Chris Blattman on War and Centralized Power | Conversations with Tyler

Hello everyone, and welcome back to Conversations with Tyler. Today I am sitting down, yes, sitting down with Chris Blattman once again. There's an earlier conversation with Chris. If you don't remember, Chris is both an economist and a political scientist. He does not let either discipline claim him. He is a professor at the University of Chicago, and most importantly for our purposes, he has a new and wonderful book out, which I have blurbed. It is called Why We Fight: The Roots of War and the Path to Peace. Chris, welcome.

Great to be back.

I have many questions about war. Let's start with demographics. So Richard Hanania made the argument recently in his Substack that countries where there was only one child per family were going to be much less likely or willing or interested to fight wars. Do you agree?

Because they'd put their children at risk and they—

But also it's a sign they don't care that much about the future, that they're complacent. They're not going to disrupt their lives just to have three kids. So if they're not going to disrupt their comfortable lives for 3 kids, they won't disrupt it for a war either.

I don't think— yeah, that doesn't resonate with me. I think that— yeah, I don't think that the average person's weight on the future and the quality of their life is like a first-order determinant. If it is, it's the kind of political system that I'm not too worried about going into war.

And why isn't how they feel about their lives, their future, their country?

So I mean, my— the way I think about this is that there's usually a narrow set of people who decide whether or not to go to war, and those people are either accountable or not accountable to lots of people. And the places that we— that in the places where they're accountable to everybody else, people have an enormous amount of stake not to go to war. And so like small gradations as to whether or not they have one child or two children or a little bit more weight on the future. I mean, the war is just so hugely painful that I think if you give those people a lot of influence, that they're going to press their leaders generally not to go to war, just in general. And so I just don't see this perturbation or this trend as like fundamentally shaping that calculus, just because there's already this overwhelming incentive not to have their leaders invade somebody.

But if you look at the marginal cases, since there are some wars, there's a bunch of cases, even if unusual, where someone is right at the margin, right? So at the margin, what are the factors that are most likely to account for the explanatory variation in whether or not a country goes to war?

I mean, for me, the one that people talk the least about that strikes me as the most important is how concentrated is power in a country. And so how— like what's holding back someone from considering all of the implications of their actions on other people should they decide to take their society to war? And so, and it's maybe the most important margin in history and it's maybe the one that one of my tribes, which are political economists, think and talk the least about. And it's the one that in journalism people leap to psychological explanations and they try to understand the psychology of leaders, but they don't try to understand the way in which they're constrained. And so it's sort of this combination of the most important and the most ignored.

So federal societies are less likely to go to war?

I think federal societies, I think— so I think of— I mean, the word I like that— the word I both loathe and love is polycentric, which is this word from Vincent and Elinor Ostrom, and it's this idea where there are many centers of powers. And so I think we're in the US, we're programmed to think in terms of federalism, which I think is a great example, but I think there's lots of other ways. We can be polycentric in the sense of having lots lots of supranational power, so we can delegate a lot of authority and bind our hands through international treaties and agreements and alliances. And we can— it's not just about— we focus a little bit too much, I think, on downward and federalist sort of polycentrism in the United States.

I know it's a small country and it's not near many other countries, but New Zealand is very non-polycentric, right? If we put them somewhere else, made them bigger, we think they would be more likely to go to war? Than say Canada?

Well, it's funny because I think of Canada as a not very polycentric place in at least a formal institutional sense. And they're also in a reasonably— they're in a place where they don't have a lot of temptation to go to war. I think they are— but I think those both societies— I've never been to New Zealand, so I'm more projecting from other places I know including Canada onto it. I think they are polycentric in the sense that there's just a lot of informal— the country is in some ways governed and restrained by informal norms and an understanding of what is acceptable behavior and what people would or would not accept. And so in that sense, I think it's a very decentralized place, but even if power in the formal decision-making power isn't very decentralized.

It seems to me the United States launches a lot of military actions. You may or may not want to call them wars. Something blows up, right? Russia pre-Ukraine launched a very large number of cyber attacks. Your theory of what makes war more likely, how is it different for an actual physical violent war versus major cyber attacks?

That's a good question. I think there's, so in one sense, I think one is that cyber attacks can be less attributable and less clear. And so one of my colleagues, Ethan Bueno de Mesquita, has thought about this. This sort of makes the world, I think, a more dangerous place in the sense that there's a lot of uncertainty about who's attacking you. And you're always worried that somebody else is attacking you and they're cloaking it in this person. You'll be duped into a war. And so I think this just leads to a lot more complex strategic calculations that might lead to worse responses. And so I think we should be wary and probably nations are worried about using it for that reason because it could spin out of control more easily. I think that's probably top of my mind.

If I had the hypothesis that the least responsible countries are most likely to do cyber attacks, and the sort of more responsible countries, even if you disagree with what they're doing, are more likely to mount military actions. So the earlier British Empire, in a way, may have done the wrong thing, but it proudly announced what it was doing, right? Does that make sense as a hypothesis?

It does, but then it also gives actors in those larger places incentives to appear to be one of these more rogue actors. And so I think the thing we have to be worried about in this context is when the ability to conduct cyberattacks is very disaggregated. So the idea that, I don't know, some CIA part-timer or consultant could potentially launch a cyberattack on his own if he was so aggrieved, or vice versa on, you know, in some other strategic enemy to the United States, I think is— we can't do that with warfare very easily.

