:21 Mm-hmm.

Garett Jones 24:22 And so if I move, I've- I ... That's my version of what I've been exposed to as open borders theory is that institutions exogenously exist. There's some places have, uh, moderately laissez-faire institutions in their country, and moving a lot more people there-... will not reduce the productivity of the people who are currently there, and they'll become much more productive. And so like, the institute, you know, the Institute for Quality is crucial. So, I mean, if you're a real geography guy, you'd be excited about the fact that Iceland is so far, so close to the North Pole-

Dwarkesh Patel 24:51 Yeah.

Garett Jones 24:51 ... because latitude is a predictor of prosperity.

Dwarkesh Patel 24:54 Um, I want to go back to the thing about w- well should we have open borders so that 20% of the popula- uh, global's, uh, world's population that comes from-

Garett Jones 25:01 Yeah.

Dwarkesh Patel 25:01 ... um, equivalent SAT and other sort of cultural traits as America.

Garett Jones 25:05 Mm-hmm.

Dwarkesh Patel 25:06 Um, 'cause I feel like this is important enough to dwell on. Yeah, you know, it seems someone is saying that once you've picked up a $100 bill on the floor, you wouldn't pick up a $20 bill on the floor 'cause you only want the best bill. Uh, the $20 bill is right there, why not pick it up? Um, so what if we have-

Garett Jones 25:20 Yeah, what if the $20 bill makes your, turns your, uh, $100 bill into like an $80 bill and turns all of your 80, $100 bills into $80 bills?

Dwarkesh Patel 25:27 But isn't, aren't you controlling for that by saying that they have equivalent scores along all those cultural tests that you're considering?

Garett Jones 25:34 No, because, um, the median ... So, so take the simple version of my story which is the median of the population ends up shaping the productivity of everybody in the country, right? Or the mean.

Dwarkesh Patel 25:44 Right.

Garett Jones 25:45 The mean skill level ends up shaping the productivity of the entire population, right? So that means we end up, I mean, I, I try not to math this up, I don't want to math this up for the, in a, in a popular book, but it means we face a trade-off between being small, a small country with super awesome, uh, positive externalities for all the workers by just selecting the best people. And every time we lower the average skill level in the country, we're lowering the average productivity of everyone else. We're creating-

Dwarkesh Patel 26:11 But what if we didn't, what if we didn't lower it? So e- e- you have to have skills that are equally or greater-

Garett Jones 26:14 Lower compared to the-

Dwarkesh Patel 26:16 ... than the median of a median American. You know, the 80-

Garett Jones 26:18 So this, this is a ceteris paribus story, right? Like, if you could... Suppose the US is at 80 now on a zero to a hundred scale, right? Let's just, just say, it's 80.

Dwarkesh Patel 26:27 Yeah, yeah.

Garett Jones 26:28 And you have a choice between being 100 and being 99. If you're 99, the 99 is making all, is ... Compared to the world of average of 100, the world of an average 99 is making-

Dwarkesh Patel 26:42 Yeah, yeah.

Garett Jones 26:42 ... reducing the productivity of all those 100s.

Dwarkesh Patel 26:45 Okay, so let's say-

Garett Jones 26:45 If we chose 90, we're reducing the productivity of all those 100s.

Dwarkesh Patel 26:48 Yes. Okay, so let's say we admit all the smartest people in the world, and that gets us-

Garett Jones 26:52 Yeah.

Dwarkesh Patel 26:52 ... from 80 to 85.

Garett Jones 26:55 Yeah, yeah.

Dwarkesh Patel 26:55 That's the new a- that's the new median America.

Garett Jones 26:57 Yeah.

Dwarkesh Patel 26:57 At that point, and but this is because we've admitted a whole bunch of like 99s that have just increased our average.

Garett Jones 27:02 Yeah, yeah.

Dwarkesh Patel 27:02 Um, at that point, open borders for everybody who's above an e- 85?

Garett Jones 27:09 Like I d- a- this is, this is ends up being a math problem that's a little hard to solve on a podcast, right? 'Cause it, it's the, it's the question of do I want a smaller country with super high average productivity or a bigger country with lower average productivity? And by average productivity, I don't just mean, uh, a compositional effect, I mean negative externa- I mean relatively fewer positive externalities. So I'll use the term relatively fewer positive externalities rather than negative externalities, right? So like, I don't exactly know where this trade-off's gonna pan out, but um, there is a case for a sort of Manhattan. When people talk about a Manhattan project, right? They're talking about putting all, like a small number of the smartest people in a room, and part of the reason you don't want like the 20th smartest person in the room is 'cause that person's gonna ruin the, ruin stuff for all, for the other smart people.

Dwarkesh Patel 27:58 I- I'm-

Garett Jones 27:58 When we th-

Dwarkesh Patel 27:58 ... agree with this.

Garett Jones 27:59 It's amazing how your world view changes when you see everybody as an externality.

Dwarkesh Patel 28:03 I- I'm kind of confused about this because, uh, th- just having w- a- at some point you're gonna run out of the smartest pe- the remainder of the smartest people in the world if you admitted all the brilliant people.

Garett Jones 28:14 Yeah, yeah.

Dwarkesh Patel 28:14 And given how big the US population is to begin with, you're not gonna change the median that much by doing that, right? So it's, it's almost equivalent to just having more births from the average American, like if, if the average American just had more kids, the population would still grow.

Garett Jones 28:27 Mm-hmm.

Dwarkesh Patel 28:27 And the relative effect of the brightest people might dilute a little bit. Um, but I- I-

Garett Jones 28:33 And that, maybe that's a huge tragedy, we don't know. Without a bunch of extra math and a bunch of weird assumptions, we don't know. So like I'm, there's a point at which I have to say like, "I don't know," right?

Dwarkesh Patel 28:42 Okay. Yeah, uh-

Garett Jones 28:43 Yeah, is diluting the power of the smartest person in America like keeping us from having wondrous miracles all around us all the time? I mean, probably not, but I don't know.

Dwarkesh Patel 28:54 But, but I guess the sort of the meta question you can ask about this entire debate is listen, there's so much literature here and it's hard to tell what exactly will happen, you know, it's possible that culture will become worse, it's possible it will become better, or it's possible it will stay the same. Given the fact that there's this ambiguity, why not just do the thing that on the first order of effect seems good and you know, just like moving somebody who's like in a poor country to a rich country. First order effect seems good, I don't know how the third and fourth order effect shapes out, let's just, you know, let's just do the simple obvious thing.

Garett Jones 29:22 I, I thought that the, one of the great ideas of economics is that we have to worry about secondary and tertiary consequences, right?

Dwarkesh Patel 29:28 But if we, if we can't even figure out what they are exactly, why not just do the thing that at the first order seems, uh, good?

Garett Jones 29:35 Um, because if you have a compelling reason to think that the, uh, direction of strength of the second and third and fourth order things are negative and the variances are really wide, then you're just adding a lot more uncertainty to your outcomes, so. And adding uncertainty to your outcomes that has a sizable negative tail, especially for the whole planet, isn't that great. Go ahead and run your experiments in Iceland. Let's run that for 50 years and see what happens. It's weird how everybody's obsessed with it running the experiment in America, right? Why not run it in Iceland first?

