People who live in large cities produce more patents per person than those who live in small towns. They also walk faster down the street, and use less infrastructure per person. Indeed, the relationships between these quantities and city population obey power laws with apparently universal exponents, reminiscent of scaling laws is biological organisms? Luis Bettencourt argues that they can be explained in similar ways, using ideas from complex systems and network theory.
Support Mindscape on Patreon.
Luis Bettencourt received a Ph.D. in theoretical physics from Imperial College London. He is currently the Lorna Puttkammer Straus Professor Department of Ecology and Evolution & Data Sciences Institute, as well as Associate Faculty and Special Friend in the Department of Sociology, at the University of Chicago. He is the author of Introduction to Urban Science: Evidence and Theory of Cities as Complex Systems.
Click to Show Episode Transcript
0:00:01.1 Sean Carroll: Hello, everyone, and welcome to the Mindscape Podcast. I'm your host, Sean Carroll. One of the most vivid thought experiments or metaphors in biology is the idea of replaying the tape of life. This was popularized by Stephen Jay Gould, who had the idea that if you went back to some point in evolutionary history and you... I don't think that he used quite these words because he was not a physicist, he was a biologist, but if you kept the macrostate of the system, so you kept the Earth and all the critters on it, but you could change the microscopic features and then you let the evolution of life go on from that point, Gould argued that you might get very different answers, or at least you could imagine you could get very different answers.
0:00:47.7 SC: And for sure, in the details, that can be true. If an asteroid does or does not hit the Earth, the evolution of the dinosaurs would be very different. In fact, we had an episode of Mindscape with my evil twin, Sean B. (for biologist) Carroll, where he emphasized all of the contingencies and the randomness and how things that you can't control can have a big impact on biology.
0:01:14.0 SC: But there's also an aspect of what you might think of as universality. There are things that are going to happen biologically whether or not you play the tape once or many, many times. Things like this we've talked about with Geoffrey West, for example, the Santa Fe Institute professor who talks about scaling in biology. The idea that if you think about the mass of different kinds of mammals, let's say, and relate them to their metabolic rates and their lifespans, they all fit on a straight line on a log-log plot, which is to say a power law, a scaling relation. And they can even explain why those power laws are there based on arguments from networks and the geometry of space and things like that.
0:02:01.6 SC: So you can have both, and it's really important that you have contingency and you also have laws of physics and constraints from natural reality that lead things to fill certain niches and take certain forms. All of this is to build up to today's podcast with Luís Bettencourt, who is actually a collaborator of Geoffrey West, another complex systems theorist who works on cities, not on biology. Probably he's done some biology in his life, but his focus is on cities as complex systems.
0:02:36.2 SC: Like biological organisms, cities are large things made of little things that interact in various ways and have specialization of roles and all this stuff. And guess what? Much like in biological organisms, cities obey scaling relations. There are certain facts about cities, whether it's the amount of crime that you have or the amount of innovation you have, the new patents or works of art or whatever it is, that are pretty dependably related to the population of the city.
0:03:09.8 SC: And this is kind of an astonishing fact. I'm speaking to you from Santa Fe, New Mexico, where the Santa Fe Institute is located. Luís is speaking to you on this podcast, or will be, from Chicago, Illinois, another city I spent a lot of my life in. And when I think about all the cities that I've spent serious time in, Philadelphia, Boston, Los Angeles, Baltimore now, even Santa Barbara and Santa Fe, they're very different. Different histories, different cultures, different people. And yet, if you know the population, you can make very accurate predictions for things like the crime rate, the rate of innovation, the amount of infrastructure you need, et cetera.
0:03:52.0 SC: And amazingly, it's not just the bigger the city... Well, it is true that the bigger the city, the more, let's say, patents you have, but it's not if you double the size of the city, you get double the patents. You get more than double the patents. And you can make a prediction for what the scaling relation is, and it comes out to be pretty darn accurate, not just right now, but even through history and in different countries and so forth. So to me, this is just the quintessentially beautiful example of why it is useful to think of complex systems as a domain of study all their own, because very similar kinds of reasoning lead you to insight about both biological organisms and about cities, which are completely different kinds of things.
0:04:37.0 SC: There's something to be said for thinking of the nature of complexity itself and how it shows up in many, many different areas of human interaction, biological interaction, physical instantiations of matter, and all that stuff. And by doing it, as we'll see at the end of the podcast, you learn things about how to make better cities and how to move things forward in useful ways. So since most of us in the modern world live in cities, this is kind of relevant to many, many people out there. Let's go.
[music]
0:05:24.9 SC: Luís Bettencourt, welcome to the Mindscape Podcast.
0:05:27.6 Luis Bettencourt: Great. Thank you. It's a pleasure to be here.
0:05:29.8 SC: So we're doing this in part... I've been meaning to get you on for a long time. We're both fellow Santa Fe Institute affiliates, and so I've bumped into your work many times. But Marc Berman was recently on the podcast, your colleague at the University of Chicago, telling us we need to go outside and look at nature and we'll be thinking better and happier. But afterward, he emailed me and said, "I really want to emphasize that I like cities. Cities are actually good, even though we were saying that going out in nature is good."
0:06:01.6 SC: So you're an expert on cities. I guess a good place to start is, for you doing research in this area, which came first? Was it cities and you wanted to come up with the best lens of thinking about them, or did you fall in love with complex systems and realize that cities are good examples?
0:06:19.5 LB: It was a little bit of both. It's always interesting to think about that. Marc is a dear friend and colleague, and so I thank him for the plug-in. But basically, I kind of had this dual personality problem that I've always liked people and the exceptional things that can happen when people come together. So that happens mostly in cities. I grew up in Portugal at a time where there was a lot of change. There was a lot of change in the city I lived in, which is Lisbon.
0:06:50.5 LB: And there was a lot of development, but it was noisy. And my parents were kind of involved with nonprofits and so on. So I was always frustrated why this is so hard. And then I love science. So you and I are kind of theoretical physicists. I'm also a theoretical physicist. Started also with phase transitions in the early universe. Not exactly what you did, but pretty close.
0:07:10.7 SC: Yeah.
0:07:11.8 LB: And that kind of way of thinking that we learn in physics is just beautiful and has this long tradition and so on. So I love science. So I, for a while, was looking for a way to do science like in some sense we do in physics, at least in spirit, but at the same time with the problems that fascinated me, which were problems in complex systems, particularly in society.
0:07:32.4 LB: And so I wanted to find a place to think fundamentally about phenomena in the world, but eventually you call them complex systems. And then cities came as a little bit of a happenstance, a happy episode, actually, at the Santa Fe Institute, where people were looking at many different types of organization and complex systems. And I had an interest in cities, particularly as places of innovation. And we started looking at a bunch of quantities using scaling, which is a very simple technique. And then we found all these interesting new things that people hadn't found before.
0:08:07.3 SC: And just so we know what we're talking about, what counts as a city? Like, is there some threshold? You need enough people to be in it? Does the geography matter?
0:08:17.0 LB: Yeah, you know, that's controversial, just like... But basically, so there's several ways to think about this. There are people, kind of structuralists, a bit like more in geography and social sciences, that think about, oh, it's this boundary, it's the administrative city of some kind.
0:08:34.5 SC: Right.
0:08:35.0 LB: But if you think more like, I think, from science and even from physics, it's sort of a bound state. It's a place where people are interacting with each other, right? And so that's also the modern definition that the US Census uses. So it's basically the definition of everybody that commutes to interact with each other. So that includes sort of central cities that we think of, New York City, Manhattan, but then includes all the other parts of the city in which people are coming in and out and so on and interacting strongly.
0:09:04.7 LB: And so there's sort of a, it's kind of a hazy cloud that forms the city as well as sort of a very dynamic, fluid quantity. And you can basically operationalize this idea through observing commuting flows and more recently even using cell phones and other things like this. But it's kind of a little bit more like the atmosphere of a place as well as its core.