Let's say we wanted a theory of which nations were most likely to mount a geoengineering war or attack, which probably we haven't seen yet, but imagine dumping iron fillings in the ocean, change the climate, change the climate of your rivals. How is the theory of a geoengineering attack different?

I guess it would depend a lot on the extent to which —this, there's sort of a—you're able to—so if I fire a missile at an enemy, I'm fairly confident that I'm going to only harm them and I'm not going to harm me. And so anytime anything you do is going to potentially blow back on you, which for example, like a nuclear bomb is one way to think about it, I think then is going to change your calculus because it's going to be less attractive relative to the alternatives. And so there's always going to be a selection of instruments that you could use. And I guess these ones, I've never thought of these kinds of attacks before. It seems to me like their risk of blowback is actually much, much higher, and there's just a lot of uncertainty. So I'm not sure why somebody would want to use them vis-à-vis one of these other more conventional things that are— Well, they're very cheap, right?

And it might be hard to trace. But in general, shouldn't a theory of war be super technologically contingent? So if I look, who's likely to launch a drone war? —of course, I think the US, but not only—who's likely to launch a soldiers' war, it's maybe negatively correlated. So how can we have a general account of who's likely to launch a war? Isn't it always like per technology?

I think it's really dependent on— there's some sort of cost— I mean, there's some sort of cost-benefit calculus that is at the root of all of these. And And so yeah, I'm struggling to think about why this would— that to me is still going to be like the first-order decision. And so I'm struggling to think about— so the technology is going to change the costs and it's going to change the potential negative externalities, which is just another kind of cost and how much it blows back to you. So it still strikes me as like a good— like a lot of economic theory, it's always like a good first-order principle for you to use to sort of structure your thinking. And then I don't— the second-order elements just don't seem to me to be particularly large.

But say in your list of the 5 main factors that explain wars, as I understand it, technology isn't one of them. Why shouldn't technology be a clear number 1? So the US does a lot in Yemen that it probably wouldn't do with soldiers, but it's clearly willing to do with drones. We do drone attacks, I don't know in how many countries, but it's a large number, right? West Africa. Sahel. So why isn't technology the number one factor? So I think—

so here's the thing. So let me use a slightly different analogy. So like when I have thought about— I'm not a particular— let's take Afghanistan and the US involvement in Afghanistan, which was a— I'm not— I have no particular expertise in the region. But when I try to understand to myself why was that such a long war, I think it's because it was not particularly expensive to the United States. The United States was saying, This is going to cost us tens of lives per year and maybe 1% of GDP. In return for that, we are going to buy reputation for not tolerating rogue states and terror attacks by rogue states. In return for that, we're going to have some probability diminishing over time of victory over this particular rogue group. And given that this rogue group will not negotiate with us and there doesn't seem to be any ground where we can actually compromise because of our mutual intransigence. This is a cheap war and we'll wage it. And that's not what they say to the public, but I think that's de facto what happens. And so I think drones and technology is like, does it make war cheap? And if war is cheap, then I think you're going to be more likely to use that as an instrument. And that's the first order that I think technology sort of gives us.

So how much does state capacity predict war. I mean, should that be in your list of 5 main factors? State capacity.

A lot of these things have to do— I think it's really important to distinguish between the things that change your bargaining power vis-à-vis a rival and things that actually change the probability of war. So if I have state capacity or if I have particularly cheap weapons or if I have particularly effective weapons, then it sort of changes what I can expect to get from my rival.— it doesn't necessarily have any effect on this basic intuition that we should be actually trying to find some solution and that is going to avoid war, which most of the time we do. As you know, this is one of the main themes in the book. And so yeah, so I would just say it's merely— we always have to— every time we say, does this cause war, we have to say, am I talking about something that changes relative bargaining power or I am talking about something that changes the incentives to avoid the cost of fighting?

Now, in our last conversation, we were both fairly bullish about Ethiopia on the grounds of its high state capacity. That now seems wrong. What do you think we got wrong or you got wrong or I got wrong?

So long run, I still have high expectations for Ethiopia. And I think short run, things have obviously gotten very bad. The thing that I think I— and you know, I have a couple of friends who work very closely there and so partly what I know, I'm channeling them. I think what was fundamentally problematic about that, like in a lot of African countries, is that in spite of the state capacity, you have power that's very concentrated in a handful of elite groups and these elite groups happen to be ethnic groups. So you have the Tigray and the Amhara and the Oromo, and they had— they'd cut a deal with one another. They'd cut a deal that for a long time heavily, heavily, heavily favored the Tigrayans, and that started to fray, and the other two groups began to sort of demand more, and they fumbled that. And I think what surprises me, and I think surprises also the leaders of those three groups, is that they fumbled it. Apparently, one thing I've heard is that the Prime Minister was surprised at the time that this suddenly turned into violence and continued to be surprised for some time. And so I think that So maybe people just underestimated the chance that these sort of concentrated elite bargains can go wrong. But I guess that's the thing I remembered that I always believed but I never really applied to that case is like that to me goes back to this first thing I said, like the fundamental thing we don't pay enough attention to is concentrated power and how easy it is then for these deals to fall apart.