Dwarkesh Patel 30:04 'Cause America's got great, a lot of great institutions, right there which is

Narrator 30:07 Yeah.

Garett Jones 30:07 But we can check and see what that Iceland's gr- Iceland's a great place too.

Dwarkesh Patel 30:11 Um, are-

Garett Jones 30:11 And I use Iceland as a metaphor, right? Like it's people are obsessed-

Dwarkesh Patel 30:14 Yeah, yeah.

Garett Jones 30:14 ... with running it in America like there's some kind of need, I don't know why. So let's try in France. Um, let's try-

Dwarkesh Patel 30:22 Are, are-

Garett Jones 30:22 Let's try Northern Ireland.

Dwarkesh Patel 30:24 Uh,

Dwarkesh Patel 30:25 are, are places with low SAT scores, and again, SAT we're not talking about the, um, uh, in case you're skipping to this timestamp, we're not talking about the college test. Um-

Garett Jones 30:34 The deep root SAT.

Dwarkesh Patel 30:35 Exactly.

Garett Jones 30:35 Uh, state history, agricultural history, tech history.

Dwarkesh Patel 30:38 Right, exactly. Are, are those places with, uh, low scores on, um, on that test...Are they stuck there forever, or b- w- is there something that can be done if you are a country that has had a short or not significant history of, um, technology or agriculture?

Garett Jones 30:56 Well, the, I start off the book with this, which I really think that, uh, one thing they could do is, uh, create a welcoming environment for large numbers of Chinese migrants to move there persistently. I don't think that's, of course, the only thing that could ever work, but I think it's something that's within the range of policy for at least some poor countries. I don't know which ones, but, uh, some poor countries could follow the, uh, approach that many countries in Southeast Asia followed, which is create an environment that's welcoming, uh, welcoming enough to Chinese migrants. Um, it's the one country in the world with large numbers of high SAT score, uh, uh, with a h- lar- with a high SAT score culture, large population, and it's enough of an economic failure s- for at least a little longer that, uh, folks can, might be able to be interested in moving to a poorer country with lower SAT scores. In a better world, you could do this with North Korea too, but the population of North Korea isn't big enough to make a big dent in the world, right?

Dwarkesh Patel 31:50 Mm-hmm.

Garett Jones 31:51 Uh-

Dwarkesh Patel 31:51 Yeah.

Garett Jones 31:51 ... China's population is big enough. Yeah.

Dwarkesh Patel 31:54 Another thing you w- have to worry about in those cases though, though, is the risk that if you do become successful in that country, there's just gonna be a huge backlash and your resources will get expropriated.

Garett Jones 32:04 Yeah.

Dwarkesh Patel 32:04 Like what happened to, um-

Garett Jones 32:04 Famously so in, in Indonesia, right? Yeah, there have been many-

Dwarkesh Patel 32:07 Oh, yeah.

Garett Jones 32:08 ... times across-

Dwarkesh Patel 32:08 Yeah, yeah.

Garett Jones 32:08 ... Southeast Asia where anti-Chinese pogroms have been, um, uh, unfortunately a fact of life. So...

Dwarkesh Patel 32:15 Yeah, yeah. Or Indians in Uganda under, uh-

Garett Jones 32:17 Mm-hmm.

Dwarkesh Patel 32:17 ... Idi Amin.

Garett Jones 32:18 Idi Amin, yeah. Yep.

Dwarkesh Patel 32:19 Yeah. Yep, yep. Um, yeah, yeah.

Garett Jones 32:21 Yeah.

Dwarkesh Patel 32:21 Uh, okay so, uh, l- l- actually, le- I'm, I'm curious how you would think about this given the impact of national IQ,

Dwarkesh Patel 32:26 um, if you're an effective altruist, w- uh, are you just, uh, handing out iodine tablets, uh, across, across the world? What, what are you doing to increase national IQ-

Garett Jones 32:34 Yeah, this is-

Dwarkesh Patel 32:35 ... in these places that desperately need it?

Garett Jones 32:35 ... this is something that I... Yes. Uh, finding ways, uh, this is what I call a, a Flynn cycle. Like I wish... I'm hoping for a world where there are enough public health interventions and probably K through six education interventions to boost test scores in the world's poorest countries, and I think that ha- ends up having, um, uh, a virtuous cycle to it, right? As people get more productive then they can afford more public health, which makes them more productive, which means they can afford more public health. I think brain health is an important and neglected part of child development. Um, fortunately we've done a fair amount to reduce the amount of environmental lead, um, in a lot of poor countries that's probably having a good effect right now as we speak in a lot of the world's poorest countries. You're right. Um, iodine, basic childhood nutrition, uh, reliable healthcare, uh, to, you know, prevent the worst kinds of just mild childhood infections that are probably, uh, creating what the, what they s- what economists sometimes call health insults. Things that end up just hurting you in a way that causes, uh, an ill-defined long-term cost. A lot of that's gonna have to show up in the, in the brain. Um, I'm a big fan of the v- of the view that part of the Flynn effect is, uh, basically nutrition and health. Uh-

Dwarkesh Patel 33:51 Mm-hmm.

Garett Jones 33:51 ... Flynn wasn't a huge believer in that, but I think that's, um, certainly important in the poorest countries. Yeah.

Dwarkesh Patel 33:57 Um, uh, uh, well I, I think Bryan showed in Open Borders that if you look at the, um, IQ of adoptees from poor countries, um, who go, uh, Sweden is the only country that collects data, but if you get-

Garett Jones 34:10 Yeah.

Dwarkesh Patel 34:10 ... adopted by a parent in, um, uh, Sweden, uh, y- the, the half the gap between the averages of those two countries-

Garett Jones 34:18 Like by half the gap. Yeah, yeah, yeah.

Dwarkesh Patel 34:18 Yeah, yeah. Goes away.

Garett Jones 34:19 Yeah.

Dwarkesh Patel 34:19 So I mean, is one of the ways we can increase global IQ just by moving kids to, uh, countries with good health outcomes that, uh, will nourish their intelligence?

Garett Jones 34:28 Oh, that's a classic short run versus long run effect, right? So, uh, libertarians and open borders advocates tend to be focused on the short run static effects. So, um, and so you're right. Moving kids from poor countries to richer countries is probably gonna raise their test scores quite a lot. And then the question is in the over the longer run, are those, uh, lower skilled folks, the folks with lower test scores, uh, going to degrade the institutional quality of the place that they move to, right? So if you close half the gap between the poor country and the rich country, half the gap is still there, right? And if I'm right that IQ has big externalities, then moving people from a, a lower scoring country to a richer scoring country and closing half the IQ gap still means on net you're creating a negative externality in the country the kids are moving to.

Dwarkesh Patel 35:18 Um, yeah, yeah. Uh, w- we can come back to that but-

Garett Jones 35:22 Yeah, yeah.

Dwarkesh Patel 35:22 ... uh-

Garett Jones 35:22 So it, it's, it basically you just look at the question, is this lowering the mean test scores in your country? And if it's lowering the mean test scores, in the long run it's on average gonna lower institutional quality, productivity, savings rates, those things, um, it's hard to avoid that, it's hard to avoid that outcome. So...

