Chris Blattman on Development, Conflict, and Doing What's Interesting

Conversations with Tyler is produced by the Mercatus Center at George Mason University, bridging the gap between academic ideas and real-world problems. Learn more at mercatus.org. And for more conversations, including videos, transcripts, and upcoming dates, visit conversationswithtyler.com. Today I am with Chris Blattman, economist, and he is at the Harris Public Policy School in Chicago. Chris has worked on most of the central issues behind economic development. Thank you for coming, Chris.

Thank you.

I want to start first with your work on cash transfers. So say I have some extra money and I want to do the world some good, and I'm thinking, Should I just send money to some individuals in a poorer country, go to the Western Union office and send it off? Is that the best thing I can do with my money?

I think if you don't know anybody deserving of help or that needs something or if you don't have a personal connection to a place, as an impersonal way to give effectively, it's probably the best way we have. I don't do really any of that myself or I do, but I'm sending cash transfers in some sense to very specific people. There's very dynamic people I know who are getting through university or or going through school. And so I— but I have this advantage of these personal connections.

So say I have connections to developing economies, I know some people, and I'm thinking of not only sending cash but packaging that with some kind of complementary benefit or intervention. What's the most effective thing I can package the cash with? And say I have some time, some resources, some expertise. What's actually going to help?

That's a good question. I would say if you were a large organization— and you're not— but if you were a large organization or a government, I would start pointing you in the direction of bigger sort of public services or infrastructure-like things or things that get delivered to groups, but that's not really what you're asking.

I'm going to give person cash and something. Should it be cash and a banana, cash and a book, cash and an Apple Watch?

I think a lot of people, and this would be true even of a young person in this country, I think don't have a good sense of what their paths and options are, and so I think if The time that I've spent helping people think through what their options are and reevaluating those opportunities suddenly changes all of their returns to investment. So that kind of advising and mentoring and connection I think is unusually powerful.

What's the last important thing you learned about cash transfers?

We recently went back to some cash transfers that were given almost 10 years ago and following up a randomized control trial in Uganda in the north and we're just Tyler Cowen, in some sense putting out those results. But what we found is the initial result after 2 and 4 years was like other places seeing big advances in income. So people get cash. They're poor. They couldn't invest in some of their ideas, but they had good ideas and so they take off. And now what we've seen is essentially they've converged with the people who didn't get the cash. The people who didn't get the cash have caught up because they slaved— they saved and accumulated slowly and got up to the point where they have the same levels of success. So they converge to a good level, but this means that I think cash transfers are much more of a temporary acceleration than they are some sort of permanent solution to poverty.

So after 10 years, it might wash out.

Right. Well, this is just one place, so we'll see. There's a lot of these things around the world, but we've based a lot of our optimism really off of a handful of transfers in Kenya and Uganda. So looking back at those, I think we're seeing convergence, and that's what we'd expect.

To the extent there's convergence at, say, a 10-year horizon, what do you think is the mechanism? The people who get the money, they work less hard or they give it away to kin who expect a form of social insurance or the people who don't get the money become more ambitious. Just what's your intuition?

One of the constraints that you see on people's success is that these are just incredibly volatile lives. So if you ask someone to recount the last 10 years of their economic activity, you'll just get this incredible litany of terrible shocks and unexpected events that are really quite amazing. And so it's amazing that anybody succeeds at all after you hear this list. I think— so a lot of people are just failing, but for the most part, people are doing well. I think it's just the fact that a lot of these people getting cash transfers are young people who have a future and it takes time to get there, especially if you have to slowly accumulate capital. And so it's just the fact that this— it just takes a longer time for everyone else.

Given your current income based on tenure, of course, how high a permanent basic income would you have to get from the world for your life to change?

Maybe you feel the same way. I probably do what I do practically for free.

But people say that, but if they had, say, a billion dollars a year, I'm not convinced it would be true.

If I had a billion—

So we work downwards from that. $200 million a year, $1 million a year.

At some point, maybe a little bit, maybe if it was a million, at some point I would quit my job and just do what I do, but as a free agent. I would probably write books and do media and try to produce ideas outside of the university, unencumbered by a lot of the administration, the teaching.— the things that just let me sort of focus on the ideas and the reading.

So for you, that would accelerate your career path. It wouldn't lead to convergence for you.

Well, I mean— I don't think— yeah, I'm not sure my— the question would be, would my output be any greater than in the counterfactual? I'm not sure if that's true. It would be different in an unmeasurable way.

Now Chris also works on violence and war and conflict, so I have a whole host of questions for him on these topics. And I'll start with a real softball. There's a survey article you did, and it showed in the data that civil wars seem to rise from 1960 to 1990, and then after that point, they fall dramatically. Why do they first go up so much in terms of their number and then go down?

I think especially in sub-Saharan Africa, but certainly all over the world, you saw this period after the First World War and after decolonization. You know, it's interesting that one of my favorite things ever written about this period is by these Bob Bates and Jeff Williamson and John Coatsworth who are these sort of famous political and economic historians and they compare this period in Africa to the late 19th century and a period of complete— of war and instability and gradual political consolidation after a period of rapid and unexpected decolonization. And so I sort of think of Africa which is the site of many of these wars as just undergoing a period of political instability followed by consolidation after a rapid and unexpected decolonization at the same time that you're having experimentation with economic systems and political organizations that turned out to be failures, which I think interacted with that to make it so much worse. And so the gradual— the end of these wars and the political consolidation that comes from essentially identifying, I think, an economic system and a political system that tends to work a little bit better explains a lot of the peace since then.

It seems though, if you look at Latin America, they've been independent quite a long time, and there were many more civil wars in Latin America, say, when I was growing up, than what we see now, or even the prospect thereof. Do you think there's some notion to there being a general, you know, spirit of the world that's either more peaceful or less peaceful?

