9° Hahn Lecture 2026

I'm wondering if I can use this. Can you hear me? Oh, terrific. Okay, I'm very bad at sitting down. So, I might even come and stand out here. Is that too outrageous? Okay. So, wonderful. Thank you very much. I'm extremely happy to be here. Thank you very much for inviting me. I'm very honored to be giving the Hahn lecture because actually I have a bit of a personal history with Frank Hahn. I'm not sure to what extent they knew that when they invited me.

But when I was a student at the London School of Economics, of course, it was Mrs. Thatcher. It was the days of that letter, you know, criticizing the economic policy of Mrs. Thatcher's government. And, you know, I was just desperately trying to understand like what the issues were and what did economic theory say about this. So, I became very obsessed with the work of Frank Hahn. And actually after I graduated, I took a job working as a research assistant at the London Business School. And I attended some of the lectures he gave at the LSE as a sort of guest lecturer on the kind of microfoundations of unemployment and how to think about monetary policy.

I have to say, you know, so here's, this is some of the work that kind of inspired me, especially his book Money and Inflation, which talks about, you know, what's the right way, what's the kind of right theoretical microfoundations for thinking about monetary policy and the impact of monetary policy on real variables. And I have to say I'm not going to talk about monetary policy. But I am, by the end of the talk, I'll clarify exactly why I am not a monetarist. And I think the reason I'm not a monetarist are sort of different from the reasons that Frank Hahn was not a monetarist, but I hope that he would have approved of the spirit of my discussion.

Okay, so he was an immensely inspiring person and I have to say that, you know, grappling with his work, his work is very challenging to read for those of you who've tried, but it's also very rewarding and it put me on a path to studying general equilibrium theory when I was a PhD student at Yale, as you mentioned. My thesis advisor was Truman Bewley, who was also a friend and great admirer of the work of Frank Hahn. Although he did, Truman did write a withering critique of Money and Inflation. So if you're going to read Money and Inflation, do look up Truman Bewley's review of the book. So anyway, just to say for those of you who haven't engaged with Frank Hahn's work, I think he was, it was immensely sophisticated and very inspiring to me. So you can see why I'm so happy to be here today. Okay?

And rather than, you know, but Frank wrote on many things other than monetary theory, and a paper which I put on the right here with Robin Matthews, a survey of the theory of economic growth, I actually read when I was an undergraduate. It was more or less my first introduction also to the theory of economic growth. So I'm going to start more from thinking about economic growth and economic development rather than monetary economics since most of my work has been on that topic. But as I said, I'm going to circle back to why I am not a monetarist. I have to say also that Frank wrote a fabulous article called The Winter of Our Discontent about capital theory quoting Shakespeare, Richard III, and it's one of my ambitions in life to write a paper with the title The Winter of Our Discontent, made glorious summer by this son of York, if you know the play.

Okay, so all right. So let's think about economic growth, which I feel more comfortable in doing than monetary economics, and think about how to think about economic growth. Okay? So we know a lot, you know, we know a lot about the facts about the enormous variation in the world in development outcomes. And, you know, using the terminology of North and Thomas in their 1973 book, The Rise of the Western World, we know a lot about, in some sense, the proximate determinants. You know, the theory of economic growth, as Douglas North said, is about the proximate determinants. It's about capital accumulation, physical capital, human capital, innovation, technological change.

And sitting behind that, we know, you know, what North and Thomas, we know something about what North and Thomas called the fundamental determinants. Okay? And a lot of my work has been about that, you know, on institutions, you know, thought of in different ways, security of property rights, you know, financial market development. I haven't done much work on that, but obviously it's important. And the topic, my kind of more, the focus of my talk today, kind of state capacity. So, there's growth theory, there's factor accumulation, there's innovation. Lying under that, there's a lot of factors which we think create incentives, create opportunities that stimulate accumulation, that stimulate innovation. Most of my work has been on that, trying to understand the role of institutions, trying to understand the massive kind of institutional variation that we see in the world and how to think about that.

Okay? But today, my main focus is going to be on, let's kind of zoom in on these fundamentals. You know, fundamentals, okay, but they're also endogenous. You know, that was why we had to do the kind of, that's why we had to adopt the research designs that we did. We wanted to try to understand the endogenous distribution of institutions in the world. And so I just want to start with this point that these things are outcomes. So to talk about like state capacity, say, what does that actually mean? You know, it means the state is good at doing things. The state is good at raising taxes. The state is good at kind of organizing society, you know, promoting, producing order in the way Professor Pagano was talking about. Okay, but what does that, how do you actually do something like that? What does it mean? Okay?

And I think we have in economics, we've adopted a kind of hegemonic paradigm to think about that, which really goes back to the work of Max Weber. We were talking this week about how the first book that I read as a student at the LSE in a class taught by Michio Morishima, another distinguished former colleague here in Siena, was Max Weber's Protestant Ethic and the Spirit of Capitalism. That was the first book that we had to read. This comes not from that book, but from his enormous treatise Economy and Society. And Weber kind of pointed out that, you know, there was this fundamental transition in modernity between what he called patrimonial authority, society organized patrimonially, and a society organized by rational-legal authority. Okay?

And I think what's happened in economics, a large amount, this originally comes from sociology, of course, political science, economics, is that they've taken this as being the kind of paradigm of how you get state capacity. So Weber talks about, you know, what's patrimonial, what's organizing authority or the state along patrimonial lines, what does that mean? The state is organized not on the official's commitment to an impersonal purpose and not obedience to abstract norms, but to personal loyalty. Power is wielded with discretion and unencumbered by rules, and this allows for the use of rents or emoluments in exchange for the rendering of certain services. So, the state is organized through relationships, patronage, without rules, kind of informal order. And he contrasts that with rational-legal rules, okay? Bureaucracy.

Now, Max Weber doesn't talk about state capacity, but if you read Economy and Society, you can see that he had something very similar in mind. When he talks about rational-legal authority, a rational state organized along rational-legal lines, he clearly has the idea that this was a very efficient machine, and it enjoys, here it is, a technical superiority over any other form of organization. So, I think this is, you know, this is what economists have come to think of as constituting state capacity. So, creating state capacity is where I look at a world, you know, and I look at a country like Denmark, you know, that has state capacity, you know, then it has state capacity because the state is organized along these rational-legal lines.

And if I look at a country like Nigeria, which does not have state capacity, then we say, "Well, you know, it's patrimonial. It's patrimonial." So, for example, in African studies or in, you know, the political science of Africa, the state would be described as patrimonial or neo-patrimonial, adopting this Weberian terminology, okay? So you might think, you might ask, you know, where does this come from? Why is the state organized patrimonially? Well, you know, that's not going to be a big focus of what I'm going to show you today, but Weber talks about this is a kind of historical thing, you know, it's like the idea of the king as sort of a patriarch or, you know, related to the organization of the family. So there's a kind of traditional, you know, he talks about time out of mind.

You know, it's also a way of organizing power. When scholars would think about patrimonialism in Africa, they think of that as a way of organizing power and kind of stabilizing power, controlling people, entering into, you know, patrimonial relations with them. And, you know, why is it that rational-legal states emerge, say, in Western Europe in the early modern period? The usual story is this is connected to the military revolution, interstate warfare. If you're going to fight a war effectively, you need fiscal resources, you better create a tax system, you need an army, you have to pay the army, you know, etc. So, it's interstate warfare that kind of forces this transition from patrimonialism to rational-legal authority, okay?