Dwarkesh Patel 35:38 Uh, I don't remember the exact figures but didn't Bryan address this in the book, uh, in the Open Borders book as well that you can, e- even if there's a, the a- national IQ, uh, lowers on average, if you were just, uh, if you're still g- raising the global IQ that that it still nets out positive or am I remembering that wrong?

Garett Jones 35:55 Well, th- notice what he's d- he, what he does is he attributes, uh, he says there's some productivity that's just in the land that's just geographic factors. So basically-

Dwarkesh Patel 36:04 Yeah.

Garett Jones 36:04 ... being closer for... And so that, basically moving people away from the equator boosts productivity substantially. And again that's a, a static result. Um, the reason I, uh, I mention that ignores all the I7 stuff that I'm talking about where anything that lowers the, um, level of innovation in the world's most innovative countries has negative costs for the entire planet in the long run. But that's something you'd only see over the course of 20, 30, 50 years and libertarians and open border advocates are very rarely interested in that kind of timeframe.

Dwarkesh Patel 36:34 Is there any evidence about, uh, the impact of migration on innovation specifically? So not on the average institutional quality or on, you know, uh, the, the corruption or whatever but like just directly the amount of innovation that happens or maybe the Nobel Prizes won or things like that?

Garett Jones 36:49 Um, no. I mean, I would presu- I think a lot of us would presume that, uh, the European invasion of North America ended up having, uh, positive effects for global innovation. It's not an invasion that I'm in favor of, but if you wanna talk crudely about-

Dwarkesh Patel 37:01 Yeah, yeah.

Garett Jones 37:02 ... whether migrations had an effect on innovation, uh, you'd probably have to include that as any kind of analysis.

Dwarkesh Patel 37:07 Yeah, yeah. Do you think that the people who are currently Americans, but their ancestry traces back to countries with low SAT scores, i- i- is it possible that U.S. GDP per capita would be higher w- without that contribution? Or how do you think about that?

Garett Jones 37:21 I mean, that, it follows from thinking through the fact that we are all externalities, positive or negative, right? I don't know what in- any particular, any one particular country could turn out to be some exciting exception to the rule, some interesting anomaly. Um, but on average, we should presume that the average skill level of voters, the average, uh, traits that we're bringing from, uh, the nations that, uh, the nations of our, of our ancestors are, is having an effect on our current productivity, for good or ill. It's just following through the reasoning. I'd say on average that's most likely, uh, but it, there could always be exceptions to the rule.

Dwarkesh Patel 37:57 I guess we see large disparities in income between different ethnic groups across the world, not just in the United States.

Garett Jones 38:05 Yeah.

Dwarkesh Patel 38:05 Doesn't that suggest that some of the gains can be privatized from whatever the cultural or other traits there are? 'Cause i- i- if these, if over decades and centuries these sorts of, uh, these sorts of gaps continue.

Garett Jones 38:18 I don't see why that would follow, right? Um...

Dwarkesh Patel 38:21 Uh, w- if everything is being, if all the externalities are just being averaged out over time, wouldn't you expect that these g- gaps would, uh, narrow?

Garett Jones 38:30 Well, I mean, I'm being a little rhetorical when I'm saying everything's literally an externality, right? I don't literally believe that's true. Uh, for instance, people with higher education levels do actually earn more than people with lower education levels, so it, that's literally not an externality, right? So, some of these other cultural traits that people are bringing with them from their, um, ancestors, nations of origin, um, could be one, are ones, likely one source of these income differences. I mean, if you think about differences in frugality, uh, differences in personal responsibility, which show up in the surveys, uh, that are persistent across generations, those are likely to have an effect on long run productivity for you yourself and your family.

Dwarkesh Patel 39:06 Mm-hmm.

Garett Jones 39:07 Let alone the hive mind stuff where we find that there's a positive relationship between test scores and, and productivity.

Dwarkesh Patel 39:13 There was a blogger who l- took a look at your 2004 paper about the, um, impact of national IQ on, um, on G, uh, GDP. Um, and they calculated, so they were just speculating. Let's say you cloned a million John von Neumanns and a- assumed that John von Neumann had an IQ of 180. Then you could, uh, let, let me just pull up the exact numbers. You could, um, you could raise the average IQ of the United States by .21 points, um, and if it's true that one IQ point contributes 6% to, uh, G, increasing GDP, then this proposal would increase US GDP by, uh, 1.26%. Do you buy these kinds of extrapolations or are they-

Garett Jones 39:55 1.26%?

Dwarkesh Patel 39:57 Yeah, yeah. 'Cause you're only quoting a million John von Neumanns.

Garett Jones 39:58 Oh, yeah. Okay, so it's about one million John von Neumann. Yeah, yeah. That sounds, uh, I mean, that's the kind of thing where I wouldn't expect it to happen overnight, right? I tend to think of that s- uh, the IQ externalities as being two, three generations. I, I lump it in with what economists call organizational capital. That sounds about right. Yeah, yeah, yeah.

Dwarkesh Patel 40:15 Yeah.

Garett Jones 40:15 I mean, I, I can't remember-

Dwarkesh Patel 40:16 But-

Garett Jones 40:16 ... where I saw this. I think I re- I stumbled across it myself at some point too. So...

Dwarkesh Patel 40:19 Yeah, yeah. Uh, by the way-

Garett Jones 40:20 Yeah.

Dwarkesh Patel 40:20 ... his name is Alvaro de Menard, if you wanna find out more.

Garett Jones 40:22 Oh, okay. Yes, yes. Okay.

Dwarkesh Patel 40:24 Um...

Garett Jones 40:24 Yeah, it's, I mean, and it's in that ballpark, right? It's just this idea that, and, and more importantly, um, a million John von Neumanns would be a gift to the entire planet, right?

Dwarkesh Patel 40:33 Yep. Yep, yep.

Garett Jones 40:34 So yeah, if you had a, if you had a choice of which country to have the John von Noi- the million John von Neumanns, uh, it's probably gonna be one of the I7. Maybe there, maybe there's a, maybe Switzerland would be ano- good alternative.

Dwarkesh Patel 40:46 What is the optimal allocation of intelligence across a country? 'Cause one answer, and I guess this is the default answer in our society, is you just send them where they can get paid the most because that's a good enough proxy for how much-

Garett Jones 40:59 Uh-huh.

Dwarkesh Patel 40:59 ... highly they're contributing.

Garett Jones 41:00 Yeah.

Dwarkesh Patel 41:00 And so you have these high agglomerations of talent and intelligence in places like Silicon Valley or New York, um, and, you know, because their contributions there can scale to the rest of the world, this is actually where they're producing the most value. Another is, you know, you actually should disperse them throughout the country so that they're helping out communities. They're, you know-

Garett Jones 41:18 Mm-hmm.

Dwarkesh Patel 41:18 ... teachers in their local community. Um, I think there was, uh, a result, th- there was an interesting anecdotal evidence that during the Great Depression, the crime in New York went down a ton and that was because the cops in New York were able to hire th- you know, they had like 100 applications for every cop they hired. And so they-

Garett Jones 41:36 Mm-hmm.

Dwarkesh Patel 41:36 ... were able to hire the best and the brightest, and th- there were just a whole bunch of new police tactics and ev- that were pioneered at the time.

Garett Jones 41:42 Mm-hmm.

Dwarkesh Patel 41:42 Anyways, so b- uh, wh- sh- is the market allocation of intelligence correct or do you think there should be more distribution of intelligence across the country? How do you think about that?

