Esther Duflo on Management, Growth, and Research in Action (Ep. 82)
Esther Duflo’s advice to students? Spend time in the field. “It’s only through this exposure that you can learn how wrong most of your intuitions are and preconceptions are,” she explains. For Duflo, it was time spent in the Soviet Union on the brink of collapse. While there she saw how Jeff Sachs used the tools of economics to advise policymakers on matters of crucial importance. To her it seemed like the best job in the world — and she began to pursue it in earnest. Now it is she who is advising governments on how best to reduce poverty, having co-founded one of the leading policy research centers in the world. That work, together with that of frequent collaborators Abhijit Banerjee and Michael Kremer, has now been honored with the Nobel Prize.
She joined Tyler to discuss that work, including how coaching increases the effectiveness of cash transfers, why she cautions against falling in love with growth rates, what France gets right about child-rearing, the management philosophy behind her success building J-PAL, how she briefly became the face of an anti-Soviet revolution, the under-looked reasons behind the decline of geographic mobility in the United States, what rock climbing can teach us about being a good empirical economist, her daily musical move from Bach to Bob Dylan, and more.
Want to support the show? Visit conversationswithtyler.com/donate. Your gift helps create enhanced transcripts like this one.
Listen to the full conversation
Read the full transcript
TYLER COWEN: Today I am very honored to be here with Esther Duflo, who recently has won a Nobel Prize in economics, the youngest economics winner ever. And today, our date of recording, is also, I believe, the publication date for her new book with Abhijit Banerjee, and that is called Good Economics for Hard Times. Esther, welcome.
ESTHER DUFLO: Thank you very much. Thank you for having me.
COWEN: My first question has to do with what, I think, is maybe your most important paper. And that’s the 2015 paper in Science with Abhijit, Dean Karlan, and some other coauthors. There you show that a cash transfer to the very poor, combined with training and coaching, has super-high rates of return. Why don’t people study that more? It seems to be the most potent intervention we have to fight poverty, other than migration.
DUFLO: I think it has been studied a lot. It has also been adopted as policy after this original study and other studies. The original project comes from Bangladesh, and there actually has been a study and a long-run study in Bangladesh by Robin Burgess and others who are now looking at households 10 years hence. And you continue to see large differences between the households who got the transfer.
And they have a recent paper that they just released to say that they continue to study it, where they are really putting a lot of theory into it to look at the data through the lens of asking, actually, is there a poverty trap in which some of these households have fallen?
They show that, in fact, there is such a thing with a threshold of wealth above which — if households can go above — they can grow to a point where they are significantly better off. And if they are below it, they go back to wherever they came from. I think that’s a very important paper that is making a big splash and show people continue.
In our case, the part of the study that I was most involved with was in West Bengal. We also continued to study them after 10 years. We are studying their children as well. We are studying the impact on migration to go through not just, “Oh, wait, there is an effect,” but how all of these effects are building on each other.
COWEN: As I understand the paper, there’s a cost-benefit ratio of 133 percent to 433 percent across six countries. That’s enormous. What exactly about the mentor makes the difference as opposed to just giving people cash? What is the mentor or the coach actually teaching?
DUFLO: Just from this particular experiment, it would be hard to tell because one could say, “Well, you should try to do cash only, and then we can see.” But subsequently Abhijit and Dean Karlan were involved in an experiment in Ghana where they just gave goats. They call it a goat drop paper. Just goats were given, without the coaching. And there, they find very clearly that the coaching makes a big difference, that people who you give just assets to have more assets, but it doesn’t serve as a springboard for more activities.
So descriptively, what the coach does is two things. Number one is avoid the temptation of liquidating the assets quite early. If you give people cows and they don’t feel confident using them, then the easiest thing to do is to sell them. And then they have cash, but they don’t have the productive asset anymore.
The second thing is to provide the complementary human capital for taking care of your assets. For example, when people are given cash — some people are given cash to start a petit business — but have no idea even how to go to the market because the poorest person in a village is also excluded from productive activities. They have not really been working. Or if they have been working, it’s in very local circumstances. They really have no idea.
So the coaches are physically taking them to the market. This is how you take the bus, this is where you buy the trinkets that you’re going to sell in the village, and this is how you bargain for your trinkets and then come back with the trinkets and sell them. So there is an amount of technical skills, if you will.
The third component, which is harder to quantify but probably very important as well, is confidence training. They are meeting in groups. Those are people who have not been involved in productive activities, often are living off alms or very petit works. They are making them confident that they can do it.
COWEN: How scalable is the coaching in your opinion?
DUFLO: That’s a good question, to which we should have an answer soonish because there is now scale-up of these programs in various countries, and in particular in India. Some state governments in India are scaling this up as part of their program.
So they are mainstreaming it as part of their program, still with the support of the original microfinance institutions that have been running this program, but hiring many, many, many people. So you will be able to see whether you could do this with the best NGO worker of the country, and you only have a hundred of those, and then that’s done. Or whether this is something that can be spread.
What is encouraging in this respect is that in Bangladesh, the program is huge as it is. It’s enormous. It reaches hundreds of thousands of families as it is. So to some extent, we already have the answer to how scalable it is because it’s already scaled.
And even the evaluation by Robin Burgess and others was done on a quite large scale.
COWEN: Have you ever thought you should just spend the rest of your career working on this intervention, and 5 percent of the economics profession should just work on this, with possible rates of return so high for the world’s biggest problem, which is poverty, right? Why don’t we do much, much more of working on this problem, on this study?
