Abhijit Banerjee on Theory, Practice, and India (Ep. 83)

Tyler’s former grad school classmate has made quite a name for himself.

Long before Abhijit Banerjee won the 2019 economics Nobel with Michael Kremer and Esther Duflo, he was a fellow graduate student at Harvard with Tyler. For Tyler, Abhijit is one of the brightest economic minds he’s ever met, and “a brilliant theorist who decided the future was with empirical work.” But according to Abhijit, theory and practice go hand in hand: the real benefit of a randomized control trial isn’t getting unbiased estimates, he says, but in testing hypotheses borne out of theory. 

Abhijit joined Tyler to discuss his unique approach to economics, including thoughts on premature deindustrialization, the intrinsic weakness of any charter city, where the best classical Indian music is being made today, why he prefers making Indian sweets to French sweets, the influence of English intellectual life in India, the history behind Bengali leftism, the best Indian regional cuisine, why experimental economics is underrated, the reforms he’d make to traditional graduate economics training, how his mother’s passion inspires his research, how many consumer loyalty programs he’s joined, and more. 

Want to support the show? Visit conversationswithtyler.com/donate. Your gift helps create enhanced transcripts like this one. 

Watch the full conversation

Recorded December 2nd, 2019

Read the full transcript

TYLER COWEN: I’m very pleased to be here today with Abhijit Banerjee, one of the recent Nobel laureates in economics. He is professor of economics at MIT; coauthor of the recent best seller with Esther Duflo, Good Economics for Hard Times. Abhijit and I also went to graduate school together at Harvard, and he is one of the most brilliant economic minds I have ever met, heard, or read. Abhijit, welcome.

ABHIJIT BANERJEE: Thank you. With that introduction, I’m going to be shy.

COWEN: Let me start with a question about economic development. Dani Rodrik has argued that premature deindustrialization is a significant problem for emerging economies, that in essence, they are moving into manufacturing at a time when automation already has happened. And this will not, for these countries, build a middle class. What do you think of this argument?

BANERJEE: Well, it’s sort of beyond my competence, to be honest. It seems like reading lots of tea leaves, but my instincts are that he’s right, and I think there’s additional problems. One of them is, the fact that there’s so much inequality in the world means that the products that are being designed now — a lot of the new money is being spent on — are products where the component of quality to it is extremely high.

They tend to be products that are — think of kids’ toys. Most of kids’ toys are a matter of checking and making sure that they have nothing dangerous about it. And the checking is often done, actually, in other places — in Germany, for example. For some of the things, the checking is done in Germany even though the manufacturing is done in China. As a result, the cost part of manufacturing is an extremely small part of the price of the good.

So what we pay in the US for the toy — a tenth of that might be the labor cost and the capital cost of it. But that means that, essentially, the things that generate large earnings for people are not happening anymore. Lots of this is just very, very well-designed products, of which brand-naming, marketing, making sure that they’re okay — all of these things are a huge part of the price. Therefore, essentially, if you think of the rest of it, which is a small amount — that means that, if you’re a worker, you’re getting 40 cents for a $10 toy or something, and that’s just not going to make a lot of rich people or even middle-class people.

COWEN: One of the early papers in your career — you seem to be arguing there was too much petty entrepreneurship in emerging economies. Do you think that’s still the case? And if so, how do those economies make the shift away from that?

BANERJEE: I don’t see a way for them to make a shift out, to be honest. This is why I think this whole literature on universal basic income and microcredit and all these ways of creating more entrepreneurs — in some ways, I think this literature is sort of stuck in this place where they’re trying to get most of these petty entrepreneurs to become a little bigger.

I think that right now, we don’t see a way to do what I think needs to be done, which is pick the best among these people and make them much, much bigger because I think the way petty entrepreneurs have been absorbed into the economy historically has been by making some people being very successful, employing lots of people and creating large enterprises.

People give the example of Italy, but if you study it carefully, even in Italy, a lot of the reason why the petty entrepreneurs can get somewhere is because there’s huge coordination by some brand name, Benetton being the classic example. It’s not obvious to me that we are, right now, even thinking properly about a mechanism for getting the smallest entrepreneurs out of what they’re doing and into the labor market.

COWEN: What do you think is the nature of the positive externality for manufacturing jobs in emerging economies? Where does it come from?

BANERJEE: I don’t know that it has to be manufacturing. That’s, I think, Dani Rodrik’s view and something that I said I’m quite willing to believe it, but I don’t know why it has to be manufacturing.

It seems to me what was important was that it was a set of middle-skilled people. The skill range in which you could move was still accessible to people who had a poor education. That was the key. The key move was not so much whether it’s manufacturing or not, but rather, it was there’s some skills involved, and the skills can be learned, and the skills were available to people who didn’t start with having an excellent education.

I don’t know. It could be that manufacturing on top of that has additional benefits. That’s the part I really don’t know how to show or whether I actually believe or not. It’s not that I disbelieve it. It’s not that I have studied it. It’s just not clear to me that there’s a reason to think that, if the same options were available in services, they would not serve the same purposes. I just don’t know whether either of that is about to happen.

COWEN: Could you imagine comparable growth to, say, that of South Korea exporting manufactured goods, but through the export of services? Or is there something different about services that makes that harder?