0:09:24.7 SC: Yeah. And when did they start? When did we have our first cities?
0:09:28.6 LB: Oh, that too is a little controversial, because there were some settlements before that people fight over if they were cities or not. But something like, there were certainly well-defined cities in today's Middle East 4,000, 5,000 years before Christ. And before that, there were a few things that are a bit older, like Jericho and a few cities in today's Turkey. There were definitely settlements where people were farmers, but they had some division of labor and some evidence of a few other things, but people fight if these were cities or not. But they were certainly proto-cities.
0:10:06.3 LB: And there are other things that have come to the fore that are very interesting, but they're probably not cities in the modern sense, which are these seasonal gatherings, sometimes with a lot of people, that we see in hunter-gatherers a lot. In almost every people, before they have cities, they have places where they come together for a while, often with constructions, but it's non-permanent. But it's kind of a pulsating city wanting to be a city. It's kind of an instability.
0:10:30.3 SC: Burning Man.
0:10:31.5 LB: Yeah. Yeah, Burning Man's an example, and people actually use this. The famous economist Paul Romer, who is also a friend, he used to go there because he wanted to see sort of some of those elements of a city.
0:10:41.6 SC: I'm sure that's why he went there, yeah.
0:10:43.6 LB: Yeah, yeah. Well, that was the story.
0:10:46.6 SC: And the cities are coming to dominate, as I understand. It wasn't that long ago that a lot of people lived in rural environments, and these days it's overwhelmingly cities.
0:10:58.3 LB: That's right. So that's another reason to study cities, because they are this very spectacular phenomenon. It's a global phenomenon. It's happening everywhere. And it's now they characterize the lives of most people on Earth. So it's kind of very interesting how fast this happened. And so again, when we go back to those very first cities, in pre-industrial societies, including Romans and really up to the Industrial Revolution, most countries were about up to 20% urbanized, which means obviously most people didn't live in cities, even small cities.
0:11:35.4 LB: But then ever since, partly because of the ability to produce more food and energy and materials and so on, we now have countries like the United States, which is almost 85% urbanized by the definitions we were talking about before. So that means basically that almost everybody lives in cities. And we see this happening now everywhere. The big transition over the last few decades was China and East Asia, but the same thing's happening now, maybe faster, in India and parts of Africa and most of Africa.
0:12:04.8 SC: So to get into a little bit more of the weeds here, the city is more than just a density of people. You've already used the phrase division of labor, right? What goes on in a city that makes it special over and above the fact that there's a lot of people there?
0:12:20.8 LB: Yes. So in your show we talk a lot about complex systems, right? And so it is a complex system in the sense that it is a bunch of people. It's denser. But even from beginning of the social sciences, one characteristic that's obvious, it's obvious in complex systems, is that the agents become different from each other. They're heterogeneous. And this is in itself kind of magical. It's something that happens only in complex systems, not in physical systems.
0:12:44.2 LB: And so that's... Traditionally we talk about that in terms of division of labor, goes back to Adam Smith and so on as a mechanism in his view of the Wealth of Nations. But it's really, I think, more fundamentally, when we think in more modern terms and with better theory, it's really a division of knowledge. It's just that the information each one of us has, as well as our behavior and our jobs and so on, become differentiated and interdependent. And this is a magical thing, is that latent in being a human, that was there all along, right?
0:13:15.1 SC: Right. Possibility of it, yeah.
0:13:17.8 LB: But it kind of... This sort of this instability that's there that you see with hindsight develop, but really comes to the fore in large cities where you really see that at work, and all the amazing things that come out of it, but also some of the problems of society that come from it. But that's basically why cities exist. So that's why they create this interdependence of different roles. And this turns out to be very productive and very creative.
0:13:47.8 SC: And I guess this is something that, despite the fact that I've been interested in complex systems for a while now, this fact that you just said has only recently sunk into my head. The idea that what's special, one of the things special about complex systems is that the individual pieces aren't by their nature different, but they become specialized to serve different purposes in the emergent whole somehow.
0:14:11.3 LB: That's exactly right. And this is, I think that's almost a good definition of complex systems. And I think opens up more this idea that it happens gradually. It doesn't happen just structurally, right? So and again, through human history, you'd see societies obviously where that happens less. It happens a little bit in human societies from the beginning, but happens much less than in modern societies and cities. But that process therefore needs explanation and has a mechanism.
0:14:37.9 LB: And so that's where this becomes really interesting. And that really is a better way to start thinking about cities and opens up a whole scope of thinking about not only human societies, but also then complex systems, as you said. So this is where it becomes really interesting to me, and that's why I really stick with it.
0:14:53.4 SC: The other thing, just looking at your work and trying to compare it to other discussions we've had here about complex systems, I mean, a city is kind of a set of interlocking complex systems, right? And in a way that I guess other things are, but maybe it's less obvious, but you have people, but you also have infrastructure, and you have communication and transportation. And it's an especially rich set of things going on.
0:15:19.4 LB: Yeah, exactly. I mean, I think of the things we know about, it's the most extreme complex system, the most complex complex system, complex squared, in that exactly as you say, it has brains, it has organisms, it has ecosystems, it has infrastructural systems, it has design at play, it has human institutions. And all these things are interplaying with each other, right? So it's a great system to study.
0:15:46.9 LB: And one of the things that's joyful from a point of view of research, but also just even pondering about it, is that you can approach it from different angles, different perspectives, but then you see how things are interconnected, how you need the other thing. And this diffuses, I think, a lot of the things that are current standing problems, for example, in social sciences, where you just see the economic aspect of a system, or just the political aspect, or just the sociological, the network issue.
0:16:13.5 LB: And you also see the role of infrastructure vis-a-vis the lives of people. And you also see innovation as sort of something that becomes interesting and how that then leads to more economic exchange and economic growth. So everything that we study more in isolation becomes interconnected. And I think the point is that it becomes much easier to understand, much, much more tangible, and those articulations become clearer. So this is kind of the journey, right, is how complex systems allow you to start connecting these things and understand them together rather than being confused as you study them in isolation.
0:16:51.5 SC: Yeah, it's... I always do have this back and forth conversation with scientist friends of mine. What's so great about complexity? There's a lot of complex things out there, but are there really any universal rules? Why should I think that an ant colony and the human brain have anything in common? And so can you give a sales pitch for that they do have things in common?
0:17:15.2 LB: Yes, of course. I mean, first, these are all interesting systems that exist in nature that we don't understand very well. So they're just challenge problems. In that sense, they're interesting. But I think what they have in common are some of the things already said, but let's say them again in light of the examples you just brought up. So the ant colony, the brain, the cities, I say, so they're made out of things that we can identify more as individuals: the ants, the neurons, the people, perhaps, the places, the buildings. But then these are in interaction with each other. The whole thing doesn't make sense if you're just thinking.
0:17:48.8 LB: A neuron alone is a pitiful thing, right? But a brain, my gosh, glorious thing. Same thing, you could say that about humans. I'm not going to go there, but it's more interesting when we come together. And probably the same thing for the ant is a very simple thing, but the ant colony, wow, right? So and as that happens, you have this differentiation of roles. You have always this, again, division of labor, division of knowledge thing that creates a differing functioning whole, something that works differently from what the individual would do.
0:18:21.4 LB: So in all these, and all these systems have a certain metabolism, a certain rate of energy harvesting and throughput, which is the physical part, the physicsy part. But then also what they have, which is less visible, at least if you don't look for it at first, but is kind of the main focus of a lot of research, including mine, is the informational part. The informational part, both structural, how each piece is orchestrated with the others, how they are interdependent, but it's also the information they contain that enables new behaviors, their model of the world, if you will. And that is the scope of what they can do.
0:18:58.1 LB: And in some systems, but not others, or at least not others easily at the same speed, that adaptation to create a new model of the world, a better model of the world, is very fast. So cities are very good at that. You throw almost anything at a city, if it's not too disruptive, and people learn how to deal with it.