Let me tell you my worry and my account of what I got wrong. I know this theory has only two data points, but I'm struck that Rwanda also has a reputation for high state capacity, which whatever their failings is in some sense true, right? So maybe high state capacity develops from a past where you had ethnic strife or war, and in part you develop the high state capacity to keep some groups down or enforce some bargains. And maybe on average it's a good thing, but it's also increasing your variance of outcomes. So there's this new piece by O'Reilly and I think Murphy in Economica, which shows the best predictor of state capacity today is historical prevalence of war. So maybe when you see high state capacity in Africa, you should also start getting nervous a bit intrinsically, and indeed anywhere, because it's a sign it was needed. It doesn't come from nowhere. Just like you needed a fragmented Europe to develop a lot of advances in capitalism, but that too had a high cost. 17th century wars and so on. So I guess I would say—

I mean, the two places we've picked here, which are Ethiopia and Rwanda, are places that have— I mean, they have high state capacity today because they had— and coherency because they had high state capacity and coherency 400 years ago as well, relative to a lot of other places on the continent. And I think this is like a little bit of a— not a false correlation. I think it's true, but I think like the thing that strong states have— and I think political scientists would talk about state systems being— is that you have lots of competing states close to one another, reasonably strong competing states, and most of the thing that is generating the state capacity is that competition. And 9 years out of 10 or 99 years out of 100 is peaceful, and then sometimes it's not.

So that's like the Hayekians' point.

Right. So the fundamental driver of that capacity are these— competing state systems and war is an occasional byproduct of that. But we only— when you run a regression like maybe they did, I don't know this paper, of current state capacity and war, there's just this omitted variable which is the competitiveness of state systems. And so I think we just ascribe a lot of explanatory power to the fighting part that we should attribute to competition. And historians would agree. They would call this like defensive modernization. There's a lot of words and concepts like this that say, well, we actually— revolutionized our society or technology in response to a competitive threat that never actually turned violent. And that's for sure. So I'm not saying competition is unlike— I'm just saying it's the actual— there's like an inefficient way to do that competition and sometimes we devolve into that, but we shouldn't ignore the fact that most of the time that doesn't happen.

Why did the Irish Troubles heat up in 1968, '69 and not earlier?

In those years in particular? Yes.

'Cause things seemed much better. A lot of the 1960s, there's a sense, well, the civil rights movement has come, there's gonna be nonviolent resistance, we'll work this out, and then it all explodes.

I mean, I think it's hard to nail these things down to specific, I mean, ex post, I mean, I think this is, if I remember the case well, I mean, ex post, I think this is when the British Army came in.

—which is also an unusual and somewhat surprising development because they had not been coming in during a lot of earlier manifestations of trouble.

I mean, so the decision to bring them— so I think the time— there's a genuinely fragile situation that could have been resolved different ways, and the decision to bring the British Army in and then the decision to conduct themselves in the particular way that they conducted themselves, I think, helps start that actual violence. You know why it actually— it's a good question why it happened that year and not 5 years later or 5 years earlier.

I don't know. What are the greatest risks to peace in Ireland long term, or is it just all over?

I don't think it's all over. I mean, I've spent a total of 6 days in Ireland, so I don't want to— I would say I think Brexit and is people feel like there's a very A decrease in integration with neighboring places, both if it finds itself caught between integration, unable to integrate with Republic of Ireland and deintegrating with England strikes me as really problematic. But that's not maybe the flashpoint I'm worried about in the near or long term.

What does a bad scenario look like then? In Ireland? In Ireland, Northern Ireland or the border, wherever. It's going to happen.

You know, my friend Richard English is a historian there, and I think he echoed what a lot of people think, which is that— or were thinking when I was, you know, before the pandemic, when I had lots of current knowledge about some of these things when I traveled— was this— I think it's this idea of in these fragile moments where it's not really in anyone's interests to go to war, there will be some small splinter groups, essentially people who for their own ideological reasons or mistakes or stupidity will set in motion a series of events that it's difficult to walk back from. I think that does happen. And so people who are living in one of these border communities and want to protest against— violently against —some change or lack of change. That's kind of what worried me.

So nested games would be my phrase. Okay. That you have subgroups playing their own games and they don't care about how the whole game comes out. They want to optimize within their game, but what they do is a move in the larger game as well.

And to me, this is just a special case of my whole centralization of power story, which is to say that there are some groups in society that are just not accountable to anyone but themselves and their own peculiar set of preferences.— can determine outcomes. And so when I think about any splinter group, whether it's some Northern Irish band of terrorists or a faction in Palestine or of the Israeli government or wherever it may be, I think fundamentally it's these groups who can act in their own self-interest because they have at least a temporary interest that can then push a situation into violence that was otherwise quite likely to be peaceful. And so it's not just this macro autocrats are more likely to to wage wars, which I think is true, but it's also these sorts of little pockets of autocracy that we create in our own social movements that are really dangerous.

But isn't that in a way the opposite of centralized power? So you have a lot of nested games, you have many autocracies. The problem is that they're interrelated and no one of them is making a move for the whole side or the whole team. So you can have your allies drag you into war. World War I in part came about because smaller countries dragged in larger countries. It wasn't very centralized. That's one of the most destructive wars of all time. And that seems to fit the nested games worry as the main driver of a lot of wars.

Yeah. So maybe not surprisingly, you very quickly brought us to the, I think, the bleeding edge of conflict research and the thing we don't understand very well. Like when I think about the one area I wish I understood better. Both the game theory, the strategic dynamics, the empirics would be the difficulties coalition formation and the fact that I think fragmenting coalitions and many decentralized actors with the means of violence is actually really destabilizing. And I think we need more— I don't know how to think about it systematically because I don't think we've had 20 people with their shoulder behind the boulder pushing and trying to understand that.