BANERJEE: Excellent question. I’m slightly out of my depth in all of this because this is sort of macro trade, and macro trade is not exactly my specialization. Nonetheless, I would say that prima facie, no, I don’t know that there is something that a small country exporting lots of services — could it be a location in even a big country like India? The software and, after that, the BPO boom brought in a lot of people who were actually relatively low skilled, who learned English and basically spent a lot of effort.

There’s lots of these classes all over India in small towns. “Learn English, work for a BPO.” And that’s exactly this idea that somebody with a high school education who’s smart, who’s enterprising, who’s willing to work hard can get trained in English, speak like an American, say, “Hello, this is John.” Or whatever you’re supposed to say and carry on a conversation.

And that training, I don’t see is any different from the upgrading of, I don’t know, a steelworker to a line supervisor. I don’t know enough about either of these things, but it’s not obvious to me that a location or a small country could not get a lot of benefits out of services. The problem is more, the services are also where automation is hitting.

On charter cities

COWEN: You discuss charter cities at some length in your new book. Why are you skeptical of them as a way of inducing more economic growth?

BANERJEE: I think that if you could have charter cities, they may very well be great. The problem is more, can you believe that the country would honor the particular commitments that it requires?

For charter cities, the idea that we are going to hand over the maintenance of our law and order to a different country, and somebody else will act as an enforcer, I think the democratic process doesn’t really . . . it’s very easy to exploit that. It’s very easy to come and say, “Look, your government has been taken over by these enemy people or these foreign people, and they’re running. How could they be the people deciding what’s legal and what’s not?” Et cetera.

This is politically extremely unstable, easy to attack, very visible targets. It’s not clear to me that is anything more than that. I think yes, if you could imagine a situation where the politics is exactly aligned right, where I would say that creating the legal and physical infrastructure for enterprises would be excellent.

But the idea that this is a country which can’t enforce its own laws, and therefore, outsiders can do it for them — there’s something in that idea that seems to me to be odd because it’s precisely the fact that this country can’t enforce its own laws because there’s something in this country which is making it hard to enforce those laws. And maybe it’s competence, but it’s often not. I mean, this is not just competence. There’s politically, the opportunities that these things offer are too tempting, it seems to me.

COWEN: It seems that British-ruled Hong Kong worked very well. Portuguese rule over Goa, much less so. How well do you think we understand why some charter cities have gone well and others not?

BANERJEE: I don’t think we understand it at all well, and the reason, partly, we don’t understand is because is it that Goa was just too near Bombay and that Hong Kong, being at that absolutely wonderful spot between Japan, Korea, the north of China, India? A lot of the opium trade and all of these things went through there. It’s a great place to be an entrepot, and I don’t know that the world can handle a certain number of successful entrepots. Goa may have been too near to Bombay to be a successful entrepot. It’s not clear that that was a great location. Again, do I know that that was the reason why Hong Kong was successful? Absolutely not.

There are only 10 examples of very successful entrepots, or maybe 15 or 20. I don’t know the history well enough to say that. Venice, for a long time, was one, and then it stopped. But is it that, essentially, your location is your ticket to success? Or is it something more than that? Can you build economies out of entrepots?

COWEN: How well do you think economists understand economic growth more generally?

BANERJEE: I think they don’t understand it. This is the point we make in the book. I think we understand some very basic insights, one being the one that my colleague, Bob Solow, made very clear, which is that growth has a tendency to run out, that once you used up your best talent and your best capital and your best locations, there’s going to be less and less so. And I think that’s probably the only thing which has been pretty consistently borne out: that fast growth slows down. China’s slowing now.

And there’s nothing tragic about it. It’s a normal way of things. I think that the question is more — the thing that we don’t understand is what makes that happen sooner or later, more or less, et cetera. Brazil grew for 20 years, extremely fast, between 1960 and 1980, and then essentially stopped for 20 years. Why do those kinds of things happen? That we don’t understand. I think we’re lousy at predicting growth in general.

COWEN: But if we have countries — take Singapore as an example. Say they invest heavily in human capital, they actually have the political will to grow, and they very broadly have some kind of neoliberal policies, but tailored to their specifics. Don’t we know if they do that, they’ll grow? Singapore it worked, Dubai it worked, Republic of Ireland it worked. Don’t we know a lot about growth?

BANERJEE: Alas, no. You have these tiny, tiny countries, basically, with, I would say, very specific political configurations. I think Lee Kuan Yew knew very well that if he didn’t deliver a very specific set of environmental conditions, and if he were falling behind, there’s Malaysia around.

Singapore was a very, very fragile project in 1960 and one that, I think, they were lucky to have him as the dictator. I think he was, in the end, committed to some things which were . . . But they could have easily had someone who had crazy views, and there were plenty of people with crazy views.

I think small cities at the edge of other, less liberal societies also are very strange examples to use. I was at Dubai. You see so many people from Saudi Arabia, Iran, et cetera, who need to escape their own societies temporarily for pleasure or for just freedom. I think lots of rich people around — Iraq less so — Iran and Saudi Arabia, probably two big countries with lots of rich people who need a playground. I don’t know that that’s replicable at any scale.

On Indian sweets

COWEN: Why does Kolkata have the best sweet shops in India?

BANERJEE: It’s a bit circular because, of course, I tend to believe Kolkata has —

COWEN: So do I, however, and I have no loyalty per se.