So, here's some evidence from a paper, a survey paper in the Annual Review of Economics by Tim Besley and colleagues. And on the vertical axis is just GDP per capita, and on the horizontal axis is a measure of kind of how bureaucratized the state is, which comes from this V-Dem project, if you know, in Sweden, V-Dem, which produces a lot of data, comparative data, kind of contemporary and historical on political institutions, state institutions, etc. So, they measure, they try to measure how bureaucratic different societies are, and this is looking at, you know, how are people, how do people get appointed to the state? You know, they appointed through examination or based on meritocratic criteria, how are they promoted, etc.

Okay, so this is actually again work that sociologists started doing, you know, that my former colleague at UC Berkeley, Peter Evans. You know, Peter Evans did some pioneering work trying to kind of code the extent of meritocracy across countries in the late 1990s. Okay, so you can see there's an extremely strong relationship here between, you know, higher on this, you know, these different indices that are averaged into some score between zero and one, you know, higher kind of bureaucracy score, much higher level of GDP per capita. Okay, so that picture is sort of interpreted as saying, okay, this is a very fundamental institution you kind of need here, you know, and it's almost necessary and sufficient to become a developed country.

Okay, if you get a state like, you know, Denmark up there, Luxembourg, Norway, I'm not quite sure where Italy is, then that's going to be associated with all good things, you know, public good provision, secure property rights, incentives, etc. Okay, so but let me start with that picture and sort of say, you know, what that picture shows really is that eventually developed countries end up with Weberian states. But let me ask, what about the transition? On the transition, what do states look like? Okay, so I'm going to show you a different picture. I realize this is not completely... Okay, so this is using exactly the same data from V-Dem but looking at the kind of historical aspect of it.

So the previous picture's just a cross-section in the world today, but this has a kind of, it has time variation. So we can look historically at what happened to bureaucracy over time. And here's the UK, and I'm going to do that at particular dates. So just take a bunch of benchmark years, 1913, 1915, 1917, 1919, 2010. So if you look at Britain, I'm going to plot on the horizontal axis here, you know, and here is, you know, think of the bureaucratic score kind of relative to England, okay? So England is at zero the whole time, you know, so everything is relative to England. So that's not telling you very much. But if you look at Denmark here, you can see that over the entire 20th century from 1913 onwards, Denmark, you know, gets richer, GDP per capita increases as it does in Britain, but it already had a pretty bureaucratized state, you know, 110 years ago.

Denmark kind of went into the 20th century with a bureaucratized state and it did very well. I want to point you to this line here. This line here is South Korea, okay? So, and what this shows is this starts in 1913, here's 1950, here's 1970, okay? So, what this shows is that eventually South Korea, today South Korea has a very, very bureaucratized state. You could say South Korea has a, has a Weberian state, but in the key moments in the transition in the 1960s and 1970s when Korea launches into this famous state-led industrialization drive, actually it doesn't have a Weberian state at all, okay?

And in fact, you know, in this period, you know, here this key period between 1950 and 1970 when this whole kind of policy is launched and implemented, not only is it very far from being Weberian, the state capacity doesn't improve at all over that period, okay? So, yeah, if we look at Korea today, it looks like a Weberian state, but the point I want to make here is that we need to look at the transition, okay? And that's particularly important if you're interested in policy in developing countries because you have to start the transition, you have to precipitate the transition. And what I'm going to try to convince you by the end is that the data actually, if you sort of look at it in detail like this, does not suggest that having a Weberian state is necessary, at least, for having that transition.

I'm also going to talk about Rwanda. So, this is, you know, I'm going to give you a, I'm giving you a big picture here, but I'm using this as a vehicle for talking about much more specific research that I'm doing. And I should have mentioned, it was on my slide, but everything I'm talking about today is part of joint research with Leander Heldring, who's a professor at Northwestern University. Okay. So, and I'm going to come and talk about Rwanda. Okay, so here's Rwanda. So, Rwanda, you know, we have fewer data points for Rwanda because the time series doesn't go back so far, but Rwanda is, you know, has been growing. It's been growing, of course, not as dramatically as South Korea, but you can see it's been growing over this latter period between 1990 and 2010, which is his last two data points, but actually, the state of anything over this period is becoming less Weberian rather than more Weberian, and I'll tell you exactly what that means.

Okay. So, so here's the lesson from the transition. South Korea might have a Weberian state today, but it certainly did not have one during this transition. And, you know, here's another way of seeing that. So, one very common way, you know, I think these measures of kind of meritocracy and bureaucratic capacity are becoming very popular as ways to measure state capacity, but in a previous generation of papers, people measured state capacity with tax revenues relative to GDP. Okay, so here's what the data looks like for South Korea, which is even kind of more fascinating in some sense, which is, you know, there is... So, this is, you know, first of all, the International Monetary Fund says that you need to raise 15% of GDP in taxes to have a serious state.

So, South Korea basically never managed that until the last five or six years. Okay? So, South Korea's always had a very small state. And you can see here, there's some improvement in fiscal capacity. It's not much. This is 11, this is 12, this is 13%. You know, we're not talking about enormous increases here. But then, during this period of rapid economic growth that you can see, more or less nothing happens. Okay? So, if we look at it through this lens, you know, also we don't see where does this Weberian state come from. Remember, the whole Tilly-esque story about the connection between the military revolution and the emergence of kind of modern states is that you need to raise taxes to do stuff, to provide public goods, to, you know, to fight.

And by the way, you know, the South Koreans do have a bit of a fight on their hands. You know, if you go into a Starbucks in Seoul, if you've ever been, they have bomb shelters underneath Starbucks in Seoul because they're, you know, they're 90 km from North Korea, you know, so the military mechanism is certainly relevant here also, which is interesting. So, okay, all right. So, none of this data is to say the South Korean state does not have capacity or did not have capacity. It did. It just wasn't manifested in this kind of Weberian way. You can't measure it in this way. Okay?

So, how did state capacity work in South Korea during this transition? All right, I'm going to come back to Rwanda, but let me start with South Korea. Okay? And here's a picture I like very much on the right-hand side. So, this is from President Park's museum. So, Park Chung-hee was the president between 1963 and 1979. He's the person who's kind of most famously associated with this industrialization drive. Some of you may be familiar with Nathan Lane's recent paper in the Quarterly Journal of Economics that looks at this kind of heavy industry policy in the 1970s and looks at the kind of causal effect of this policy on exports and productivity.

And this is what President Park did. He's sitting at the far end of the table there. This is what President Park did at the last Friday of every month. Okay? He convened the export promotion meeting. So he initiated this almost immediately he came to power and he carried on doing it until 1979 when he was assassinated. So he sat at the end. Who were all these people here? These are the CEOs of Samsung and Hyundai and Daewoo and stuff. So they were all summoned every Friday. And the fascinating thing about this is this was all recorded. So there's minutes of all of this. So the Korean Development Institute, you know, actually digitized all of this and stuck it on their webpage. So you can actually download the discussions of this.

I mean, there was no doubt there were things going on behind the scenes, but what was happening in public, you can download it over the whole span of the time and kind of, you know, what was going on here? I'm not going to show you any results on that. But it's basically enormously informal negotiation and discussion of like, what should we do? What's the problems? Where are the economic opportunities? You know, the Americans just invaded Vietnam. Shouldn't there be lots of business opportunities? Can't we, don't they need like ports and roads and we can do that, can't we? You know, like can't you do that, Samsung? You know, it's like this is incredible sort of discussion going on.