0:19:14.3 SC: Yeah, okay. Yeah, that's a very good point. And in fact, that's connected to the other thing I was going to say, which is that it seems that compared to some other complex systems like an organism or a brain, the cities don't necessarily reach a steady state and just stay there, right? The fact that they're always changing is a feature.
0:19:34.4 LB: Yeah, and I think we don't have quite a very clear-cut taxonomy, but allow me just a quick job at that. I think it's interesting to think about complex systems that are limited in that sense, that are not open-ended. So an organism, we die, right? So we age and we die, for example. And so typically an organism therefore is not, does not, cannot keep going. Whereas if you look at the ecosystems are a higher level, and if you look at a city, again, a high level relative to individuals, even, then you have a system that's open-ended.
0:20:08.3 LB: The people that you'll see 20 years later are not the same. The buildings may have changed, many of the pieces will have changed, but the city's still there, right? So we have cities that have been there, and ecosystems that have been there for many millions of years, though they change slowly, and cities certainly for thousands of years in some cases. So that's kind of interesting. It means that the pieces individually are not essential. The roles they play relative to each other can change a bit, but they are more essential. But those can be filled by newcomers and exiters and so on.
0:20:38.6 SC: Okay, so we have some context for the cities as complex systems, but cities are also places where human beings live. Can you give us a little bit of a feeling? I think that probably many people have their own opinions, but what are the good aspects of living in a city? What are the bad aspects? What do we need to keep in mind here?
0:20:56.7 LB: Well, I think everyone has their opinion about that. So I encourage people to write the pros and cons in their little cheat sheet. But I think that what's interesting, let me kind of dodge it a little bit and try to answer it in a way that's maybe not so parochial. One of the things is that I think it's very easy to answer what's bad about a city, right?
0:21:20.2 LB: So let's go to a large city. It tends to be expensive. It tends to be a little constrained spatially. Your house may be not as big as you'd like. It's congested, maybe it's polluted, maybe noisy. Depending on people's attitudes, you may be hard to deal with that and so on. It may be dangerous in some cases, in some situations. Okay, so the list of bad things is very long and very clear. And this has probably been true always, actually.
0:21:48.0 SC: Okay, yeah.
0:21:48.3 LB: Thinking about old cities, like the Romans wrote a lot about Rome, and it's just not very different. Okay. So it really begs the question that's very interesting. If there are so many costs, so many disadvantages, what are the advantages, right? What is binding people to come to these places, particularly as we were talking about before, that more and more people want to live like this all over the world? And the answer is that we actually don't have a very good answer to that. You know, everyone should actually express why they think they live where they live. I can do that. I live in Chicago partly because of my job, but I do like to live in a city because I like sometimes to go out, I like to walk in the city, I like to see people, I like to get a cup of coffee, I like to see the news, I like a certain diversity around me, I like the arts. So and therefore it's very hard for me to not live in a big city.
0:22:40.5 LB: But having said that, it's kind of a bunch of little things, and it's not really a hard cost-benefit analysis. And some of it often is aspirational. It's about the things that could happen here that will not happen in a small town, right? That I can learn something, my job may improve, or I can meet someone, or I can develop my career, whatever. People at different stages of life have different priorities, but that's kind of very interesting.
0:23:09.5 LB: So I think what you see is that, for example, we tested this with data of many kinds, and we are investigating this better just to have better answers. But for example, if you look at just people's expenditures versus what they make in terms of salaries, there's some data on this, wages and so on. Basically, larger cities are more expensive, but they also, people earn more. So it's basically you break even, you just spend more money, make more money kind of thing. So it's not that, right? It's just basically something else. It's kind of this appetite for speed and diversity and possibility that seems to happen in cities.
0:23:48.7 LB: So again, it's something that's hard to quantify very well. And I think it appeals to different people at different times. Large cities are particularly attractive, modern cities, cities that we experience today, are particularly attractive to some people and not others. So this starts answering the question a little bit in a more impersonal way. So larger cities tend to be very attractive to people that are young, young adults who are more educated. So this goes a little bit what I was saying, because you're prepared to maybe take better advantage of what is there in a city, but you also need to develop your potential, you need to develop what you've learned perhaps into a network of people, into a work experience, into something where you can have choice.
0:24:31.0 LB: Another demographic that's particularly attracted to large cities are foreign migrants. So this is kind of topical right now, but it's interesting, it seems to be quite universal. So we think this is because large cities have many different pockets of culture, including a little bit of people from their own point of origin, but also other people that are not necessarily a homogeneous culture, and they're therefore culturally more places to connect and you don't stand out so much.
0:24:56.9 LB: So there's sort of a certain cultural mixing that is more generative, and this is true, of course, of people in the arts and people who are more not in the mainstream so much. So there's a bunch of things like this that we're just starting to understand that have to do with these mechanisms that allow you to transform yourself in some sense towards realizing then structurally what we perceive as division of labor and division of knowledge, but really what is the process of self-expression and perhaps trying to improve your life in some sense.
0:25:27.8 SC: That is very interesting. A long time ago I did a podcast episode with Will Wilkinson. I don't know if you know him, but he's a kind of political commentator. And he was looking at the density divide, as he called it, the political polarization between people who live in cities and people who live elsewhere. And the interesting angle he took on it was everyone knows that cities tend to be blue, rural areas tend to be red in the United States, liberal-conservative axis.
0:25:54.6 SC: He looked at personality traits, he looked at the Big Five personality inventory and kind of found like what you just said, that the people who tend to go to cities have higher openness, and I don't know what the Big Five is, but they want to explore new, more new things. And I guess part of what you're saying is the higher density in cities gives you more opportunity to do exactly that kind of thing.
0:26:17.2 LB: Yeah, that's right. And of course, it's not a sure thing. It's sort of an exploration thing. And so the other thing that cities allow you to do, that go to this fact that we discussed more fundamentally just a little bit while ago about them being more open-ended, is that if it doesn't work out, you can go away again and go somewhere else. Whereas the cell in the organism can't quite do that, right? It's kind of stuck there and so on. There are, of course, pathological situations where that's true of people as well in the place they are, and that was obviously more common in the past and so on, but it's still nevertheless a problem.
0:26:50.5 LB: But you kind of see these issues also as you start understanding conceptually what's going on and you can ask these questions. So a lot of the problems, for example, of poverty and other issues that are social, they have to do with this being close but not being able to access the network that cities are in principle creating, but they're not available to everybody. So that's another set of issues, but it's kind of important because it has to do obviously with very important social and economic problems.
0:27:19.4 LB: But it's also, if you allow me to just take a step back, it's also, if you will, the city being formed not as a place but as a network, and the people that can join it and the people that cannot, or the people that can join it to a greater degree in different dimensions. So you see kind of this mechanism at play.
0:27:40.1 SC: Good. I think this leads us right up to the juicy part of what we have to talk about, because this is more than just words. There's math and there's data here also. So one way of asking the question is, is a big city just a small city times a few, like just bigger extra, or does something different happen quantitatively as the city gets bigger?
0:28:06.2 LB: Yeah, so this is kind of what in complex systems often we call scaling. It's a way of thinking. It's also a bit of math, but that's sort of an entry point. But it's very easy to say. And this has a long history as an idea. I think it was done by engineers trying to have models of buildings and models of ships and trying to think about if you build a real thing, what you need to change. But we also do it in science, for example, in obviously in physics we do this a lot. For example, if you think about a gas and you double the size of the box, how do the properties change depending on whether you double the amount of gas or you don't, you just double the volume. So you can tell how some things change, if the pressure changes, for example.
0:28:45.8 LB: So there's this idea that you can compare what you think is the same system at different scales, at different sizes. In this case, usually for a city, it could be its spatial extent, but usually it's just its population. So you compare New York to, like Chicago is about half the size of New York, for example, and then you can go down another. And so you can compare things that people measure all the time, like, I don't know, it could be the number of homicides, or it could be, just to be grim, or it could be the GDP of the city or the wages, the average wages that people make or whatever. This is very easy to do, and that's why it's a nice probing tool.
