Will you be in London on September 18th? Join Yascha Mounk and special guests for a live recording of The Good Fight Club at 6:30pm at the Sekforde! You can find out more and sign up here—we hope to see you there!
Daron Acemoglu is an Institute Professor of Economics in the Department of Economics at the Massachusetts Institute of Technology. His books include (with James A. Robinson) Why Nations Fail, and (with Simon Johnson) Power and Progress: Our Thousand-Year Struggle Over Technology and Prosperity. In 2024, he was awarded the Nobel Prize in economics.
In this week’s conversation, Yascha Mounk and Daron Acemoglu discuss the impact of colonialism, the role of culture in civil society, and China’s strengths and weaknesses.
This transcript has been condensed and lightly edited for clarity.
Yascha Mounk: Congratulations on the Nobel Prize, which you won last year with James Robinson, among others. One of the things that’s interesting about that is that your approach to economics was not particularly mainstream or well-trodden when you started out: using differences in historical development to explain the vast differences in how wealthy countries are today, among other things you tried to explain. How did you start off on that methodology, on that way of thinking about economics, at a time when that went a little bit against the grain of the mainstream of the profession?
Daron Acemoglu: Well, there are really two answers to that. One is that I actually was drawn to economics, perhaps somewhat mistakenly, because I was interested in these historical questions and I thought economics would have the answers. It turned out that was not quite the case. Second, one of the things that has always fascinated me is how amazingly different human societies are, both from each other today and over time.
You can think of that as big evolutions that have taken human societies from hunter-gatherers to big empires, from big empires to city-states, from city-states to absolutist monarchies, from absolutism to some other mixed regimes, and today, something quite different. For now, we will see where we end up. That sort of big transition is historical, so you cannot understand it by just looking at the latest inflation statistics. History has got to be your data set. From the very beginning, I was drawn partly to history because of that. Also, these big events really have so much color and interest.
I remember one of my professors at the London School of Economics, a wonderful industrial organization economist, John Sutton, once said, “Econometricians throw out the outliers, but I think the outliers are the most interesting observations.” History is like that. History has a lot of outliers in it, but they are the most interesting things if you want to understand where we end up.
Mounk: Interesting, and just to explain to the audience, why is it that econometricians throw out the outlier? Because when you do a statistical analysis, you are worried that the correlation you find, the cause and effect you find, is over the top. There may be one random country where something else might be going on.
Acemoglu: Exactly. That is not a crazy thing to do, especially when you take into account that many datasets have mistakes, so outliers might actually be a reflection of those mistakes. But there is something to John Sutton’s statement that those very unusual cases have a wealth of insights for us.
Mounk: Makes sense. I was just traveling in East Asia. Japan, I think, is a cultural outlier in all kinds of ways. I can see why in many aggregate statistics, when looking at 180 countries, when you are not really able to think carefully about why a place is an outlier, you might want to exclude that. But if you want to understand something about the varieties of possibility of human culture and society, saying, “Japan is a really interesting outlier. Let’s go and try and understand why Japan is so different from China, Vietnam and from other neighboring countries, let alone from the United States and from Germany,” would actually tell you a lot.
You set out to try and understand why France today is so different from France in the 18th century and France in the third century, but also why France is such a different place from Japan, or from Nigeria, or from somewhere in Latin America. Of all the different kinds of historical factors that might help to explain those things, how did you narrow that down into a quite compelling central causal historical thesis?
Acemoglu: Well, I think you have to start somewhere. If you want to understand why France is so different from Nigeria, you probably want to start three or four thousand years ago, perhaps two thousand years ago. But James Robinson, Simon Johnson, and I were motivated at first by a slightly narrower question, which is why there are such huge differences in prosperity, for example, as measured by gross domestic product per capita and income per capita today, or at the time we were writing in the late 1990s.
Thanks for reading! The best way to make sure that you don’t miss any of these conversations is to subscribe to The Good Fight on your favorite podcast app.
If you are already a paying subscriber to Persuasion or Yascha Mounk’s Substack, this will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!
And if you are having a problem setting up the full podcast feed on a third-party app, please email our podcast team at leonora.barclay@persuasion.community
That is an unmanageably big question as well. One potential line of attack that seemed attractive to us was: if you want to understand the role of institutions, look at a big subsample where institutions were relatively recently reshaped through channels that we understand to some degree—the colonial subsample, the countries that used to be European colonies. That is a very interesting starting point. It turns out that the degree of rich-and-poor gaps that exist in the whole sample are replicated in the former colony sample. So it is very representative in terms of understanding why there is so much poverty and why there is so much economic development.