Now versus then?

Because you see independent regions, the number of civil wars seem to be declining, and whether there's any specific theory that would account for that. Or is there some kind of global contagion possibility?

I think there's always the chance of a global contagion. I usually think of the incentives as aligned for peace in the sense that— so war's sort of the puzzle to be explained. I guess I don't think of the— how would I say this? I guess I see this as maybe what happened in your childhood was a reaction to a series of political and economic shocks. And a search for equilibrium after that. And it's possible that we'll see very dramatic political and economic shocks to destabilize that many countries again, but I don't see them coming very soon.

What does the median economist/global development thinker fail to appreciate regarding order and conflict and violence? Someone who knows a lot about the topic, but you feel you have a key or core insight that they don't.

So there's sort of— if I think of two types, if there's sort of the academic economist or political scientist, I think they tend to ignore the human and the emotional and the erroneous aspects, the things that drive conflict that are mistakes. And then at the same time, I think that those passions and mistakes and greed are vastly overestimated by— I won't even say the casual observer. I would say the very, very deep and experienced policymaker, and so they actually look at the problem. They're both right. They both have like a piece of the puzzle, but they both only have one piece of the puzzle. And so they're not— one's not looking at it with enough reason and one's not looking at it with enough emotion.

Why don't economic risk factors matter more for predicting violence and civil war?

I think they do a good job of predicting how easy it is to mobilize somebody into a conflict. And so if you have a of time, if there's been some drought or there's been some collapse in trade or the one commodity that your country produces collapses in price and there's a lot of people out of work, then I think it's a lot— it's much easier to tempt people away from whatever middling economic opportunity they had to join a rebellion and you can get more of them for your money. That's a very basic intuition, but that doesn't tell you why the war broke out in the first place. Just because the pie grows or shrinks because of economic growth or collapse doesn't necessarily mean that you have more or less incentives to fight over that pie. You're still trying to bargain over that pie.

But wouldn't it be like a standard model? There's a country, it sells a natural resource, the natural resource feeds the budget, makes it possible to buy off special interest groups. The price of the commodity falls, there are no longer the bribes for all the different interest groups. Someone attempts a coup d'état or a civil war. For what is partly an economic reason, that doesn't show up very strongly in the data in your papers as I understand it. What do you think that model is missing? What's the intervening factor or what is that model overestimating?

Well, I mean, I think all of these models don't really think about— the way that a lot of economists and political scientists think about conflict is that it's a bargaining process. So if the pie has shrunk and it's true that there's less transfers, but then were you to capture that pie, it's— you're going to— one is that pie is smaller, so Perhaps it's less attractive to capture in the first place, so that's going to be a countervailing force. But the war— if you're thinking about the cost of war, these wars are so long and so costly that you have a strong incentive to try to find some kind of bargain, and so people typically find bargains. This is why I think you see these broad coalition governments. You see lots of different ethnic groups included in a particular regime, very stable regimes, bring together lots of people and pay them off. All of that bribery is a way to— keep a peaceful bargain. And when the pie shrinks, there's a risk of conflict, I think, because there's a— it initiates like a new round of bargaining, and that's hard, and things can happen. But most people still have incentives to try to find some kind of peaceful solution.

As you may know, Moscona, Nunn, and Robinson have a recent paper, and they argue that a big risk factor for conflict and violence is when people have many extended relations who need to be defended, or their honor needs to somehow be held up, and this makes it easier, in their opinion, for conflict or violence to spread. What's your take on that?

So all of this talk of— in this case, I think they talk about segmentary lineages, and I've never seen it. It's one of the things that resonates the least with me. Any description of African society or African culture or conflict or politics that relies on this, I have to believe it's there because so many people have written books, but it doesn't seem to me to be a first or even second order determinant of what's going on when I think about the conflicts I know and the politicians I know and the decisions that get made. So—

Could it be a non-African or Middle Eastern effect? So if a lot of people have married their cousins, then family lineages are maybe tighter and they also reach further and there's an honor culture of some kind?

My intuition would have said that the more horizontal ties you have to other parts of society, that the more you're going to internalize some of the cost of conflict and so in some sense you're going to be— less likely to— I mean, if I weren't presented with this empirical fact that those things seem to be related to more conflict, I would have actually predicted the opposite.

Say I'm a child soldier in northern Uganda. What's worse for me, being abducted or having to fight in a war?

Well, you're usually being abducted to fight in a war.

Sure. Which of those two experiences is the one that does me the most harm?

I think it's the transition period. It's the preparing. I mean, it varies from— it's probably different. It's certainly different in every in every rebel organization. The one I worked with or studied actually was one where it was the transition from abduction to becoming a fighter where— because the training usually tried to dehumanize and desensitize you into committing violence. And so you were often— not always, you were often asked to commit pretty heinous acts. And those had this double— it wasn't just to desensitize you, though it was to try to convince you that Your outside option was really terrible. The idea was to sort of convince you that there's no way you can go home. We're going to make you do a set of terrible things right now that will mean that you can never really go back, and so like it or not, your future is really with us.

And later on, my earnings will be lower?

There's this famous paper on Vietnam veterans in the US where they find that being conscripted into fighting in Vietnam had positive effects on the wages of of blacks and negative effects on the wages of whites. And the reason was it was really down to like what was your alternative labor market and training experience in the absence of this war. And we found something similar in Uganda, something sort of eerily familiar, which is that the women economically weren't so worse off than— I wouldn't say they were better off, but they weren't necessarily affected adversely in an economic sense or adversely affected in other ways 5 or 10 or 15 years down the road while the men were. And it was sort of spoke to just how terrible women's options were. I mean, being conscripted and abducted to be a rebel wife to some degree wasn't that different than what your marriage opportunities looked like if there wasn't a war. And so— but for men, it just meant that you were out of the civilian labor market getting a bunch of skills that had turned out not to be very useful. And so it was bad for them. But a different war, different context, and a different labor market, and that could switch.