DUFLO: One should certainly continue to do more working on this study and working on the questions you were asking about the scale-up and working on how various modalities of it want. Already, the study you mention in Science was already a sign that there is at least some interest in the economics profession to focus on this question because it was the second paper on this. The first paper was the evaluation of the mothership program in Bangladesh.
People didn’t say, “Oh, that’s a nice program. Let’s go and study something new.” Instead, Dean Karlan led a group of people to replicate this same evaluation in seven different countries, which was a very major undertaking involving lots and lots of researchers and NGOs. So the fact that people were willing to do that, to write a 12-page paper in Science at the end of it, suggests that there is already recognition that it’s a very important topic.
Now, should everybody do that? I guess there are probably diminishing returns to effort on the particular problems, and there are other interesting problems to look at. So I’m glad that people continue working on this. And I think the work is only going to deepen, both in understanding the mechanism better, understanding variations around it, and understanding how to make it adopted as policy — because I completely agree with you that of everything that I’ve done, studied in my life, that this is one with the most important, spectacular results.
On Microcredit in Hyderabad
COWEN: You have some well-known papers with coauthors about microcredit programs in Hyderabad, India, and I’ve long had some questions about how those programs actually work.
As you know, often the borrowing families are paying annualized rates of 50 to 100 percent a year. And they keep on borrowing. They’re not just borrowing once to bring the kid to the doctor. They’re in a steady borrowing program. And at the same time, they’re not just drained of all their wealth. Does that imply there are rates of return on investment in those communities of 50 to 100 percent a year? And if that’s true, why aren’t the growth rates higher? Do you get my question? How does the whole thing fit together?
DUFLO: Well, it actually does not imply that, because the plurality of these households, maybe the majority of these households, does not actually invest in any productive activity with the money, so most of the interest are paid from people’s regular jobs.
COWEN: But don’t they just get drained dry? If I borrowed every year at 80 percent —
DUFLO: Oh, it’s not the same households that keep borrowing and borrowing and borrowing. In fact, the number of repeat customers is less than you might think.
People join microfinance, take a loan, buy something. It can be a durable good, for example, a fridge or a cycle, and then use their regular income to repay it. For some households that’s a very, very expensive way to save in some sense because what they’re looking for is accumulating enough money to buy this durable good that they want for the household. But it’s very difficult to save because there are no saving instruments and there are a lot of temptations, et cetera.
So the microcredit officer serves as a commitment once you borrow the money. The incentive of the credit officer is very strongly to make sure that everybody reimburses. That will create a way for you to get your assets and then progressively reimburse. That seems kind of a funny way to save, but I think it’s a reasonable response in the environment.
And then there are some people who are borrowing for business. We have a recent paper where we look at the same setting again, but now eight years after microfinance was introduced, and in fact, two years after it’s gone because there was a huge crisis of microfinance. We look at the group of people who had a business before microcredit even got in, so they are people who actually borrowed to invest in their business often. And for those guys, we see that some of these people are actually at very, very high rates.
So there are some people with very high rates of return, as you were saying, 50 percent or 60 percent. But there are not very many. I think that is why the . . . First of all, there are not very many, and they are not necessarily the one that have access to money. That’s one of the big reasons why the cost is lower, is that there is a mismatch between the investment opportunity and who has the money. And that’s what my colleagues call misallocation.
COWEN: Your randomized control trials — they’re very large. They employ large numbers of people. I even once visited the Hyderabad project, was very impressed how much was going on. Do you have a personal philosophy of management or leadership? Esther Duflo the manager, not the economist — tell us a little bit about her.
DUFLO: I kind of stumbled onto this because, as most academics, I was not born to be a manager. I didn’t bring a premade philosophy to this business, but I developed one, and it has several elements to it. The key is that I’m not a micromanager. The key is that you want to give people a pretty clear vision of where we’re heading and then give them a lot of ownership for how we are getting there.
That is true for the whole organization of J-PAL, which is the sum total, the network of all these researchers, where from the very beginning, with the first executive director, Rachel Glennerster — it was really her thing to lead, and she put her mark to it. And we were behind as kind of enablers. That’s true at that level, and it’s still true today, I think.
And it’s true at the smaller level of the projects where everybody has a piece that they own, and they run with it. They’ll make mistakes, and then we have to correct the mistakes. But the ownership of the piece of the project means that you get a lot of effort and a lot of creativity of everyone.
And then, I must say that I’m in a very good business — no credit to me — because the people who get attracted to this work tend to be very committed and passion driven, so you get very good material to start with. You just need to give them a little bit of guidance, and they move.
COWEN: And which is the input that’s the factor of production that prevents you from scaling your work more?
DUFLO: Well, I think it has scaled quite a bit. If you took it —
COWEN: But you’re not doing 5x of what you’re doing, right? Something stops you. It could be money. It could be talent.
DUFLO: I always try to be behind the money. I think a lot of problems of some centers — sometimes you get some money to do something, and then you have to have the idea to match the money. We always had some idea before, and the money comes. So at some level, the money is always the constraining factor at any point in time. You could say, “Why don’t you get more talent to get more money?” But there are some limits to that.
Another thing is, I’m not sure I’m aiming at world conquest in the sense that I’m more interested in — and I think that that very much applied to the three of us — we were very much interested in creating an ecosystem where that kind of work becomes easier because policymakers are interested, funders are interested, talent is interested, and because there are some things that are eased.