BANERJEE: I think largely because Kolkata actually also — which is less known — has absolutely amazing food. In general, the food is amazing. Relative to the rest of India, Kolkata had a very large middle class with a fair amount of surplus and who were willing to spend money on. I think there were caste and other reasons why restaurants didn’t flourish. It’s not an accident that a lot of Indian restaurants were born out of truck stops. These are called dhabas.

COWEN: Sure.

BANERJEE: Caste has a lot to do with it. But sweets are just too difficult to make at home, even though lots of people used to make some of them. And I think there was some line that was just permitted that you can have sweets made out of — in these specific places, made by these castes.

There’s all kinds of conversations about this in the early-to-mid 19th century on what you can eat out, what is eating out, what can you buy in a shop, et cetera. I think in the late 19th century you see that, basically, sweet shops actually provide not just sweets, but for travelers, you can actually eat a lunch there for 50 cents, even now, an excellent lunch. They’re some savories and a sweet — maybe for 40 rupees, you get all of that.

And it was actually the core mechanism for reconciling Brahminical cultures of different kinds with a certain amount of social mobility. People came from outside. They were working in Kolkata. Kolkata was a big city in India. All the immigrants came. What would they eat? I think a lot of these sweet shops were a place where you actually don’t just get sweets — you get savories as well. And savories are excellent.

In Kolkata, if you go out for the day, the safest place to eat is in a sweet shop. It’s always freshly made savories available. You eat the freshly made savories, and you get some sweets at the end.

COWEN: Are higher wage rates bad for the highest-quality sweets? Because rich countries don’t seem to have them.

BANERJEE: Oh no, rich countries have fabulous sweets. I mean, at France —

COWEN: Not like in Kolkata.

BANERJEE: France has fabulous sweets. I think the US is exceptional in the quality of the . . . let me say, the fact that you don’t get actually excellent sweets in most places —

COWEN: Sure, hardly anywhere.

BANERJEE: Hardly anywhere. You pay extra through your nose in France. We spend lots of time in a small village in Provence. It’s a really small village, not one of these villages where Albert Camus used to live. It’s a no-name village at the edge of the Luberon. And the sweet shop there produces better sweets of different kinds than any place in Boston that I would go. I think just the average quality of what they produce is so wonderful. It’s a no-name. There’re two bakeries; one is cheaper and less good.

The high-end bakery, you pay 10 cents more for a baguette instead of 95 pence — whatever cents you pay — one euro and five cents for a baguette or something. And the baguettes are always excellent, and the cakes are excellent. I actually think that it’s an interesting question, why the US is so exceptional in this. In Italy, it’s the same. I don’t love the bread, but the pastries, excellent. In Spain, it’s the same.

COWEN: Are Indian sweets, in particular, so labor intensive in their construction that you can’t sustain them?

BANERJEE: No, actually, they’re not more labor intensive. I cook a lot. I actually can make a fair amount of them. In a sense, they’re somewhat more forgiving than French sweets, which require, if you’re making a mille-feuille or something — those are extremely labor intensive, and they require extremely mindful labor.

Indian sweets are actually relatively slow cooked. They don’t involve baking, which is the thing that goes wrong because once you stick it in the oven, it’s like God is your only guide. So I think Indian sweets are relatively forgiving. I much prefer to make Indian sweets than French sweets. I love cooking, and I will make anything if I have the time, and mostly because I’m greedy.

COWEN: What country, other than India, has the best Indian food, not counting Pakistan or Bangladesh?



BANERJEE: Absolutely, no question. These days, the best Indian restaurants in the world happen to be in Bangkok, but I have not tried Bangkok Indian food because when I go to Bangkok, I want to eat Thai food, which is excellent too. But the UK has excellent, excellent and trashy and horrible and kind of gloppy — the whole range but including the high end, which is really quite excellent.

On Bengali intellectual influence

COWEN: There’s a long-standing tradition of left-leaning Bengalis being especially important in Indian intellectual history. Why do you think that is?

BANERJEE: I guess why Bengalis — again, it goes back to the fact that Bengalis were the first group of people, maybe a little bit with the Parsis, as being the comprador classes we Marxists one day, once upon a time, used to call them, of the British. The people who worked for the British, with the British, and made a good living.

And I think that class very quickly started absorbing English writers and Scottish writers they read, by 1800 already. This is just 25 years after you’re starting to see people who are writing in Bangla about Hume and about Rousseau. They’re reading European writers mostly in English, but they’ve mastered English. They’re reading all of these things. Contact with the British created this.

This is very controversial, and I’ll get killed by many, many people who have strong views on this, but I think that it’s very hard to imagine exactly what happened then. Maybe something else could have happened that was wonderful. I don’t know the counterfactual, but what happened then was very much people absorbing English 18th-century liberal and conservative discourses, using them.

There’s a bunch of historians who studied this period, 1800 to 1840, in Bengal, and it’s a really fascinating period. They’re up in arms, they’re fighting the brahminical tradition. The brahminical tradition is fighting back.

The one fact that I like to emphasize is the first, I think, purely secular institution of higher education in the world is in Kolkata in 1817. What now is called Presidency University was set up as Hindoo College, but Hindoo College was never for Hindus because the first famous professor there was Derozio, who was a Christian. He was the first and perhaps the most visible sign of . . .

It was funded by the Hindu community, and the idea was each community will fund one, but in fact, that didn’t happen. So Hindus, Muslims, Christians, everybody went there. This is 1817. Jefferson was very proud of having set up University of Virginia as a liberal arts university with libraries the central point — that’s in 1825. So this is eight years before that they already set this up.