0:29:23.9 LB: And so we use this, this is one of the ways we came in to look at cities. Our famous colleague Geoff West had done this for organisms with Jim Brown and Brian Enquist and so on, to see their metabolism. And so cities are different. They just have this weird property in which, as we already said a little bit, if you look at money quantities, in larger cities you have more money being made. So whether it's GDP, so wages or profits make up GDP, or expenditures, how much money people spend on various things and so on. And so these increase per capita.
0:30:00.5 LB: So this basically just means that people spend more money and make more money the larger the city. But this happens on a continuum. But this continuum increases the amount of money as we go. And if you think about this for a second, physicists like to think about this in terms of dimensions, right? So what are the dimensions of this? It's money per unit time. So it's kind of a speed. It's a rate of change in time. So money is being made and it's being expended. And so that's kind of a clock for the system. So what you find mathematically is that basically when you double the size of the city, there's a sort of an exponent that says it increases by 16% more or less. It's crazy.
0:30:38.2 SC: Sorry, sorry, what is it that increases by 16%?
0:30:41.4 LB: So for example, the GDP per capita or the wages.
0:30:44.1 SC: Okay.
0:30:44.6 LB: That somebody makes, on average, okay, so this is a big average. There are people that make a lot more money, people make a lot less money. But usually when you use these tools, it's just about the average. So that's kind of interesting. So you make more money, spend more money. But it turns out that almost anything that's a social behavior, whether it is, so we looked at lots of different things. We looked at rates at which people catch new diseases like AIDS or COVID in the beginning. So that's just, people have models for this, but basically that makes it contacts between people. That's the idea that these things are transmitted that way.
0:31:15.3 LB: Or we looked at patents, we looked at crime, we looked at many things that tend to have the same character that requires a person-on-person, so a transaction or interaction. And these things all have this feature that they accelerate in time. So you get more of them per unit of time, both good things and bad things, as the city gets larger. So again, this just quantifies what people, already what we talked about, what people expect. That larger cities are more expensive, but people make more money, that there's more invention, but there's also more crime as well, for example. But also they're more susceptible to health crises, which again is the history of cities, a lot about those. So that's interesting.
0:31:59.5 LB: So it's kind of this... And the other thing you find is that for a different class of quantities, you have a mirror effect. It's exactly a mirror effect, it turns out, which is to do with, if you will, the volume of infrastructure. So for example, all the roads, the surface of all the roads, roads are a surface. If you measure the surface of all that, it decreases per capita by about the same amount, 16%, as you double the size of the city. And that's true of pipes and a bunch of other things. It's just like, what the heck? Who asked for this? When we started measuring this. And so what's going on? So then, we need an explanation. But basically the intuition here is that whatever's being produced socially, it's a product of interactions. People are interacting more with each other because it's denser and they move around more. That's what a city is about.
0:32:49.6 LB: So the city is more about the interactions than just having people concentrated. It's really about them interacting. This is very natural for a physicist, right? That's what you'd expect. But then the spatial part is just telling you the density part, that actually there's less space per capita and there's more interactions per capita. And so these things have to go together. It's just if you've done any calculation about a cross section and somebody moving through a medium, that's exactly the math, is that if you go through a city, there are more people to interact with, more businesses because it's denser. And the amount by which it's denser is the rate at which the number of interactions per unit time increase. So that's basically the effect.
0:33:31.7 LB: So what is a city? So what is magical is that these things kind of go together in such a way that the cost-benefit, so the cost of moving around this infrastructure and the benefits of making this amount of money, are exactly sort of compensating so that you can keep on having larger and larger cities that are ever faster and more expensive, but at the same time have more supportive infrastructure of just the right amount. This doesn't always happen in time. We know when we have a crisis, when the highway closes or whatever. But sort of, again, it tends to happen when you compare cities over large periods of time.
0:34:07.1 LB: So it's kind of this amazing thing that humans invented. I call it a social accelerator, a bit like a physics accelerator. You might like that. Because basically it's a way of bringing people together on essentially a daily basis to interact, to produce all the things you produce socioeconomically and to create a system that can be scaled up or down.
0:34:28.5 SC: So just to...
0:34:29.9 LB: Yeah, so there's a little bit more I wanted to say about space and social interactions. Should I just say that?
0:34:35.1 SC: Yeah, just say it.
0:34:36.4 LB: What is also very interesting is just because these things combine, there's an actual trade-off that I think is very intuitive, which is that when you go to a larger city from a smaller city, you have more interactions, more transactions, and so your social space has more choice. It's bigger. So there's an expansion of your, if you will, social freedoms, socioeconomic freedoms, choices in that sense. But your physical space is getting smaller. If you go in the other direction, you get more physical space in a small town, but your network has less.
0:35:12.7 SC: Right.
0:35:13.1 LB: So it turns out this is sort of what physicists call a bit of an adiabatic invariant, that basically you increase one thing and decrease another, but this gives you something that gives you then the scale invariance that can exist in many different scales. But there's a trade-off.
0:35:26.7 SC: That's very good. Yeah. So just to make it super simple in my mind, I'll try to repeat it back and then you can tell me if I have it right.
0:35:32.9 LB: Yeah, please.
0:35:33.8 SC: So if you double the size of a city, you might very naively think you're going to double the size of its outputs, its patents or works of art or whatever. But you don't, you more than double. Like each individual person in the city is more productive in some general sense, just by being in a denser environment, a larger environment. And that's kind of interesting, but maybe you can understand it just in a hand-waving way, like you said, there's more people to bump into. But there's a number, there's a sort of scaling that is kind of universal. And not only is it a certain number per phenomenon, it's the same number for all sorts of different phenomena. Like you get an extra 16% boost in all these different things. That is kind of amazing.
0:36:19.8 LB: Yeah. And what is interesting, so why does it happen? You could think it's just a speed effect, as we just said. But we already said before, before we got into the math, that something else was happening, is that people themselves are changing internally. So if I may push, I'm going to shamelessly push this idea of the accelerator, right? In physics, when we collide these particles, right, we get to see what's inside them if they're not fundamental. Right? I mean, this is pushing that idea, but that's how we discovered all these other fundamental particles that are inside other particles.
0:36:52.4 SC: Yeah.
0:36:53.4 LB: When we bring people together, you actually reveal something that's inside them that's forced to change. And this ability to learn. So you basically are forcing people to adapt to each other more. And what happens is that they have to learn to be useful and to also use the skills of other people around them. And so this is what leads to this division of knowledge, division of labor. But turns out that learning as a phenomenon, which we're starting to understand also now in AI, is a nonlinear phenomenon. And if you stick with the same thing, you get much better at it than being a generalist. It's not just per unit time again, there's another nonlinearity in this. And this basically means that a great scientist, a great lawyer, a great musician is much better than a generalist person who also plays some music.
0:37:43.5 SC: Right. Right.
0:37:45.9 LB: And so the ability to specialize and depend on other people creates knowledge and value in these networks in a way that can't exist before, cannot exist in other ways. So that's the part that's kind of inside the individual, and that's what creates the fact that these things stick.
0:38:05.4 SC: Right.
0:38:05.7 LB: Is that now you've changed people. And what's interesting, not only you see this, of course, in professions and knowledge and personality types, as you said, it may be that some of this effect may be sorting, there may be people of different dispositions that prefer different environments, but some of it also seems to be learned, that people just in that environment acquire that. So, you know, early on we saw that people walk a little faster in cities, for example, that are larger. So that's interesting.
0:38:29.2 SC: I want to... Sorry, I want to dwell on that because that is my favorite of all the different outcomes. People literally walk down the street faster in the city just because. Why? What is the explanation for that?