When we started even the Hyderabad project, we were still learning as we went, and we made mistakes. We’re still making mistakes but fewer now because there is an infrastructure and some learnings. This infrastructure can then be given to other, younger people that have more imaginative great ideas to run with. I’ve always been more interested in doing that and using my own work as a little bit of a testing ground, trying new things that are quite risky because it doesn’t matter to me if that fails. I have tenure.
COWEN: Do you worry much that the randomized control trial [RCT] method centralizes authority in too few institutions? You need a certain amount of money. You need some managerial ability. You need connections abroad. It’s not like running regressions — everyone can do it on their PC. Is that, in some way, going to slow down science? You get more reliable results, but there’s much less competition of ideas, it seems.
DUFLO: I think it would be the case if we had not been mindful of this problem from the beginning. And it might still be the case to some extent. But I actually think that we’ve put a lot of effort in avoiding it to be the case.
When you take an organization like J-PAL, just in India we have 200 staff members. And we have, at any given time, 1,000 people running surveys. I say we, but these people are not running my project. These people are running the projects of dozens and dozens of researchers. When I started, I couldn’t have started without having the backing of my team because it was such a risky proposition that you needed to be able to easy risk capital kind of things.
But at this point, because of the infrastructure, it’s much more normal sense. People can get in with no funding of their own, in part because one of the things we are doing as a network is raising a lot of money to redistribute to other people widely. J-PAL has 400 researchers that are affiliated to it, or invited researchers, many of them quite, quite junior.
So that sort of mixture — it was very important to us, and I think we’ve been quite successful at making the tool marginally available. It’s never going to be like running a regression from your computer. But my philosophy is that if you have the drive and you’re willing to put in your own sweat equity, you can do it. And our students and many other students who are not at top institutions are doing it.
COWEN: In your book, the new book, Good Economics for Hard Times, you talked quite a bit about a revolt against the elites in the United States. But what if I say, “Well, the elites are underrated. They’re pretty good. Who else runs RCTs other than the elites?” What’s your response? Do you side with the revolt or do you side with the elites?
DUFLO: First of all, I think that it’s good if it’s not just the elites who are running RCTs, so back to my point of trying to make it a general tool that’s the elite of the academic movement. But you could say, “Well they are still academics, so in the broad scheme of things they are still elite.” My problem is not with the elites. My problem is with the attitude of the elites with respect to everyone else.
I don’t even know what elite means in this context. But if us economists, as a profession, spend more time trying to understand what are the problems of people who do not have, say, the comfort of an academic career behind them, then we can make progress in understanding the issues and therefore, potentially, thinking through policies that might help solve these issues.
But if us economists, as a profession, spend more time trying to understand what are the problems of people who do not have, say, the comfort of an academic career behind them, then we can make progress in understanding the issues and therefore, potentially, thinking through policies that might help solve these issues.
It would be a bit hypocritical of me to be against the elites given that I am the elite [laughs] in some sense from when I was born. But I think it comes with some amount of . . . It’s the type of problem that you choose to put your attention on that matters.
COWEN: In your new book, you seem, at times, quite skeptical about how well economists can understand what drives economic growth. But what if I just looked at a few cases? I looked at say, Singapore, Ireland, Poland, which have done remarkably well in the last few decades. And in each case, you had leaders who more or less followed the prescriptions of economics. They had the will or the ability to do so.
I watched all of those experiments unfold. I felt they were going to do pretty well. They did extremely well, actually, better than I thought. Don’t we, in fact, have a pretty good idea of what does drive growth? It might be different for each country. But you go to a country, you look at it, you have a good sense. Many countries have done it, and it’s worked. Why is that wrong?
DUFLO: Then I could give you South Korea, and it did pretty well, too, with an entirely different set of policies.
COWEN: Oh, sure, policies can be different. But given a country, it seems the leaders can know, and if they have the will to do it, it happens.
DUFLO: Well, if the policies are very different from one country to the other and are a complicated function of the context . . . In Ireland, you need very low tax rates, but in South Korea, you need industrial policy. That a little bit reflects our difficulty as economists to have come up with guidance for the next place.
Take — I don’t know — Ethiopia, and they want to grow. The question becomes, do we have anything in the experience of the past country that will help us for sure predict the set of steps that Ethiopia needs to take that is going to lead them to grow? My firm belief is that we don’t. But if you do, then I think you have a side gig in helping Ethiopia, and that would be great.
COWEN: But Ethiopia has grown at about 10 percent for 10 years running. They invest about 38 percent of their GDP. It may collapse for political reasons. But if they stay on the track that they’re on, isn’t that further evidence that they figured out what they need to do to grow? They had the will to implement it. So far, it’s working, politics allowing.
DUFLO: That’s why I picked Ethiopia, because in a sense, they have. Again, they have their own path. But the thing is, their own path becomes so particular. I’m not saying that good policymaking is not feasible because we have lots of examples of good policymaking and lots of examples of bad policymaking, actually. But the good policymaking is so varied that at the scientific level, it becomes very quickly “Good things are good.”
Because the path of Ethiopia — even the steps they had to take to carve out a place in international trade, for example, is very different from what a country like China had to do, just by virtue of China existing, et cetera, et cetera. I’m certainly not saying that good policymaking is not feasible. I’m just saying that we haven’t found — it seems to me, empirically — we haven’t found a series of prescriptions that we could give to countries, saying, “If you do this thing, it’s going to work out.”
Similarly, now growth is slowing down in India. I don’t think . . . at least, I wouldn’t know what to tell them to do, what they are supposed to do now. I don’t think they themselves know. So that’s what has turned out to be difficult empirically.