Garett Jones 41:50 Yeah, I mean the m- the, the market signals aren't terrible. Uh, but uh, this is my, my inner Paul Romer kicks in and says, uh, "Innovation is all about externalities and there's market failures everywhere when it comes to innova- the field of innovation."

Dwarkesh Patel 42:04 Mm-hmm.

Garett Jones 42:04 And so, you know, I, I personally, I mean, I, I like the idea of finding ways to allocate them to, to STEM sty- STEM style technical fields, and, uh, we do a fair amount of that and maybe we do the, maybe the US does a pretty good job of that. I don't have any huge complaints at that, at the cr- at the crudest 50,000 foot level, um, for our, the w- you know, the fact that people know that there's, uh, status games they can play within academia that are perhaps more satisfying or at least as satisfying as the sorta corporate hierarchy stuff. So...

Dwarkesh Patel 42:40 Yeah.

Garett Jones 42:41 Yeah, I, I, you, you don't want 'em all just, I wouldn't encourage them to solely follow market signals, right? I'd, I'd encourage them to be more Hansonian and, uh, play a variety of status games because the academic, um, and intellectual status game is worth a lot, both personally and then it leads to positive spillovers for society.

Dwarkesh Patel 42:58 But how about the geographic distribution? Do you think that it's fine that there's people glea- uh, smart people leave Kentucky and go to San Francisco or, uh-

Garett Jones 43:08 Yeah, I'm a big conglomeration guy. Yeah. I'm, I'm-

Dwarkesh Patel 43:10 Yeah, yeah?

Garett Jones 43:11 Yeah, I'm a big conglomeration guy, yeah.

Dwarkesh Patel 43:13 Talk-

Garett Jones 43:13 I mean, the internet makes it easier, but then like, still being close to people is... Being in the room is important.

Dwarkesh Patel 43:20 Um-

Garett Jones 43:20 That, there, there's something, uh, both Hansonian and Girardian here about like, we need to find-

Dwarkesh Patel 43:25 (laughs)

Garett Jones 43:25 ... role models to imitate and that's probably important for productivity.

Dwarkesh Patel 43:30 Um, ar- are there increasing or decreasing returns to national IQ?

Garett Jones 43:38 No. I think, um... You know, my findings were that it was all basically log linear, and so log linear looks, uh, crudely like increasing returns, right? So, yeah, it looks exponential, right? So yeah, there's increasing returns to national IQ, yeah.

Dwarkesh Patel 43:55 Are- are you talking about-

Garett Jones 43:56 'Cause h- but, but this is, this is a commonplace finding in a sense because so many, uh, like huma- all the human capital relationship I'm familiar with end up having something like a log linear form, which is exponential, so.

Dwarkesh Patel 44:09 Why is that?

Garett Jones 44:10 Um, yeah, there's something multiplicative. That, that's h- what I have, that's all I have to say is like, it's something, somehow this all taps into Adam Smith's pin factory, and we have multiplicative, not additive effects when we're increasing brain power.

Dwarkesh Patel 44:27 Um-

Garett Jones 44:27 I ha- I suspect it does have something to do with, uh, a, a better organization of the division of labor between people, which ends up happening, something close to e- to, uh, exponential effects on productivity.

Dwarkesh Patel 44:39 Uh, ar- uh, are you a fan of genetic selection for intelligence, uh, as a means of increasing national IQ or do you think that's too much playing at the margins?

Garett Jones 44:49 If it's voluntary, I mean, people should be able to do what they want and, um, after a couple day- decades of experimentation, I think people would end up finding a path to, uh, government subsidies or tax credits or something like that. I think vol- people voluntarily deciding what kind of kids they want to have, um, is, uh, a, a good thing and so by genetic selection, I assume you're meaning at the most elementary level, people testing their embryos the way they do now, right?

Dwarkesh Patel 45:17 Yeah. Yep.

Garett Jones 45:17 So I mean, we al- we, we already do a lot of genetic selection for intelligence. Um, anybody you know who's uh, uh, in their mid-30s or beyond who's had amniocentesis, they've been doing a form of genetic selection for intelligence. So it's a widespread practice already in our culture. Um, and, uh, welcoming that in a voluntary way is probably going to have good effects for our future.

Dwarkesh Patel 45:40 What do you make of the fact that GPT-3 or I think it was ChatGPT-

Garett Jones 45:44 Yeah.

Dwarkesh Patel 45:44 ... had a measured IQ of 85? Do you think that-

Garett Jones 45:47 Yeah, I've seen a few different measures of this, right? You might have seen multiple measures. Um, yeah, I think it's, I think it's a sign that basically, and, and when you see people using non-IQ tests to sort of assess the outputs of GPT on, um, long essays, it does, does seem to fit into that sort of not quite 100, but not, not off by a lot. Yeah, I mean, I think it's, I think it's a sign that a lot of, uh, uh, mundane, even fairly complex, moderately complex human interactions could be simulated by a large, uh, language learning model.

Dwarkesh Patel 46:19 Mm-hmm.

Garett Jones 46:20 And I think that's, that's, uh, gonna be rough news for a lot of, uh, people whose life was in the realm of words and dispensing simple advice and solving simple problems. That's pretty bad news for their careers. I'm, I'm disappointed to hear that, so.

Dwarkesh Patel 46:36 Yeah. Yeah. Um-

Garett Jones 46:37 At least for the transition.

Dwarkesh Patel 46:38 ... uh, ho-

Garett Jones 46:38 I don't know what the, I don't know what's gonna happen after the transition, but.

Dwarkesh Patel 46:41 Yeah. I'm hoping-

Garett Jones 46:42 Yeah.

Dwarkesh Patel 46:42 ... that's not true of programmers or economists like you.

Garett Jones 46:46 I mean, it might be, right? I mean, it's, if that's the way it is, I mean, I, the, I mean, the car put a lot of, uh, people who took care of horses right out of, out of work too, so.

Dwarkesh Patel 46:55 Yeah. Yeah. Um, you, okay, so let's talk about democracy. That, I thought this was also one of your really interesting books.

Garett Jones 47:01 Oh, thanks. Yeah.

Dwarkesh Patel 47:02 Uh, e- even controlling for how much democratic oversight there is of institutions in the government, there seems to be a wide discrepancy of how well they work. Like, the Fed seems to work reasonably well. I, I, I don't know enough about macroeconomics to know how, the object level decisions they make, but-

Garett Jones 47:18 Mm-hmm.

Dwarkesh Patel 47:19 ... you know, it seems to be a non-corrupt like, uh, technocratic organization. Um, the, the-

Garett Jones 47:23 Enough, yeah.

Dwarkesh Patel 47:24 But it, yeah, yeah. Uh, if you look at something like the FDA, it's also somewhat insulated from democratic processes. It seems to not work as well.

Garett Jones 47:31 Mm-hmm.

Dwarkesh Patel 47:32 What determines, c- controlling for democracy, what controls, like what, what determines how well an institution in the government works?