So there’s a long-winded way of saying, I think this is a moment in history which seems to me comes at the encounter with the British, the sense that there is this extremely successful culture out there. Whether there’s good or bad, we don’t know, but we can’t . . . And people are saying that, and they’re writing that. “We need to know what these bastards are up to,” or whatever. They’re often quite negative, but they’re saying, “How could we not? They’re successful. They’re ruling the world. We really need to know what they’re doing.”

So they start reading up on all this stuff, and they’re writing about it. And this is by 1820s, actually, even before the 1820s. Ram Mohan Roy was the first, I think, great Bengali intellectual who’s, I would say, not left leaning. I think he’s very —

COWEN: He’s kind of a Benthamite, wasn’t he?

BANERJEE: Benthamite.

COWEN: Yeah.

BANERJEE: Exactly. Benthamite, but not so much. But he’s a contemporary of Bentham’s. He dies in 1825 or something. So he’s a contemporary of Bentham’s, but he’s absorbed all this.

He writes in Persian a lot because he’s from the old aristocracy, which was trained in Persian. But he writes about all these very important liberal ideas. He’s a liberal in a very important sense in which it’s much more than a leftist. He’s a liberal. And I think that liberal tradition becomes extremely contentious immediately, with the conservatives pushing back against them. And the liberal tradition then has, actually, a hard time not moving to the left.

So it does move to the left, but partly in reaction to a set of reactionary and completely meaningless reactions to the liberal tradition. They were a bunch of people who you can speak about in hallowed voices. I could talk about them, but they didn’t . . . Their views seem to be mostly, “Come on, this is foreign stuff. We don’t like it.” And so, I think, the liberal tradition grew partly in reaction to conservative pushback. Conservative means conservative. I don’t mean Right or Left.

COWEN: Right.

BANERJEE: Conservative pushback.

COWEN: Wasn’t the first IIT in Bengal, in Kolkata?

BANERJEE: Not in Kolkata, in Kharagpur…

COWEN: Kharagpur, yeah…

BANERJEE: That’s about a hundred miles from Kolkata. But I think those were . . . Nehru already . . . By this time there is a national plan for these things, and even though the first one gets built, there’s one in Bombay, one in Madras, one in Delhi, one in Kanpur. So every part of the country gets one — the first five IITs, and I think that was already conceived of as a single plan.

But going back to the question of Left, I think the Left comes actually later. The liberals become Left later, partly in reaction to just the horrors of the last years of British rule. I think the partition . . .

No, Bengal starts the first piece of that in 1905. But most importantly, the Bengal famine just completely moved everybody to the left because the Bengal famine in Bengal is seen as being egregious completely, a callous government with, maybe, some sympathy for the people who control the rice supply. Three million people die. Churchill sends ships that the Americans had sent with grains to India — doesn’t let them come to India — sends them to help British troops elsewhere.

And it creates, I think, an extremely strong basis for the Left. The Left really becomes very powerful starting in the ’40s. The Bengal famine is a defining event of that period. This is one of the biggest famines ever. And it’s such an egregious tragedy that a lot of people just feel that there’s something deeply wrong with the world.

And that’s when from Tagore, who was basically a liberal, we moved to the left very quickly in the ’40s. Tagore dies in ’41, and after that, the Bengalis move quite sharply to the left. It started before. It’s not that it didn’t — there were elements of Marxist influence before, but I think the general move of Bengali intelligence to the left starts in the 1940s.

COWEN: How would you place your own father in the Bengali intellectual tradition?

BANERJEE: My father was a liberal, He was not a leftist, I would say. Relative to the Bengal spectrum at the time, I would say he was on the right.

COWEN: And he was an economist?

BANERJEE: He was an economist, but I think more importantly for him in his self-presentation, he was a liberal, and he really believed in all the liberal values: freedom of speech. He had taught himself British 20th-century philosophy quite well. I think it was [Bertrand] Russell and G. E. Moore that he looked up to most. Russell is Left. G. E. Moore is more center. It’s more those — that was his intellectual tradition rather than the Marxist tradition. He was very much more of a liberal than a leftist.

COWEN: Your colleague, Daron Acemoglu, and his new book, The Narrow Corridor, argues that the Indian caste system has held back Indian economic growth quite a bit. But Pakistan and Bangladesh are poorer than India and don’t have a similar caste system. Do you have a perspective on this whole debate?

BANERJEE: No. I think that maybe he’s . . . We don’t know the counterfactual. Maybe other things went wrong. Certainly, the fact that there was a liberalism that was built into the Nehruvian state encouraged a certain type of person to stay Hindu, Muslim, Christian, and that self-selection was important. I don’t know all the ways in which these countries are different. It’s a very interesting thought. I don’t know whether I actually believe that . . .

I certainly believe the caste system is awful. I don’t know whether I would think its first-order effects are on growth directly or because they just happened to be a way of ethnic organization. So whether, without the caste system, we would have had like in Pakistan . . . Pakistan is a war zone between the Punjabis, the Pashtuns, and the Sindhis and the MQM. The kind of movement of Karachi is exactly the . . . All of these people are fighting the Muhajirs or the people who come from India, so it’s a bunch of ethnic violence.

Whether India would have had other axes of violence . . . In a sense, right now, the Indian political economy is moving, I would say a bit in the direction of trying to unify the Hindus against everybody. Whether that works or not, I don’t know. I’m not sure whether that’s a desirable thing or not. I tend to think that diversity is helpful, actually.