How many northern Ugandan child soldiers have you interviewed?

Oh, at least a couple hundred, maybe more. I think maybe, you know, depends if you count someone who's involved for like a month versus 2 years, but the sort of the long, long-term soldiers who are there for many, many years are few, maybe only a couple dozen.

And those contacts, those conversations, how have they changed your outlook on life emotionally or intellectually, otherwise?

Well, I mean, they— I'd never worked on conflict before I went to northern Uganda. I followed a woman there who later I would marry, and we— I followed her because we had this idea for a research project. I'd been doing some work on children, and after that I just could never do normal economic development again. It just— I've really only worked in violent context since then because it seems so So first order important and just so neglected.

And overall, your take is that children, you now see them as more resilient than you used to believe or less?

Certainly more. I mean, if you're a psychologist, you just know that people are subjected to horrible things and most people are incredibly resilient. This is sort of their big stylized fact, but I never knew that before doing this. And then you certainly see it firsthand over and over again. So I updated since then. I think I've, you know, there's a lot more nuance and I tend to think we're maybe a bit glib about saying that people are resilient because I think there's a lot of pain or problems that don't necessarily show up in our data but turn out to be really important.

The Stockholm syndrome. Some people who are kidnapped or abducted start identifying with those who took them. How true is this in your experience?

Well, I think it takes work. I think if I look at the example of the Lord's Resistance Army in Uganda or some of the organizations in Liberia that I know people who are part of, they— I think the organization— this really happens when the organization really, really dedicates itself to trying to do that. It works better with children too, but they were quite ingenious. The Lord's Resistance Army spent a lot of mental energy coming up with— they had a lot of time to experiment, a lot of people to experiment with trying to actually indoctrinate. And they were, they were good at it.

There's now, as you know, a growing literature on what I call microviolence. So studying scenes of crowds when a fight starts and using maybe video evidence to figure out why did the fight start. And I was chatting with Mark Levine, a researcher in this area, not that long ago. And he told me in his research, it seems to be the case that crowds actually are a force for peace, on net, that it's not that the crowd eggs people on. That bystanders try to stop conflicts. They don't always succeed. And this to him was a kind of revelation, how individual the violent act is. Your take on that?

It's interesting. So I don't know this. You probably know this literature better than I do. I think of, I mean, when I, what I've, what little I've read sort of thinks of those reactions as very individual instinctive reactions to a threat and they invoke the fight or flight mechanism. And I can imagine a set of circumstances where a non-hostile crowd reduces that sense of risk and personal danger, but I imagine a lot probably depends on the context and what that crowd is doing or who that crowd is and whether or not you perceive that crowd as part of your in-group or your out-group or part of the threat or not.

True or false, most humans are bad at violence.

Oh, I think they learn quickly. Probably they're bad at first.

And the microevidence on violence and the more individual-level evidence, and then finally macroevidence, like will there be a civil war? Do you think there's ultimately an overarching theory that ties these all together, or are they just separate levels of investigation where you have empirical results and they stand somewhat separate and they'll always be distinct areas? I mean, how optimistic are you about a kind of grand unified theory of violence?

So I think these individual how I react in the moment fight or flight type mechanisms are quite distinct from the way that small groups or large groups or nations go to war. But once you get beyond that to the level of small groups and larger groups and nations, I just see a lot of unity in the theory and I think this combination of this more game theoretic view of bargaining where we see conflict is arising from uncertainty commitment problems and melded with a set of more psychological issues around emotions and levels of self versus more social interest and the mistakes people systematically make. I think that framework for me explains an astonishing amount of violence. Every— it continues to explain more phenomena as I learn about more phenomena, be it drug gangs or urban disorder, and I see it in more and more places, so I'm optimistic.

René Girard, as you know, is one of the best-known theorists of violence, and one of his hypotheses concerns the use of violence for purposes of ritual sacrifice, which he then traces throughout myths and literature. What's your take on the René Girard worldview?

So the only reason I'm aware of the René Girard worldview is because I glanced at the "What should I ask Chris Blattman?" questions and I saw that name and I thought, "Oh, I don't know who that is." And my 10 minutes of investigation suggests like I just find nothing about this idea resonates with my personal experience in particular wars or— but this is the danger of you also blogging your questions or potential questions. But I didn't go study so much that I have a coherent answer other than to say it didn't strike me as somebody who had— spent a lot of time actually in or with people who commit violence.

What do we know about the causes of the Peloponnesian Wars?

So this is something that I'm brand new to. What you see is a lot of political scientists like to write articles about grand theories of war, whatever theory that may be, and they like to cite Greek historians talking about things like the grand— you know, first of all, the Spartans and the Athenians, you couldn't— Hollywood couldn't imagine two more ideal types to sort of talk about and two different— more different worldviews. And then for them to embark on this, you know, war to end all wars for the period, it— people use it to explain— to think about something that we think of as bargaining and the commitment problem, which is that massive power shifts can lead to war. Partly because it becomes difficult for the temporarily powerful person to be paid off not to sort of exercise their advantage and through first strike eliminate. And so this becomes potentially a model of that, but then no one has ever gone beyond this and actually I think really tried to see how it works. And so I've been reading it and I'm more and more persuaded that it's actually a good example of this, but it's a little bit more complicated.