Sometime around 1500, Europeans started having an oversized effect on the institutions of these colonies. They did not impact their geography, but they completely transformed their institutions in many places. Look, for example, at Australia or what became the United States, which were largely empty or very sparsely settled. In some places, they took existing institutions and reshaped them, such as in Latin America, or completely erased and rebuilt them, such as in the Caribbean. That gives you at least a sample where you can have a chance of understanding some sort of, quote-unquote, exogenous variation in institutions. That is what we were after. Once you read the history, it becomes very obvious that asking a question that was still not so popular, but investigated by several scholars—“What’s the effect of colonialism?”—is not very well posed, because there is not just one kind of colonialism.
The colonialism in Australia and the northeastern United States was very different from the colonialism of, say, the Mexico Valley. So we said there is a big variation here. There are institutions often associated with Europeans settling there and building institutions, often not top down, but sometimes with conflict between colonial authorities and the people who went there as indentured servants or convicts, as in Australia. In other places, they set up extractive institutions in order to take the gold or the silver or put people to work. Those were very different kinds of things. That is what we focused on: that variation. We tried to find an exogenous source of variation, and therein lies the origin of our first foray into these topics.
Mounk: So one way of thinking about this is that, in the year 800, it would have been very difficult to come up with that kind of unified historical causal explanation, because Nigeria, present-day Chile, and present-day France were shaped by such different forces that it would have been very difficult to find one.
Acemoglu: Exactly. They evolved over time: in the parlance of social science, endogenously. Which are the consequences and which are the causes is a bit harder to determine.
Mounk: Then you have to exploit this kind of historical bottleneck where a huge swath of the world is deeply influenced by the imposition of institutions from the outside. Therefore, you can look at how these institutions subtly differ from each other. My understanding is that, starting in a famous paper in 2001 and then in the bestselling book Why Nations Fail, you argue in particular that the key difference is that in some societies the institutions were inclusive and in others they were extractive.
This was not because some countries were more enlightened or more well-meaning. It had to do with who the settlers were and how they were trying to exploit those resources. So tell us about that set of differences and why that is so historically impactful.
Acemoglu: Yeah. In some sense, some commentators, including Winston Churchill, in his History of the English-Speaking Peoples, said, “The British were much better colonizers than the French.” We do not find any evidence for that.
Mounk: An old Indian friend told me that in high school they unfortunately had an examination system that was not very conducive to free thinking. They had to make ten points in an essay on any topic, and those ten points were pre-given. They had to underline each point so the examiner could at a glance know what the ten points were. The ten points about colonialism involved nine points about how horrible colonialism was in India, reasonably enough. The tenth point was, “Thank God we were colonized by the British rather than the nasty French.”
Acemoglu: Yeah. It brings to mind an apocryphal saying attributed to Gandhi. When he was asked what he thought of European civilization, he said, “It would be a good idea.”
We do not find religion or the identity of the colonial power to be that important, but the conditions on the ground are very important. Several of them seem to matter, and we have explored them. The one that we focused on in that 2001 paper you mentioned, Yascha, is that a big difference is whether Europeans could settle. If the lower strata of Europeans went there, they could demand rights, et cetera. Versus, if Europeans just sent some administrators to exploit the local population. One important determinant of that is whether Europeans could settle or not.
If you look at the history, that is not a trivial consideration. Europeans many times tried to colonize and settle in Africa, and it often led to essentially 100% of the people who went there dying of diseases because they did not have immunity to the diseases that were prevalent there, especially yellow fever and malaria, but also some gastrointestinal diseases. In Australia, by contrast, it was actually healthier than tuberculosis-infested European cities. So the disease environment really varied across places.
That is the source of variation that we exploit in the parlance of economics. We use that as an instrument, meaning we focus on the variation coming specifically from that disease environment, and we try to document that it is not the direct effect of diseases and that it is not correlated with other things. That was the big methodological innovation of that paper, which we then built on and developed in various different ways.
Mounk: This is a very neat explanation, because you are saying, “Look, these colonial powers are trying to extract resources from these societies. In some places, they are going to send citizens from their own countries to dominate these societies and establish those institutions, because they are linked to the motherland in various ways. Because they are seen as culturally, and perhaps at the time ethnically, in some way, related, they can make much different kinds of demands.”
So you end up, in one case, with the relatively inclusive economic and political institutions of the United States—not without struggle, and certainly not without exploiting others in extreme ways—but that sets you up much more for economic development, as opposed to places where you do not have it.
Acemoglu: Yeah, that’s right. At the end of the day, sometimes people think of an alternative story, which may have a small grain of truth, but just very small: “Europeans settled and brought their own institutions.” It’s not actually that. In the end, what happened is that when people in the Mexico Valley made demands for better treatment, they were treated harshly. When European indentured servants made similar demands, they were somewhat more protected. That is where that cultural and ethnic overlap came in. The conditions there were also conducive to them running away, because it was sparsely settled. So the balance of power was very different in the places where Europeans settled. That is how institutions emerged.