It’s not obvious to me that, if the Hindus were not as divided, India may have gone in a much less liberal direction easily because there are other minorities, and it could have been much more organized on anti-minority lines. Part of the division, the fact that the Scheduled Caste don’t trust the upper caste might be actually — not necessarily — it might actually be a way of creating enough countervailing power, if you like.

COWEN: So a kind of stabilizing property.


COWEN: In the short run at least.

BANERJEE: Yeah, so it’s not obvious to me what the counterfactual is. That’s the point that I’m making.

On Indian classical music

COWEN: If I look at the earlier history of Indian classical music in Bengal, to outsiders, Ravi Shankar is best known, but there are many top performers. And more recently, Bengal in Indian classical music seems to be less important. Do you agree? And if so, why is that?

BANERJEE: Well, I think Bengal was never the place for vocal. As a real, I would say a real addict of vocal Indian classical music, I would say Bengal is not, never the center of . . . If you look at the list of the top performers in vocal Indian classical music, no one really is a Bengali.

In instrumental, Bengal was always very strong. Right now, one of the best vocalists in India is a man who lives in Kolkata. His name is Rashid Khan. He’s absolutely fabulous in my view, maybe the best. On a good day, he’s the best that there is. He’s not a Bengali. He’s from Bihar, I think, and he comes and settles in Kolkata. I think a Hindi speaker by birth, other than a Bengali. So I don’t think Bengal ever had top vocalists.

It had top instrumentalists, and Ravi Shankar, Ali Akbar Khan, Nikhil Banerjee — these were all Bengali instrumentalists. Even now, I would say the best instrumentalists, a lot of them are either Bengali or a few of them are second . . . Vilayat Khan and Imrat Khan were the two great non-Bengali instrumentalists of that period, I would say, of the strings especially. And they both settled in Kolkata, so that their children grew up in Kolkata.

And the other great instrumentalists are these Kolkata-born. They went to the same high school as I did. There were these Kolkata-born, not of Bengali families, but from very much the same culture. So I think Kolkata still is the place which produces the best instrumentalists — sitarists, sarod players, et cetera.

COWEN: Why is the better vocal music so often from the South?

BANERJEE: Oh, that’s a separate question. Separate. The South actually has its own vocal tradition, which is completely different. I actually don’t know enough about it to say. I think the best vocalists typically, right now, come from Maharashtra and North Karnataka. That’s sort of the middle southwest.

But the deep South has a completely different tradition. They sing in entirely different ways. The tradition that I listen a lot to is the tradition of Maharashtra and North Karnataka. And that’s where, to me, in my mind — the great home of classical vocal music right now is that part of India, the area kind of starting at Bombay, going down to Bangalore. That chunk of India is where the great classical musicians are.

COWEN: Like Pandit Kumar.

BANERJEE: Yes. Or . . . I’m going to blank on the names of . . . Certainly Venkatesh Kumar, but also . . . I’m going to blank on the name of this young woman artist who I just heard whom I thought was just fabulous.

COWEN: Amonkar? She’s not young anymore, but —

BANERJEE: No, she died, Kishori Amonkar. She died a few years ago.

COWEN: Oh, I didn’t know.

BANERJEE: She was the best, I think, till I would say . . . I heard her a few times live, and when she sang, well, I don’t think anybody . . . I think what is difficult in Indian classical music is to keep the variation going and keep the musicality running. And some people do lots of variation, but they lose the sense of the overall music. She was amazing at her best. In fact, I heard her first time live — maybe not first — second time at Harvard, when we were students together, in 1985.


BANERJEE: And she sang an unbelievable concert in the old Fogg Museum courtyard. It was probably the best concert ever.

But no, much younger women who are in their 30s. There are really some very talented ones, but I can’t remember their names right now. I am blanking on the names. I will remember them as soon as I walk out of the room. But there’s actually lots of talented people in their 30s, 40s. Indian classical music, being the kind of thing it is, you never become a top artist till you are in your late 30s or something.

COWEN: Do the economics of Indian classical music have a future?

BANERJEE: I think so.

COWEN: Through concertizing or overseas work or how?

BANERJEE: I think through concertizing. I do think that if you look at the top echelon of these people, their fees have gone up a lot, and they’re in demand. It’s not clear to me that this is any worse than it ever was. There’s always a little bit of “You have to do it if you love it.” You don’t do it because you could make money.

And the movie industry has always been quite generous. Ravi Shankar also was a music director for many films, and he composed music for many films, et cetera. Bollywood is actually a pretty good client for good classical musicians. And Bollywood is one of India’s great successes.

COWEN: With the tradition of training Indian classical musicians through the family, being passed down father to son — will that persist, or is that an anachronism?

BANERJEE: I think that’s an anachronism. I think the training schools like the SRA [Sangeet Research Academy] in Kolkata are very good. They train you, make you listen to many different people. That’s actually a good thing, for musicians to learn how to distinguish their tradition from other people’s tradition. I think that the old norms were a little bit too parochial.

On things under- and overrated

COWEN: In the middle of these conversations, we usually have a segment, underrated versus overrated. I’ll toss out some names, some ideas, and you say if you think they’re underrated or overrated. But of course, feel free to pass. First one, Balzac.