This is not me, actually. I’m not a macroeconomist, so I would not have the arrogance to say that we haven’t figured it out. But that seems to be what the macroeconomists have . . . the conclusion they’ve come to is they haven’t gotten a set of recommendations, even context-dependent recommendations, that they can run with.
COWEN: As you well know, there’s a number of papers that suggest human capital is a factor of overriding importance for economic growth. I think Mankiw, Weil, and Romer attribute 80 percent of growth to human capital. What’s your view of those theories?
DUFLO: It seems as a gross accounting matter — which is, what’s the correlation between how much of the growth is explained by differences in human capital across country? — it does seem to be that human capital plays a big role, even though Mankiw, Weil, and Romer figures have gone down since then with the work of Khazali, for example. But it is important.
Some of it is correlation, rather than cause and effect, that countries that have been good at increasing education have been good at doing other things. Also, countries that have grown fast also generate an incentive to invest in education. That’s an old paper by Bils and Klenow that nicely makes this point. But the bottom line is that I’m totally ready to believe that human capital, broadly defined — education in particular — plays a big role in growth.
To tell you the truth, even if it didn’t, it seems it’s not a big bet to invest in education anyway because worse comes to worst, people would be educated, which seems to be a good end in itself. It’s not just a means to an end. I will totally agree with you that human capital is super important.
In fact, it is what I think was the original impulse of Michael Kremer when he worked on the first RCTs. He was coming from growth, and he thought, “Wow, human capital plays a big role in growth.”
Now, what’s human capital? Once you said education is important, you have to start unpacking this question. Is it years of education? Well, not really. Clearly if the kids learn nothing, that’s not so useful, so it has to be learning. Then you can start asking yourself questions on whether it’s cognitive skills or noncognitive skills or a mix of the two. Heckman is telling us that noncognitive skills is such a very important part of human capital.
And then you want to ask yourself the question, how do cognitive skills get produced? How do noncognitive skills get produced? And that’s what willy-nilly lands you to much more specific policy advice and hence, potentially, things that you can start running RCT on.
COWEN: To return to what I think was your master’s thesis, how well did the USSR’s first five-year plan go?
DUFLO: This was actually my master’s thesis in history. And I was interested in the time in propaganda, not so much the first five-year plan per se. But how over this period the talk and the use in propaganda of the big construction sites evolved.
This is also the period where Stalin consolidated power. At the beginning there were a lot of ways that the Soviet Union could have gone. It could have gone a more market-friendly way, which is what Lenin was hoping at some point. It was going a completely utopian everybody-live-together kind of crazy way. And then there was the Stalin thing, which was finally the form that they adopted, which was boring, inefficient materialism. All of this played out in how they represented their construction sites, like the big dam on the Dnieper.
So this is how those economics and the narrative of economics were used to consolidate power and to consolidate grip on society.
COWEN: You didn’t like economics at first, I read in one interview. Why not? What happened? What soured you on it?
DUFLO: I was exposed to the typical econ 101 course, the French version of it, with the indifference curves and all that jazz. I thought it had no relevance, that it was so far removed from any real person’s life that it could not possibly be useful.
I also was skeptical that it was extremely ideological. That if you want to show that minimum wage is bad, then it’s very easy to write a supply-demand curve and the horizontal line. Low and behold, minimum wage is bad, and redistribution perverts incentives. I thought the conclusions were a little bit too directly coming from the assumptions that were put in.
I had only done economics initially because I wanted to do economic history. And then I thought, “Okay, I’m done with this whole economics business.” But at the same time, I was a little bit dissatisfied with history for different reasons, in particular because I thought it was not very immediately useful.
So that’s why I decided to go to Russia under the pretext of writing my master’s thesis, but really to get some exposure to the world. This is where I saw economists in action because I was the research assistant of the research assistant of Jeff Sachs, whom I never met in this occasion, but I was hustling for them.
And there I saw I didn’t necessarily agree with everything they were doing, but I saw that they had the luxury to think about things and then to be heard by policymakers when they had something to say. And I said, “Wow, this is the best job in the world.” And I still think that.
On living in Russia
COWEN: On the internet, there’s a photo of a teenage Esther Duflo — at least it looks like you — protesting against fascism in Russia on top of a tank, is it?
DUFLO: That was a bus, and it was me. It was me. So that was in 1991. This was not when I lived for one year there. I lived one year in ’93–’94. But this was in ’91. I had gone to Russia about every year since I was a teen to learn Russian. I happened to be there the summer where there was this putsch against Gorbachev. That summer.
COWEN: That’s the year my wife left as a refugee.
DUFLO: We actually watched TV. And I do remember those GKChP they were called, like fascist Communists, old, half-decrepit, half-dead, making the coup. We were a little bit shocked. With my friends, we decided that we would go out in town to see what’s happening. We went toward the White House, and we couldn’t see very well because there was a lot of people.
So we saw this trolley, and I said, “Let’s go up. We are going to have a good view of what’s going on.” And we went to the top of the trolley. It was amazing because there was an immediate fraternization of the troops and the people. In retrospect, it was not a very reasonable thing to do. But at the time it sounded like a completely happy moment. There was a truck that was delivering bread. It was the end of Communism.
And someone gave me that fashizm ne poletit placard and asked me to hold it. And I’m like, “Sure, I’m going to hold it.” So I’m holding my placard. We stayed there for a long time when things were happening. Next time I saw in the evening, my parents called me, “What are you doing?” Because it turned out that that image was on all the TVs in the world. [laughs] And that’s how I very briefly became the face of this revolution.