I asked Charles C. Mann the same question. Why are so many parts of the New World this hemisphere so relatively violent compared to so many places in Asia, also in Africa, as you know, where you can walk around without any thought of violence whatsoever, but really a pretty significant part of our hemisphere, that's not the case. Canada, of course, an exception, but nonetheless, I'm sure you've noticed this difference. What ultimately is your theory at the macro level?

But I think you're talking about personal security and crime.

Personal security, crime, kidnapping, someone pulling a knife on me. In cities.

That's a very— a lot of that's not particularly violent. So I think—

But murder rates are fairly high.

Murder rates are fairly high. So first, I'm a little skeptical of that. I don't totally accept this as a stylized fact. I think there are other aspects of life. Probably the actual levels of violence in society are mostly going on in the home, and I don't know if— so we're talking about a very specific subset of violence that has more to do with rogues deciding that they can extract something from a stranger. And so my instinct is to say that has a lot to do with policing. It has a lot to do with the level of policing and the concentration and the dissuasion from a very specific kind of violence. And—

It still seems odd there's this commonality across so many countries.

Well, you know, most of the cities where I think you see some of the highest murder rates, Latin American cities have very few police per capita when you compare them to American cities and certainly if you compare them to Asian cities. So I don't like to be that crude because I'm actually a little bit of a skeptic of just this sort of knee-jerk to policing and we need more police and we need more investment in security resources. But as like a first order approximation, I think that's a big part of it. But I think you also then have subcultures. I think once you have cities that— and groups within cities that begin to do this, I've seen this in Uganda. It used to be that Nairobi was always a very violent place, and so— but people would sort of copy it. It became like a thing you could do. I now had a model, and this was a career or a vocation that existed in Nairobi, but it didn't exist in Kampala or it didn't exist in Dar es Salaam, and people just didn't do it. And then once you had people sort of prove that it could be done and it could be profitable, then you had this relatively small group who sort of professionalized it and do it, and now it becomes a thing., and it's kind of entrenched. But so it's sort of a multiple equilibria story.

I've read that murder rates in New York City have fallen by a factor of about 20 from their peak, which was within my lifetime. Do you think the research you've done gives us any insight into that, or is that a fully separate phenomenon?

I think of it as a fully separate phenomenon. I think of the stories that people tell when they have data or when they tell something convincing just sound very different from the kinds of stories that I'm used to thinking about for the kinds of violence I study. So some of the stories that are plausible and have some good evidence are their successful efforts to get guns off the street, the police per capita, and just the— and then just also the— going back to the old idea of just many eyes and many people on the street in more neighborhoods. So I— that just— none of that really helps me understand civil war violence or recruitment of child soldiers or any of the things that I've worked on in the past.

20 or 30 years ago, the nation of Colombia was much more violent and dangerous than it seems to be now. What has changed there? I know there are superficial answers you could give. Well, they're fighting each other less. Civil war has wound down. But at the most fundamental level, how would you explain the significant decline in violence?

So I'm still new to Colombia, and I spend most of my time in the cities thinking about the kind of violence that frankly not even most Colombians think about, and less time thinking about the guerrilla violence and the paramilitary violence. You know, I think of that, I think there was actually a relatively short period where the violence was very intense and you had a relatively well-organized guerrilla group, the FARC, mount a credible assault on the state with a reasonable belief that they were going to win and maybe take over, and a state that was slow to respond and eventually under this existential threat actually pulled it together with some US support and managed a counterattack and managed to effectively defeat this force or get it close enough to push them to the bargaining table.

But it seems more deeply rooted in their history. So there's a period starting in 1948 where 300,000 people die, la violencia in Colombian history. 19th century, maybe we don't have numbers, but it has a reputation for being a violent place.

Well, I'm trying to draw a distinction between a situation where you have a lot of low, relatively low-level persistent violence that might have to go with an insurgency like that in Colombia, but you could have said the same thing about Northern Uganda. It's not actually that violent in the large scheme of things and just draw a distinction between that and the much more high-death, high-intensity, much more violent sort of conventional war that was fought for a period of time. And so those aren't like distinct phenomena. One existed as a precursor to the other. They didn't have to. So I think of them distinctly. In that case, like I think— so Medellín is a city I've spent a lot of time in and so I don't know that this is true, but in my head the analogy is that in Medellín you had these elites. You had the Colombians in the city at the bottom of the valley with a very nice city. It was the industrial center. And as people started to flood in from the countryside into the periphery of the city and they built up these slums on the sides of the mountain slopes. The state essentially decided not to govern. And so in the absence of deciding not to govern, in the absence of any kind of government, in the absence of public services, those communities started to govern themselves and often that meant criminal groups and it could have been paramilitaries or guerrillas or eventually it's street gangs and mafias grow up and they provide order and they provide certain services. And then eventually the elite— and there's a lot of violence and then eventually the elite wakes up and realizes that this is an existential threat and has to fight it. And so that in like a microcosm feels like the last 100 years of Colombian history where you had elites in the cities essentially allowing the periphery to sort of be quasi-governed by fairly violent actors and not being particularly concerned about it until it got so bad that it threatened them in their cities and then they got serious about it and were fortunate to be able to eradicate it to some degree.

Tyler Cowen: You have a recent paper where you suggest that people being exposed to violence can make them in the longer run more cooperative because they're somehow scarred or scared by the violence. And so they want to set things right, they cooperate more. Do you think in general there's reversion to the mean when it comes to violence, and thus we should be optimistic about Honduras, El Salvador, and other places, and Colombia fits into that bigger story, or no?