And when you look in detail at the kind of secondary literature or the qualitative literature, it's extremely consistent with this very informal approach to policy implementation. So probably Alice Amsden's book, Asia's Next Giant, is one of the most famous books by a Western academic about this. You know, she endlessly emphasizes the kind of informality of this. Perhaps the most famous example talking about Hyundai is the creation of the shipbuilding industry. So, President Park in 1967, he created a steel industry because he wanted to have an armaments industry and he put a military officer, former military officer, in charge of the shipbuilding at the steel industry. And then he said, "Okay, now we need a shipbuilding industry." So, he went to...

He went to Chairman Chung, who was head of Hyundai, the founder of Hyundai, and said, "You know, you're a guy who knows how to get things done. You're good at building a shipyard." So, Hyundai hadn't built a rowing boat, you know, at that time, let alone a big ship, you know. And he said, "Can't be done. No way. Don't know how to do it. There's no financing. There's no market. Can't be done." And this is in his autobiography. You read his autobiography. And Park says, "Look, it has to be done, okay?" And he turns to the prime minister and says, "Take away all the government contracts from Hyundai. Like, take away all their contracts." And Chairman Park says, "Ah, let me look. Let me—I'll look at this again." And he comes back. They give him some land in Busan. He builds a shipyard. He goes to Scotland, he gets to—and Greece—and gets two contracts. He gets a loan, you know. And so, the rest, as I say, is history.

And you know, 10 years later, the Koreans take over the world shipping industry. Okay? So, again, completely sort of informal. And you know, we started—Leander and I started trying to understand like what the heck is going on with this implementation. How did it—it's not very Weberian, if it's not meritocratic, if it's not bureaucratic. Like, how does it actually work? And there's a beautiful book by a Korean scholar about this. And he points out, "One of the notable characteristics of Korean society is the intricately webbed nexus among state, non-state, official, non-official sectors. These networks are mainly woven through blood relations, school ties, or acquaintances by locality." So, in the Korean literature, there's an enormous amount of discussion about the social networks connecting people in the state, the economic planning board, different ministries, and the chaebol, these big large industrial corporations. Okay, so there's lots of kind of facts about that.

So what we've been doing for the last six or seven years is actually constructing measures of kind of network connectivity between people working in the state, working in ministries, and people working in senior management in all of these chaebol—Hyundai, Samsung, etc.—to just try to understand that connectivity and try to understand how that is connected to policy implementation. Okay, so the Weberian kind of image, or this image of the transition from patrimonial to rational-legal authority, is, you know, you have to destroy these networks. Okay, you have to replace these networks with proper procedures, with meritocratic hiring, and you know, these are patrimonial, they create rent-seeking or whatever. Okay, but the hypothesis we started working on is this: is actually how policy was implemented in South Korea. These social networks were used to kind of get people to do things and to make people accountable to the state.

You know, Alice Amsden says, "So much of Korean industrialization has involved rewarding the same set of government friends with favors for expansion. Repeated support was exchanged de facto for good performance." So she, you know, it's interesting when you—if you read this book in detail, like if you went to the World Bank—the World Bank just published a big report, you may have noticed, on industrial policy. You know, that they have in mind, you know, like what went on in Korea? Well, that's obviously—we know what, you know, we need state capacity. Of course we need state capacity. What does state capacity look like? Well, it doesn't look like this. Okay, but it's interesting. So in some sense, you know, these arguments are already out there in this case study and qualitative literature. What we're trying to do is take them seriously and kind of quantify them. All right? So, and the words, you know, that we want to use is the state is embedded in society. Okay? This is going to be important when I come back to why I'm not a monetarist dressed, you know.

I think you know, the argument—you anticipated the argument brilliantly. So the state is embedded in society, okay? And you know, this is a word that, you know, the great—I'm not really sure what he was actually—Polanyi, Karl Polanyi had this idea of how, you know, you had to think of the economy as embedded in society. So that's the word we want to use. And these, you know, it's connected through these networks which can be used—sure, they can be used for rent-seeking and corruption. And indeed, you know, three former presidents of South Korea ended up in prison, found guilty of corruption. Most recently, President—well, not actually most recently, but the one before last was President Park's daughter even. But they can also be leveraged to get things done. Okay? I think that's the main kind of insight that you're going to see, I hope, today. Okay? So I'm not going to get too much into the econometrics and the data. I'm just going to give you a flavor of it. You know, but you can see here like where the hypothesis is going to come from. The hypothesis is we construct all these social networks. How do we do that? Well, at the moment we're using very detailed information on basically universities in Korea, looking at what majors people did and when they attended that. Okay? So if you're working in the economic planning board and you went to Seoul National University, and you majored in economics, and you started in 1970, and you're working in Samsung, you went to Seoul National University, you majored in economics, and you started in 1970, then we say you're connected. Okay. So that's how we use it. Basically using not just Seoul National University, using other universities, but and you know, including whether you went to Harvard or the University of Siena also, but I don't think we have many people in our data set that went to the University—we have some who went to Harvard. So, you could form a link if you were in the same class at Harvard also, but you know, but mostly it's Korean universities. So, just to give you a sense of how we create these links in the network.

And we're interested in, you know, both in some sense, you know, do these network connections predict who gets targeted with policy? You know, like why pick on Hyundai to implement this particular policy? Why not pick on Samsung? Or why not pick on Daewoo? So, we'd like to know who gets targeted, and we'd also like to know, you know, who succeeds in some sense. Okay? Meaning, you know, who makes a transition to exporting, for example? Or who starts doing something that they've never done before? So, most of these transitions—I don't quite know where this insight comes from, but of course you see most of these dramatic transitions always involve, you know, people doing things they've never done before. Okay? People starting to produce things they've never produced before. Like Hyundai starting to build ships. Okay? That's a pretty kind of amazing example, but of course it's true. And all of these dramatic transitions—true in China, it's true in Taiwan, it's true everywhere. You get these suddenly people start doing these things they've never done before. So, we can also—I'm not going to show you these results, but we can like look at some of these other things about you exporting, sort of start producing in an ISIC kind of sector that you never produced before, okay?

And you know, what I'm going to show you so far is just, you know, this is just the set of firms we have, you know, we have a panel between 1968 and 1982. We have about 82 chaebol, so there's, you know, there are these big industrial conglomerates and these industrial conglomerates have many subsidiaries. So we have data on all of that. And this is just to show on the horizontal axis, you know, this is a partial scatter plot, so controlling for, you know, firm fixed effects and time effects and things like that, you know, we can show that on the horizontal axis is the kind of residuals of connections between these firms and the government, and on the vertical axis is revenues, okay? So these connections are very closely associated with revenues just like as a kind of reduced form, but not with profits, okay? So if I show you this picture, you might say, "Well, this is completely consistent with a kind of rent-seeking. Of course, you're making a lot of revenues because you're managing to charge the government high prices for contracts or whatever it is," but we actually find nothing on profits, okay?

So I, you know, I wasn't going to show—I wasn't sure how much detail I should get into, but you kind of—I just wanted to get the idea across that actually, you know, the way that South Korea under President Park, and subsequently in the early 1980s, although that changed a bit, implemented policy was in a very non-Weberian way, but it was terrifically successful. So this is one of the most dramatic experiences of economic transition in kind of world history. It was not achieved in a Weberian way as the data shows. How was it achieved? By leveraging all these social connections between the state and society. Okay. So that's the—that. Of course, you know, I say here, you know, that state capacity was non-Weberian but patrimonial hardly seems the right word here. So that's why I'd prefer to talk about like the embedded state, you know, it's not patrimonial, doesn't seem to be the right way to think about it. Okay. So now let me say, okay, so maybe that's one case, maybe that's an isolated case. Let me talk about Rwanda. Okay, let me return to Rwanda. Okay.