Garett Jones 47:38 Well, I, I, I think, um, in the case of the Fed, it really does matter that they, uh, the people who run it have a guaranteed long term and they print their own money to spend. So that means that the, basically Congress has to really make an effort to change anything at the Fed. So they really have the kind of independence that matters, right? You know, they have a room of their own. And, uh, the FDA has to come to Congress for money more or less every year. And the FDA, uh, heads do not have any kind of security of appointment, they're appoi- they serve at the pleasure of the President.

Dwarkesh Patel 48:14 Mm-hmm.

Garett Jones 48:14 So I do think that they don't have real independence. Uh, I do think that they're basically, um, they're living in this slack, this area of slack to use the sort of McNugget poli sci jargon. They're living in this realm of slack between the fact that the President doesn't want to me- m- uh, muddle with them, uh, meddle with them, excuse me, and the fact that Congress doesn't really want to meddle with them, but on the other hand, I really think that, that the FDA and the CDC are doing what Congress more or less wanted them to do. They reflect-

Dwarkesh Patel 48:49 Mm-hmm.

Garett Jones 48:49 They reflect the muddled disarray that Congress was in over the period of, say, COVID.

Dwarkesh Patel 48:55 Hmm.

Garett Jones 48:56 Uh, that, I think that's of first order importance. I mean, I do think the fact, it's the fact that, uh, FDA and CDC don't ha- uh, seem to have that culture of, um-... raw technocracy the way the Fed does. That, I think that has to be important on its own, but I think behind that, some of that is just, like, FDA, CDC creatures of Congress much more than the Fed is.

Dwarkesh Patel 49:17 Should the power of the president be increased?

Garett Jones 49:20 Uh, n- no. No, like, the power of independent committees should be increased. Like, more of Congress should be like the Fed. Uh, my plan for a Fed re- for an FDA or a CDC reorganization would be about making them more like the Fed, where they have appointed experts who have long terms and they have enough of a long term that they can basically feel like they can blow off Congress and build their own culture.

Dwarkesh Patel 49:42 Mm-hmm. So, uh, y- the European Union is an interesting example here because they also have these appointed technocrats, but they seem more interested in creating anno- annoying pop-ups on your websites than with dealing-

Garett Jones 49:56 Yeah.

Dwarkesh Patel 49:56 ... with economic, the, you know, the end of economic growth on the continent. I- i- is this a story where more democracy would have helped or how do you think about the European Union in this context?

Garett Jones 50:05 No, in the EU, like, uh, the European, European voters just aren't that excited about democracy, or excuse me, aren't that excited about markets overall. The EU is going to reflect that, right? Uh, what little evidence we have suggests that, uh, countries that are getting ready to join the EU, they improve their economic freedom scores, their sort of laissez-fairness-

Dwarkesh Patel 50:24 Hmm.

Garett Jones 50:24 ... uh, on the path to getting ready for, uh, joining EU. So, and then they may increase it a little bit afterwards once they join. But basically, it's like, it's like, uh, when you're deciding to join the EU, it's like you decided to have your Rocky training montage and get more laissez-faire. And so-

Dwarkesh Patel 50:41 Well, right.

Garett Jones 50:41 ... the EU on net is a mes- it pulls in the direction of markets compared to where, uh, Europe would be otherwise. I mean, just look at the nations that are in the EU now, right? A lot of them are, um, east of Germany, right? And so those are countries that don't have this great, you know, uh, history of being market-friendly and a lot of parties aren't that market-friendly, and yet the EU sort of nags them into their version, like, as much markets as they can handle, so ...

Dwarkesh Patel 51:05 What do you think explains the fact that the Europe- uh, Europe as a whole and the voters in there are less market-friendly than Americans? I mean, if you look at the sort of deep roots analysis of Europe, you would think that they should be the most, uh, mo- mo- most in favor of m- I- I don't know if the deep roots, uh, actually, maybe the-

Garett Jones 51:22 I mean-

Dwarkesh Patel 51:22 ... planet as such, but-

Garett Jones 51:23 Yeah, compared to the planet as a whole, they're pretty good, right? So, um, I- I'm, I never get that excited about, like, the small little distinctions between the US and Europe, like these 30% GDP differences which are very exciting to pundits and bloggers and whatever, and like, 30% doesn't matter very much, that's not really my bailiwick. What I'm really interested in is the 3000% between the poorest countries and the richest countries. So like, I can speculate about Europe. I, I don't really have a great answer. I mean, I, I think there's something to the, the naive view that, um, the Europeans with the most, uh, what my dad would call gumption are those who left and came to America, some openness, some adventurousness, uh, and maybe that's part of what trans- uh, made we, so basically there's a lot of selection working, um, on the migration side to, uh, make America more open to laissez-faire than Europe would be.

Dwarkesh Patel 52:14 Does that overall make you more optimistic about migration to the US from e- uh, anywhere, like, you know, the same story of

Dwarkesh Patel 55:15 If you look at the median voter in terms of their preferences on economic policies, it seems like they're probably more, um, in favor of government involvement than the actual policies of the United States, for example.

Garett Jones 55:29 Yeah.

Dwarkesh Patel 55:30 What explains this? Shouldn't the median voter theorem imply that we should be much less libertarian as a country than it actually is?

Garett Jones 55:35 Yeah, that's a great point from, um, Bryan Caplan's excellent-

Dwarkesh Patel 55:38 Yeah.

Garett Jones 55:39 ... book, The Myths of the Rational Voter, right? Yeah, I think part of it, I mean, I think his stories are right which is that, uh, politicians facing re-election have this trade-off between giving voters what the voters say they want and giving the voters the economic growth that will help the politicians get reelected, right?

Dwarkesh Patel 55:53 Mm-hmm.

Garett Jones 55:54 Um, so it's a, it's a version of saying like, you know, "I don't want you to pull off the Band-Aid but I want my wound to all get better." So, then so the politician has to, it's the politician's job to handle the contradictory demands of the voters. And by delegating authority to them, uh, to the voter, to the elected politicians, you get some of the benefits of elitism, um, even in a so-called democracy.

Dwarkesh Patel 56:19 Um, over the long run, should we expect any of the tensions or all of the tensions of ethnic diversity to fade away? Like, nobody today worries about the different Parisian tribes in France butting heads at the workplace, right? So over time-

Garett Jones 56:32 Yeah.

Dwarkesh Patel 56:32 ... we shouldn't lose this-

Garett Jones 56:33 And, and you're right, like, that, uh, the j- uh, anti-German ethnic sentiment in the US, totally gone, right? So-

Dwarkesh Patel 56:39 Right.

Garett Jones 56:39 ... yeah.

Dwarkesh Patel 56:40 But over time, so then this is, this is another one of those short-run effects that you, uh, emphasize we should focus less on, right?

Garett Jones 56:46 Yeah, that's a good point. Um, it's possi- y- you don't know which w- I mean, the problem is that ethnic, ethnic conflict has been a hardy perennial. It's not the only conflict that people can ever have. I don't, I don't know to what extent, uh, they'll, these things will fade away. As I emphasize in The Culture Transplant, the ethnic diversity channel is actually the m- least important of any of the channels I discuss. Um, and so I'm open to this thought that what you're saying will actually happen, and maybe we'll just find something else to get mad at each other about, like, um, uh, social media tribes or, uh, religi- religious groups. Um, I mean, it hasn't happened yet in all the documented human history we have. People seem to find some ethnic, uh, bounds for conflict. It is worth pointing out that the one, one, um, study that I report, uh, I think it's a Wagsia coauthor piece, find that when the, the real, uh, the source of ethnic conflict happens when private values are correlated with, um, ethnic groups, right? So if, uh, cultural values are basically uncorrelated with ethnicity, then basically there's nothing to fight over, and that's really what's happened with a lot of old ethnic battles in the US.