So exposure to a violent act for some reason— I think one reason might be a psychological reevaluation. It might be a set of interactions you have with people within your group that sort of happen to be— that are cooperative and that reinforces future cooperative behavior. I think the evidence suggests— we don't know— the evidence suggests that you're much more cooperative with your in-group, so people who are like you. We have less evidence on cooperation with your out-group, but the little we have suggests that you're potentially less cooperative with your out-group. And so if you're talking about community government in a city or decisions that you have to make with people in your region, maybe you're slightly more pro-social, but that doesn't necessarily mean you're more peaceful overall if you're on net less pro-social at a larger political entity where you've got to interact with other groups. So it's not clear that there's reversion to any mean there.

Is it possible that the fading of the memories of World War II and also World War I are a reason why previous levels of international cooperation can no longer be sustained, that reversion to the mean is kicking in again, but now against us? And thus we're headed to a GZERO world and World Bank, IMF, international order as we knew it will become progressively weaker because of this mechanism.

This mechanism that we are, that we're reverting to the mean.

We're not scared anymore.

We're not scared anymore.

So the Nazis come, Stalin, we do wonderful heroic things. We cooperate with the British, the French, even the Soviets actually with Stalin. And then that sticks for a while. We're all committed to the UN, the World Bank, the IMF. But now, my goodness, things have been so good for so long and pretty peaceful, those are all going to decay.

Well, so the richer we are, which we're much richer, and the more costly war is, and war is now much more costly, I think the stronger the incentive we have or not, we have not to have war. And so I think it actually makes it much less likely. I think if war happens, then it's going to be that much more catastrophic, which is maybe a pattern we've seen. Over the past century. So, but I think it's actually less— so I, again, but it comes back to this calculus of not feeling like there's some equilibrium level of violence that just happens because violence happens, but that this is something that is often strenuously avoided because it just pulls people so far back.

Back to economic development. What in your view is the most underrated national development story? —The people should pay more attention to? Which country or region?

—Well, including me, Botswana, in the sense that this is the growth miracle of the late 20th century and there's a few explanations, but I think there's really only a handful of people who've paid any attention to what happened there. And so there's a few pat stories that say, well, they idiosyncratically established a parliamentary government before they discovered their diamonds. And through some magical set of circumstances, maintain this good governance and the diamonds just fed growth and prosperity. But that just doesn't seem very persuasive. So there's another story that's waiting to be told, and I don't know the answer to it.

Many people say that for numerous parts of Africa, national borders are a problem, that they don't correspond to ethnic groups or tribes in the right way. If you look at the data,— as I think Jeffrey Herbst has pointed out, among others, there are quite a few fairly small countries: Lesotho, Gambia, Guinea-Bissau. They're not actually doing that well. It's not clear they're doing better than the big countries. Does this refute the misdrawn border story, or do you still think there's something to that?

So I think, you know, going back to this question you had earlier about why was there so much instability in the years after the war, I think a set of misdrawn borders and trying to find a political equilibrium where all of these groups have to live together is— this is one source of that political shock and instability. That said, you know, and maybe I'm not understanding your question correctly, but I think that a lot of countries have been very successful at building national identities in a short period of time. So Jeffrey Herbst is very fond, I think, of the idea that Rwanda, which is a very strong state and can exercise and project its power into Eastern Congo, maybe we should experiment. And I don't know if he'd advocate this now, but 10 years ago I think he thought this was just a useful idea to propose that maybe Rwanda should be governing Eastern Congo. At the same time, I'm not sure there are many places in Sub-Saharan Africa where you would go to a peripheral place and you would find a stronger national identity than Eastern Congo where they really think of themselves as Congolese. And so there's a bit of a paradox here that people— so the national identity project has been very successful in a pretty short period of time in a lot of these countries, which makes me think that these borders are no longer so unnatural as they were, say, in 1950 or '60.

There's a much longer historical continuity of the state in Ethiopia than in almost any other part, maybe any other part of Africa. How much does this help Ethiopian development, and does this make you more bullish on Ethiopia than most other parts of Africa economically?

So it helps a lot, and it makes me much more bullish.

And how does it help?

If I had to pick one variable for a cross-country analysis, it's interesting that nobody really does this cross-country analysis. But if you ask me, like I get to just make one bet and to go back into 1800 or something and one bet about where you're going to see development, I'd probably pick historical levels of state capacity. And so first I can just, my personal experience, whether it's the little regression I get to run by working in places as diverse as Liberia and Uganda and Ethiopia and Colombia. Columbia, that correlation is more apparent than any other correlation. Why do I think it's there? I think it's there because— I mean, partly it's just a proxy for a general level of social organization that we as a society have figured out a way to solve certain collective action problems, that we've figured out a way to make certain types of collective decisions in a way that does not have to devolve into conflict and as a result were able to just in some sense do everything more effectively. So it's like a general purpose technology.

Where else in Africa would you be especially bullish about given what you just said?

So I think Rwanda. I think, you know, you can look at Rwanda and you can see one of the most longstanding and coherent kingdoms that its borders happen to match. Its borders today are somewhat close. I think part of Uganda's part of its success has been even though it's a collection of— it's many other things but it includes a collection of these historical kingdoms that also had like a continuous existence up until the colonial period. I think part of its reasonably effective state is effective. That's— I'm pretty— I'm not sure how Uganda and/or Rwanda for that matter will manage their next political transitions. That to me is maybe the biggest— The difference between them and Ethiopia is they don't have political parties or political systems to manage these difficult political transitions, so they could fall apart. But if they manage those transitions, I'm very bullish.

Former American slaves, as you know, settled Liberia in the 19th century with quite a utopian vision of what a wonderful place it will be. Mm-hmm. That experiment seems to go badly. What's your account as to the fundamentals as to why the evolution of Liberia as a modern nation state failed?

Well, until the mid-1970s, it was an experiment that was going pretty well. I think it tended to be wealthier and better off and on a slightly healthier trajectory than other parts of Africa.

Not dramatically so. Say compared to Sri Lanka, Ceylon, or the Philippines. It's not obvious it was going, I mean, that well.