African states, of course, as I mentioned, are taken to be canonical examples of patrimonial states. You know, typically, if we look at data on taxation, I could look at that data on kind of bureaucracy, they're very un-bureaucratic relative to the rest of the world. This is just data on, you know, again, you know, income taxes as a proportion of GDP is here's a whole bunch of African countries including Rwanda, Kenya, Ghana, Sierra Leone, etc. Okay, so Rwanda definitely doesn't stand out from that perspective and what I showed you earlier in terms of like these measures of bureaucracy actually over the last 35 years that measure has deteriorated in Rwanda rather than anything else. Nevertheless, over this period of stagnating tax revenue and even deteriorating ratings of bureaucracy, Rwanda has been an economic miracle of a sort. Of course, it was an immensely poor country. This is the period of the genocide in 1994. Since the period of the genocide, there's been extremely rapid economic growth. Of course, Rwanda is still a very poor country and what I want to point out is that at the heart of that economic growth has been very effective government policy. Okay, here's a measure of this from a recent World Bank study just looking at electrification. So, you can see here in 2005, you know, what was the proportion of people in Rwanda that had access to electricity, you know, close to zero. Where is it today? Close to 80%. Okay? So, the light blue line is the expansion of the electricity grid. The dark blue is like mostly solar power. Okay? But again, created subsidized by the government. So, this is an enormous expansion of public good provision over this period. Okay? So, that's, you know, this is part of the kind of, you know, the proximate explanation you could say for economic growth in Rwanda over this period. All right? So, this is, you know, this is about the effective implementation of public policy. So, how are they implementing public policy in Rwanda if they're not doing it bureaucratically or with large amount of fiscal resources? Okay? What's the nature of state capacity? So, to get into that, I have to tell you a little bit about the history of the Rwandan state and the history of the state more generally in Africa because Rwanda is a kind of classic example of what social anthropologists call the segmentary state. Okay? And what is a segmentary state? It's a state which is fundamentally based on kinship relations. So, traditionally Rwanda was made up of a number of clans, of extended kinship groups. And indeed, the traditional state was called the Nyiginya kingdom. So, the Nyiginya was the royal clan. Okay? The king came from the Nyiginya clan. And it was based on a royal clan. And it was formed probably around about 1620 according to oral history by a gentleman called Gihanga who migrated from the north. So, I'm not going to go into all the details here. There's many of the kind of classic features here of segmentary state such as, you know, rule from outside. We decided yesterday this is a podestà. You know, you had the podestà in medieval Siena. That was like your executive. He came from outside. So, that's very common in African traditional histories of states. What they call a stranger king. So, you bring a king from outside. Why do you bring a king from outside? Because he's outside of all the conflicts and disputes. And he doesn't own land. He's intrinsically weak in many ways. Maybe he provides useful services like rain making or mediating with the supernatural powers. But, you know, this is a very odd state, the Nyiginya kingdom. It's a very odd state. There was no fiscal system. There was no bureaucracy at all. The only centralized administration was basically like ritualists, you know, meaning priests. And they had this Ryangombe kind of cult that was spread all over the Great Lakes region. And the way the state was organized was patrimonially, often through use of cows. So, there was an enormous system of patronage which basically revolved around cows. So, I have cows, and Gihanga came with lots of cows in the oral history. I give you a cow. And if I give you a cow, you're now my person, you know, and you owe loyalty to me, and I can call on you to do things. But now you have a cow, and you can keep them. It's kind of my cow, but you can keep the milk, and you can keep the calves. And so that was the sort of—that was one of the ways in which, you know, patrimonialism, you could say, was constructed historically not just in Rwanda, but all over East and Southern Africa. Okay? So I'm just telling you a little bit about the state to emphasize it was based on this kind of clan structure. Because by the end of the talk, I'm going to kind of emphasize, you know, the sociology of society is very important for thinking about the potential of state capacity. Okay, there's a beautiful book about the construction of the Rwandan state in the second half of the 19th century by Catherine Newbury who was a historian and what she looks at is how the state spread into the southwest of Rwanda and how the state spread. I'll show you a map of these clans in a second. How the state spread was, you know, they would integrate, you know, integrate groups into different clans. They would intermarry of course and they constructed a kind of hierarchy which was based on kind of analogies to kinship systems, you know, so this is, you know, this is her example of, you know, the Mwami was the king, the Rwandan king in the Nyiginya kingdom, Mwami, and then the Mwami would, you know, would appoint different, you know, different types of chief, provincial chiefs, so-called hill chiefs, Mutewa chiefs who were responsible for particular types of tributes, you know, and these were these people in these boxes on the hill were kind of pre-existing people living there, okay? And they were integrated into the state by integrating them into the clans and the lineage system. So here's some data we found in Tervuren, in Brussels, the colonial archive of Belgium is now in Tervuren in King Leopold's palace in the suburbs of Belgium which if you haven't been to is absolutely kind of surreal experience. Anyway, we found a census in 1960—we haven't got the micro data, we only have the aggregate data—but we found a census in 1961 where people actually were asked their clan identity. So we can plot here, this is just the proportion of—at the sector level, the proportion of people who are members of this Nyiginya clan. And what's fabulous about this is that this area—so here in the middle of the map here more or less is Kigali or, you know, or a very close Nyanza which was the traditional capital, and this very dark red area here, the darker the red, the heavier the proportion of people who are in this Bigogwe clan, exactly shows this area where Catherine Newbury was studying, where, you know, the state spreads through the clan spreading and integrating people into the clan. Okay, so it's a very interesting sociological structure. Okay? Of course, from my perspective, this is another example of sort of embeddedness. You know, the state—the king is part of a clan, the state is constructed around these clans. To understand how the state works, you have to understand that clan structure. You have to understand the sociology of the state. You know, that's another example of embeddedness. It's different from the Korean, but it's another example of embeddedness. So, let me ask, you know, how does a state organized around social networks like that work? How does state capacity work? Okay? So, let me give you an example which we've been studying and I can show you some regressions on.

In 2006, about the time when President Kagame really sort of took control of the state, he was at a meeting with a bunch of mayors. And the mayors were kind of promising, "Oh, we're going to do this. We're going to do that. I'm going to be the best man." Kagame says, "You don't go around promising things you can't deliver on in Rwandan society." And he invoked a pre-colonial principle called imihigo, which were really applied, as I understand it, you know, it really applied to a kind of warriors. So, warriors would go and say, "I'm going to capture that hill" or "I'm going to capture 100 cows and bring them home." And if they didn't succeed, they were kind of humiliated or ostracized. So, he invoked this principle of imihigo, and he said, "You have to kind of make a promise of imihigo to deliver policy." And he started kind of institutionalizing this thing. And now the main public policy implementation—that rural electrification that I showed you—is all based around this principle of imihigo. So, how does imihigo work? It starts at the village level. That's the most decentralized kind of place. And the village comes up with three priorities, okay? A group number of villages are grouped into a cell. The cell has to aggregate those priorities. And up it goes to the sector, you know, and at the sector it meets government priorities coming down through the districts and the mayor. So, there's this huge process of kind of aggregation of preferences which we're studying at the moment.