That photo was then in a book. The guy never knew who I was. Recently, he put two and two together.
COWEN: Other than missing your family, what was the hardest thing about living in Russia for a year?
DUFLO: Oh, Russia was extremely, extremely hard place to live in that time. It was extremely violent, just the day to day. It was the end of Communism, but the mafia was out there replacing it. Some people were really extremely poor. They were given these vouchers, but they quickly exchanged them.
You saw this woman queuing near the subway with one fish to sell, and the whole thing was so hard. Everything was hard. It was a difficult environment. I assume it still is, but I haven’t really gone back since then.
COWEN: Do you have a favorite Russian novel?
DUFLO: My favorite Russian author — not very original — is Dostoevsky. Crime and Punishment probably.
On French education
COWEN: Why is the French education system so good? Or would you challenge the premise?
DUFLO: The French education system is very good at producing a very small elite of very educated people, and at reproducing it.
COWEN: But productivity at the median is quite high in France, right?
DUFLO: At the median of the productivity of the median worker?
COWEN: Yes, so it’s not just training an elite.
DUFLO: Yes, although I wonder whether, if we look in a generation on the current kids who are in school, will we find the same thing? Because I’ll tell you exactly why: the current PISA and TIMSS results show France exactly at the median in terms of the scores, but with a lot of heterogeneity, with excellent results at the top and really bad results at the bottom.
And the second thing that these test scores also show — I think it’s the TIMSS — the math scores are the strongest correlation of the OECD country between your socioeconomic class and your score. This is the place where being in a bottom quintile of socioeconomic class, parental income, is the most predictive of being in the bottom quintile of the math performance.
So I do think we, historically, have an excellent system. I do think it has roots in 19th century France and the Republic of the Teachers and the both inclusive and demanding system at once.
I think to some extent, I’m the product of that. But I also think that we’ve a little bit let it crumble. I do hope that we can recover it.
COWEN: Does child-rearing in France strike you as more sensible than child-rearing in the United States?
DUFLO: Oh very much so, very much so.
COWEN: And why?
DUFLO: You know that book, Bringing Up Bébé?
DUFLO: I think she picked up on something which rings so true to me, which maybe is a marginal point about the US versus France. In France people are reasonably content to just go with the flow and do what everybody does. Every kid eats the same thing at 4:30, has dinner at the same time, has gone through the same experiences, learned the same songs, and everybody thinks they are totally free. But in fact, they are all on this pretty sensible railroad. And also, they don’t agonize about it.
In the US, child-rearing is one more occasion to make a statement about your identity. You’re the kind of mother that carries the baby, or you’re the kind of mother that puts the baby in a stroller. And somehow it almost can predict what you’re going to think about Donald Trump. That’s crazy. Some people are so concerned about what they do. Not only they feel that they have to invest a ton in their children, and they feel inadequate if they are not able to, but also, exactly what they do creates them as people.
In France that’s not there, and I think that makes everybody so much more laid back, children and adults.
In the US, child-rearing is one more occasion to make a statement about your identity. You’re the kind of mother that carries the baby, or you’re the kind of mother that puts the baby in a stroller. And somehow it almost can predict what you’re going to think about Donald Trump. That’s crazy. Some people are so concerned about what they do. Not only they feel that they have to invest a ton in their children, and they feel inadequate if they are not able to, but also, exactly what they do creates them as people.
COWEN: Why are the biggest and best French companies all so old? I think they all predate 1980 or so.
DUFLO: That’s a little bit above my pay grade. Not that your other questions were not above my pay grade.
COWEN: [laughs] We’re pay grade elevators at Conversations with Tyler.
DUFLO: Yes, I appreciate that. But that might be above my pay grade even with your elevator. What economists seem to say about that — and I’m just going to parrot, say, someone like Augustin Landier — the idea that in the US, failing is quite all right; bankruptcy is not really a problem. Bankruptcy laws are reasonably easy to navigate, and there is not much stigma associated with failing, and so there is more churn in the system.
That produces a bunch of horrible companies, but those guys — we don’t see them. And that produces some good ones.
The French system is less encouraging of failing, and that is less conducive to lots of thousand flowers blooming. Hence, there is no new things. But I’m repeating something which is not deeply original and has some ring of truth. But honestly, I’m not going to claim expertise there.
On French cinema
COWEN: What is important to you from French cinema?
DUFLO: Oh, so many things. That’s not above my pay grade, and I am so much inside my pay grade.
I grew up watching movies. The French system is such that in the first two years of undergraduate, you work like a mad person, and then you can pretty much take vacations for the rest of your life. I decided against the rest-of-my-life thing after going to Russia and all that. In between I had this one year where I used to go to the cinema about three or four times a week, so I had a chance to get groomed in the American classics and French classics.
COWEN: And what do you recommend to us?
COWEN: For me, it’s Godard and Bresson.
DUFLO: Yes, I would also put Bresson there. It’s just unbelievable cinema.
On growth rates
COWEN: On the wall of your home in Boston, according to one newspaper, is a slogan from the 1968 Paris Riots. And it reads, “Nobody ever fell in love with the growth rate.”
DUFLO: It’s not on the wall of my house in Boston, but it’s in a book of quotes that were on the wall of Paris. It’s a little red book that was in my parents’ bookshelf, which I read as a kid. I kept in my mind that I knew we had that and was looking for it. It could come in handy at some point. It has a lot of these quotes from 1968, some of them more well-known than others, like “Sous les paves, la plage.” It’s one of the most well-known. “Under the pavement you have the sea.”