It is not as if they had templates for good institutions from Europe and just imported them. In the 17th century, the institutions that European settlers—English settlers—tried to build, sometimes successfully, sometimes unsuccessfully, in the northeastern United States were much better than what was there in the UK.
Mounk: In a follow-up book called The Narrow Corridor—which I do not want to touch on in too much detail because your co-author James Robinson was on the podcast talking about it when the book came out, and people should go back and listen to that conversation—you make a slightly different argument. The argument is that there is a narrow corridor, as the title suggests, between places where the state completely dominates and places where society completely dominates.
What you need is a kind of balance between those two things. Explain that idea and explain how it is related to the central idea in Why Nations Fail. To what extent is it the same or different idea?
Acemoglu: Well, before I do that, let me say one other thing that shows the progression from the papers to Why Nations Fail to The Narrow Corridor. In the academic papers we wrote, we were looking at a given cross-section as the outcome we were trying to explain, and the institutions likewise. We were interested in modern institutional variation. One could get the feeling—and we did not do much to dispel that feeling in the early paper—that there are these institutions that, once you put them in place, they stay. That was never our view. Our view is that institutions, especially what we refer to as inclusive institutions, are a process.
You build them, they have to vary over time, they have to evolve over time, and they are constantly challenged. We start making that point in Why Nations Fail, where we talk a lot about the evolution of institutions, critical junctures, threats against institutions, et cetera, but still not to the level necessary for understanding that dynamic evolution of institutions. In some sense, The Narrow Corridor is one more step in that direction. That is the way in which it builds on and extends the Why Nations Fail agenda. It is about that evolution. It is about the process through which those institutions are built.
Another point, which in some sense builds on but significantly departs from or adds to Why Nations Fail, is that in Why Nations Fail we also note that state capacity matters. How centralized the state is, whether there are many groups that can pulverize state institutions easily, is an important part of inclusive institutions because you need third-party enforcement, you need public good provision, et cetera. That is what we take on much more seriously in The Narrow Corridor. That is what we see as one of the very important determinants of those dynamics, because the critical role that civil society and popular mobilization play, for example in democracy, is something that emerges and evolves over time.
Especially in the struggle between states and elites that try to dominate society, and society trying to find new ways—with collective action, with civil society, with culture—of living with those state institutions or sometimes resisting them. That is the dynamic that defines and happens in the narrow corridor, when there is a balance between state and civil society. We also point out the more usual case where that balance is absent and one of them dominates. When the state dominates, as in China, you have a despotic set of institutions where civil society is either non-existent or very dependent. You can also have situations in which state centralization does not happen at all. In those cases, it is all collective action, norms, traditions, and local adjustments, but without any third-party enforcement or public good provision. That is the overall picture that emerges from The Narrow Corridor.
Mounk: Thank you very much for this concise overview. I want to touch on one way in which I think your work is very optimistic, and then one way in which it is potentially pessimistic. The optimistic piece is that you are saying that it is much better to have inclusive than extractive institutions, which goes against the instincts of many people. In the United States, for example, many people want to argue that slavery is at the root of American wealth.
It is interestingly an argument often made by people who want to emphasize how horrible slavery was. They somehow think they are serving that argument by saying it is really because of slavery that all of American wealth was built. I think it would be horrible to think that we have a choice between having extractive institutions like slavery and fast economic growth, or having inclusive institutions where we treat people well but at the cost of less economic growth. Your argument implies that slavery obviously made some people very rich, extracting resources from one group of people at terrible cost to give to others. But overall, that is not why America became wealthy. Presumably that is an optimistic insight.
Similarly, I think the core argument of The Narrow Corridor is optimistic. We do not have to choose between having an active civil society and having a state that is able to accomplish important goals like public goods or some amount of welfare state. In fact, those two things can be complementary when they are both strong and keep each other in balance. So tell us whether I am right to read those two points as rather optimistic about what human institutions are capable of achieving.
Acemoglu: Well, I think you are broadly right, but there is a progression here. Why Nations Fail was written largely in 2009 and 2010. I think those were more optimistic times. Your optimistic take for Why Nations Fail, as well as your pessimistic case, is broadly right. I would just add the caveat that we always emphasize that centralization of resource allocation can sometimes work. It is not that centralization always fails.
You can have centralized allocation that pours resources into established firms or forces people to work for very low wages for certain periods of time, and that can give you a boost. Look at Soviet industrialization, or the mid stages of Chinese development over the last 40 years, or Prussian industrialization in the 19th century. They can work, but they have the seeds of their own destruction within them. That centralization, especially when bolstered by extractive institutions, is going to lead to inefficiencies and block innovation. That was the argument of Why Nations Fail.