Aha. No, I was thinking relative to other sub-Saharan African countries. I mean, though, I put those— if we just use the state and the strength and cohesion of society as an example, you're comparing a place that had absolutely no central organization or coherency as a region to places that had hundreds of years of civil service and incredibly developed social structures and integration with global markets, not to underplay the general integration or state development in Africa, but if you had to pick— if you had to put your finger on a point in the— on the continent where all of those things were least together, you could do worse than put your finger on Liberia in 1821 or whenever they first landed the ships. And so I think it was just a very naive— it was just a naive development project about what could be transferred and built in that kind of environment.

As you may know, in all of these conversations, there's a segment in the middle, overrated versus underrated. I'll toss out a few candidates. Of course, feel free to pass. Religion as a development force, overrated or underrated?

So I don't think of it as having really— probably overrated. I don't think of it as having any real strong predictive or explanatory power for long-run differences in income.

So if a region in Nigeria becomes either Christian or Muslim, you don't feel that changes the long-run forecast of per capita income?

I don't have a strong prior for the direction of that effect. Fair trade coffee? Probably overrated, but I don't really have a very strong opinion of fair trade coffee. It strikes me as one of these things that is more of a marketing device than really an effective and large-scale strategy for improving the welfare of people.

The Canadian political system.

Overrated. It strikes me as one of the most centralized— in terms of the way the political institutions are set up, it strikes me as one of the most centralized in terms of power advanced democracies. If you have a— the Senate, the upper house is almost a rubber stamp, is relatively powerless. Not entirely, but it's unelected. You have a prime minister who rules the lower house, and if he has a majority, essentially everybody needs to toe the line.

But say you have a very open economy. Don't you want more centralization? Because the openness of your country is itself the extra check and balance. So New Zealand is pretty centralized. They're quite an open economy. US, of course, is not so open, and we need all these layers of nonsense.

For me, the check and balance there that probably makes this work isn't the openness. It's the fact that they think they're extremely strong informal social norms that govern what the government can and can't do. So this extremely centralized powerful political figure, the Prime Minister or his cabinet, is very constrained by the larger populace who have shown an ability to completely wipe out an entire party if necessary within one election cycle if they really overstep. That's this magical, elusive, ill-defined thing that is protecting it.

Overrated or underrated, United Nations peacekeeping?

I'd say underrated. I think a lot of people are either cynical about United Nations peacekeeping and its effectiveness or just genuinely unsure. And then you have certain incidents like what's gone on in Haiti with the the spread of— now I'm forgetting the disease, but— Cholera. Cholera. That have tarnished the name and you've— a lot of accurate assessments of the impact on the local sex work trade and women in places where there's a lot of peacekeepers. So these are all problems, but I think they, you know, if you look at the data, they're overwhelmingly effective and if I had to say the one thing that really made me a stronger believer in this whole bargaining approach to conflict and the kinds of problems that actually lead violence to break out. The peacekeepers, in some sense, the everyday job of a peacekeeper is to solve all of these problems. It's to create commitment. It's to reduce uncertainty. It's to temper emotions, and they do that very well. I've seen it. You know, I was in a northern town in Liberia trying to do work, and a murder of a girl in one little village led to Christian and Muslim hostilities in that village, which turned into Christian-Muslim riots in the entire district, to the point where we're holed up in a hotel and the building down the street, which is where the Vice President's giving a talk, is being stoned by an angry mob. And a Pakistani battalion, which for some reason took an extra 24 hours to act, but nonetheless sort of came in and made that all go away within about 3 or 4 hours. And the ability to to do that over and over again. There's just 100 stories like that in Liberia. We wouldn't have peace there if it weren't for those forces right now. Battlestar Galactica. I guess underrated by anybody who hasn't watched it.

And what do we learn about violent conflict from that show?

What I liked about it the most actually was less about what we learned about violent conflict than what we learned about authoritarian politics. Mostly the fact that I think it got the average American reader to get on board with election stealing and terrorist violence by putting them in the position of either the elite or the subjugated group, it depended on the season in terms of what they were willing to do, what sort of violent and non-democratic acts they were willing to do to achieve their political objectives.

The standing desk, over or underrated?

Overrated by people who have standing desks and underrated by everyone else. And do you still do it? I do both. I sometimes I sit, sometimes I stand. I probably, I would like to stand more than I do, but I don't always sleep as well as I once did because of my children. So I sit more often right now.

Let's say that I'm coming to you and I'm a seasoned traveler. I've been to a lot of countries, but let's just say that I've never been to Africa before and you're giving me your 2, 3-minute pitch on how to think about where to go in Africa and how to do it.

What's your take? If you can go and spend a long period of time, like a couple of months, which most people can't, then I think trying to find an excuse to live in a village is an incredible experience. We have— there's more ways to do that than you'd think than just voluntourism. Barring that, I would say— how long do we have? Do we have like a week or 2 weeks? 2 or 3 weeks.

So I would— per trip maybe there'll be more if you give— Right, exactly.

So I would say I would spend 1 week in Kenya and Tanzania seeing a mix of the large popular and small never-visited wildlife parks, partly because they're amazingly well-run and they deserve to be rewarded with tourist dollars, but also because they're incredible experiences. But I wouldn't spend more than a few days on that, and then I would probably send someone to see the thing I haven't seen, but even in spite of working in Ethiopia for many years, I never went and properly saw this 800-year-old state as you've noticed, but seeing a state and its architecture in a culture that you've probably never thought about before and seeing that firsthand is not only spectacular, but I think you change some of your priors.

Ethiopia, you have a famous paper on sweatshop jobs in Ethiopia. One thing you found, I believe, is how often people quit these jobs or they become unhappy with the jobs. Why are they unhappy with the jobs?