0:38:43.9 LB: Well, the explanation is that in some sense you need to fit a certain number. Time is not compressible, right?
0:38:49.8 SC: Oh, okay.
0:38:50.5 LB: So all these things we were talking about, you're compressing space, but you cannot change time. So what you're going to do is fit more things in it.
0:38:58.7 SC: You have to move faster.
0:38:59.8 LB: You move faster as well. Yeah. You have more interactions and you have to move faster between them.
0:39:04.8 SC: And it's the same 15%?
0:39:06.1 LB: It doesn't work for everything, but in the behaviors available to people, walking, we kind of regulate the speed, right? We have more highways, we regulate the speed in some ways, but we have subways that go faster. So there are a bunch of very curious adaptations, but a lot of this, as you can see, there's a lot about space and time and then has to do with this ability that people, I mean, biology more generally, but people have to adapt to such circumstances by acquiring different knowledge.
0:39:29.8 SC: So let me ask about the robustness of these scaling relations. So you double the city in size, it gets 15%, 16% more productive in various ways, 15% less infrastructure is needed per person. Is that true around the world? Is it true historically? What do we know about the limits of these ideas?
0:39:52.2 LB: So, yes, two things should be said. These numbers actually, let me just say where these numbers come from. So this 16% or something. When you think about scaling, I'm going to take a step back and then try to answer with the data, but just take a step back. First, you need to understand where these things come from. When you think about scaling, scaling is like... So we're talking about, in some sense, we're talking about a city as a bunch of interactions that can fit into a physical space over time. And time we're imagining that every day is a little bit the same for now. Of course, over long times that changes. But so there are two things going on, and scaling is all about essentially one thing inside another and the dimensions of these two things having to become commensurate.
0:40:43.3 LB: So what's happening here in a city is that you have a social network that to begin with could be anything, it could have any degree and so on, and to some extent has a lot of variety. But on large scales, it depends on the built spaces of the city. And so you're fitting these interactions into space. So this depends on the dimension of space, which the 16 comes from the space being essentially two-dimensional.
0:41:09.4 SC: Okay.
0:41:11.0 LB: And it comes from people traveling in ways that basically have to do with visiting a few places. So this is what... I don't know if listeners know about this, but it's called the fractal dimension. If you think about how you draw a line, for example, over a sheet of paper, you could draw just a bunch of points and that would have fractal dimension zero. It behaves like a point. It could have the fractal dimension 1 if it just really looks like a line, a straight line or maybe a circle, but something that looks like a line. Or it could fill the space, you actually color it all in, and it could have fractal dimension of 2.
0:41:45.8 LB: It turns out that people tend to explore space in cities in this way that looks like a line. You go from home, you go to work, you maybe go to a grocery store, go to school, and do this. So that fractal dimension being 1 and the dimension of space being 2 are actually important. And it's a combination of that that gives you one-sixth in a way that I could try to explain. But it's basically 2 and there's a third that comes from 2 plus 1, and it all ends up being one-sixth. So if you had cities in space in three dimensions, for example, or like in a sci-fi movie, we've seen some of those, then this number would not be one-sixth.
0:42:24.5 SC: Ah, okay.
0:42:26.0 LB: So it'd be different, and you can calculate what it is, and you could try to engineer a transportation system to make it different and so on. So just to say, this number depends on things that we could go and measure. So that's one aspect of it. It's a fundamental number given a few things, like how people explore space in a city and the space in which that object is embedded in real space. Okay. The other aspect is that, of course, we went to see when this thing breaks. So does it break if we go to China? Does it break if we go to the Romans in history?
0:42:58.7 LB: And there are a bunch of really beautiful things that you find in data. So let me just say a few. The short answer is that these numbers are very robust. We observe them in almost every nation, in contemporary nations. In China, it seems the data get a little rough before 2000, but ever since, basically this number emerges. And before that, we don't know if it's data or something else. But in Europe, most European countries don't have a lot of cities that are large, but again, we observe the same number essentially. We observe the same number in Brazil, in South Africa, and everywhere we measured, basically the number is similar.
0:43:39.5 LB: There are countries that have a small number of cities dominated by one where the number is a little bit less, more uncertain. Like, for example, the UK has London being very big and so on. But basically this number is similar. And then we went through history. So there's a beautiful experiment, like what people now call natural experiments, which is when the Spanish get to Mexico in the 1500s. And of course, pre-Columbian people had cities. And so the idea is that this is sort of an independent experiment, because the people that came to the Americas did not have cities. That's the idea. They were hunter-gatherers. So they invented cities independently. They couldn't have copied them from Eurasia or Africa.
0:44:28.9 LB: So basically, so we had a colleague called Scott Ortman, who's an archaeologist, anthropologist. He had a very good survey of the cities of Mexico before they got built up by Mexico City and so on. And so we tested that idea and again we get the same number. What is interesting... So, but I should qualify that in that it's easier to measure spatial things because they live there in archaeology. But we struggle to find what are the equivalent of GDP or wages.
0:45:02.7 SC: Okay.
0:45:03.1 LB: Partly, actually, because the Aztecs didn't have money proper. They used cocoa beans and cloth and so on. So what the heck were they doing? But remember that we think that this is actually expressing general social capacity and rhythms of things. So what Scott thought of doing was to measure basically monument construction. Turns out people at different times care more about doing different things.
0:45:25.8 SC: Okay, yeah.
0:45:26.6 LB: So that works very well. So that is also superlinear in this very same sense. So we've done this then in a bunch of other episodes, including Rome and Greece and so on. The Romans seem to have cities that had... So it works very well. The exponents for Roman cities are slightly different, and I think they have to do with the fact that the infrastructure networks don't go all the way to each house and so on. So they have these large blocks and they're very dense. And their cities, even though they had tall buildings, they don't have two-dimensional infrastructure, sorry, three-dimensional infrastructure. They don't have highways and tunnels as much as we do. But basically you have a very similar scaling. It's more like one-third.
0:46:09.4 SC: I think that we're going to allow ourselves to ask the question, where does that one-sixth come from? I mean, I just love the idea that in a space colony where we're truly filling a three-dimensional volume rather than mostly living in a two-dimensional surface, innovation will scale in a different way. So let's do the numbers. Tell me where the one-sixth comes from.
0:46:34.2 LB: Nice way of putting it then, it could be a whole other thing. Yeah.
0:46:38.1 SC: So what is the formula that gets us to 16%?
0:46:42.0 LB: Ah, okay. So the one-sixth is a little difficult to be the first thing to explain. So I have to take it one step back.
0:46:51.2 SC: Sure.
0:46:51.5 LB: The first thing that you get mostly the effect by getting an exponent that's more like one-third. And the idea is the following. So this is a little bit like the kind of calculation physicists like to do. So I hope that we can visualize this. But suppose you have a city that's not like such a complicated city, just has a bunch of houses over space and they're kind of scattered. But you can define more or less a place... We put a circle around it, right?
0:47:20.4 SC: Right.
0:47:20.7 LB: So we put always spherical city, right? Not spherical, circular city in this case. By the way, economists and geographers do this in order to create models. So this is a long... It's not just physicists. So you have a certain characteristic size for what that city is. So you can define a radius. Okay. So now you can say two things. You can say basically that travel around the city, if you go to visit your neighbors and so on, or maybe you live somewhere and you want to go to the center of town, is more or less proportional to the radius. So there's a cost of moving and that cost is proportional to this radius. Okay.
0:47:54.5 LB: On the other hand, if you move through this space, there's a certain probability you meet someone. And this is kind of basically... Or even if you come to the center, it's similar, it just has a different geometric factor. So basically the rate of interaction is proportional to the density, which is the population divided by the area times some factor that has to do with the shape. Okay, this is just basic cross-sectional behavior for the physicist, but that's the way it looks.