So we're talking today at the very beginning of March, and we don't know what's going to happen with Russia and Ukraine, but we know a lot has happened. And there's one group of people that very confidently predicted a kind of disaster. Kasparov would be one. Really many other people, I would say the mainstream, that until the attack actually came, they just thought it would be some kind of bizarre limited police action in the east and not amount to much. What difference in theoretical approach accounts for who got it right and who got it wrong?

Well, I mean, I think sometimes any hugely confident prediction of military action is almost— is more often wrong than not. And so I'm always a little— so I'm not sure that there's something theoretically that drives the hugely confident predictions, because I think people— I think in those instances, some people are simply too confident. I think the only reasonable way to sort of look at the situation 3 months ago or 2 months ago was that this is at best even odds. And I speak also as someone who was less than even odds. And then it's very hard to evaluate these claims after the fact. Like, if— was the person who said it was 80% likely to happen correct? Was the person who said it was 20% likely to happen correct? Like, we don't know. But the thing that I think the more that you— weight. I don't think they differed theoretically. I think they may have differed in empirical beliefs and the things they weighed. And so I think the more that you are convinced that Putin has a personal stake in this military action and in taking Ukraine at all costs, and the more that you believe he's and his cabal and his military apparatus are are capable of complete misjudgment, then I think the more likely you perceived a risk of war. I mean, in retrospect, the thing I think people should have used to predict a risk of war, but that I very seldom heard anyone talk about, was the degree to which you thought the Ukrainians would be unflinchingly intransigent on any compromise, uh, and the degree to which you thought the U.S. government would be intransigent on any compromise whatsoever. That's mainly what I didn't predict, but I don't think anybody was talking about that, including me, a month ago. Let me tell you my sense.

It may be close to yours, but tell me what you think. The people I know who took seriously the fact that Putin believed in his own stated ideas tended to predict this particular outcome better than the people who thought it was a kind of game theoretic calculation.

Well, but I mean, so I don't really think we should place a lot of value on speeches that politicians give as their actual intentions, as a general rule. I just don't think that's gonna lead us to accurate predictions. And so if it— it might have been consistent with what happened in this case. I don't think it's like a useful predictive tool.

But there are clearly cases it applies to. So Hitler more or less said what he wanted to do. In fact, he did it or tried to do it, right? That would be a case where if you pay attention to his ideas and speeches, you would do very well. Though just in the abstract, you might think it's crazy. What guy would ever do this? Well, he did it.

I mean, this is an empirical— I mean, if we just took all the leaders and all the things they said and then whether or not they did the things they said they were going to do, I guess my prediction is they would actually— the correlation would go in the other direction, particularly with military exercises. But that might be wrong. I've never seen someone run that analysis. Yeah.

I think of political speech as more informative. So China is insisting that within the 100th year anniversary of the revolution, which would be 2049, that it will bring Taiwan back into the fold. Now, they may not succeed in doing that, but I strongly believe as a result of their words, that's what they intend to do, and they will try very hard to do it. Why is that a mistaken inference? Why would they say it? What's wrong with the simple— they say it because they're selling the idea to their public and they believe it, and a lot of their public believes it, and they want to do it.

I'm not saying there's no information. I'm just saying they I mean, you often— I mean, most of the time you have two parties that— I mean, we have the word brinksmanship for a reason, because people bring it to the brink and then they walk back. And the language that is necessary to establish a bargaining position— I think— I mean, let's put it in a totally different situation. You're negotiating in a market for a carpet. And you have to sort of stake out— I said, oh, I would never pay that. And our whole ability to get a deal is actually contingent on me knowing that you have an incentive to actually say something that's false about your own intransigence because it strengthens your bargaining power. And so maybe that's why I don't believe political speech because the whole act of bargaining with an adversary is it's necessary for you to pretend that you will give less than you want. We all know this. We all do this instinctually. Usually, and it's a ritual, and more often than not, you find the deal. You know, you don't walk away, and eventually you buy a carpet.

But how's this for a theory? Political speech really matters in cases where there's no Solomonic solution where you can just split the baby. So you can't give Putin half of Ukraine, Taiwan can't give China half of Taiwan, and then in those cases, it actually indicates there might be a likely conflict. Another case, there's just plenty of bargaining, right? Right.

I mean, I think what's the problem though is actually the fact that we create a set of institutional or ideological rigidities that keep us from compromising. So that hasn't happened in Hong Kong, right? There's all these gradations of Chinese control over Hong Kong because there aren't as many bright red lines. And I think that's meant that— that's why there's not— actually, that's why I'm not particularly I wouldn't predict violence in the case of Hong Kong. And so to the extent that you're right, I don't think it's the statements that are the problem. I think I'd look for these institutional rigidities that cannot adapt to changing relative bargaining power.

Given your understanding of the psychology behind decisions to go to war, what can we do to lower that risk other than just make people nicer? It's sort of tautological answers, but substantively.