By the time we were writing The Narrow Corridor, largely in the mid-2010s, we were already somewhat more pessimistic. That is why it is not The Corridor, it is The Narrow Corridor. You really need this fragile and constantly challenged battle between state and society, elites and regular people. There are other attractors. In some sense, we were also reacting to “end of history” type narratives that were already being challenged by others by the mid-2010s, but were much more dominant in the early 2000s: the idea that there is a natural progression. Let China get rich, trade with China, and ultimately China will look like the United States, will respect human rights, and will be democratic. We were saying there are other attractors.
You can end up with a very stable—indeed, very, very stable—despotic Leviathan. Or you can end up with no state institutions whatsoever, and that is stable as well. In that sense there is greater pessimism in The Narrow Corridor. By the time I took on the other important topic of technology and its direction, I think I became even more pessimistic, because technology’s direction centrally interacts with institutions. At least the way Simon Johnson and I express it in Power and Progress, it is another big set of traps into which human society can fall, and from which it will find it very difficult to escape.
Mounk: I very much look forward to returning to those questions in a later part of the conversation. You slightly teed me up for the potentially pessimistic aspect that I was going to ask you about, which is that I know Why Nations Fail and The Narrow Corridor were particularly influential in many developing countries around the world where people were thinking, “Well, what lessons can we take from this for how to succeed, for how to get closer to the economic performance of those countries doing the best?”
In a way, the historical approach is more optimistic than some other ones. I mean, some people try to explain differences in development by things like geography, by things like climate, saying, “Unfortunately, if you’re stuck in the wrong climate, there’s really not very much you can do.” You might imagine some people on the far right trying to explain it for biological reasons. Obviously, saying, “No, it’s because of what happened in the 16th century or the 17th or the 18th century in terms of how these institutions were set up,” is a more appealing story than those alternatives.
Of course, on the other hand, you might also read that as somewhat fatalistic, to say, “Well, you’re not screwed because there’s anything wrong with you, you’re not screwed because there’s nothing wrong with your climate, you’re just screwed because 500 years ago people set up the wrong institutions.” But that can in some sense feel just as disempowering.
Acemoglu: Yeah, absolutely. Let me clarify that last part, then I’ll say a few more things about the first part of your comment. That is a very important point; some people make that. Let me clarify: we never say that what happened 500 years ago is inescapable. It’s just an influence. In the parlance of the instrumental variable strategy, it’s a predictor. It’s not a full determinant. It determines perhaps 30% to 40% of the variation. There’s a lot of agency there—not individual agency, but collective agency. There are many countries, like Botswana, which suffered the most exploitative, most negligent type of colonization, and then ended up with very different institutions. Obviously, we get into the issue of the equivalent of the discussion of free will versus determinism in biology and individuality, and how much of that was because of other conditions that enabled them, etc. There is certainly agency, broadly construed.
Mounk: What does that agency look like? If you’re in a place that has an inheritance of very extractive institutions, what is required in order to successfully reform those institutions and make them more inclusive and therefore more conducive to economic and other forms of success?
Acemoglu: Certainly, leadership plays, the quality of leaders does play some role. More importantly, in my opinion, is culture. Meaning different cultures provide different menus for the things that you can do. That menu, of course, doesn’t need to translate into action. By menu, I mean there are many possibilities, but having those possibilities and then choosing the right ones is critical.
Organization of civil society plays a very important role. Britain was not destined—absolutely not, and I emphasize that in every book—Britain was not destined to be a democracy. Forces that made it authoritarian, top-down, hierarchical, were extremely strong throughout the 16th, 17th, 18th, and even 19th centuries.
There was demand from the bottom up, so this organization of civil society is part of the agency. Then ideas matter. If we did not have ideas related to liberalism and liberal democracy and popular sovereignty, we would not have templates to provide alternatives to those that were being proposed by absolutist rulers, like the divine right of kings in England. The right ideas, again, and then we have to get into where ideas come from, how they spread, et cetera. That’s what makes social science interesting.
If I can say, Yascha, just in response to your earlier question of optimism from history: the way that I’ve come to express that in the context of technology is that I’m not an optimist, but I’m hopeful. There are several things to be hopeful about. We are apes, and we’ve come so much further than apes. One of the things that I think chimpanzees or gorillas would find completely impossible to understand is how we build this amazing power.
We create concentrations, and then we use them mostly for decent things like providing public services, even in the United States, which has now morphed into dysfunctional politics. Most of the power of the state is still used relatively benevolently. That’s just an amazing achievement. We’ve come from Egyptians, Pharaohs, and tyrants to some version of democracy. That’s an amazing achievement.