Dwarkesh Patel 57:56 Mm-hmm.

Garett Jones 57:56 And, um, so you're right. Some of these things will fade with time. The problem is that human beings are, one of our great evils is that we are always looking for a focal point and we can, people will use visible appearance as a horrifying focal point around which to, uh, peg their conflicts. It's an easy one because our brains are looking for visual patterns, and I don't like that but it's something that will probably keep happening.

Dwarkesh Patel 58:22 One of the interesting points you made in that chapter was that the, uh, benefits of diversity are greatest when search costs are lower and the costs of vetting are lower. What do, h- h- how do we make sure that that is true of non-elite professions? So if you're looking for a plumber or if you're looking for a carpenter, how, how do you make sure that you can vet them easily?

Garett Jones 58:42 I mean, I have to say that this has to be a case where, like, you know, Yelp and Google and all these online ratings have given us tools for checking these things out. We know we have to be skeptical, of course, but, uh, for people who know that they're good at something, the cost of entry into a new field has to be much lower than it was a few decades ago because, you know, 10, 20 good Google reviews, um, and you can actually enter. So yeah, lower, basically not, not banning, not banning disclosure of data, I'd say that's the most important thing we can do.

Dwarkesh Patel 59:14 Mm-hmm.

Garett Jones 59:15 Um-

Dwarkesh Patel 59:15 I think, um-

Garett Jones 59:16 ... you sometimes hear that medical doctors, I haven't checked up on this in a long time, but apparently medical doctors often, uh, make it very risky to give bad reviews. Sometimes you get laws, you get a lawsuit or something.

Dwarkesh Patel 59:26 Mm-hmm.

Garett Jones 59:26 So making that a lot harder is worth it. You know, we know that some negative reviews are gonna be malicious and inaccurate, but the benefits of information flow seem really high.

Dwarkesh Patel 59:38 Yeah. I thought one of the really interesting chapters in the, The Ten-Percent-Less Democracy was the chapter on, you know, bondholder democracy.

Garett Jones 59:46 Yeah.

Dwarkesh Patel 59:47 And I'm curious, so I mean, corporations are ex- uh, obviously an example to use here where they do have bondholders who hold them accountable.

Garett Jones 59:54 Mm-hmm.

Dwarkesh Patel 59:54 But the average lifespan of a corporation is ten years, I believe. So do you think it'd be, it would, um, it would be even shorter if bondholders had a lesser say on corporations or, you know, w- what is the, what does the transience of the corporation tell us about their controls?

Garett Jones 1:00:11 Oh, that's a good point. Um, well, we, we can suspect that the average person who's investing in a corporation makes money, right? Because, uh, otherwise people wouldn't be doing it, right? On average, people-

Dwarkesh Patel 1:00:20 Right.

Garett Jones 1:00:20 ... it must work. Um, so this is one where I, you actually have me stumped here. So can you rephrase the question again? I'm trying to think through what the-

Dwarkesh Patel 1:00:30 Sure.

Garett Jones 1:00:30 ... what the question is there. Yeah.

Dwarkesh Patel 1:00:32 If bondholders do extend the longevity and the long-term-

Garett Jones 1:00:37 Yeah.

Dwarkesh Patel 1:00:37 ... thinking of the organizations on, on whom-

Garett Jones 1:00:39 Oh, yeah.

Dwarkesh Patel 1:00:39 ... they hold bonds, why aren't corporations who give out bonds, why don't they tend to live longer than, I think, the average of ten years? I- would it be even shorter without bondholders or uh-

Garett Jones 1:00:50 Oh, I, I'd say, I, A, I'd say it'd probably be shorter without bondholders or any-

Dwarkesh Patel 1:00:54 Yeah.

Garett Jones 1:00:54 ... kind of financial monitor. But second, most corporations just shouldn't live that long, right? Most corporations-

Dwarkesh Patel 1:00:59 (laughs)

Garett Jones 1:00:59 ... they're ideas that you try out and then it, like, you find out it, it doesn't work or it should be bought up by somebody else or the IP should be sold off. And so having a lot of companies, uh, fade out is actually on net a good sign. I think this is really part of the John Haltiwanger line of research, the, the sort of modern version of creative destruction research...... which finds that, you know, uh, low productivity firms exiting, you know, uh, just as naive laissez-faire predicts means that those workers and that capital can get reallocated over to a more productive firm. So, the alternative is, uh, you know, stereotypically Japanese zombie firms, right? They're kept limping along by banks that are perhaps under political pressure to lend and so a lot of a- a lot of ca- human and physical capital gets tied up in low productivity projects.

Dwarkesh Patel 1:01:45 Yeah.

Garett Jones 1:01:45 So yeah, um, uh, a brutal bond market is, uh, a good way to send a market signal to move capital from low productivity to high productivity projects.

Dwarkesh Patel 1:01:55 Um, why are yields on 30-year, uh, fixed, um, treasuries, why are they so low? Because theoretically, investors should know that we have a lot of liabilities in the form of, you know, Social Security to baby boomers-

Garett Jones 1:02:09 Yeah.

Dwarkesh Patel 1:02:09 ... and that we've radically inflated the money supply very recently and may do so again. Uh, do you think the investors are being irrational with the, uh, low yields or wh- what's going on?

Garett Jones 1:02:19 No, no, I'm a fan of the view that the bond holders are gonna win in the long run and that, uh, inflation, any kind of super, like, super, super high inflation is not gonna be the path, uh, of the future. And what's gonna happen is that, at least thinking about the US, um, the way the bond holders are gonna win is that there's gonna be a mixture of tax hikes and slower spending growth, especially hurting the poor, and that's how the US is gonna close its fiscal gap. I don't know particularly what paths other countries are gonna go through, but the US has this basically, um, this one tool, this one superpower sitting, um, sitting in the room that it hasn't used yet, and it's, uh, it's a VAT, right? So, the US could dramatically increase its tax revenue through either an overt or disguised value-added tax, and that would raise a ton of money just like it raises in Europe and that would... That's the easy way to close the US fiscal gap. We probably won't even have to get to that. Just making Medicaid worse, slowly over the long run-

Dwarkesh Patel 1:03:20 (laughs)

Garett Jones 1:03:20 ... um, maybe making Medicare worse over the long run.

Dwarkesh Patel 1:03:24 Right.

Garett Jones 1:03:24 Um, that, that by itself would close a lot of this fiscal gap. So basically, I think they'll balance the long run budget on the, kind of, the backs of the poor and the middle class. That's probably the most likely outcome.

Dwarkesh Patel 1:03:34 So are you expecting-

Garett Jones 1:03:35 Hence no, hence no hyperinflation, so.

Dwarkesh Patel 1:03:37 Are, are, but so you're expecting the welfare st- state to shrink either in quality or quantity over time?