Okay, like just to give you a sense of what this looks like. So, here's the Southern Province. Here's the district Muhanga. Here's the sector, you know, this is the level at which we observe these policies at the moment. The lower-level village priorities are all kind of—they're not digitized, so we're trying to collect those at the moment and digitize them. But here you can see bringing electricity to the sectors of etc. etc. Construction of the road linking Kigaga to Nyagatovu. Construction of the road, exactly. So, these are like the projects, okay? So, from this huge process of aggregation, they choose these projects and then the people involved have to sign these imihigo agreements. They have to promise we're going to deliver on this, okay? So, what we have from them—we've been working with the National Institute of Statistics in Rwanda since 2019, they've been doing these randomized audits to basically investigate whether those projects that I showed you have actually been implemented or not, okay? So, we can compare those scores, you know, the scores which come from the project evaluations with development outcomes. So, linking up a lot of household surveys which have information on access to public goods, access to electricity, you know, schools, consumption, all sorts of things. We can look at a whole vector of development outcomes. So, what you see is imihigo success is strongly correlated with development outcomes. Okay, so effective implementation of imihigo is strongly connected with development outcomes.

So, how do I know that the effectiveness of imihigo has anything to do with the social structure? Okay, so the image I'd like you to have in mind is, you know, you're appealing to this pre-colonial principle. It's almost like reciprocity as Alice Amsden was describing in Korea, you know, like we're going to get you—I'm promising I'm going to, you know, I'm going to do this and the government is saying, "Okay, you know, we're going to give you resources." And, you know, it's not quite as simple as that actually because in Rwanda there's many systems of kind of voluntary contributions. So, people contribute labor, even resources. Like in Kigali, the capital for example, if a neighborhood in Kigali doesn't have a surfaced road, if they want to get the road surfaced, the people in the street have to come up with 50% of the cost of the road. And if they come up with 50%, then the mayor will match that and then they can get a surfaced road in their neighborhood. So, there's a lot of voluntary contributions everywhere. I don't think we can really measure that. We'd have to do a survey to measure that. That kind of stuff is not really reported in anything that I've seen. But again, so why—so here you see there's a lot of variation in implementation success. Okay, implementation success is strongly correlated with development outcomes. There's a lot of variation in implementation success. So, why would—how would—how do I want you to think about that variation in implementation success? I want you to think that this strategy of reciprocity is going to be much more successful in some parts of Rwanda than other parts. And a natural hypothesis would be, well, the parts of Rwanda where traditionally the Nyiginya clan, the royal clan, was stronger or more embedded is actually going to create—the reciprocity is going to be much more effective there because those people were much more integrated into this segmentary state historically. So, a simple hypothesis is that there's going to be heterogeneity in some sense by the density of the royal clan, okay?

One thing I haven't said so far, you know, I was talking about South Korea, you know, doesn't have any resources and Rwanda doesn't have any resources and you're thinking, well, how does that work? Well, how do you—don't you have to pay these people? No. No, you don't actually. In fact, most people working for the state in Rwanda don't get paid anything, okay? So, this is—we're still kind of working on this. This is a reconstruction from a Dutch anthropologist called Bert Ingelaere who kind of tabulated. So, these are the—this is like, this is the village here. The—I'm not—my Kinyarwanda is not very good, but the—this lower level is a kind of sub-village administrative unit of like 10 houses. So, you know, talking of state capacity, if any of you have been to Rwanda, you know that Rwanda is unbelievably intensively administered by the state even if it doesn't have resources. I mean, this is one of the things that set us off on this. So, this is a unit of 10 houses. This one—this is the village. This is the cell. This is the sector. I don't want to go into this in detail, but most of these people don't get paid, basically. So, that's why it's cheap. That's why you don't need tax revenue, contra Tilly. Okay. All right. So, let me just kind of—I, you know, I gave you the idea, you know, in some sense like the hypothesis here would be there's heterogeneity by these traditional clan structures. And that's exactly what you see. There's not a very elegant regression table. I think this is the only regression I'm going to show you. I couldn't find a nice figure of it at the weekend. So, what you see is—so now that implementation kind of the success I had on the horizontal axis plotted against development outcomes. Now I'm trying to explain the variation in that kind of success. And I'm kind of—I'm trying to explain it by the fraction of members of the Nyiginya clan in the 1961 census. So, you can see that's extremely positively correlated with that mark. And I can look at other clans. So, this is pretty cool. The Sindhi clan, okay? The Sindhi clan is very interesting because that's where the queen mother came from. So, like many pre-colonial African societies, the Rwandans had a kind of dual—what they called a dual political system meaning there was male authorities and there were female authorities. And there was a queen mother who was not related to the king. She was not the queen. She was not the mother of the king or the sister or anything. She came from a different clan. She was the queen mother and she had also like enormous amounts of power over women. So, she had kind of parallel administrative institutions which just dealt with women, women's disputes and things like that. So, that, you know, but again that's something very kind of closely related to the traditional state. And the Ega and Gesera clans were the clans of the ritualists, of the priests, of the Ryangombe cult. So, you can actually see those clans which are—So, these are just looking at the clans that are very associated with the traditional state. They're all positively correlated. Not so powerful. Not, you know, the coefficient is about half of what the Nyiginya clan is, but they're all very positively correlated. And these other clans, not so much, okay? So, that's—I think we think that's kind of very interesting. All right. So, what's the lesson from Rwanda? I'm almost going to shut up. Okay, am I doing okay?

Okay. All right, I'm always worried I'm talking too much, you know. When I was a student, you know, I went to a seminar at Yale by the great economist Robert Solow, and he started with a joke which has always stuck in my mind, which he said like, "You know, I've been going to seminars for 40 years, and I've never been to one that was too short." So, here we are, 40 years later, I'm still anxious about that. Okay, so what do you see about this policy implementation in Rwanda? You know, Rwanda has not become good at implementing policy by becoming more Weberian, quite the opposite. If you remember that data I showed you to start with, it's actually less Weberian according to V-Dem than it was 35 years ago. But, it's used these traditional mechanisms to very effectively implement policy, okay? Now, you're thinking, "Okay, South Korea, pretty interesting example, hard to ignore South Korea, you know, though many development economists succeed, but for me it's hard to ignore South Korea. Rwanda, you know, but like, okay, fine, you know, but that can't possibly be true in European countries, can it?" Oh, yes, it can, okay? So, you know, we haven't written up the Korean or the Rwandan data, but Leandro and I have written up the English data. So, this is a paper with two students, David Kedrowsky and Matthias Vigand, which is on my webpage, if you're interested in the English case, the success of the embedded state in England. Okay, so, now you're thinking, maybe we should do Prussia, but let's—here's England, okay? We can observe England, you know, so what's the fact about England?