0:48:21.5 LB: So if you now equate these two things, you say, well, these interactions must be net positive because this thing is here. So if it were terrible, people wouldn't be here. So you say, well, so there's basically a factor proportional to density, N over A, and this must equal something that goes like the radius, which is area to the one-half. So if you kind of do the factors of area, you get that area goes like the population to the two-thirds.
0:48:49.6 SC: Okay.
0:48:50.9 LB: Just have the area, there's a one-half plus 1, and then you put it on the other side, there's the three halves. Put it on the other side, you get two-thirds. That's very simple.
0:48:59.4 SC: And so secretly, a big role is being played here by the fact that there are real-world constraints on what human beings can do. No one's going to live in a city where their commute time is eight hours to and from work.
0:49:10.9 LB: Exactly, exactly. Exactly. Very important. So, and this is kind of something that in practice people discuss 'cause often when cities become very big and very expensive, people commute for two hours. But that's basically the, if you want to say, it's the limit of the bound state, right? It's the escape clause that at some point it's not worth it anymore. So when you do these whole calculations, there are several ways in which a city can fail, which is precisely that if moving around becomes too expensive, then it kind of becomes Balkanized, right? A bit like LA, that you cannot travel across LA anymore so much, but you become sort of more local.
0:49:49.8 LB: So that happens a lot. Or cities that are very undense, like maybe sprawling in Florida or something. So that does happen. And then it looks like a city 'cause it's continuous build, but as an interaction system, it may not work that way. But then you also can have the opposite, which is that the city becomes so dense it jams. So that's more characteristic of cities that are developing that don't have yet the infrastructure that allows the highways and the subways that allow you essentially to tunnel.
0:50:18.8 LB: So this is where you go from one-third... So that's the two-thirds for the area. And then if you ask again, that density, N over A, now goes like basically one-third. And the total interaction is N times N over A, goes like four-thirds. So that's now superlinear. That's the amount of interactions realizable in this space.
0:50:43.1 LB: Now, if you think about what happened from where I started, you just have a bunch of houses and they're getting denser by these equations if you increase the population. So this system's going to jam. It will have a limitation at some point when it does hit the size of the buildings and the space in between and so on. So that cannot be a solution that... By the way, these kind of systems do exist and we measure them in archaeology, very small towns.
0:51:07.3 LB: But then as a city gets larger, it needs to start channeling movement. So you need to basically create streets... We call them streets, right, or subways, or pipes. But places of movement need to be dedicated, otherwise the system jams. And this gives you a different exponent. It works a bit more like a parallel circuit where you now have different routes.
0:51:29.1 SC: Okay.
0:51:30.2 LB: But you need also to build a hierarchical system. So in a city, if you go to a city that's just developing, that's just building its infrastructure, it has local streets, but it doesn't have the highways yet. So movement is very slow. So in order for the city to stay connected, you have to build these ways in which you can move faster across the city, okay. When you put all that together, you create a parallel circuit that gives you a different resistance. The cost of movement is no longer a constant, it's now reduced, actually, because of this infrastructure. And that ends up giving you the one-sixth number that's magical. So it takes a little bit of calculation, but that's basically observable.
0:52:16.2 LB: There's an experiment I always tell my students to do, which is very special. What this predicts is the following. If you fly over a city, we all fly over a city at some point, probably, right, and you look down, you will see the little streets, the local streets, and you'll see the highways, right, obviously. And you see the ones in between. What all this predicts is that the density of traffic, so you could see number of cars per unit space, in a small street is smaller. There are fewer cars per unit of space. The highways are occupied more densely. Not always, but when they're occupied, you see. And traffic moves, of course, hopefully faster in a highway when things are flowing.
0:52:56.4 LB: So what you have is that you have to build this hierarchical infrastructure in a way that has where traffic is denser but moves faster as these levels of hierarchy increase. And this turns out to give you just such a way that gives you a scalable system. It's kind of interesting, but that's where that comes from. So when you're flying, you will see the traffic flow being denser and traffic flying more quickly, faster in the highways than you see in the cities. And this is why highways are very susceptible to jamming.
0:53:32.9 SC: Yeah, okay.
0:53:34.0 LB: Because they're already dense and they rely for not jamming on a fast flow. So the moment that flow is slowed down, you see this, for example, at toll booths, toll plazas, then the whole thing jams, right?
0:53:47.5 SC: Right. And I guess...
0:53:48.9 LB: So there are a bunch of things you can start predicting that are also interesting.
0:53:51.7 SC: You're revealing your formative years as a physicist, I think.
0:53:55.5 LB: Yeah, hopefully.
0:53:56.3 SC: The physicist always wants to throw away as much as they can while still capturing something interesting. So I wasn't completely sure, was there a step in there where you... I mean, the density of people in the city seems to be relevant if I'm just encountering people at random.
0:54:16.9 LB: Right.
0:54:17.2 SC: But in my life, I don't really think I'm doing that, encountering people at random. So how good is that kind of assumption?
0:54:23.6 LB: It's good only on the average for the entire city. It's what we physicists like to call mean field approximation.
0:54:30.6 SC: Yeah.
0:54:30.9 LB: So you take just a big average, but of course, you and I and everybody else is not the average. There's no one who's well represented by the average. So this now requires that we look closer as to how people actually interact with others and the good and the bad that comes from that, the richer and poorer and so on. So this is important for real cities. And there are general things we can still say that usually people with more interactions tend to be richer, more educated, and so on. So there's sort of a knob there that still persists, that this idea that more interactions is a pathway. It's both enabled, but it's also associated as an outcome with more choices and more access and better outputs. So the condition of being isolated or segregated is associated always with poverty. So we see this, right? It's not abstract.
0:55:32.2 LB: So that already is telling you something interesting, but that now is in distribution. It's both structural, but then it's also functional in terms of what we said cities were for, which is that if you cannot realize the advantages of being in this place, then you're just suffering the bad consequences but not enjoying the benefits, right? So that's kind of where cities fail. And they fail both at the edges of people traveling very long times, but also inside the city when people don't have access to that network, are excluded for one reason or another.
0:56:07.2 LB: And so there are all these ways in which this mechanism is failing and this has consequences for individuals and for neighborhoods and for then the city as a whole as well, because this creates conflict and so on. But what's interesting about cities is that again, this pushes the system to work better. Now it's a social mechanism by which people try to reduce poverty, increase inclusion, improve public services. Of course, for anyone living in the city, you know this doesn't happen smoothly, but over long periods of time, that has been what has been happening mostly.
0:56:47.1 SC: And it's, yeah, I hadn't really thought about the fact that there are of course different subpopulations within the city and some of them are going to almost live in the equivalent of a isolated small town as far as their daily lives are concerned. But their existence, I mean, they're still feeding into the ability of the city as a whole to specialize and to allow for people who have more opportunities to really have superlinear sets of opportunities.
0:57:19.3 LB: Yeah. So this is another thing that's important when we think from the point of view of the social sciences, right? Because there are different modes of interaction, and it's not like it's a homogeneous thing. So for example, if you've been to any city, you know that all those people are not your friends, right? So these large-scale interactions are mostly economic, jobs and transactions. So that's a channel where you don't need to be friends necessarily, though in some cases you need trust and so on, that comes from repeated interactions or the reputation of the individual or the business.
0:57:54.9 LB: But nevertheless, those allow for people to be different. In fact, they require people to be different for there to be a transaction. So that's good in a certain way, but it's kind of not necessarily a very friendly, very close, very intimate relationship, right? And the same thing is also another channel is politics, which is very important. So politics, again, is supposed to bring everyone together towards the same sort of administration of the city. Certainly in most cities, politics and democracy, of course, are urban inventions. Politics, that's just all it means, right? It's city stuff, right? That's what it means.
0:58:30.0 LB: So but it's supposed to be sort of something in which people need to participate, be part of, right? So that's another channel of large-scale interactions between strangers. But then, of course, locally, depending how you live and so on, you also need privacy, both in your home, of course, and family life. But then there's an intermediate scale which is also very interesting, particularly in sociology, to do with neighborhood organization or community organization, which has both good aspects and bad aspects, aspects of segregation by race, ethnicity, culture, but also aspects of cultural diversity and the maintenance of that and mutual support and so on.