Garett Jones 1:03:43 In relative terms, uh, compared to the trend. I mean, if America's still getting, say, 1%, 1.5% richer each year, um, then that by itself, uh, means that you'll, you know, that adds a certain level of quality to healthcare over time and so basically, if it, if it ends up staying the equivalent of, say, 0%, um, that by itself... If you get healthcare spending in, in real terms to be 0% over time, um, that, that would end up closing the gap when you compound it out long enough, right? So-

Dwarkesh Patel 1:04:15 Yep, yep.

Garett Jones 1:04:16 Yeah.

Dwarkesh Patel 1:04:17 Um-

Garett Jones 1:04:17 That kind of thinking.

Dwarkesh Patel 1:04:19 So, when Liz Truss tr- in the UK tried to implement, uh, tax reform-

Garett Jones 1:04:23 Uh-huh.

Dwarkesh Patel 1:04:23 ... there was a bond holder revolt and, you know, she was ousted.

Garett Jones 1:04:26 Yeah.

Dwarkesh Patel 1:04:27 Uh, this is-

Garett Jones 1:04:27 Bond holders won right there. Yeah. (laughs)

Dwarkesh Patel 1:04:29 So do you, do you, are we, do we already live in the bond holder utopia or (laughs) what, what do you-

Garett Jones 1:04:33 Oh yeah, I, I mean, I think that that was a nice, uh, reminder that basically contrary to sort of the MMT view, um, and the sort of pop MM, the pop Keynesian view that debt is no barrier at all, um, I think that showed the bond holders, uh, are actually paying attention to long-term signals of fiscal policy credibility and they'll take action, yeah.

Dwarkesh Patel 1:04:56 Mm-hmm.

Garett Jones 1:04:57 Yeah. So.

Dwarkesh Patel 1:04:57 Wh- wh- what are the deep roots of Mormonism? Why do they have such high trust and, uh, uh, such tight-knit communities?

Garett Jones 1:05:04 I mean, I think part of it is that they are, they reflect a lot of this sort of, uh, what for them was Western pioneer culture, upstate New York, Pennsylvania, then Ohio. Um, I think those communities tended to be, uh, high trust communities necessarily because of the difficult, uh, environment that they were living in. Um, and there was a lot of selection. There was a lot of selection over the first few decades of Mormon history where those who were willing to sort of trust the group stayed in, and those who weren't willing to trust the group, um, ended up leaving. And not just trust, but trustworthiness. I think a lot of p- people probably got weeded out because they weren't contributing to the common good. So, I think that basically by the time the Mormons got established in Utah, uh, they, they'd already selected for a strong culture of, um, a kind of, a kind of in-group prosociality, and I think that's, um, helped them, that helped them weather the storms of the 19th century. For the whole 19th century, people were only joining, uh, during this, this is during the era of polygamy in Utah, right? Um, if they thought that they were willing to put up with this, right? And you know you're, you're signing up for some kind of deep sociality with a mixture of unco- a lot of unconventional stuff. And that foundation, uh, really helped. And the fact that Mormons since then have stayed as a religion that requires a medium high level of s- of commitment also weeds out people who just aren't willing to make that kind of commitment.

Dwarkesh Patel 1:06:36 Mm-hmm.

Garett Jones 1:06:36 So, I mean, I was, I was raised Mormon. Um, you know, I wasn't ultimately willing to make the commitment and maybe part of the reason is 'cause I'm too much of a free rider, so Mormons who are left to pry better than me.

Dwarkesh Patel 1:06:46 Um, the Mormon church has, I think, more than $100 billion in assets.

Garett Jones 1:06:50 Uh-huh.

Dwarkesh Patel 1:06:51 W- what is it planning on doing with all this money? That's a, that's tremendous, (laughs) that's a tremendous sum.

Garett Jones 1:06:56 I don't know. I mean, maybe they're planning to hand it to the savior when the second coming happens, right? It's, uh, there's, there's gotta be a great argument for this option value of just having the wealth there, right? It do- it must give them a kind of independence from the world when various storms come along, cultural, political storms. Um-

Dwarkesh Patel 1:07:18 Mm-hmm.

Garett Jones 1:07:18 ... you know, I don't actually know what their plans are for what they would do with all the money. But I do know that, like, normal economics tells us that just pe- being frugal is good for the economy overall.So, Steven Landsburg has a great, uh, essay, In Praise of Scrooge. It's especially appropriate for this time of year. And, um, being frugal means you're building up the capital stock and you're giving a- a sort of invisible gift to future generations. So Mormon frugality is basically helping build up the US capital stock and indirectly the world's capital stock, and helping make us all more productive, which I think is something that fits in with Mormon values.

Dwarkesh Patel 1:07:52 Yeah, yeah. I think people have pointed out that people should be spending more money given the fact that they have so much leftover by the time they die usually. But I u- uh, yeah-

Garett Jones 1:08:00 Yeah. As a- an-

Dwarkesh Patel 1:08:00 ... the individual level-

Garett Jones 1:08:01 ... individual level, if you care about your own wellbeing, that's true. Yeah.

Dwarkesh Patel 1:08:03 Right. Yeah, yeah. But okay, so if you're e- e- it just is interesting, like leaving a large inheritance is, uh, is so- is socially valuable.

Garett Jones 1:08:10 Yeah, leaving a large inheritance means there's, uh, you're leaving ... you're- you're- you're producing a lot, but you're not consuming very much-

Dwarkesh Patel 1:08:15 Right.

Garett Jones 1:08:15 ... so that means you're building up the capital stock, so.

Dwarkesh Patel 1:08:17 Yep, yep. Um, and, uh, wh- what it w- there's also I think, um, a- a large amount of, uh, multi-level marketing schemes that proliferate in Utah.

Garett Jones 1:08:28 Yeah.

Dwarkesh Patel 1:08:28 Is that one of the downsides of high social trust?

Garett Jones 1:08:31 Yeah, the ... 'cause people predate upon it, right? It s- strikes me as-

Dwarkesh Patel 1:08:33 Right.

Garett Jones 1:08:33 ... a total rent seeker sort of thing. I mean, I have to say, the knives are really good so I'm a-

Dwarkesh Patel 1:08:39 (laughs)

Garett Jones 1:08:39 ... uh, everybody I know who's ever had CUTCO knives ends up using them for decades. So that's one of the popular multi-level marketing schemes that actually gets to- to men. A lot of them are targeted at- at women through cosmetics, as you might know, so.

Dwarkesh Patel 1:08:50 Mm-hmm. Yeah.

Garett Jones 1:08:51 At least the men's one works

Garett Jones 1:08:52 out.

Dwarkesh Patel 1:08:52 If you had to implement the immigration system from scratch, would you actually g- uh, would you actually consider these SAT scores and these other deep roots scores as part of somebody's admittance or would you just consider the individual level, um, your personal skills and education and things like that?

Garett Jones 1:09:08 No, I'd wanna launch a 10-year, maybe 20-year research project of figuring how to turn the deep root scores into something useful. So I, like I've, uh, I think right now with the deep roots literature we're about where Milton Friedman's monetarism was in the late '60s. You know, Friedman said, "Hey, I figured out where inflation comes from and we'd be a lot better off if we just grew the money supply at 3% a year." And n- ultimately nobody thought he was right about the 3% a year thing, but they did think that he had a still had a lot of good advice. So Friedman ended up having a lot of good ideas, but they weren't policy ready. And I think that's about where we are with the deep roots literature right now, which is, um ... I mean, at most one would use it as a s- like a small plus factor in a points system, but I don't even know which points I'd use.