The English state, okay? So, you may be familiar with John Brewer's book, The Sinews of Power. So, the historian John Brewer wrote this book in the late '80s, The Sinews of Power, which is about the rise of the kind of modern state, particularly the fiscal state in Britain after the Glorious Revolution. And that book is all about tax systems and bureaucracy and things like that. Great. Brilliant. Turns out that's a very small fraction of the people who are working for the English state at that time. So, what we found from 1835 is a parliamentary report, a 4,000-page parliamentary report, which is an investigation of about 300 borough governments. So, the 300 biggest cities in Britain, the Parliament appointed a commission of lawyers who went out to collect information on the structure of the borough governments. Like, who worked for the governments, what were the—who worked for the boroughs, like the cities, what were their positions, what were their jobs, how good were they at doing it, and whether or not they were paid. Okay, so here's the facts, like looking at basically, you know, look at 1835 here, there were 95,000 people working for parishes. So, this is like the smallest administrative units, about 14,000 parishes at that time. We don't have data on parishes, but none of them were paid. None of them were paid. What about the boroughs? We have in our data, there's also county people working for the counties. Again, we don't have systematic data on that, but 90% of them are—this is just using fragmentary estimates and assuming it's representative. But, this is the number that I really really believe, which is the borough government. So, we have in 1835—sorry, I was looking at the wrong column there. So, focus on 1835. There's about 12,000 people in our data set working for boroughs, and 71% of them were unpaid. Okay? So, it's a bit like the Rwandan state. Okay? So, it's true that people working in London and the people John Brewer looked at were paid. Okay? But, that was about 20,000 people. Okay? So, there were far more people working for local government. And, of course, local government was responsible for enormous amount of public good provision, for the roads, for police, for jails, for maintaining the harbors. That was not the national state that was doing that. Okay? So, let me show you a fact which we find really interesting, and then I'll try to explain the fact, and then I'll come back to monetarism. Okay? Which is—So, you know, here's—I mean, we have a way from this report, this parliamentary report, to actually evaluate—So, we know who was paid, who was unpaid. And, we have the report evaluates how effective they were. So, it evaluates whether public goods were actually provided, you know, whether the jail was satisfactory, whether the harbor was dredged, what was the condition of the roads, terrible, adequate, very good, very good, etc. So, it has all these indices basically. So, we can create an index of the effectiveness with which different tasks were performed. If I just look at the average, and I split that between paid and unpaid, meaning we know who was responsible for that task. So, we know what the chamberlain was responsible for, or what the alderman—it says their kind of job description in this thing. It's kind of amazing. It's an amazing source. So, we can actually match individuals to tasks, to whether those tasks were effectively implemented and we know whether they were paid or unpaid. So, what we see is just looking at the means that people who did unpaid tasks were more effective. Okay? Now, before you get too excited about eliminating salaries of professors, I'm going to explain how that works on the next slide. Okay? You know, this is just the regression version where I can control for all sorts of fixed effects and stuff like that. You know, borough fixed effects and all sorts of things. There's a very strong relationship between—this is the residuals now, plotting the residuals of being unpaid against performance. So, people who were unpaid performed better. Okay? So, what the heck is going on? It's all about who is doing unpaid jobs and who is doing paid jobs. Okay? So, for example, let me break this down and look at different jobs. Like, here I'll, you know, I can classify these to do with governing, sort of the people who are like really in charge of like the mayor, the mayor—not every urban area had a mayor, but most of them did. The police, you know, people running the jails, kind of clerical work. And you can think of this as a probability that they were unpaid. Okay? So, the people at the top who were in charge of governing, they were very likely to be unpaid. But if you're a policeman or running the jail, you're much more likely to be paid. Okay? Most policemen or people running the jail were paid. Okay? And most very senior people were unpaid. Okay? Why is that? Because very different people were doing those jobs. You couldn't get the landed aristocrat or a gentleman or an elite to be a policeman or run the jail. But you could get an elite to be an alderman or to be the mayor, okay? So if you look at these different things, you know, maybe the simplest thing to think about is gentry. You know, so it used to be the case in England that there was this very clear status distinction between like a gentleman and somebody who was not a gentleman. It's a very weird concept and it's now completely disappeared in British society, interestingly enough. You know, if you called someone a gent, it's like—it's a satirical thing. You'd be kind of making fun of them for being pretentious or something. Like, "Oh, he's a gent." You know, like so—but in 1835, it at least, you know, it seems to be an important thing, okay? So you see gentlemen are much more likely to have these high governance jobs, very unlikely to be in the police, okay? You can't—Why is that? Because you could acquire status from these jobs. You could get—There were career concerns, meaning we can look in the paper, for example, at who ended up becoming members of Parliament, all right? So becoming the Lord, you know, becoming the mayor of Birmingham or something was a stepping stone to becoming a member of Parliament. A member of Parliament did get paid and that was powerful and influential, okay? So there was those types of career concerns, there's status associated with these positions, which is not true about running the jail or being a policeman, okay? There's also a very interesting—right, you know, prestige—that even prestige is, you know, it's being 1835 Britain, they even discuss what—in the report they discuss which positions were more prestigious. So the idea here is, you know, you can get people to run the government and actually implement policy without paying them by again sort of exploiting the structure of society, exploiting what kind of activities do people get prestige from, you know, what do they get status from, and maybe, you know, what does the hierarchy look like in terms of how do I move up that hierarchy, okay, what I was calling career concerns. All right, so two more slides.

So I think this is another example of the embed—what I'm going to call the embedded state. You know, of course you can see these examples are extremely different, meaning, you know, sometimes it's the nature of the embedding in Rwanda and South Korea and Britain is very different, okay? But in all cases, you kind of exploit the structure of society to implement policy without having huge amounts of resources or without even being kind of Weberian, all right? So we think, you know, this is very different from the Weberian tradition, of course, which is, you know, like it's all over the place in social science. I was just randomly putting up here Henry Sumner Maine's idea of from status to contract, you know, this idea of kind of modernizing and, you know, moving along this kind of path from, you know, backwardness to modernity or something. Turns out you don't need to do that, you know, to have a development miracle. And of course we think that has very rich policy implications, you know, if you're thinking about that in Africa or developing countries. Okay, so let me end now going back, what's the—what the heck has this got to do with monetarism? Well, when I think about Frank Hahn's work, you know, when I was trying to understand Frank's work, I think, you know, Frank was one of the kind of most profound thinkers on what I call the disembedded economic theory. You know, if you think about the kind of work he did on general equilibrium theory, I think of his book, you know, General Competitive Analysis with Kenneth Arrow, which is kind of one of the sort of definitive summaries of this kind of model of what I'm calling the disembedded economy. You know, the idea that somehow there's something called the economy that exists separate from politics and separate from society. You know, and he worked on the theoretical properties that, you know, the existence of uniqueness of equilibrium, stability. You know, he wrote a lot of papers on these kind of tâtonnement, different types of, you know, dynamics, the welfare properties of equilibrium. I think like that wave—for me that wave of economic theory was let's try to really understand how this kind of disembodied—not disem—not disembodied, disembedded—economy functions. You know, and I think for me people of my generation, you know, I don't know what you're going to do, but like people of my generation, we wanted to embed—you know, like most of my work if I think about it has been about embedding the economy, getting it back into institutions, getting it back into politics, getting it back into society, and kind of thinking about those interactions. You know, you could say a lot of my earlier work on institutions, you know, we did, you know, in some sense was trying to think about like, you know, under what circumstances does an economy emerge that looks something like what Frank studied. But I guess what I've been talking about today is more like, you know, actually, that's not the neces—that's not a necessary path for prosperity. There's many different ways you can leverage the structure of your society or your cultural specificities to be successful economically. And I think that's a very hopeful—and you know, I think you know, so that leads I think to, you know, like I don't want to, you know, I don't want to start getting into like, you know, the homogeneity of excess demand functions, you know, but I think it leads to a very different sort of critique of monetarism that Frank had in mind in the sense that monetarism, you know, is a classical example of, you know, a kind of public policy and, you know, set of theoretical propositions based around a disembedded economy. Once you put the economy in the context of society and politics, you know, as you were saying earlier on, then monetarism, you know, may emerge in some circumstances, but seems like an even less likely set of policy implications than it was in the theoretical approach that Frank took. So, it's a very different critique of monetarism, you know, than the one that Frank developed, but one that I hope he would have approved of. Thank you.