0:59:10.3 LB: So it's tricky, but you see again different mechanisms at different scales and the city being structured differently at different scales. So that's another aspect of a good complex system is that it has a somewhat different dynamic at different scales, but it still hangs together in terms of being a whole system, but has sometimes costs and benefits that play out differently at different scales. That's what keeps it dynamical.
0:59:37.5 SC: I do want to emphasize how dramatic a claim it is that there are these essentially universal scaling relations. I mean, you're telling us that the history of a city, which might be centuries or millennia old, and the local politics at the moment, Democrats, Republicans, or whatever your local political parties are, kind of don't matter that much for a certain set of really important quantities you might want to calculate.
1:00:04.7 LB: That's exactly what I'm saying. And I think it's amazing and it gives you so much intellectual opportunity when you think this way, because we've been conditioned to think that socialism or capitalism, Western democracy versus some East Asian thing, these things are fundamental.
1:00:23.6 SC: Yeah.
1:00:25.2 LB: I'm saying at some level that they're not. At least from the point, cities are more fundamental, right? And that cities exist as you change political system, as you change economic system to some extent and so on. No, there are things that exist in cities that are then unifying if you start thinking this way. So I think this is a great opportunity, actually. And it's not by accident that almost all the social sciences started by people looking at cities because they were flabbergasted by all this complexity and trying to make sense of it.
1:00:55.6 LB: So at the beginning, we talked already about Adam Smith and division of labor. Beginning of sociology is all about this, how are all these strangers hanging out together? This must be awful. Is it awful? What's happening? How people's behavior is changing, how is their cognition changing? Goes back to Marx and so on. So all this is happening and is it good or bad, right? But what this says when you look at an actual city, right, it's any city that we have that we live in, your city, in Baltimore, right? Or in Chicago. It's neither socialistic nor capitalistic, right? Has elements of both, obviously. Right, it's a mixture.
1:01:29.6 SC: Yeah, sure.
1:01:30.9 LB: And the same thing, is it politically also it's not a democracy all the way, right? There are firms and organizations where people make command and control decisions and so on. So I think it just shows you, gives you an evidence, proof, if you will, that societies actually are more complex but more subtle and different from what we've been assuming in a lot of our ideology, certainly, but also some of our basic ideas about society. And we have an opportunity, particularly as we see that these modes are more pervasive than, again, the geography or the culture or the political system that we have around the world, to have a common ground for understanding society, but also possibly to running them better in a way that's more, in many ways, more commonsensical, but also truer to the needs of people. So there's a lot happening.
1:02:23.5 SC: Yeah. Is there an unfortunate consequence of this if there's some like attractor mechanism that puts cities into a certain kind of distribution of various things going on? There's more poor people than rich people in a typical city. Is that a law of nature? Or I mean, is it always going to be true but we can raise the level of what counts as poor, or what can we conclude here?
1:02:50.8 LB: That's a great question. I think that's a question for the ages. I mean, there are several answers. I obviously don't have an ultimate answer, but I think we've been starting to understand even poverty and wealth differently. Let me just try to say that without, I hope I won't say anything that is insensitive, but just in the last... So one point that's interesting is that just in the last, with Sustainable Development Goals, which is 2015, there was this agreement that says we're going to end poverty worldwide forever.
1:03:29.1 SC: Right.
1:03:30.1 LB: Why did this happen? Why were we confident that this was happening? First, you know, there was less poverty, but a lot of what is happening in the background was the urbanization and economic development of East Asia, which took more than half a billion people out of poverty quickly in one generation. So there's this sense, which is a broad sense of development and poverty. So one of the things I'm studying now is, if you will, cities as engines of what we call social development or socioeconomic development in society. So you never have a society that develops without urbanizing, never.
1:04:05.6 LB: So, but cities, as you then point out, at a different scale, let's now go a different scale and look inside a city, right? Then you do see indeed greater, typically greater inequality because you have very rich people, right, in a lot of cities, but you also have poor people. And therefore this seems to show that in some sense the situation perhaps got worse, right? And there are certain episodes where that is true, particularly episodes where people are poor and excluded and therefore don't have the ability or the support to improve their own living conditions.
1:04:40.1 LB: So this does happen, for sure. But there's also a sense in which the way poverty happens in cities, it's kind of another interesting thing that cities tend to have a lot of poor people. And when this has been studied in detail, it's not only people that have been there a long time, but it's also people who come. So a lot of our work at the moment has to do with informal settlements or slums, particularly in developing cities. So it's not our cities here, but there's a parallel history here too, but probably too long to tell.
1:05:11.6 LB: And so there are a lot of poor people that are coming to cities and living in what we would identify as poor circumstances. You can ask, we're doing actually a survey about this, "Why are you there? Isn't this a terrible situation to live like this?" Most people say, "No, no, no, I like it. I live much better than I lived in the village, and my life can improve." So we have to try to understand if this is a permanent condition or a temporary condition where people are acquiring, not to say it's perfect, but acquiring more agency and more possibilities, more pathways to improve their life. That's one aspect of it, which is more the individualistic perspective.
1:05:48.5 LB: But the other aspect of it is that in cities we invented historically a lot of the social supports, which mean that being poor in a city is typically not being completely unsupported because there's social services. I'm going to rattle a list, and of course, there are insufficiencies about this in every city, but there are social services, there are shelters, there's public housing. There are a bunch of things that allow you, there are public spaces, there are streets. So there are things that you can enjoy and have that are not mediated by money, right?
1:06:23.6 LB: You don't have to have money to get them. This goes back to the Romans and the Greeks. They already had some mechanisms doing this for anyone who loves their history and so on, because if you don't do that, cities will explode, right? But the fact that you're doing that means that the money inequality that you see is somewhat mitigated by social interventions, social services, political interventions and so on, that tend to force the system to equilibrate itself more.
1:06:51.9 LB: And so this is the origin of social democracies, of social programs and so on. They're always things that people in cities understand and support, going back to politics, but that people in other environments think you don't need or can be done through personal charity or something. But in cities, they become institutionalized. They're necessary. And not to say that these solve the problem of poverty, but they change a little bit its character. And when these things work well, in the countries where they have worked well, you find reductions in poverty and social mobility.
1:07:24.9 LB: So I think this is a complicated answer, but it's just to show you that if you think about it as a series of dynamical mechanisms, I think the picture is a little bit different, and it starts giving you access to how poverty can be mitigated and perhaps ultimately, maybe eliminated.
1:07:40.5 SC: Well, I think there's sort of two things going on as far as I could tell there. I think I understood and appreciated the idea that there are more mechanisms for alleviating and maybe even reducing poverty. But this idea that in people's minds, they prefer to be in the city because it's more a prospect thing, right? If I can rephrase it as, in the village, I was poor and I knew I was going to be poor forever. In the city, I'm poor, but I have a chance, as far as I can tell. And I wonder both how important that is and also how accurate that is on the part of the people thinking this.
1:08:19.7 LB: Yeah, we don't have comprehensive hard data on this. We have a lot of surveys and a lot of ethnography. So that's the kind of evidence one gets. And there's also where we have hard data is what's the profile of the migrant. So there's almost a universal law of demography that the migrant is a young person. It's somebody between 18 and 30, and again, people move for various reasons, but that's the age at which you're building your adult life. You're building, in the modern world, a career, but also your social life and so on. And so it's the people that feel... As you just said, it's not everybody, but it's the people that feel they want to try it, give it a try.