Yeah, I mean, I think the answer is more organizational than psychological in the sense that I think we— they're related because— but I think we make decisions as groups and we make decisions as bureaucracies most of the time. And very— only in the most centralized countries do individuals make this decision. And even then, I don't think it's individuals. I think there's still this apparatus. And so I like to pay attention to the errors that we make. As a result of that maybe accentuate our psychological problems to be— our psychological tendencies to probably process too little information and to be a little bit too— just succumb to something like availability bias, which is the opinions in the room. And so any type of organizational procedure— so I think a smart White House or a smart Kremlin or a smart military, you know, defense ministry would actually try to foster a way of actually reducing availability bias in the room and actually getting these red teams and these sorts of other types of ways of getting devil's advocates into the room. That's— yeah.

What did Shakespeare understand about war that current social scientists do not?

So I'm one of these people who didn't really read books until I was at the end of my college career and missed the induction into Shakespeare. And so I feel I might be the last person.

But you talk about this in your book. You talk about Henry IV, right?

Uh-huh. Uh-huh. So maybe not in any— I don't know that— yeah, so since I can't name the particular insight that Shakespeare had, I will— I mean, that to me is an instance of— and this was I think true of a lot— again, I mean, I don't want to beat this down because I do think there's a lot of other sources of instability, but I mean, that period of European history was a period of concentrated autocratic power and a set of monarchs who could indulge not just any private interest they have in war but actually ideological or glory-driven preferences for status and status competition amongst themselves. So that's a reading of history that echoes very loudly in my ears some of the patterns I see today.

What do you think of the Steven Pinker thesis that basically war is diminishing over time? Consistently and for structural reasons?

So I think it's— I mean, I think it's a plausible hypothesis that seems not to be true according to some of the people who I think have done the careful analysis of interstate and civil wars. And then I think if you look at other types of violence, particularly violence within societies, within cities, you look at homicide, it's unassailable. Unsaleably true. And so I think you have to evaluate that depending on the type of violence. Interpersonal violence— interpersonal violence that can be controlled by the state, definitely going down partly because we have stronger states that control violence. Violence between strategic actors, different polities or factions within a polity, I would like the trend to be downwards and it might go down, but I don't think that's where the evidence points.

What do you think of the model where maybe there's a downward trend for frequency,, but there's an upward ratchet in terms of destructiveness. So from World War I to World War II, well, it wasn't that long, but the second was more destructive. Now it's been a long time since World War II. Weapons are much more powerful. If another world war happened again, the standard line, World War IV will be fought with sticks and stones, right? Is that the best model we have?

Uh, I think that's, uh, I think it's, yeah, I think it's true that as we It's not always true, but I think as a first-order prediction, thinking that war is going to become less frequent but more deadly than it is, especially when it involves nuclear-armed powers, that's— yeah, I think that's basically right.

So you're a pessimist for your grandchildren? Yeah. Because the next one will be a doozy, right? We don't know when it's coming, but under that hypothesis, there's a great deal to worry about. You have to think it will be worse than World War II.

So I think it will, but worse than World War II and nuclear Armageddon are two very different things. And then the truth is going to be somewhere in between. And so I think the probability of limited use of nuclear weapons in the next 50 years is extremely high. And I think the use of all-out mutually assured destruction is, I mean, impossible to assess. I don't see any reason to think it's particularly high, but the fact that it's small and possible is a little bit terrifying.

Where do you think is the most likely next target for a nuclear weapon?

Does this include dirty bombs and— Not dirty bombs, but a major explosion. So a state-sanctioned use of a warhead. Doesn't have to be state-sanctioned, but— It's a good question. I would be more worried about North Korea or using one, or Pakistan or India than I am of Russia or the United States. I think we differ on this, but I— maybe not. In the next 3 weeks, anything's possible, sadly. And so this will come— so this will look very dated. I think it's unlikely, but I do think we're recording this when it comes to Russia at a moment of just tremendous uncertainty. We're just 3 days, 4 days, 5 days into an invasion, depending on when you want to count the beginning. Anyways, this is— but ignoring that, which is kind of impossible, I think it's probably in one of these smaller countries.

If Putin thinks he can survive, I certainly would infer from that that the chance of him launching a nuclear attack is very small. But if he thinks he's not going to survive, I would expect he would be willing to incur a strategy that just mixes up what's happening, probably lowers his expected value but boosts the variance. And then I would think a nuclear attack is at least a few percent, maybe more, maybe up to 10%. Am I thinking about that wrong? Is, is that the right distinction, whether he expects he can survive or not? How would you think about that?

So even ignoring the nuclear calculus, in some sense, I think the whole basis for this war or the whole basis for the aggression and the aggressive bargaining is Putin's desire to preserve and advance his control and his vision and his— or maybe that network and that elite group's control is my— and not being a Russia expert, I mean, the whole— what possible threat does NATO or capitalism or democracy have to the average Russian? I mean, even— I don't see it. And even if it existed, I don't see a threat that's worth this kind of aggressive posturing and brinksmanship and invasion and possible use of nuclear weapons with the Ukraine. So it's all predicated on the interests of this group that's in power. And so all of the action and how far will they go to preserve that and how far will people around him let him go and how personalized is power actually? I don't think we really know. Or at least I don't know.

Why is it you're more likely to get randomly beat up on the street in England than say in Canada, America, or New Zealand?

Are you more randomly likely to get beat up? I think so.

At times I'm slightly afraid. Is it an alcohol issue or a content issue?

Well, we live on different streets. You live in Virginia and I live on the South Side of Chicago, so— Then it's not true, but— So, I mean, no one's gonna beat me up, but is someone gonna come and knock me down and grab my phone? Maybe.

That's a very different kind of violence.