BANERJEE: I think rated about right. I think Balzac is great. I don’t think in France anybody doubts that he’s great. He’s sort of the canon, so I think he’s not overrated or underrated.

COWEN: Experimental economics.

BANERJEE: Underrated.


BANERJEE: I think people haven’t quite understood how well it connects to the rest of economics. It seems to me that people think that this is some boutique exercise that’s happening elsewhere, but in fact, it’s changing economics. It’s changing the way we do theory and changing the way we do policy analysis and the way we think about the world.

COWEN: Tolstoy.

BANERJEE: I think rated about right again. I think Tolstoy is wonderful, but I think people think he’s wonderful. They don’t read him, but that’s a separate question. [laughs] Is he rated high? Yes, sure, he’s rated high. He’s the canon.

COWEN: And the Beatles — underrated or overrated?

BANERJEE: Neither. They’re the canon.

COWEN: They’re the canon.

Economic growth prospects for Pakistan — it seems to me 30 years behind India. Will it catch up, converge, or not?

BANERJEE: I don’t know enough to . . . It’s growing fine. It’s actually doing fine. And Indian growth is a bit faltering, so maybe it’ll catch up, but I don’t know. I wish I knew more about what makes Pakistan grow. After all, it still grows at 6 percent, which is not nothing. So I feel like, who are we to sniff at 6 percent? For many, many years, it’s been growing at 6 percent. That’s nothing to laugh at.

COWEN: Who is an underrated Indian writer of importance to you?

BANERJEE: Any number of the vernacular writers who are not read. I would say among novelists, I only really read Bangla. But that’s not to say that there aren’t other similar people in other languages, but Bibhutibhushan Bandyopadhyay, Tarasankar Bandyopadhyay, Manik Bandopadhyay, Satinath Bhaduri — just amazingly good novel-form literature from the ’20s to the ’70s, and also amazingly good poetry, both of which are just not read by anybody.

COWEN: Is it translatable into English or —

BANERJEE: Yes it is translated, even some of it’s translated.

COWEN: But it can be translated but not translatable, so to speak.

BANERJEE: I think it’s translatable. I do think that it needs somebody with lots of talent to do it, and somehow that incentive hasn’t . . . I think now there’s a little more interest in doing it. In fact, Penguin has a series where they try to publish high-profile translations. Given incentives, there is just so much great literature.

COWEN: Other than Bengali food, what do you think is the best Indian regional cuisine? For me it’s food of Chennai, but —

BANERJEE: There’re in my mind only two in competition. The competition is either Chennai or Hyderabad. I think Hyderabad has an even broader range of food in the sense that it goes from very meat based and different styles of meat based — the Hyderabadi, Nawabi cuisine and the Hyderabadi . . . Uttar Pradesh is full of people who are not vegetarian, the Hindu — not vegetarian; they eat lots of meat.

The South actually is where meat-eating is more common than in the North, bizarrely enough, though we associate the South with vegetarian food. And Uttar Pradesh has fabulous food across the range. If you go to small towns in Uttar Pradesh, the meat, the vegetables, everything is stunningly good. Fish is stunningly good.

I would have picked Uttar Pradesh, but Tamil Nadu being the other possible. Tamil Nadu is where Chennai is. Tamil Nadu has enormous variety. Again, going from meat, the Chettinad food all the way to very, very proper vegetarian food, and all that is really great. And that part of the southeast, the eastern South has great, great food.

COWEN: Why is the quality of vegetables so much higher in India than anywhere else?

BANERJEE: Vegetable cooking?

COWEN: Cooking, but the vegetables themselves — if you eat a fairly plain dish, it’s still wonderful.

BANERJEE: I think partly the —

COWEN: Short supply chain?

BANERJEE: Yeah, short supply chain. Even now, people really pull out the vegetables in the morning and bring them to the market. If you go to the Gariahat Market in Kolkata, there’s this part of Gariahat Market where you buy basically — old ladies whose sons have sent them from their village to vend the okra that they’ve picked this morning. Just freshness is such an enormous part of taste, and also not designing food for durability. In the US, tomatoes are designed for durability, and that takes away from other things.

COWEN: Johann Sebastian Bach, the cantatas — overrated or underrated?

BANERJEE: My wife will kill me when I say what I told her — boring.

COWEN: You have a 2015 Science paper with Esther, Dean Karlan, some other coauthors about how cash transfers to the poor, combined with training and coaching, have very high rates of return, over 100 percent, up to 433 percent. What exactly do you think the coaching is adding in those RCTs [randomized control trials]?

BANERJEE: So these people — we’ve actually done it without the coaching —


BANERJEE: — in Ghana, and it doesn’t work. So we are reasonably confident that the coaching is doing something interesting. I wouldn’t say it does it necessarily for everybody, but the people targeted in this are the poorest of the poor. They’re among the poor.

And for them, I think confidence is an enormous issue because they’ve never actually done anything in their life successfully. They’ve been living hand to mouth, usually begging from people, getting some help. What that does to your self-confidence, your sense of who you are — I think those things, we haven’t even documented how brutal it is. People will treat you with a little bit of contempt. They might help you, but they treat you with a little bit of contempt as well.

This is the kind of people — at least the one that I was a big part of studying was the one in India and also one in Ghana. Especially the one in Bengal. These women — they were living in places where nobody should live. One said, “Oh, we get snakes all the time.” Another one said, “I’m now vending knickknacks in the village,” basically kind of cheap jewelry, that cheap stone jewelry or plastic jewelry.