I think when I first started studying these jobs, I thought they were jobs that were a little bit like winning the lottery. I thought you got paid more than your alternatives, as miserable that pay might be. More importantly, I thought you got full employment more naturally than your alternatives, and I thought that there was upward trajectory. Maybe that was the most part, that it would lead lead you on this virtuous path towards more productive work and better pay. And instead, I think they turned out to be more like McDonald's or Walmart jobs.

And why does anyone take them at all then?

So a large number of people in Ethiopia, I think, were taking these jobs. The factory jobs were their social safety net. So they were able to make more money doing their own business or working for a firm or working even in temporary wage labor. But all of these things were risky and volatile. And so when that business went south or when the wage labor temporarily went away, with those 3 months of your life, you'd go work for the factory if you happened to need cash. And then you would go and you'd be looking for something better and you would find it. And after a few months, you would leave. And this was the social safety net. It was working factories.

The fieldwork I once did in Mexico, I also found most people didn't want these jobs. Two factors I found that drove some people to take them. The first was just extreme boredom with rural life. The other was women being quite unhappy with all of their options where they were, in some cases possibly even being hit or abused and just wanting a way out. This was one possible ticket. Are those possible motives for Ethiopians?

I think those were possible motives. I wouldn't say they were the most common motives in this group, but most of these farms weren't in really rural rural areas. So I think a lot of the women— and it was 80% women in our sample— had a lot of other options. So maybe the thing I updated on most of all was the high quality of people's options. I had a very low opinion of their options and so the thing that I learned as a result of this was mostly that the options for these young women were much better than I expected.

In a lot of typical economic development stories, you industrialize, there's manufacturing, then you export.— and there's some kind of virtuous cycle maybe with increasing returns. So if an Ethiopian takes a factory job, though many of them won't like it as you say, do you think the social return to that decision is higher than the private return or the other way around? So at this—

at their level of industrialization, I don't think it matters very much. I would say that—

It's because you don't buy into the virtuous cycle theory? You don't think they're increasing returns? Something else?

Well, I think that you would need— at this point, it's really quite a minor sector and the missing ingredient that's leading to any— let's suppose there are increasing returns to the existence of this sector. The mystery factor that needs to be there to sort of make them successful has almost nothing to do with the availability of labor. I'd say it's actually the availability of management talent. The availability of financial— like high-quality financial service, things that you need to sort of buy and split and merge and grow businesses like banking and accounting and— I think these are binding constraints on industrial development that— going to Ethiopia to be an investment banker or an accountant would probably be a far more individual development strategy than sending somebody cash transfer.

—this idea of premature deindustrialization, the notion that because of automation, manufacturing now creates so few jobs that many parts of Africa perhaps can never build a middle class because they could have factories maybe, but it would be done by machines and there's no wage differential reason for pulling your factory out of China, bringing it to Africa the way maybe there had been in the past.

What do you think of that? Are we seeing premature deindustrialization? I mean, there's— how many industrial workers are there in China? Like maybe 50 million or something?

Data in China seem to be bad, but you look at Brazil, India, South Africa, they seem to have declining shares of manufacturing employment before they're really fully developed nations the way Denmark, US, and Germany are.

So, you know, I come from Canada and if I cut up Canada into 20 independent parts, I would find all sorts— I would find like 18 prematurely deindustrialized places, I think, and then I would find a couple of places where you have, you know, like Southern Ontario and where you have a lot of industrial development. And so I don't think the fate of like every nation in Africa is to, absent premature deindustrialization supposing exists, actually be highly industrialized. So I would think that I can see a lot of Africa just sort of ending up like Newfoundland and irrespective of any of these forces that you're talking about. Minus the puffins. Minus the puffins. So yeah, I don't think the technological change is the first order thing. One way, I don't know this very well, but I would've suspected that one thing that I think might be true is that there is a peculiar set of circumstances that exist in East Asia and Southeast Asia that have made it hypercompetitive in this particular sector and you've seen whatever— a lot of industrialization that existed elsewhere in the world, even if it's from other developing countries like the Maquiador region in Mexico, relocate there. Until the rest of these countries catch up in terms of, I'd say, their state development, which will be a very long time, and until real wages rise in China, I just— that to me is like the first order explanation of why there's less industrialization in a place like Africa, and it's not this other set of forces that's talked about.

Paul Romer's charter cities, would you want to try them anywhere? Can they work?

Paul Romer and I had long debates, some in private, some in public about this. I always worried about the fact that there was some risk, maybe a large risk that these would be the Cabrini-Greens, like the housing projects of the 21st century. So there was some risk that they would go awry. So it's just an incredibly— I just don't like— I'm very suspicious of high high-risk development strategies, especially when you're playing with someone else's life. It just struck me as there's a lot of ways for that to go dramatically wrong. And so an attractive feature of a lot of development strategies is maybe low downside risk.

As you know, if you think about Africa, parts of it are quite underpopulated. You could find an area that had not so many people in it, set up a charter city, let people choose to migrate inwards if they wanted to. Other people could leave, but it wouldn't cover that many people. And if continuity of rule is a binding constraint that makes you more bullish about Ethiopia than other places, why not import some of that? I mean, what's the risk really?

Right. So a depopulated coastal area close to trade networks that has the potential to be like a Singapore as an experiment, that strikes me as having— that's a fair point. That wasn't typically what he was The concrete things that they were looking to do looked less like that than actually trying to take much more established settlements and try to turn them into charter cities. So I guess I wonder about the practical feasibility of redoing the Singapore experiment of building a city in a swamp because they didn't ever seriously seem to have a lead on doing that.

What's it like co-authoring with your spouse? Passes are allowed, of course. No.