That's a very different type of—

In England, if you were beaten up, it might just be for sport.

By a bunch of— Yeah, I think— I mean, I would say— listen, if there's bands of, you know, hooligans going around beating up people for sport, I would hypothesize that's a social norm that exists in that society that somehow says we're going to tolerate this particularly egregious set of behavior. We tolerate— we just tolerate different bad behavior in our own communities here. That's different. So it's just that we tolerate— like some communities to some extent We'll tolerate one kind of violence maybe in— on the South Side of Chicago, and we kind of tolerate and have become accustomed to maybe something different in England. That would be my suggestion.

We used to have generations of American men that they'd fight in wars, they went away, they came back home. Maybe they were hailed as heroes, but they probably had a lot of PTSD. That's really not so much the case anymore. How has that changed American society? Well, they—

sorry, they probably did have PTSD. Sure, yeah, a lot of them.

And now there's less of that. Obviously some veterans coming back from Afghanistan, Iraq, but much more.

So there's less people having this martial— well, I would say like the experience, the first, the most important thing about this experience probably wasn't the exposure to trauma and PTSD. I would have thought the organizational experience and the bonding, but maybe A small fraction would have PTSD, but everyone would have come back with the socialized— the military socialization, for example, and maybe either depending on how that war conducted itself, potentially very poisonous and xenophobic views or potentially the opposite. I mean, I think that would be very idiosyncratic to the particular war. I would say— Maybe like a— we're probably a society that is less socialized into large bureaucracies and obedience, would maybe be my first response.

I would think we're more socialized into large bureaucracies. More people work in big companies. Government employment, if you count contractors, is up. Society to me feels much more bureaucratic than when I was a kid.

I see. I would say, well, maybe it's that particular— I mean, that sort of group project, maybe that's— I don't know. I work in— maybe I'm just projecting my own experience because even though I'm in a large institution, I'm this very autonomous individual. And so I'm not seeing that. I do think there's something qualitatively different about being a maybe temporary member of a corporation for the next 3 years of my life to socialization into sort of a mass war move, even if it's also for 3 years. I think it's probably a very different psychological experience.

Especially for the younger members of Congress, is it a good thing that so few of them have served in the military, or should we be concerned?

You know, there is this— I mean, we shouldn't make too much of these correlations, but it is interesting that the number one— if you look at— I think some of the people who have run correlations of what kinds of leaders are most likely to end up taking their country to war, it's ones who have been trained been in the military but never fought in a war. So you can imagine that there's sort of an inverse U-shape, that no military experience or military experience with actually fighting a war both lead you to be averse, and then there's that middle ground. That seems intuitively plausible to me. At the same time, I'm not sure that's on my top 5 list of things that would or would not make a congressperson go to war.

Now you're doing quite a bit of work in Colombia recently. Would you go long or short the country and why?

I go long on the country on democracy and civic participation and stability to a degree, which is to say I think it's unlikely to get worse than it is now and it's quite good. I don't go long on economic development. I see them maybe as getting stuck in the middle income trap. And what's their problem there? So I keep asking people who are in government or in industry, and it's striking to me that nobody has an answer and people haven't been thinking about it. So the fact that there's no— the fact that it's not an energetic topic of discussion and focus is in itself maybe a problem. I think it's also a country that just faces like some severe cost disadvantages at being either an industrial or service powerhouse. And so my whole view is that most places in the world are doomed to be part of a middle-income trap and the lucky ones will find themselves in political union with a place that's not. And so Newfoundland would be one of the poorer countries in the world if they didn't have the good fortune of joining Canada. 70 or 80 years ago. And so if Colombia was in a political union with Mexico, I would be much more bullish on Colombian economic prosperity because I think they would siphon off a lot of the success that I would predict for Mexico.

Does Colombia have the problem that I associate with much of Central America, that market size is actually quite small, the country appears moderately populous, but the different parts of the country are not well connected. So it's a set of semi-isolated islands. And then their ideal major trading partner, the United States, is fairly far away. And maybe the best scenario for Colombia is to think of Panama as part of Colombia, which of course it once was, and Panama is doing great. And the rest of Colombia is the hinterlands, and you have the hinterlands with different cutoff parts with all small market size and no, no great neighbor nearby at that. I mean, I think that's—

I think it's a general strategy for the Columbias and the Ugandas and the— all of the countries of the world that are not obviously going to be a center of agglomerations of industry or services or tech or ideas but are pretty good, um, would do better by— in the long run, they might not do better in other dimensions, but they would do better in terms of economic development by seeking more political union, maybe not with the Panamas, although Panama is important because of the trade networks and things, but I would see political union with the economic, the places with those agglomerations.

But for Colombia, there's no natural partner, right? It's certainly not Venezuela. Right.

I mean, if, yeah, the tragedy for Colombia is that Brazil doesn't speak Spanish, I mean, as their principal language. And so otherwise that would be the natural. No, Colombia is, I think, at a geographic disadvantage for long-run economic development.

Why does the country have a relatively high-quality bureaucracy given those problems?

You know, every single researcher who works there asks themselves why they have this good fortune of working with amazing technocrats. It's, you know, maybe it's the— I mean, it's a very unequal country in terms of access to education and in terms of access to these positions, but that's also true of many other Latin American in African countries, so I'm not sure why Colombia seems better. I'm not sure that's entirely it. One self-serving hypothesis I have is that in the United States, your path to political power and your generic degree is in law. Maybe in China, I'm told it's engineering, and in Colombia, it's political science and economics. So it might just be that that idiosyncratically creates— it takes these great people who might have become engineers, who might have become lawyers in another system and turns— and and then teaches them a technocratic science. And that's like the elite language.