1:09:07.3 LB: Then often you migrate, and tend to migrate to a city in this. And then if it doesn't work out... You see in the modern world people are going back and forth a lot. So we see this in COVID, there was all this... So it's not like people come and are stuck. That can happen, but it's rare. But people can try it out and go to another city or go to a small place again and so on. So we did once upon a time a very interesting study. It was a bit anecdotal, but we had the cell phones of people that were migrating to rural areas and cities in Kenya. And so we could see who they were calling.
1:09:38.1 LB: And so we could see that the people that got from the rural area to the city and they were calling people in the city, they were predictably going to stay. But the people that were still calling their moms back home were going to go back. So there's sort of an element of selection, experimentation. This is not to say... So there's a lot going on. But the fact that it works for a lot of people means then that we see this macro phenomenon that cities grow and that phenomenon takes a life of its own.
1:10:08.2 LB: But also what it does is to create this demographic imbalance where cities remain younger with people that are more motivated, who are more on the make. And then that is missing sometimes from the rural area, particularly at times of strong migration. So there's sort of a whole bigger demographic picture that's interesting as well. But it has to do sort of what happens across the urban hierarchy, as we say sort of euphemistically. But also, it has to do now... This becomes important with problems also of population decrease and so on.
1:10:42.9 SC: And if these scaling relations with the one-sixth in them, to go back to that, a lot of the explanatory mechanism that you told us about, like we said, relies on the fact that space is three-dimensional and in fact area on which we build is two-dimensional. Therefore, in a world where there's more technology and we can phone people and send emails and whatever, are these dimensions changing because now we have more longer-range interactions?
1:11:13.3 LB: Yeah, yeah, exactly. So yes, probably. Certainly now we have these new channels which have to do with cell phones and internet and everything else that we now have, telecommunications, as it used to be said. But so when we... And these are also networks. So to take a step back, to just refocus again, the central problem of humans, if you will, is the ability to create these networks where we can become interdependent and specialize. And this leads to more knowledge in the network, a greater ability of groups of people to do more things. And this leads to gains in economic productivity, but also just the general innovation across many fields, but also to a group of people being able to hold more culture, more knowledge.
1:12:00.5 LB: Okay, that's the picture. So therefore we need to find ways to build these networks. Or if we do and the networks are good, these things are going to grow. So the old-age way to do it is cities, or settlements and then cities, through space and time. But increasingly now we have these ways of making networks not through space, right? So I think these networks are good for many things, but they're not so good at others, right? So you cannot get a haircut on the internet yet, right? Or a delicious meal, maybe increasingly you can call it in, but still within your city, right? It's not going to be yet... I mean, unless you're crazy rich, it's not going to come from across the world.
1:12:41.8 LB: So there are many of these things that are kind of changing the relationship to space, but essentially reinforcing a similar dynamic. But I've done some studies, for example, of similar networks on the internet, and the exponent is different. It's closer to 2, as you'd expect. So 2 for... Instead of being seven-sixths, so the 1 plus one-sixth goes to become a larger exponent. So these networks have more scope, but they have more limited set of capacities for now.
1:13:15.6 LB: But obviously, there's this deeper dynamics of creating these networks and what they can and cannot do that is just playing out. I think at the moment, given the kind of creature that humans are, how we communicate with each other, the fact that just even looking at each other and being in the same physical space remains important. We had this big experiment with COVID, right? It seems that this will remain to be important. Now, an interesting question is, imagine it's all AIs, no people anymore. Will they have space, the AIs? And the answer probably is yes, but they may not be in space, right?
1:13:49.6 SC: Great question. Right, right.
1:13:54.1 LB: Because they don't need the embodiment and the kind of look and expression, looking people in the eyes and all this stuff.
1:14:02.9 SC: They don't need haircuts.
1:14:04.3 LB: But they probably will have to have some sort of signaling and so on. So these are all kind of crazy questions, but kind of maybe interesting. But the idea is there's some fundamentals here, that this idea intelligence is always collective, it's expressed in networks.
1:14:20.5 SC: Yeah.
1:14:21.1 LB: And you need to find networks that essentially can give you open-ended ways to building these things up.
1:14:26.7 SC: I guess that's something that we've let slide by and maybe is worth emphasizing. I mean, there's a bit here that undercuts the great man theory of science, right? Like, you're quantitatively reminding us that things happen when you get a lot of people talking to each other, even if it's a small number of people who come up with the final product, everyone else had a role to play somehow.
1:14:52.4 LB: That's right. I think there's some agency in being in a rich environment full of opportunities and connecting things, but those ingredients are there, and if not you, then somebody else will, right? So you find this a lot in scientific discovery, right, in the history of science, that it happens, two or three people came up with it and some get more famous than others, right? Particularly in invention, right, more practical things to do with technology. So I think that that is true. I think you need to have ingredients and latent connections and latent assemblages, if you will.
1:15:25.8 LB: But the agency of the individual in seeing those, perhaps in the future it'll be AIs, but of the agent in seeing those, connecting them, manifesting them, is still important. So I think in retrospect, we probably reward that role a lot because it is important. But yes, the whole substrate of the ingredients being there and it being latent, I think is absolutely fundamental. And this is why, again, invention and so on is always associated with large cities. It becomes much more probable, even if it doesn't tell you which individual's going to do it, it just tells you that somebody in that place is likely to do it.
1:16:01.4 SC: And I guess we can wrap up then with the most practical questions here. Like, have we learned anything that helps the people who are trying to make cities better, the politicians or the planners or the activists or whatever?
1:16:16.1 LB: Yes, in my mind, we have learned a lot. I think again... You know, this could be another podcast episode, right, just on that issue. I think we think about cities in many different ways, and many of them are quite wrongheaded, quite pathological even. I think a lot of people still think that cities should not exist because they see all the disadvantages. Couldn't we live differently? Couldn't we all just be online and not have crime? I mean, there's still crime online, right? We bring it with us. But nevertheless, be different.
1:16:51.7 LB: And so I think all this has made us begin to appreciate, there's been a big shift, what happens in cities that is good. All these social interactions, the idea that we become interdependent, this produces knowledge, it produces economic growth, but it produces many other things. It produces a different disposition in people that's expressed in politics, but also in many, many different ways. It's much more future-oriented and so on.
1:17:20.1 LB: And so it is in some sense associated with, again as I said, development or positive change. It's not to say that... It's partly because the problems are clear and present. So it's just you're facing the problems that are there, therefore people impel you to move. So if you start thinking about the city in this way, I think you can start understanding problems of segregation and poverty, as we did, as lack of connections or being excluded from the network much more deeply than it's usually done. You also start understanding issues even of infrastructure, that it must be enabling of this kind of dynamics rather than just being a thing you do.
1:18:01.6 LB: And I think you also understand that part of what cities create is not just the material cost-benefit, but it has to do with the thing you were emphasizing, which is this idea of an open future, right? More open futures, more possibilities, which are both individual, that an individual feels that, but also collective. That both the city and the nation, perhaps, and the world can have more futures that are more possible and some of them better than others.
1:18:31.8 LB: So without that latent invention, which is possible in these environments, then we're kind of stuck and things cannot get better. So I think if you think about a city more that way, which great mayors sometimes understand that intuitively, but it had not been science, I think you have something that's a different kind of complex system, something that has coherence across time and space, as we said, but at the same time you understand its failure modes and what needs to be improved and so on at the level of social policy and individuals and so on.
1:19:04.6 LB: But also you start understanding its scope into the future, including in three dimensions and future space stations and so on. And I think you understand something really fundamental about humans and sort of the kind of creature we are, why we create these systems and what could happen in the future.
1:19:23.3 SC: I always like to end up the podcast on an optimistic note, and I think you've already done it. We humans like to interact with each other, and it gives us a way to keep the future open to many, many possibilities.
1:19:33.3 LB: It's the most difficult thing we do all at the same time.
1:19:35.7 SC: And yet we can. All right, so Luís Bettencourt, thanks very much for being on the Mindscape Podcast.
1:19:39.7 LB: Great pleasure. This was fun. Thank you. Take care, Sean.
[music]