And this one, which I found when we were working on the book, and I wanted to . . . It was kind of the epigraph of one of the first drafts of the chapter on growth. Then somehow, it didn’t make it in the final cut.
COWEN: And you think we should fall in love with growth rates.
DUFLO: No, I don’t think one —
COWEN: Why not? The power of compound returns is remarkable, right?
DUFLO: I think one should have a healthy respect for growth rates and treat them as useful companions and people that you have to make work for you. I think we should think of growth rate as chief of staff, not something I think we should fall in love with.
I think one should have a healthy respect for growth rates and treat them as useful companions and people that you have to make work for you. I think we should think of growth rate as chief of staff, not something I think we should fall in love with.
And I think one issue the countries have had when they’ve fallen in love with growth rates per se, as opposed to why is it you want growth in the first place, is that going back to the early conversation, it’s pretty hard to manipulate growth rates for politicians, especially in fluctuation growth rates. Therefore, if things are not going your way, and you try more and more crazy things to try to make growth come back because that’s what you’ve fallen in love with, then you can put yourself in pretty bad corners.
One example is Japan. I think they did fall in love with the growth rate instead of trying to think, “Maybe there is a reason why the growth rate has fallen down. We should be comfortable with that and do the best with what we have.”
COWEN: Could Japan have done better? It grew very rapidly. It caught up, basically, to France, then stopped at a certain frontier short of the United States. It’s managed its decline really quite well in per capita terms. Its growth rate is okay, not spectacular. Social order is at a premium. What have they done wrong?
DUFLO: Well, they’ve built themselves a pretty big debt level, which could become a debt crisis in the making, which was entirely unnecessary because when growth fell down in Japan — in retrospect, of course, you always have 20/20 hindsight — but when growth fell down in Japan, there was a reason for that, which is the population was aging. The working population, as a share of the overall population, was therefore becoming smaller.
So it is not surprising that there would be a slowdown in the growth of GDP per capita. And for a while it was compensated by the fact that there was a lot of TFP growth. But that wasn’t necessarily going to come forever because we don’t really know where TFP is coming from.
At this point, they could have said let’s do exactly what you said they should do: “We should manage expectations for the downfall. We’re already quite rich, and we have good standard of living, which we could make, in a sense, even better by addressing some of the issues that still remain in Japanese society.”
They did some of that, but on top of that they also said, “Let’s try and make growth come back.” There was this sort of Keynesian spending spree that I don’t think did have an impact on growth but put them in a not particularly desirable macroeconomic situation at the moment.
COWEN: Why do you think, in the United States, geographic mobility has fallen so much? You discuss that at length in the new book.
DUFLO: I think there are a number of reasons. Some are strictly economic, as in financial. And some are a little bit broader. On the financial side, real estate is a big part of it. That has been documented by other people.
COWEN: But you can move to the South, right? There’s freedom to build in the South. Plenty of people move to Texas. You won’t move to the Bay Area, but it’s a big country.
DUFLO: Exactly so. That explains why people aren’t moving where a lot of the jobs are, if you will. But you are right. People could decide to move 200 kilometers away, not to a city but to another place that happens not to be shocked by a shock.
Now there is the issue that places where you would expect most people to move from are places which have been hit by economic shock, which is often clustered in each area because industry is clustered. When a region is hit by a shock, that hits not just one company when everybody else is doing fine, but all the companies because everybody was doing furniture or clothing, or everyone was doing something that is being replaced by robots or something like that.
You have an equilibrium fact that the places that are affected by the shock get hurt. Then the places that are not directly affected by the shock get hurt, too, because the guys and girls, actually, who were working are less likely to buy McDonald’s, and therefore, all the McDonald’s business goes down, et cetera, which then leads to a decline in the real estate prices.
People find themselves underwater on their market. I think that prevents mobility because at least they have a house to live in if they don’t go. But if they sell that house, they have nothing to get themselves another place. So that’s the purely economic factor.
Then on top of that, I think childcare is an issue because a lot of people have childcare provided by their extended family. You cannot move the entire family, again, maybe due to this list of things. The childcare options are pretty bad in this country in general, especially for small kids. Once they start going to school it’s a bit better.
So those are some of the economic reasons. Then I also think there is one social reason, which is, people like to be part of a social network. I think it’s profoundly human. We are not made to be lonely. We are meant to live with our friends and derive some amount of meaning from our place in the social network. And moving makes it difficult, especially if you’re going to move to something that’s completely unknown, and you don’t exactly know what you’re moving to and why.
Then finally is . . . and that’s one thing that we think we should start to put a little more in our economic model, is the fact that people are . . . It seems in surveys, people keep declaring one thing: to derive meaning from their job. But they don’t always do. And they are more likely to be deriving meaning from their job if they’ve held this job for a long time and if they’ve grown in expertise and in ranks in doing this particular job.
So here, the lack of geographic mobility is coupled with the lack of sector mobility, where people aren’t just not particularly excited, or not excited at all, to leave their job making furniture to take another job 200 kilometers away being a security guard in a furniture company.
COWEN: Why is any of that so different from the 1980s when mobility was quite high? Even for people without children, mobility now is much lower than it was in the ’80s. What’s the difference?
DUFLO: It’s about half now what it was in 1948. The decline started in the ’90s.
COWEN: Yes. So it’s very high in the ’80s, and then something happens.