Yeah. Yeah, so that's a great point. So, you know, in some sense, I had this on a bullet point, but I didn't mention it, which is, you know, for me, what's distinct about Rwanda is, you know, is not that you have all these social networks or kinship systems or whatever. You have that everywhere in Africa and indeed in many other parts of the world. What's distinct is what they're doing with them, how they're using them. Now, why are they doing that? You know, if you look at Rwanda before the genocide, for example, all those networks were there, but there was no developmental state whatsoever. You know, there was a very predatory state when, you know, Habyarimana or Kayibanda was there. So, I do think—why, where does this impulse to kind of develop the economy come from? I think that is connected to the genocide. It's connected to the kind of trauma that that had on society, and the kind of passion, you know, I know many kind of, you know, ministers and senior policy makers in Rwanda. There's just everyone is on message, and everyone is like, we have to change this society. We can't let that happen again. So, I think I didn't explain here, in some sense, where this impulse came from. I'm just focusing on like how they did it. You know, I didn't explain why President Park decided he wanted to industrialize South Korea, either. I'm just focusing on how he did it, you know, but I think...

Like, of course, that's a fabulously interesting question. And maybe that's what's — that impulse is what's missing in other parts of Africa. I would say one more thing, however, which, you know, yeah, traditional institutions, you know, and that again, that's without, you know, if there was an impulse in Botswana, um, you know, uh, it came from the threat of white colonialism in some sense. So, you know, Botswana was surrounded — when Botswana gained independence, it was surrounded by white South Africa, white Zimbabwe, by Namibia, which was controlled by South Africa. So, it was completely surrounded by white settler colonies. And so, you know, uh, there was a similar idea that we have to — this country has to succeed. You know, we have to make this country succeed. Otherwise, we're going to be eaten up by these white settler colonies.

So, but you know, I — so one thing I would say, which I find absolutely fascinating, is that, you know, there's a lot of variation here, which is very interesting from a research perspective. So, Leander Heldering and I, we've been — we went to — we've been to Somaliland, you know, to try to understand like what is going on in Somaliland. So, Somaliland, you know, that's another experience like Rwanda in the past 35 years of very rapid kind of state building, public good provision, economic growth, very poor place, but like really kind of interesting. And clan structures, famous kind of the Somali clan structure, the Isaaq clan family. But then you see immediately like what's possible in Rwanda is very different from Somaliland because of the nature of — not just the sociality, but also the economic system. So, Rwanda was an agrarian society. So, so many of these institutions, you know, were related to farming and agriculture. Somalis are herders, you know, it's a herding society. It's completely different. So, that turns out that, you know, when we discuss this with people in Somaliland, it was really interesting that we tried to make these analogies to Rwanda and they were like, 'No, you can never get people to do that.' There's no — you know, like — so what — that's an interesting interaction with the economy there also. So, there's a lot of variation there, which is fascinating.

So that's also a great, a great point. So there actually — there actually is a traditional clan system in South Korea, but that has certainly kind of atrophied today, meaning people — if you ask like young Koreans, they're kind of aware of this, even some of them know, but it seems like irrelevant. I don't know if it was irrelevant in the 1960s, but we don't really have a way of measuring that. And as you say, it seems — but you know, this is not quite like the clan system or the British hierarchy. So I'm not sure like we have a great way at the moment of like saying how to think about that. Yeah, because that example suggests you could build — you can actually build these networks. You know, you can build these networks through kind of influencing social interactions. And you're right that that's qualitatively distinct and maybe that's even more interesting from a policy perspective because it gives you more freedom to actually construct things. Yeah, but you're — yeah, I don't know. We've wrestled with that issue. Yeah.

Yeah, I mean, you know, one way of thinking about what we're saying is, you know, there's a beautiful paper in economic history by Alexander Gerschenkron, you know. So, Gerschenkron wrote this essay in the '60s about economic backwardness in historical perspective where he sort of showed that there's many ways to become industrialized actually. You know, the British became industrialized through one way, but the Germans didn't have the same sort of institutions. They didn't have a stock market. And also, they went fast. You know, if you go first, you go kind of slowly. You know, but if you catch up, then you have to accumulate capital very rapidly, and you need financial institutions for that. And so, the Germans used these kind of coalitions of banks and business. And then his third example is Russia. That Russia, they also went late, but they didn't have all these financial institutions. So, the state had to come in. I've always found that like an incredibly inspiring article. So, there's a relationship in some sense between what we're saying and that, which is like maybe in the end, you know, like several people, you know, told me this in Korea. Like, 'Oh, but you can't run Korea today the way General Park ran it in the 1960s or 1970s. It's this huge country. It's way too complicated and stuff. You just like — you can't — this kind of personalized, you know, like that bit I showed you, that photograph of — just you can't manage the economy like that anymore. It was much simpler 50 years ago.' So, so that's to say, yeah, this is about the transition. This is about getting the transition going and you kind of work with what you have to get the transition going. That's sort of exactly what Gerschenkron said actually.

I, you know, if you look at these boroughs, if you look at the franchise for example, so we have all the details. We know like who could vote and things like that. They were very oligarchic. In fact, one of the reasons that we have this report is that Parliament, you know, this is — the timing of this is not a coincidence. This is right after the 1832 Reform Act. So, there was enormous — this is when they eliminated all these rotten boroughs. They took the franchise away from all these rotten boroughs. So, there was a renewed focus on kind of problems with institutions, you know, with oligarchization and, you know, with old corruption. Uh, and Parliament launched this report because they didn't really think they knew what was going on to try to collect systematic information. So, there's wonderful information and discussion of corruption for example. So, there's not just like, you know, did the harbor get dredged or whatever, but there's also examples of like nepotism, you know, they actually have nepotism. Like did people employ their family and things like that. So, so I think, you know, I'm sure, you know, the people who were running these things, they were elites. They were interested in protecting their property rights, but I think also, you know, there's a lot of qualitative evidence or historical evidence to suggest there was a lot of public good provision going on as well, you know. So, and, you know, I think if you read the Webbs, so Beatrice and Sidney Webb, you know, who wrote this enormous nine-volume I think history of British local government, they kind of discuss a lot like the culture of service. I don't know how seriously you want to take that, you know, but their idea was this obligation of service and whatever, you know, at this local level. And you know, I'm sure that they did want to protect their property rights, but at least, you know, if you take the evidence seriously in this report, and I think you can triangulate it in lots of ways. If you look at the paper, we try to look at kind of other historical sources and like what do they say? I think it's credible that there was a lot of public good provision going on. Yeah.

I mean, it's interesting, of course, to have a completely different source of information other than the report. We also look at that. Yeah.