So I wish we could keep doing it. We went off at certain points with slightly different interests. Well, we were co-authors I guess before we were spouses, so it's a good selection mechanism. I'm not sure what we would do if we were both economists. I think it might have been too close. I think it's because she was— She's a political scientist. She's a psychologist. A psychologist. A practicing psychologist. And so it's very useful for that other person to be a complete expert in their domain and so you can— Essentially defer on all of those decisions to them on that and vice versa. So you're complementing, not competing in the same space.

You mentioned on your homepage when you were younger, you worked 2 years, quote, cooking chicken at a vaguely militant KFC outlet. What did you learn doing that?

So I had a boss who sort of literally ran the place like a drill sergeant. He was this hulking guy who had a big—

This is in Canada now?

This is in Canada. Yes. So the KFCs in Canada are a little— they're run by a different company. They're a little cleaner and nicer. Not much, but a little cleaner and nicer than the US. This big hulking guy with a Hulk Hogan style mustache and would do the white glove test in the kitchen on the ceiling under the machines and you sort of had a— so I guess I'd look at that as sort of taking this— a kind of lazy teenager who hadn't worked really hard before and getting them to do gruesome things and sort of realizing that you can grit your teeth and do some disgusting tasks and get through it. And that was maybe a useful lesson. It either selected me— it demonstrated my perseverance to me or created my perseverance. I don't know which one.

What does the University of Chicago do right that's especially important?

I think the number one reason I was excited to move to Chicago and I— when my wife and I were deciding whether we wanted to be, the one thing that was most appealing to it is that it has this intellectual atmosphere of people are just constantly talking about ideas. And you would think you could take this for granted in any university environment. So I— but I, you know, I was living in New Haven, in the small community around Yale, and in New York, in the small community just surrounding Columbia, which were tremendous places full of brilliant people, but very seldom were people always sort of pushing themselves intellectually and pushing you intellectually in conversation. And now how they've created that culture, I don't know. I haven't been there long enough to guess, but the fact that they've created that and it persists is to me the most exciting thing about it. It's what I thought every academic atmosphere was going to be before I started my PhD, and I was surprised that it doesn't really seem to exist in any places. I think it survives here a little bit at George Mason, but maybe not too many other places.

What's your favorite Australian TV show?

It's a comedy called Please Like Me, which is a comedy about mental illness and struggle by a young gay comedian in Sydney, and it doesn't sound like the most promising premise for a television show, but it's genius. Maybe it just appeals to my— I do think those of us in the Commonwealth share a certain kind of sense of humor that's different than the American sense of humor, but I think it travels well.

Final question, and I ask this of many guests. It has to do with what I call the Chris Blattman production function. So you're still young, you're at one of the very best schools in the world, some would say the best. You're famous. People ask me about you all the time. You're very active on social media. You have a huge stack of very well-cited papers. You're writing now a major book on war and violent conflict. And if someone was asking you, like, what is the insight you have into how to do this or how you did it? That maybe other people wouldn't immediately see other than just, "Well, I worked hard," or, "I went to a good school." What is in the Chris Blattman production function that maybe we don't see?

So one part of that, so there's an answer to what I think my production function looks like, but the answer for most people, when I was in my maybe third year and you do something, as you know, in academia called your third year review where your senior colleagues debate what you're doing right or wrong and then give you basically no feedback. But I persistently— I plowed into a senior colleague, a guy named Ian Shapiro, who was sort of a normative political theorist at Yale, to give me the feedback. He refused to tell me and refused to tell me. He said, "Well, you just need to do what you're interested in. You're clearly doing something you're interested in. You're excited by it. I'm not going to tell you what other people think you should do because it's different and that's not the secret to your happiness or to your success." So on some level, my production function was always doing the thing that really interested me regardless of whether or not the people around me told me it was a good idea or not. And in my case, that just happens to have been deciding that I like going to a place and spending a lot of time there. I think I've done it 3 times. First I did it in Uganda and then I did it in Liberia and now I'm doing it in Colombia. And each time I just went somewhere and I spent maybe 6 months just hanging out and talking to people and figuring out how I think things work in a very qualitative way. And then usually it was studying something where there was no data on it. And that's been my shtick. Like, most people look for existing data to analyze, and I sort of look for data I can create. And so then I would go and I would find a way to create data. And then having created data, I would try to answer How do I answer this question that interested me? In my case, it's usually something that sort of is the intersection of economics and political science and sociology and public policy. But then often I would try to find an experiment that I could run just to sort of have an excuse to be there. It usually wasn't a very good experiment, but it was evaluating something that somebody was doing that was obviously common, so it was useful to have some data on it. But then my strategy in every place is just to go around and— Try to find some community organization or bureaucrat or somebody who has a good idea, who's a true believer, who really believes in whatever cockamamie thing they're doing to make the world a better place. Most people, least of all me, don't really believe it should work, but then there seems to be something going on and if it works, it challenges how we think about the world. The first time it was cash, just giving cash to poor people, and the second time in Liberia, it was somebody doing something that turned out to be— cognitive behavior therapy that was trying to take adults and just have them— these criminal adults and totally turn their lives around. In Colombia, we've just stumbled upon this little group that's basically trying to take back governance in their communities from the gangs nonviolently just by relentlessly organizing their community and connecting it to the city government and then trying to scale that up and test it and see if it actually works and then usually— As a result, I just changed how I think about not just that question, but the whole business of violence and development.

I recommend that you all follow Chris Blattman on Twitter. And Chris, thank you very much. Thank you. Thanks for listening to Conversations with Tyler Cowen. You can subscribe to the podcast in iTunes, Stitcher, or your favorite podcast app. And if you like this podcast, please consider rating it on iTunes and leaving a review. This helps other people find the show.