If someone has 2 weeks and they want to take the perfect trip through Colombia, what is it?

I enjoy renting a car and doing my own road trips, which is something that very few people do even though it's perfectly safe. There are very few, if any, parts of the country where you would ever That's 10 or 20 years in the past where you'd worry about that. And then my general strategy is to pick medium-sized towns and sites rather than— I avoid the big cities. So I would avoid— you probably have to see Bogotá and Medellín, but I would actually try to just skip over them. And I would pick a quarter of the country and I would drive around that for 2 weeks. The ones I've chosen, we did a 2-week road trip from Maybe one of my highlights is from Bucaramanga through Colombia's Grand Canyon, Villa de Leyva, Spanish colonial towns up in the mountains, down to Bogotá over 2 weeks. And that was fabulous.

Should I be optimistic about the future of violence in Mexico?

I am just starting to answer that question for myself and the few days that I dipped my toe into it in person, and now I'm going back, makes me in this paradoxical way very bullish on Mexico in general, but wondering if, how they will ever solve this problem of, I think, I'm not sure how they will ever solve the problem of violence because I think solving the problem of violence requires recognizing that these armed groups and criminal cartels have an immense amount of power, and that they have to find a way to live with them. And I think the— and I'm not sure that's the right choice. And I think Mexicans have chosen— they've sort of said that's unacceptable. We refuse to do that no matter the cost. And so we're going to be engaged in a long, brutal, and possibly unwinnable war.

—But a lot of particular states will let the groups be.

And so I think you're going to see— yeah, you're going to see a mix of— and you see that everywhere. You see that in any country, including the United States. Different municipal governments and different state governments agree to be conciliatory to their criminal gangs to different degrees. And one of the challenges is that— is I think that tends to be democratically unpopular. And that's certainly been true in Colombia. My first read is that might be what's true at least nationally in Mexico. And so as soon as you have transitions of government, I think there's just a risk of every X years of that conciliatory nature and that peace between the state and the gang falling apart. What about the Wagner's Law catch-up hypothesis?

That for a country of its income, per capita income, Mexican government is relatively small and that if the share of government in GDP went up, say, 4 percentage points, that over time they would just have more resources and beat the gangs. And Mexico would still be violent, but more or less converge to where a lot of other countries are at. Is it that simple? They just need more money?

Well, no. I mean, that's like saying if we arm more, do we expect to have less violence? I think once again, I think it's about we're going to change the bargaining power. It's one of these things. You're not changing the probability of war. You're changing the bargaining power of the Mexican state versus the gangs, but what you might not have changed is the fact that they're unwilling to compromise. Able to become overwhelmingly more powerful than these gangs and cartels and thus concede either very little or just have an incessant war but on the margins, which is a little bit like the war on drugs in the United States, right? We've decided we're just going to go to incessant war against these almost impossible to eliminate gangs, but they're now very small and confined because we have an immense amount of bargaining power vis-à-vis these very weak gangs. I just don't see Mexico ever getting there because they're the principal channel for drugs in the United States. Now, if we develop other channels of drugs in the United States, or if we somehow discover how to grow coca in other regions that lead to different transport networks, that would totally change it. But otherwise, that's the sense in which I'm not bullish on Mexico. I'm not— yeah, I'm bullish— I'm not bullish on Mexico. Okay, so you're Canadian.

Who predicted the Ottawa truckers convoy and what is it they saw that other people did not?

Well, I don't know if anyone predicted it. I mean, I think anybody who— I would say most Canadians even, because it's not— I don't think it's nearly as polarized a place, although I haven't lived there in something like 20, almost 25 years, recognize that there's like a strong conservative and anti-establishment sentiment in the country, slightly fewer than in the United States, but not dramatically different. And then my distant view is that because there are very few ICU beds and because of the particular nature of the system, and for some reason they didn't invest in them over the last 2 years, the government has to be super cautious and really had to crack down harder than in the United States where we have many, many more ICU beds. And so, and they may have just overstepped what people were willing to tolerate. So I don't know. So I don't think people were surprised at the movement. I think they might have been surprised that it sort of emerged in this sort of raucous, kind of amateurish, kind of violent. And some of the weirdness at the fringes was extra weird for Canada.

And why didn't the Ottawa police just arrest the truckers? And as a follow-up to that, how much is the decision to arrest like the decision to fight?

I mean, I don't know. I grew up in Ottawa, and that has given me precisely zero insight into 21st century policing in Ottawa, I would say. But I mean, so the first part of your question, I don't have an answer. I think— I mean, I just— in general, I think it's— causing a— violently disrupting or closing down protests is just an inherently risky political strategy. And if you don't think you have to do it, you might just try to wait them out a little bit, especially if it's minus 20 degrees. But is it like fighting? I think it is. In some sense, that's the equivalent.

The world of ideas. What did von Clausewitz get wrong?

You know, I think the basic insight that we get from him, which is maybe the only one I've internalized and remember, I think is basically right, which is that— and which is the thing we often forget. And so I'm not answering your question because I actually don't know. Tyler Cowen. Presumably we don't talk about 99% of what's else in that book because it wasn't particularly insightful. Might be part of the answer. But I think this idea that politics and fighting are sort of two sides of the same coin is a very deep insight that is surprisingly easy to never learn and to forget in spite of everyone being able to, you know, repeat his famous phrase.