Yeah, so I mean, you know, the boring answer to that is, you know, we can control for all of that. So, actually, in that very detailed information I just tried to give you a sense of, of course some of those projects are being funded by the World Bank or, you know, by — well, I was going to say USAID, but they're not being funded by USAID anymore, but by, you know, China Aid or something like that. So, we can actually directly control for that. I mean, one thing I can't actually tell you because I don't know the answer is, you know, we could ask, is it more likely that donor-funded projects get implemented? You know, I mean, one of the obvious things we're trying to look at here is, you know, the sort of the thousand-and-one-dollar question, you know, about Rwanda and what's going on in Rwanda, is, you know, if you talk to people in Rwanda, you know, they have this narrative of like accountability, you know. So, President Kagame would sort of say, like, 'Okay, you know, this is not a Western country. We don't have a Western political system. You know, there's other ways in which accountability works in this country, you know.' So, we actually — this data that we have, we can actually look at that directly, meaning we have these data, we have the preferences of locals, you know, like what people want. Now, you know, I don't know to what extent that's kind of orchestrated, you know, like you could tell all sorts of conspiracy theories, um, but like at least in my interaction with people in the Ministry of Local Government and the National Institute of Statistics, like all I can tell you is people seem genuinely interested in like whether — why things don't work and like how do we get that to work and like what people want and things like that. So, that's just my personal experience. I don't have any insight onto like President Kagame or whatever, but there is this narrative of, you know, so what we can actually look at that. You know, for example, we can look at, you know, if people get the kind of project that they actually asked for, is that much more likely to be successfully implemented? You know, that's a kind of interesting question as an addition to the aid. I, you know, there is aid, but you know, one thing that I find actually wonderful about Rwanda, this is true about Botswana also, is like the government basically just like says, 'These are our priorities. You can fund them or not. If you don't want to fund them, there's the airport.' You know, kind of thing. And so, they convene — like you can't run about in Rwanda as an NGO. Like in Congo you can just run about like doing whatever you want and nobody notices. You can't do that in Rwanda. The government has to agree. So, I actually think that's a great thing. I think every country in Africa would be better off if they just took control of that and said, 'This is what we want. Like help us or like go back to, you know, Britain or wherever.' I mean, I think that's kind of healthy. And one of the things I like working in Rwanda is this kind of confidence. I wish everywhere in Africa had that kind of confidence.

Now, you know, again subject to the authori- — yeah, but I don't have anything to say, you know, about the authoritarianism. I would say, you know, again thinking about Rwanda or thinking about Somaliland or Botswana, it's very interesting, you know, in the sense that like even in Botswana, so Botswana, you know, creates this modern democracy in 1966 kind of based on these traditional institutions. The kgotla — if you were in my lecture yesterday, I showed you a photograph of a kgotla I was doing field work in March in Maun. And you know, the funny thing is that Botswana was basically a de facto state, a bit de facto one-party state. You know, the same party was in power until last year, between 1966 and last year. It was a one-party state. So, why was there accountability in Botswana? Because the accountability was not at the level of the national state or the level of presidential elections. You know, in the United States we think like everything hangs on presidential elections. You know, if we can't get rid of Trump and his cronies, you know, in the next election, you know, we — you know, like everything is hanging on that thing. You know, in Botswana nothing is hanging on that. Why is that? Because all the accountability is at the local level. That's where the action is in these kgotlas. It's not — you know, ministers have to travel around the country and appear in these kgotlas, you know, explaining what they're doing and that's where accountability is. And that whole system is so fascinating. And nobody votes on anything. There's no voting. Like nobody votes. They make decisions, but nobody votes. You know, so — and I — but then I look at Rwanda and I say, 'Well, Rwanda was not like that.' You know, that system I described is not like a kind of Western state, you know, the Buganda kingdom. But it was also not like Botswana. Somaliland, same thing. So I think personally I would say we should take much more seriously the political culture in these African countries if we actually want to understand like what works and what doesn't.

Yeah, that's a good question. You know, here's two ways of thinking about that, you know. One way is, you know, I used to say, you know, when I show these regressions from colonial origins or reversal of fortune or something, that — you know, if you look at the first stage, you know, where we're kind of modeling the variation in institutions, then these variables which capture kind of colonialism, they — you know, like the R-squared, you know, if I put settler mortality on the right-hand side, the R-squared is like .26, you know, in the most — so it explains 26% of the variation. That means it doesn't explain most of the variation. You know, you can put other variables on the right-hand side connected with colonialism, density of indigenous population and stuff, and the R-squared goes up, you know, to a half or something. So, one way of thinking about it is, okay, you know, we told this story, and what we were really interested in is kind of estimating the causal effect of institutions on kind of income per capita. But when we modeled, you know, to do that, we just need a source of variation in institutions which has a particular statistical property. But what the model shows you is actually there's many other — you know, there's lots of space there for other ideas about how institutions vary, you know. And so, one way of thinking about this is to sort of say, what I've been exploring myself, you know, in the last sort of decades is kind of understanding that better. Like, where do these other institutions come from, and what do they look like, you know? Um, and you know, maybe there's different paths to inclusive institutions. You know, we told this story about settler colonialism and how the dynamics of settler colonialism led to particular sets of inclusive institutions in North America or Australasia, whatever it is, and you know, and then other types of initial conditions in South Africa or in Latin America led to very kind of extractive societies. But, I guess one way of thinking about what I'm trying to think about now is, you know, maybe there's other ways to inclusion, you know. There's — I mean, there's got to be other ways to inclusion other than, you know, creating some white settler colony, you know, without indigenous people like Australia or the United States. So, you know, that's a very specific channel towards, you know, inclusive institutions. You could say kind of, you know, we're just exploiting that particular historical epoch to like kind of get at the second stage, but you know, I always thought, you know, the first stage was more interesting to me personally. Um, so that's, you know, and I also think like as I've, you know, I've been working in Africa for 30 years and um, you know, I guess I've just come to sort of understand African society much better, you know, albeit highly imperfectly. Um, and sort of understand, you know, people's normative ideas about institutions. In some sense, that's what I was talking about yesterday, you know, I've come to think that like a lot of things about Africa have to be understood, you know, on the basis of people's like normative commitments, you know, that's just, you know, and that may have economic costs, you know, but you know, that's something that maybe someone like me or someone like us, like scholars, we can clarify those costs or we can estimate the size of those costs, but that's not for me to make decisions about those. That's for the Africans. That's for those people themselves to make decisions about that.

If you look at Botswana, for example, this is also like fascinating in the 1960s. You know, if you read, for example, the speeches of Seretse Khama, who was the first prime minister and president, he was extremely aware that there were lots of traditional institutions in Botswana that were really going to be difficult to combine with a kind of rapid economic growth. And they renegotiated many of those things away. For example, they took the allocation of land away from chiefs and they created these kind of bureaucratic institutions to allocate land. So, there was a much more kind of bureaucratic process for getting access to land. And so, they did sort of change — yeah, but that was very much a kind of — they debated it. They deliberated. They talked about it. The Africans themselves kind of took the initiative and sort of said, 'Okay, like we need to change this stuff. Here's why we want to do it. Here's why it's necessary.' Like that. You know, so that — you know, but that's the same as Britain. Like think about Britain, you know, like I — you know, Britain — you know, this is something that Leander and I also studied, these enclosures, you know, these — Britain had huge amounts of traditional institutions in the early modern period in the rural sector. We had common fields. We had this crazy system of open fields with like people having strips of land scattered all over the place. We had commons. All of that was kind of privatized. You know, like what you see in rural England today is something that was created in, you know, the 18th and early 19th century mostly. How did that process work? That was a process whereby English people kind of deliberated. They debated. There were lots of economists, you know, like Adam Smith and people who actually wrote about this and wrote about the inefficiencies. So, there was a big debate. And then there was a political process whereby people could agree to enclose. You know, somewhat oligarchic for sure, you know, landowners got much more say in this, you know, and people were expropriated. People lost their rights, you know, to access to the commons because the commons was privatized and divided up basically. And you know, there was contestation. You know, there's a great book by the historian E.P. Thompson about that, you know. There was contestation. And but it wasn't Pareto improving. But, you know, it's hard to find Pareto improvement in reality. I don't really have a normative view on that. You could probably say on average, you know, on average things are much better, but you know, some people suffer. So I'm not answering your question well probably, but I've just become much more respectful, I think, you know, of people's institutions and people's kind of own systems and values. And I think like as a development economist we have to understand that, to take it seriously if you want to do anything useful.