People, Marxists, and extreme leftists never tire of criticizing market economies, and they have a point. But look at the 35 years or so from 1945 to the mid-1970s, where you had very rapid growth, tremendous growth of wages, and relative equality. Inequality declined, wages increased for low-skilled manual workers. We’ve achieved things that the greatest pessimism would say are impossible, but we haven’t achieved them consistently. That’s why I’m not an optimist. There isn’t a natural set of forces that will always take us towards everybody being happy and prosperous. But we’ve made big leaps in showing that those are possible from time to time.
Mounk: I fully agree about that. One of the places which you might say has accomplished remarkable things in the last 30 years is China. You mentioned China briefly earlier. It is a country that went from being desperately poor as recently as the 1980s to being a middle-income society, with parts of the country and stretches of the country that are very affluent. I recently spent two weeks in Shanghai and some of the surrounding areas, and it is just a remarkable miracle of humanity how the life of the average Chinese person has transformed for the better over the course of those decades.
In a sense that is surprising because China’s economic institutions, and certain political institutions, remain extractive rather than inclusive. So I guess the obvious question is whether you think that China eventually is going to run into the same trap as something like the Soviet Union. The story about the Soviet Union is that they were able, through central control and investment and other things, to make a real leap in development at a very high human cost. For a few decades many Western economists kept predicting that the Soviet Union was going to overtake the United States in its GDP per capita, and central planning in the 1950s in many Western countries was seen as the wave of the future. Then it turned out that the Soviet Union got stuck at a certain economic level.
Obviously, China’s economy is much more complicated than a central planning economy and involves many elements of the market. Perhaps you might argue that, therefore, its core economic institutions have become much less extractive, and perhaps that is why it could continue to grow. What is your prediction about where China is going to go? If the basic economic model of China remains roughly what it is now, is it potentially able to catch up to the GDP of the United States, or as Martin Wolf recently argued in this podcast, of a country like Portugal? Or do you think that it will remain far behind those levels because of the extractive elements in its system that remain in place today?
Acemoglu: That’s a really critical question. We all have to struggle with it. I think there are several points to make. One is to absolutely reiterate what you said. It’s a big achievement. We can quibble with a few aspects of it. Inequality has skyrocketed, wages haven’t grown anywhere near the same rate as GDP, and they were stagnant for 20 years. But it’s still a tremendous achievement. The first part of that achievement was exactly what you said at the end, which is that China, with great potential, was hugely underperforming because it had some of the worst extractive institutions.
Chinese agriculture, which is the reason why China has a billion and a half people—it is very productive agriculture—was completely decimated by repression, extraction, lack of property rights, and distorted incentives. The first phase of liberalization in the 1980s and 1990s was agricultural. That’s where the first big bout of growth came. As my colleague Yasheng Huang has documented and argued, that’s when wages and living standards in the countryside, and some in the cities, improved.
That’s a very classic transition from extractive to inclusive. The second phase is much more complicated. That’s what Jim and I would call extractive or despotic growth. It’s much more centralized and top-down, especially after Tiananmen Square. That was still an achievement because many other countries have tried it, including the Soviet Union, and they haven’t done it as successfully. China did it very successfully. It was able to attract foreign investment, foreign companies, and their technology. It was able to get people into cities and jobs in a relatively controlled manner. It was able to boost savings and channel all of that money to state-owned enterprises and then to large privatized companies. That was a success, but it was distorted growth with many inefficiencies.
Now we have the third phase, which I think is even more complex—technology-based. My research and my overall take is that there are two forces here pushing in opposite directions. One is that China has some advantages. Going back to before the Tang dynasty, but especially with the Tang dynasty, it had a relatively meritocratic, sometimes distorted, but still relatively meritocratic way of selecting people to the top that has persisted across different guises. That is what generates the largest cadre of engineers and generally well-trained engineers in China. What we are living through now is digital technologies, AI, robotics, and other advanced equipment, where you need a lot of engineering talent. China has that in plenty.
Mounk: Part of it, by the way, is a relatively meritocratic selection procedure for the top of society, despite political connections and party membership playing a huge role. Another part is that it drives broad-based education and skills acquisition throughout society.
Acemoglu: Right. One hundred percent. That’s what I was trying to say. It’s a continuation of that imperial education system, which was at times inefficient because it emphasized the wrong skills, but it was always focused on selecting the best to serve the ruler. That is how the education system still functions. It is perhaps more meritocratic than the imperial education system was at some points. Going against that, the top-down system also creates many inefficiencies. You see that in the way capital is allocated, which builds up a lot of misallocation of funds and companies saddled with inefficient investments they cannot get rid of. Second, you see that even in the innovative sectors of the economy, career concerns, political influences, and other factors interfere with research. Some of the work I’ve done, for example, looks at the Chinese academic system and traces the implications of these inefficiencies. Innovation is very difficult under such challenges.