DUFLO: And then something happens.
COWEN: And the ’90s are good years, right? You would think mobility would be rising in the ’90s. Now there’s the internet. You find jobs through the internet. It’s counterintuitive that mobility has fallen so much.
DUFLO: Not only that, but it’s also the years where you start having a lot of geographically concentrated shocks because the ’90s and 2000s is when trade with China picks up. It’s where entire towns — it would seem that they should go and pick up. Precisely at the moment where the factors you point out would have made geographic mobility easier and there was also more of a need for people to move — this is when we see people move less.
We don’t have the full answer to this question. But I think it’s a little bit correlated to this meaningful-jobs question. This is also the time where the geographic movement would be not associated with finding a job in another furniture factory 200 kilometers away, but would be finding a job in a completely different sector where you would be starting from scratch, and you’re taking a job you think is demeaning.
COWEN: In your recent New York Times op-ed, you seem to argue that incentives are, in general, not that powerful in motivating people economically. And that surprised me because, if I think of most J-PAL papers, incentives are pretty powerful.
Your own paper on Indonesian roads — you put in much better roads, many more parents sent their kids to school, right? It’s a big response. The recent paper on drunk driving in India with Abhijit — police set up new road blocks. People respond to that incentive. They take other routes. So there are all these papers you’ve done, your lab has done, showing incentives are powerful. Why in the op-ed say incentives don’t matter so much?
DUFLO: What the op-ed says is that financial incentives are overrated.
COWEN: But getting educated or not being pulled over for a drunk-driving traffic ticket — they’re closely related to financial incentives.
DUFLO: The traffic ticket, I think, is much more than the financial incentive. If you get stopped by a traffic officer, you lose your car right this instant. I think that it’s likely inconvenience is much, much bigger. The point that we are trying to make is not that people are not sensitive to incentives, because we can always construct incentives. If you construct the incentive boldly enough, then it’s almost always true.
The point is also not that people are not responsive to financial incentives whatsoever. The point is that the response to financial incentives is less than what typically we assume it is, both in terms of people seem to be pretty inelastic to taxes at all levels of income from the studies we have. People seem to be pretty inelastic to receiving money for sure that doesn’t lead them to go on vacation.
COWEN: But that’s just an income effect, right? It’s not a price effect.
DUFLO: Yeah, that’s an income effect. But you would think the income effect also create a disincentive to work. At least that’s in our models. That’s why we’d be wary of welfare. But you take the negative income tax experiment — in fact, probably the first RCTs in the social science — as early as the ’70s, what they were doing was kind of a Milton Friedman idea to simplify welfare.
But what they did in this experiment is to guarantee people a certain level of income, but it was such that it was taxed away at a pretty steep rate. So people were facing about a 50 percent marginal tax on their income. And even there, the labor supply response was so low.
COWEN: But take that French work many fewer hours a week than Americans. Isn’t that mainly a tax effect, a price effect?
DUFLO: I don’t think so, no.
COWEN: No? What would it be?
DUFLO: Well to start with, there is a regulation —
COWEN: But there were all these part-time jobs. In the 1970s, French and other Europeans worked more than Americans. Tax rates in Europe were lower. The ’80s — expenditures and tax rates go up.
DUFLO: It’s always difficult to compare two countries, but my bet, based on extrapolating from the micro study, is that that’s not the financial effect. I think it comes much more from regulation. How much can you work anyways? And from the availability of jobs, the availability of part-time jobs. There’s actually much less, partly also due to regulations and less to individual effort response because they are thinking they are going to be taxed away. That seems to me counter factor.
COWEN: In the new book, why not have more on social conservatism? Take something like single parenthood. It seems to be a big problem. There are many new studies showing it may be a bigger problem than we had thought. It’s a cause of poverty, persistent poverty. It’s kind of a right-wing issue, for whatever reasons. Why not have more right-wing issues in your book?
DUFLO: We do have these issues as outcomes. I don’t know why it’s a right-wing issue to worry about single parenthood because, as you were saying, all of the data suggest that it’s not a very good thing to grow up as a kid of a single mom in her teens. And it’s not a good thing for the single mom in question, either.
So it seems to me that it’s not a right-wing, left-wing issue. In fact, the person I know who is the most left-wing probably is my mom, and that’s one of her key battle cries — reduce teenage pregnancies in El Salvador.
COWEN: And your father was a mathematician, right?
COWEN: And did you learn math from him?
DUFLO: Yes, but I learned it at a very fundamental level because he’s a very fundamental mathematician. It left me, I think, with some good instinct.
Going back to the issues — why I don’t think single motherhood is a left-wing or a right-wing issue — I think what would make it a social-conservatism issue is to approach it from a kind of moral failing way. And what makes it a progressive issue is to approach it as the outcome of conditions in life that are potentially changeable. Or maybe it’s the same thing.
We do talk about it a bit in the book. In particular, for example, one of the consequences of the fact that people are not mobile and industries are clustered, so when there is a shock, everybody suffers together, is that in those towns that are hit the most, you do see a kind of erosion of these social norms. People are less likely to get married. They are more likely to have children out of wedlock, et cetera. So this is one of the possible consequences of people’s unhappiness.
COWEN: What do you learn from rock climbing, or how does it inspire you?
DUFLO: Many things you can learn from rock climbing. One is that it’s a bit like empirical economics. You look at the wall, and you have a decent idea of what might work. Think of it as a theory or an intuition. But then when you try it out, sometimes it works, sometimes it doesn’t. That’s kind of the data going in a way.