Dwarkesh Patel 1:09:55 Yeah.

Garett Jones 1:09:55 But something along those lines is worth thinking about. I would never use a cutoff ... I would never use any, uh, quotas or hard cutoffs. Uh, if you think about points-based systems, 10, 20 years of further research and maybe you'd find a way to put the deep roots into a points-based system.

Dwarkesh Patel 1:10:12 Yeah, yeah. Um ...

Garett Jones 1:10:14 How was the SBF thing?

Dwarkesh Patel 1:10:16 What do you mean?

Garett Jones 1:10:17 So did he just say like, "I'll fly you out?"

Dwarkesh Patel 1:10:20 So I was there for the, um, EA Bahamas.

Garett Jones 1:10:23 Ah, okay.

Dwarkesh Patel 1:10:24 And then, uh-

Garett Jones 1:10:24 Okay.

Dwarkesh Patel 1:10:24 Yeah, yeah, yeah. And then while I was there I wa- I, um, I talked to somebody who knew him and I'm like, "Hey, I would love to interview him-"

Garett Jones 1:10:31 Yeah, yeah, yeah, yeah.

Dwarkesh Patel 1:10:31 "... just things I would ask him."

Garett Jones 1:10:32 Yeah.

Dwarkesh Patel 1:10:33 Yeah, yeah. It was, uh, it was one of the ones where I act- actually I f- feel like I would really want to redo that one because I was aware of some things back then that were kind of ... that would be-

Garett Jones 1:10:42 Yeah.

Dwarkesh Patel 1:10:42 ... worth asking about in retrospect. And of course, it's all in retrospect, but-

Garett Jones 1:10:45 Yeah, yeah.

Dwarkesh Patel 1:10:45 ... um, I should've poked harder at that rather than asking these sort of philosophical questions about effective altruism.

Garett Jones 1:10:51 Yeah. Do you think I ... I mean, is it as simple as like he was co-mingling fun, uh, he- he lent a bunch of FTX money over to a hedge fund and then they lost it?

Dwarkesh Patel 1:11:02 Um, I- I- I ... yeah, I'm guessing.

Garett Jones 1:11:04 Brute approximation?

Dwarkesh Patel 1:11:05 I think it's ... Yeah, yeah.

Garett Jones 1:11:06 Yeah, yeah.

Dwarkesh Patel 1:11:06 Yeah, um-

Garett Jones 1:11:06 So it's like it's old-fashioned financial fraud, partly driven by, uh, not having a really good board over em or a good oversight.

Dwarkesh Patel 1:11:14 You know what? I'm really curious to ask of you this because you know, uh, you talk in Hive Mind about the fact that higher IQ people on average, um, are more cooperative in prisoner's dilemma type situations.

Garett Jones 1:11:25 Yeah, yeah, yeah.

Dwarkesh Patel 1:11:26 And I just interviewed Bethany McClean on the podcast. It hasn't been released yet, but you know, she wrote The Smartest Guys in the Room which ... about Enron, right? And then so-

Garett Jones 1:11:34 Yeah, yeah, yeah.

Dwarkesh Patel 1:11:35 ... th- there is this thing where maybe they are less likely to commit fraud on average, but when they do, they're so much more successful at it.

Garett Jones 1:11:40 They're really good at it, right? Yeah.

Dwarkesh Patel 1:11:41 Exactly, right? Um, so yeah, I- I don't know. Maybe- maybe one of the downsides of having a high IQ society is that when people do commit fraud, they're super successful at it. Uh, you know, I don't know.

Garett Jones 1:11:51 Yeah. I mean, the- the evils of- of super smart people are obviously a huge risk to all of humanity, right?

Dwarkesh Patel 1:11:59 Right.

Garett Jones 1:11:59 I don't have to worry about humanity being wiped out by a bunch of people with sticks and stones, right? I have to worry about-

Dwarkesh Patel 1:12:04 Yeah.

Garett Jones 1:12:04 ... humanity being wiped out by nuclear weapons, which could only be invented by smart people.

Dwarkesh Patel 1:12:09 Right. Yeah, yeah.

Garett Jones 1:12:10 Yeah.

Dwarkesh Patel 1:12:10 Um, are- are smart people c- more cooperative in a sort of very calculating sense that, you know, this game is gonna go on so I wanna make sure I preserve my relationships? Or even in the last turn of a iterated prisoner's dilemma game, even in the last turn they would cooperate?

Garett Jones 1:12:28 No, in the-

Dwarkesh Patel 1:12:28 You know what I mean?

Garett Jones 1:12:28 ... last turn they walk away, yeah.

Dwarkesh Patel 1:12:30 Yeah, yeah.

Garett Jones 1:12:31 Yeah, I think it-

Dwarkesh Patel 1:12:31 Yeah, but-

Garett Jones 1:12:31 ... I think the, um, there is no correlation between, uh, intelligence and say agreeableness, right? Normal psychological-

Dwarkesh Patel 1:12:39 Mm-hmm.

Garett Jones 1:12:39 ... agreeableness. And I think that's a broader principle. Or conscientiousness for that matter, right? Psychological conscientiousness. So Machiavellian intelligence I think is what's driving the link between IQ and cooperation. So in repeated game, that Machiavellian intelligence which a lot of intelligence researchers will talk about turns into Coasian intelligence where people find a way to grow the pie. But it's a very-

Dwarkesh Patel 1:13:02 Right.

Garett Jones 1:13:02 ... cynical, self-interested form of growing the pie.

Dwarkesh Patel 1:13:06 Yeah, yeah.

Garett Jones 1:13:07 And so I- I don't think that has any, um ... I don't think it's driven by inherent prosociality. I think it's- it's endogenous prosociality, not exogenous prosociality and that's a reason to worry about it.

Dwarkesh Patel 1:13:19 What i- what happens to these higher IQ people when, if society goes into a sort of zero sum mode where there's not that much economic growth and so the only way you can increase your share of the pie is just by cutting out b- uh, bigger and bigger slices for yourself?

Garett Jones 1:13:32 Yeah, like-

Dwarkesh Patel 1:13:33 And uh-

Garett Jones 1:13:33 ... then you wa- gotta watch out, right?

Dwarkesh Patel 1:13:34 Yeah, yeah.

Garett Jones 1:13:34 It's like the Middle Ages right there, right?

Dwarkesh Patel 1:13:36 Yep. I see what you're saying.

Garett Jones 1:13:37 Yeah, yeah. Yeah.

Dwarkesh Patel 1:13:38 Um, yeah, interesting. Um, awesome. Uh, Garrett, thanks so much for coming on the podcast. This was interesting.

Garett Jones 1:13:45 Hey, thanks for having me.

Dwarkesh Patel 1:13:46 I really enjoyed it. I-

Garett Jones 1:13:46 It's been fantastic.

Dwarkesh Patel 1:13:47 I'm reading the books.

Garett Jones 1:13:47 Yeah.

Dwarkesh Patel 1:13:47 Yep, excellent.

Garett Jones 1:13:48 Thanks for reading my books. I appreciate it.

Dwarkesh Patel 1:13:49 Yeah, of course.