What special insight do you think you have into Hayek?

None whatsoever. But what do people either overrate, underrate?

How's your take on him different?

I have 4 volumes by Hayek. The thinnest one I read 25 years ago. So maybe the fact that I suspect a lot of people are like me in that they have bought the volumes and aspire to read but have not fully read. So probably underrated. But that's— yeah, that's—

What's the thought of Albert Hirschman that's influenced you most?

I think— I mean, Development Projects Observed is probably like a whole set of volumes written by people of his— of that generation and maybe the next generation. It's, I think, an idea that development is a pro— the idea I take from it is that development is a process of trial and error or in some sense it's almost a process of creative destruction that requires lots of trial and error. I think it's unfortunate that our general attitude towards international development and the median— or the general attitude in most development organizations is to absolutely ignore and remain institutionally and individually ignorant of the importance of sort of Trial and Error and Creative Destruction in Policy for decades.

Jeff Sachs is a development thinker. Overrated or underrated?

I would say— I mean, Jeff Sachs was one of my teachers and letter writers and things. So with great respect to Jeff Sachs, I think overrated only because I think there's a handful of view— or properly rated, but so many other ideas are just not read. Jeff had a popular book and a clear message, as do a handful of others. All of those people whom are friends and I respect are all overrated because everything else— it's read too much and everything else is read too little.

For bouldering, what's the special skill that you need that most people don't appreciate?

I think it's the fact that you have to not like other sports because it's the kind of thing that rewards intense focus and And I think practice, practice, practice. And so the extent that you like other physical activities, I think it— maybe all sports do this, but I think bouldering in particular rewards single-mindedness. And so a lack of interest in other physical activities can be a benefit.

You wrote some long while ago, maybe it was even 2007, that people who work for development agencies should not expect to fly business class. What's your current view of that hypothesis?

So I still live by that hypothesis and I don't know if I was at all influential in this. I'd like to think I get like 2% or 3% of credit, but within a few years, the World Bank did severely limit the amount of business class travel or the conditions under which it's possible. And given how many of my World Bank friends previous to and after this were mad at me is why I'm sort of conceited enough to think I might have had a tiny bit of influence. Mostly I hope it was just makes good sense.

And you think it's correct to save money or to remove the sense of entitlement or some other reason?

I mean, to the extent that— listen, a lot of the development discourse and my own feelings are a little bit utilitarian in the sense of people are saying, listen, we can make choices here where we can spend $1,000 per person or we could spend $500 on 2 people and we'd like to decide which is better. In terms of what can have more impact. And secondly, we'd like to find a bunch of innovations that are really— have low-cost impact and so that we can take this limited and sometimes increasingly limited amount of development money and just do the most good with it. And so the antithesis of that is just paying 4 times more for a slightly more roomy seat so you get a slightly better sleep when you could do the same thing with a 5-cent Valium pill.

What's important about the thought of Harold Innis?

So I was really influenced as a college student by these economic historians of Canada and South America, particularly Southern Cone, of whom Innis was very important in the importance of specific commodities and the characteristics of commodities in determining a country's development path. I think it's like a really underrated view of understanding different countries' development paths. So Innis would say, "Well, if we really want to understand why Canada looks the way it does today, we would have to look at cod or we would have to look at furs." In Argentina, you would have to look at cows and in Colombia, you would think you have to look at coffee, but you actually have to look at all of the things that Colombia cycled through before they stumbled upon coffee as something where they had like a global cost advantage. So it's just the idea that if we want to understand the economic geography and political paths and political stability of a country, we need to understand what commodities they started out with. And my own personal— and what this led me to do and how I got started on the path of studying war actually is by realizing that some commodities are just hugely volatile in price for various circumstances and some are not. So if we wanted to understand long-term paths of development and why some places weren't— it's because they lost the commodity lottery and they got a volatile commodity rather than a stable one.

Not unrelated to Russia, of course. Last question. So you teach at the University of Chicago. It still seems to me the place has a special ethos, a mix of seriousness, people teaching things to themselves, showing up meaning business, free speech, some cluster of ideas. How has the University of Chicago maintained that? What's the secret there?

So I selected this place 6 years ago largely— in large part, it was one of its chief attractions. And I've asked myself that question. I think— I mean, there's a cheap answer, which is sort of cultural persistence. I— but then the question is why.

So many university cultures have changed, right?

I think the fact— it may be that the fact that— maybe I'm gonna go back and contradict myself, the political speech point is— I mean, the fact that I think the university created its— it built its social identity around being that. And more than— the reason you can point out this thing is not— even if you didn't spend any time there, you would be able to say that is the identity of this university because that's what they say they are and that's what they— they show up in the news because they're the university that does that. And so, yeah, I think what's maybe— here's what's more surprising to me is why— the thing that— it shouldn't be surprising that a university has done that. The question is why haven't more universities done that? And I was faculty at Yale and I was faculty at Columbia and they were many great things, but they were not places that self-consciously define themselves as places to be about ideas and about ideas at all costs. And so it's only in retrospect that I wonder why not. Thank you very much, Chris.

Again, Chris's new book, Why We Fight: The Roots of War and the Path to Peace, is now out. Highly recommended. It's been a pleasure.

Thank you.