It’s also something you do socially. You do need another person to catch you if you fall. And you need to pay attention to the other person. And a lot of economics is very social in this way. It’s not competitive, or at least not the way I do it. It’s making slow and steady progress toward some goal.
COWEN: Do you love South Indian classical music?
DUFLO: I like it. Love it would be a stronger statement than I should because I am not a connoisseur. Abhijit, my husband, is really into it. In fact, South Indian and North Indian both. We listen to it at home, and we go to concerts whenever we are in India. Chennai is one place where we do a lot of work and J-PAL has its headquarters, and it has some of the best South Indian music. So I’m kind of an unenlightened amateur.
COWEN: I’ve been to the December festival there. It was fantastic. Took my daughter. What in classical music interests you most?
DUFLO: In Western classical?
DUFLO: This I truly love. Indian classic I also love but with no expertise whatsoever. Western classic I grew up listening to forever. What interests me the most . . . I don’t even know if interest is the word because it’s very visceral.
COWEN: And what is it? What do you put on? What’s your Desert Island Disc?
DUFLO: Oh, that’s Bach. I listen to Bach almost every morning when I’m walking, in particular to Bach cantatas. That’s my morning music. Then in the afternoon I move on to more frontier stuff like Schumann or something like that. It’s only after dinner that it moves to Bob Dylan. So definitely it would be Bach, and it would probably be one of the Passions. Or if I could have two, I would take two Passions, the St. John and St. Matthew’s Passion.
COWEN: Yeah. What advice do you give to your talented undergraduates that differs from the advice your colleagues would give them?
DUFLO: I give almost all of them the advice to take some time off, in particular if they have any interest in development, which is generally the reason why they come to see me in the first place. But even if they don’t really, to spend a year or two in a developing country, working on a project. Not necessarily inner city. Any project spending time in the field.
This is what Michael Kremer did before he went back to graduate school, and I think that was a determinant to him. This, to some extent, is what I did with my little Russia experience. And this was clearly very central to me.
It’s only through this exposure that you can learn how wrong most of your intuitions are and preconceptions are. I can tell it to them till they are blue in the face to not let themselves be guided by what seems obvious to them. But until they’ve confronted what they think is obvious to something entirely different, then it’s not clear.
Whether or not this is different from what my colleagues would say, I don’t know. Maybe some of my colleague would say that you’re better off staying home and cranking some data and getting these types of skills.
COWEN: Obviously it hasn’t stopped you any, but do you worry that development economics in some ways puts women at a disadvantage because some emerging economies are especially dangerous for women? India, in particular, that may be the case.
DUFLO: I think development economics is the most gender-balanced subfield of all of economics. In fact, if anything, there are more women and more powerful women in development than men. So whatever it is, it does not stop them.
And there are many, many women working on India. Of course, you don’t do silly things. And the advantage of working, of having this infrastructure that I was talking about before, is that it makes it possible for people to always be in groups and to work in a context where they are safe. At the same time, they can get the experience of living in these environments. I think that makes people often very aware that gender rights, for example, is something to study very deeply.
For example, one of my graduate students, Maddie McKelway, worked in rural Uttar Pradesh, one of the parts of India which has some of the most extreme gender inequity. And she conducted her project like a master with a little group of largely female research assistants and got everybody to go along with her.
COWEN: Last question — if you could institute a reform to increase opportunity for women in economics, what would you do?
DUFLO: I think there are two things that one could do. I don’t know if they are reforms per se. One is to change a little bit the climate in the economics profession. It’s not particularly anti-woman per se. Some people argue it is, that people are, for example, more likely to interrupt women than men, et cetera. But even if they weren’t, they are, just in general, likely to interrupt each other. We are not a very touchy-feely culture in economics.
It doesn’t bother me, it turns out. So that’s not something that has stopped me in any way. But it bothers many people, and it tends to bother women and, I think, minority more. Changing the culture, of course, is not a reform per se. But there are small things you can do that can help.
One thing we’ve done at MIT — very simple — is to institute the rule that someone who gives a seminar gets the first 15 minutes uninterrupted. And that has, for our seminars, the development seminars, I think it’s actually improved the experience for everyone, the speaker and the audience. So that’s one. That’s not huge, but think of this as one — I’m not the magic-bullet type of person, and I don’t think there is one there — one of the silver bullets you could do there.
The other is something we could do outside of the economic profession, which is to change a little bit the image that economics has to the outside world. I think for a lot of high school students, college students — they are a bit like me when I was considering economics. It’s like, number one, it’s talk about interest rate and central banks, and it’s kind of boring. And number two, it uses incredibly crude tools.
In fact, both of these statements are wrong about the state of our discipline. In fact, we study any number of issues, including those, but also extreme poverty, discrimination, inequality, and we study them with lots of very sophisticated tools and relevant empirical tools, et cetera.
I think if people were more aware at an early age, before they choose a field, there would be more women to start with choosing to study economics. And then hopefully, more would continue, which, I think, is something we really need.
Not just women but also minority because economics is a social science, and as a social science, we do need diversity in experiences and points of view, et cetera, because people with different backgrounds, including gender and life experiences, will be interested in different questions and also will bring a slightly different perspective to those questions. So it would make the field more vibrant.
COWEN: Esther Duflo, thank you very much. Congratulations again on the Nobel Prize, of course. And the new book with Abhijit Bannerjee is Good Economics for Hard Times. Thank you.
DUFLO: Thank you so much.