“Before the people from this NGO showed me where the market was to buy wholesale, I had never taken a bus, so I had no idea how to go there. They had to literally put me on a bus, show me where to get off. And it took a couple of times because I had never taken a bus. I couldn’t read, so if it’s a number X bus number, I don’t know what X is, so how would I know I’m getting on the right bus?”

All of these things are new. If you start from a place where you really never had a chance, I think it’s useful to have some confidence building. You can do it too. There’s nothing difficult about it.

COWEN: Do you think the coaching almost serves as a kind of placebo? You don’t have to teach them so much, but just show that someone else has confidence in them?

BANERJEE: That’s an interesting question. I think it’s a bit more than that. It’s also saying, “You can do it, and here are the steps.” Turning things into a set of processes is important. Otherwise, it looks like an unlikely proposition that I can do it. I’ve never done it. I’ve never bought and sold things. In fact, I’ve never sold anything, and how do I do it?

It’s a bit more than that. Turning things into process is important also, that here is how you get on a bus. You go there, you pay this much money, they give you something, you bring it back. One of the things they are doing is also turning it into a set of procedural steps, which is very different from saying, “Go do that.”

COWEN: How scalable do you think the coaching is?

BANERJEE: Scalable? I don’t think it’s difficult. One thing this organization does well is it teaches people to be quite sympathetic. The one we worked with in Bengal, the one I know well, the people we worked with were basically able to be sympathetic. It doesn’t take much. It takes sympathy, but I think it’s a matter of motivating people to do it, making sure that they feel . . .

Lots of people work for NGOs, after all. And they are often people who want to do things for the world, who are positive people, so I don’t imagine it being that difficult. It will require investment in training a particular kind of person, but I don’t know that it’s any different from training somebody for collecting money in a microfinance organization. It’s the same level of skill, the same kind of person, same set of places we did this people will come out of.

COWEN: This seemed to work in six countries: Ethiopia, Ghana, Honduras, India, Pakistan, Peru, as you well know. Why isn’t the whole world doing this?

BANERJEE: I think lots of people are. Somebody told me 43 different countries — they’re experimenting with it. I think this is the next stage, where we’re a bit proselytizers for this, and we get called in.

But actually, the story is kind of doing the rounds of the governments. Many state governments in India now want to just do it, and that’s my exposure. In fact, today we were a part of a long email exchange on exactly how it should be done in the state of Orissa in the east of India that wants to do it. I think it’s having some traction.

COWEN: Your randomized controlled trials — they’re huge. They employ large numbers of people. Do you have a personal philosophy on management or leadership? What have you learned? You started out as an academic, and now you run big things.

BANERJEE: I’m a lousy manager.

COWEN: That can’t be, right?

BANERJEE: No. Well, I think I can be because I have lots of faith in people, and I’m good, I believe, in giving responsibility to the right people.

COWEN: That makes you an excellent manager.

BANERJEE: I’m only good at that, but I do go long on both investing in those people, making friends with those people, enjoying their company, and letting them tell me when things are going wrong and help advise me when I’m thinking wrong. I get a great deal out of working with wonderful, brilliant people. J-PAL is full of such people.

COWEN: How many consumer loyalty programs do you belong to?

BANERJEE: Two, I think.

COWEN: And you still think they’re collusive?

BANERJEE: Yeah, the fact that I belong to it in equilibrium — I should belong to it.


BANERJEE: Absolutely. I think that the airlines are completely turning the screws on you.

COWEN: Do you think there’s a regulatory solution?

BANERJEE: Maybe not. It’s not clear to me that the welfare losses are that large. In essence, the airline industry is . . . The problem with the airline industry is also what, in economics, we were taught as Bertrand-Edgeworth competition. You pay your fixed cost and then you compete. It’s known to be unstable. We were taught that. You and I were taught that.

COWEN: Right.

BANERJEE: Bertrand-Edgeworth taught us that they were unstable. And I see the airlines industry as being . . . the first-order problem is that, which is that it is designed to be unstable. The marginal costs become very low. You make a big investment. Then you drive everybody else out. Then, of course, the prices go up.

You see absolutely the Bertrand-Edgeworth pattern, and you’re seeing it right now in India where a bunch of airlines are going out of business because their prices were just too low at some level, and the competition was too hard. Jet Airways is gone, and Kingfish is gone. India’s basically now down to three and a half airlines.

COWEN: You have a 1999 QJE paper with Aghion and Thomas Piketty. It seems to be an early precursor of secular-stagnation theories. Do you still think about this issue?

BANERJEE: Yeah, I like that paper. I think it was a little too maybe cutesy for the macro people to like it. They like, properly, all the bells and whistles that make macro models. They like it. We went for a very different idea with basically theorists who were trained to write models that just deliver what you wanted to deliver and not much else. In that sense, I’m very proud of that paper. I think it’s a good exercise.

It never flew because I think people really wanted the bells and whistles that we didn’t really provide. Paul Krugman for a while did some advertising on not that paper, but the sequel of that paper, which was on open economies. But I think most people, the most macro people, real macro people didn’t see it as being . . . they thought it was too toy — the model.

COWEN: Do you miss being a theorist? Or maybe you still do theory in secret?