There is also a lot of instability in the system because everything is top-down and driven by the Chinese Communist Party’s will. Think of what happened during COVID. You stick with a bad policy, and then overnight you change that policy. That sort of reversal is very costly.
It’s the battle of these two forces. For example, you see a company like DeepSeek, which compared to its U.S. counterparts, with a shoestring budget, came up with an AI model that rivals Anthropic and OpenAI. That’s a big achievement. What is that achievement? It is engineering. They made highly skilled engineering advances. But when you look at where the technology frontier is, many companies are still behind the United States and Europe.
That battle is going to play out. My prediction is that China will continue to grow, but whether it will reach Portugal-level income per capita is an open question. I don’t think it’s going to reach U.S. or German levels anytime soon. I think the contradictions within the system are going to become more important before that happens, despite the fact that I expect more DeepSeek versions of engineering marvels to come out of China.
Mounk: Yeah, I would love to speak in much greater detail about China. Again, I just spent some time there and came away with a deep sense both of the strengths of the country and the system, and of its weaknesses.
Acemoglu: What would you add to what I said in terms of weaknesses?
Mounk: Well, I think on the strength side, first of all, twenty years ago there were still people saying that China is not a free society and that free societies cannot innovate. The existence of things like DeepSeek, as you said, clearly indicates that this was wrong. China is very innovative in artificial intelligence, electric cars, and complex manufacturing processes across virtually every product in the world. One of the interesting things I did while I was in China was go to Yiwu, which is not a great manufacturing center, but it is the biggest wholesale market in the world. Among other things, something like eighty percent of the world’s Christmas decorations change hands there. But it’s not just Christmas decorations; very sophisticated goods are also displayed in the electronics district of its huge market.
At the same time, I think China has not been very good at creating good lives for its citizens. Part of the perception of China came during a brief moment when TikTok was briefly banned, or something like that, and people ended up on Xiaohongshu, a Chinese social media app accessible from the United States. Most of those exchanges showed people saying how great life in China was, and Americans bought into it naively. I think that is clearly not true. You have an educated class that is incredibly squeezed by the high costs of rent and property in Beijing and Shanghai.
Chinese complain about that at least as much as graduates of good universities in America complain about the rent in New York City and San Francisco. Work culture is incredibly punishing. Working six days a week, twelve hours a day, is absolutely standard. Many of the things seen as the strengths of China economically actually come from its still having a giant low-wage sector. It’s not that China’s delivery apps are some miracle of engineering that allow great comfort for people. It’s that you can order something for a fifty-cent fee because the person delivering it is living in a dormitory with middle school friends on the outskirts of the city, working nonstop for very low pay.
I had a lot of conversations with cab drivers who lived that kind of life. They were all internal migrants, or at least eighty percent of them in a place like Shanghai, living in makeshift dormitories with their middle school friends.
Acemoglu: I think those two points are very important. Low-wage labor, repressed labor, has been one of the things that enabled Chinese industry to do so well with an export direction. The fact that quality of life suffers is also significant. Once you have a middle class, they are going to want other things, including more information and freedom of speech, and that is being repressed. People internalize that, but it does reduce their quality of life.
Mounk: One of the interesting upshots of all of this is that it is at least broadly discussed within China. It’s hard to tell to what extent it is true, but there seems to be a decline in ambition. Fifteen years ago, people in their twenties desperately wanted to go to the big centers of innovation and advance their careers. Now many of them have the ambition of going back to the third- or fourth-tier cities where their parents live, moving in with their parents, or living in an apartment they may have from real estate speculation that they cannot sell because the mortgage is now underwater. The goal is to have a sort of common man’s life, where you have a state-sponsored job that works you a little less hard. If that lack of ambition spreads, it will affect economic development as well. I don’t want to make this only a China podcast. I wanted to set this as a preamble because I want to talk more about China, but there are many other things we need to cover. One of those is the opposite, which is to say the strand of work about when and how democracy emerges.
Tell us a little bit about the conditions in which democracy tends to emerge and what that tells us about the future prospects of democracy. For example, if we think about democracy as the balance between the costs of repressing demands for democracy and the costs of acceding to democracy, some of the technological changes we are now seeing might not make it easier to continue refusing demands for democracy, whether in China or in other places around the world.
Acemoglu: These are really critical questions for our present moment. Of course, there are millions of factors because there are thousands of different versions of democracy. But I think one factor is overwhelmingly important: demand for democracy. You don’t get democracy by design from the top. Sometimes voting rights are granted, but durable democracy is much more likely to arise when there is a bottom-up demand for it.
Going back to Greek city-states such as Athens, or to the early phases of British democracy in the 19th century, people—middle classes, working classes, sometimes women—who were outside the system demanded rights. That is a very good path to democracy. But the converse is true as well.
Liberal democracy will have a very difficult time when the population loses its demand and support for it, when people think democracy is not delivering, whether correctly or incorrectly, and when their aspirations are dashed. That makes it very difficult for democracy to survive.
Mounk: Was the time in which people felt that the system was really delivering for them a strange historical economic anomaly? You mentioned earlier the trente glorieuses, the Wirtschaftswunder, the decades of the 1950s, 1960s, and 1970s in which there was a tremendous transformation in living standards, life expectancy, and levels of education. Even the experience of social mobility at that time did not require an ordinal change, which is always painful because it not only involves one person rising up but also another person falling down.
You could go from being a poor peasant family to a family of factory workers, to a family of university-educated white-collar workers without changing your relative position in society. We may not get that moment back. Even if we achieve big economic growth that provides more things, better entertainment, and perhaps more travel, it will not be the same as going from living on a farm, worrying that if the harvest failed there would not be enough to eat, using an outhouse, and lacking antibiotics, to the life of a middle-class white-collar worker in Western Europe or North America.
If that is the case, is the background satisfaction that people felt with democratic and economic institutions for some part of the last sixty or seventy years simply never going to return? Are we destined to go back to more restive, more unstable political systems that are perhaps more typical of human history more broadly?
Acemoglu: Well, I think the question of when democracy fully delivered has a very simple answer: never. That is not just because nothing is ever perfect in the world. It is also because democracy’s promises and aspirations are very lofty. More importantly, going back to our Narrow Corridor discussion, you should think of it as a process. If people are not making more demands and are not somewhat dissatisfied with their lot, democracy is not going to advance. At every stage of democracy’s history, you will see people complaining about it, and rightly so.
By and large, this is where the trente glorieuses—the postwar miracle decades in Germany, France, the UK, and the US—come in. During that period, the three core promises of democracy were largely satisfied. Shared prosperity: economic growth happened, there was more prosperity, and it was shared across every group and every education segment. Men, women, and different races took their share, some more so than others, meaning preexisting inequalities closed. Public goods: much better services, better health, better education, better roads, better social insurance, especially in Europe, but to some degree in the United States as well with the War on Poverty. And voice: foundational to democracies is that people feel they have a voice. That was imperfect, and they always complained, rightly so—it is representative democracy, after all—but by and large people participated, for example in the United States in local politics and in the national political process.
Over the last forty years, democracy has not measured up to its own aspirations. My work shows that democracy is better than authoritarianism, and over the last forty years democracy has indeed been better than authoritarianism. But relative to its own promises, it has failed. Inequality has skyrocketed. Many demographic groups have seen their positions deteriorate, let alone catch up or keep pace with those at the very top. The quality of public goods has become pitiful in all dimensions.
There is also a sense—whether real or manufactured is open to debate—that political elites, bureaucrats, and politicians do not speak for the common person anymore. All three of these failings have become major points of contention. The fact that people are turning their back on democracy should not be surprising. In fact, in statistical work, my co-authors and I find that when democracies deliver on these promises, as well as on other things like controlling corruption, their support grows. When they do not, their support declines.
Mounk: I know that you’re working on a book that is of special interest to me and to the kind of things I think about, and that is on liberalism. I think you’ve started to give us a preview of the arguments in that book, if I am interpreting the answer to the last question correctly. One of the ways you frame this is that you have real concerns about establishment liberalism and that you want to return to a more radical, change-demanding version of liberalism.
Tell us a little bit about what you mean by establishment liberalism and how those of us who want to defend and advance philosophical liberal ideas can return to the more radical roots of that philosophical tradition to make change and preserve those values under very challenging circumstances today.
Acemoglu: Well, thanks for asking that, Yascha. It is a multifaceted argument that I try to make in the book. Let me give you what I think are the most important parts of it.
First of all, we should define what liberalism is about because the word means many things. By liberalism I mean a left-leaning version of liberalism: a commitment to individual freedoms, equality before the law, a belief in progress, and the idea that progress is feasible. It is not as though we can never improve relative to our existing lot. Liberalism is also a commitment to helping the disadvantaged, the discriminated against, and the weakest members of society. That left-leaning is a critical part of liberalism. You see it not in every thinker labeled liberal, but certainly in John Stuart Mill, for example. Social democratic liberalism is very much committed to that.
I believe the best justification for liberalism is that it is the best arrangement for producing collective knowledge, which makes progress possible, especially in service of other aims. If you want to provide better healthcare, you need social knowledge about how to do that. That is a problem of individual-level and community-level experimentation, and liberalism is the best way of producing and sharing that type of knowledge. That requires, however, a very different type of liberalism. A top-down liberalism would not work because you need bottom-up participation. You need community—something that has become antithetical to establishment liberalism—as a central part of both community-level experimentation and the sharing of knowledge. After all, the best way of sharing knowledge is through trusted networks of people who have enough shared characteristics that trust and belief in the possibilities of doing things together become central.
If you think about it this way, you see another element of the equation that runs against some prevailing views, which sometimes pit liberalism and democracy as conflicting ideals, such as in the ideas of illiberal democracy or tyranny of the majority. If you think about it differently, self-government—individuals taking part not only in the governing process but, more importantly, in local governance and community decision-making—is a very important part of liberalism’s agenda. This does not mean there is a seamless unity of liberal democracy, but you cannot have liberalism without some form of self-government.
That is the normative justification for liberalism. In the book I also argue that, because of both political economy reasons and the freshness of liberal ideas, we practiced some of these principles. Liberalism in opposition was very successful in making these aspirations real. That is how people started demanding rights, how women demanded the vote and better treatment, and how laws protecting them in marriage and elsewhere were enacted.
Those were liberal ideas. But just as importantly, during both the early and mid-20th century, self-government became a reality for the labor movement. Local communities organized and became more cohesive, and markets became better situated within communities. That was an important step toward shared prosperity and better delivery of services.
The post-industrial period created big problems for liberalism. Digital technologies and automation severed some of the links between productivity growth, market production, and shared prosperity. Equally importantly, the rise of the college-educated elevated them into a more coherent political force. They adopted their own version of liberalism—establishment liberalism—which they tried to impose on the rest of society unsuccessfully.
Therefore, going back to the roots means returning to what I call a working-class liberalism, where communities and people from different backgrounds are brought together and given more self-government rights, always within the constraints of respecting basic freedoms. Jobs and shared prosperity must again become a central tenet of liberalism.
Mounk: I agree with many parts of this analysis. I agree that liberalism has been captured by the tastes and predilections of a certain kind of affluent, college-educated social milieu, which has both tried to impose its values on the rest of society in the name of liberalism and acted in its own material interests, cloaked behind a layer of fancy rhetoric.
I agree with the fact that many rules and regulations, inspired in some senses by liberalism and in others by a broader progressive ideology, have made it much harder to build and deliver public services. All of that is at the core of the frustration with our political system that then leads to the election of dangerous demagogues like Donald Trump. I also agree with you, and I made a similar point in my last book, The Identity Trap, that liberalism is by its nature a radical and progressive creed. It started out as an ideology looking at a deeply illiberal world and insisting that realizing liberal values would make the world a better place.
Now we are in a halfway house where some of our values have been realized, and we are rightly defending that status quo—for example, defending the rule of law and the separation of powers in the United States at a moment when they are under attack. But we have sometimes become too complacent about pushing forward and asking what vision of the future would be more liberal than what we have now. On all of those points, we are broadly agreed.
I am despairing a little bit about the ability of liberals to put that into practice. When I look at broadly liberal political parties and movements, I think they are stuck in a way of doing things, in a language, in a set of assumptions about the world, and in a broader worldview, much of which has been disproven in the last twenty-five years. Much of it has alienated people, and they are not capable of letting go of it enough to sound as if they are making a fresh political proposition.
I very much like your emphasis on a more working-class liberalism, but I still struggle to see what that would mean in practice. You are a distinguished economist, not a campaign strategist, but tell us a little bit about how we should talk about this. What would that mean for governing projects? What would that mean for how politicians should speak? How can we explain this sophisticated set of ideas and turn it into something compelling for ordinary voters who, out of frustration, are tempted to defect to illiberal candidates and parties?
Acemoglu: Well, I think when you started your statement I thought I was going to say, “It’s fantastic we agree on ninety percent.” After your “but,” I realized that the remaining ten percent I agree with as well. I don’t know how the campaign strategy for this should go. What I say at the very beginning of the book—if it remains in the final version—is that the pendulum has swung. Now the energy is with post-liberal, anti-liberal ideas. We do not even realize it, but that is where the energy lies at the moment. It takes an idea to beat an idea.
The first step is for liberalism to articulate a set of ideas that are more aspirational. The campaign strategy should follow, and I am hoping to contribute to that. A liberalism that turns its back on social engineering and top-down imposition of values, that adopts community, heterogeneity of communities, and diversity of communities—from the conservative end to the very different end—as virtues rather than things to be erased, and that makes good jobs for people of all skill levels a key tenet, could be transformative.
Finally, as part of that, replacing top-down technocracy with a greater commitment to self-government could make a real difference if we can achieve it.
In the rest of this week’s conversation, Yascha and Daron explore how artificial intelligence will change the world, the impact it will have on the job market, and how we should prepare. This part of the conversation is reserved for paying subscribers…