BANERJEE: No, not even in secret. It’s more that I do theory on things that, in a sense, I was doing 30 years ago in social learning. So doing theory, but maybe a little more inspired by behavioral economics, little bit inspired by the fact that maybe people don’t make rational choices. But I have a 2019 paper on how people choose good rules for social learning. What are good rules for social learning? Will people choose them?

COWEN: What’s the story of how you came around to doing randomized controlled trials? Having started off writing with Eric Maskin, right? He was your supervisor.

BANERJEE: He was my supervisor. I think randomized controlled trials are . . . in a sense, they appeal very much to a theorist, I would say, because what’s nice about them — and this is what I always tell people — is that it’s really not about getting unbiased estimates. That’s sort of a red herring. Maybe you can get unbiased estimates in many different ways.

What you can do is design your treatment to follow what’s inside your head. And I think what’s very nice about randomized controlled trials is to say, “Look, if I want to keep digging into something, why is it not working? I can keep varying the treatment exactly based on my current hypothesis.”

In some ways, I think this process of working on it even more convinced me that this idea that we’re going to do theory which is kind of place- and time-independent is not helpful. I think what’s the best thing with economics is that it gives you a toolkit. You interrogate things, and you say, “Oh, well it couldn’t be that. It couldn’t be that. Could it be this?”

And that training that we get, which is an excellent training of thinking through multiple hypotheses — that’s what theory is about, in a sense, building ways to accommodate multiple hypotheses into the sparest narrative possible.

So you want to see if these two forces — which of them win out under certain conditions. And that kind of thing is great because it trains you to think about all the forces at play in a particular context. I think of economics as theory as being excellent training for the mind. It gives you a way to ask, “I think it’s this, but could it be that?” And that training is still what makes me excited. When I think about a new RCT, I think about maybe we can try this, maybe we can try that.

Now I teach PhD behavioral economics, so I’ve been educating myself in those things. The hypothesis set are maybe a bit wider, but the same idea, which is that theory trains you to ask questions about what you could get wrong or where you could get confused or how to interpret a particular fact, maybe. I think, naively, that this means that, but it’s not actually because we think about it.

That was the great insight, I think, of the Chicago tradition. They used it often, maybe, to defend the status quo, but the insight was not wrong. I think the way they used theory was very appropriate, which is to be interrogative.

COWEN: How should we reform graduate training and economics?

BANERJEE: We’re trying very hard at MIT, actually. I was on a committee which recommended that we get rid of the general exam and we just, instead, have much more encouragement to start doing research early.

I do feel that we sort of stun people with “This much theory and this much macro is essential.” And there’s so much. We tease them so much that they don’t learn anything or they don’t look enough because I think some level, what maybe we got by having been at an early point is that we got little portable insights. And portable is the key word, which is that I want to take the insight and apply it to something immediately rather than know that, yes, there is this enormous edifice and the edifice can be built.

That to me is less useful than knowing the insight of why complete markets deliver efficiency. Understanding that is very useful, not because it describes a world, but because it gives you a very clear understanding of what is going wrong, and we don’t teach it as such.

We teach it as “Here is a theorem about the world.” We don’t teach people to wonder about, “Here are 17 examples. What’s going wrong in this one? Spot the place where the assumptions are failing.” Or “What’s incomplete in this market? And why does that potentially have big effects?”

I don’t think we do a good job of making these things get under people’s skin so that they feel, “I understand not just the theorem, which I can write down and prove, but also that this is, if I had to take it in my hand and use it to analyze a particular question, I’ll feel confident in doing that.”

COWEN: What did you learn from Eric Maskin, your doctoral supervisor?

BANERJEE: I had two, actually. I had Eric Maskin and an even maybe more pure theorist, Andreu Mas-Colell.

COWEN: Sure.

BANERJEE: Two absolutely fabulous supervisors. I can’t imagine having better. What I learned from both of them, and especially from Eric — Eric is interested in the world. Andreu Mas-Colell also, actually, eventually went into politics, but at that time he didn’t . . . What he did was, he taught us that in the end, there’s lots of verbiage, and there’s a few ideas that are key. Just focus on that.

And he would, when he taught advanced course — I don’t know if you took one — he would just draw a picture and say, “Look, this picture contains all the things that can go wrong, and this whole theorem is to show you conditions under which this doesn’t happen, but think about the picture and you are…

I always think of that as being the key thing that I learned from both of them. Just have the discipline to pinpoint. Don’t get lost in the complexity. The complexity is often man-made and sometimes unnecessary. Find the key place where the argument is failing or what gives leverage to the argument in both directions. And that’s what I learned.

I think it’s a fabulous way to be trained because you don’t . . . with Eric it’s unnerving, in a sense, because he’s so sharp, and if you say something that there’s three extra pieces to the argument and those are not needed, he’ll spot them immediately. You get trained to say, “Look, this is the key point. Rest of it is — we can build the infrastructure over it, but it’s not going to change the core insight.”

COWEN: What did you learn from your mother?

BANERJEE: To be angry about things that are wrong in society. My mother is very passionate, and she will just go on a tirade. Even now, she goes on a tirade about all kinds of things that she thinks are wrong.

And in a sense, her passion is not unrelated to the economics. She thinks of the economics and says, “But this economics is backwards. It can’t possibly be that this is what the market needs to do to function. So, where is it wrong?”

She doesn’t necessarily always articulate the whole talk, but her passion, combined with a certain commitment to thinking it through — I think she’s just a fabulous example for growing up. She would never be passive about the bad things that are happening in the world.

COWEN: And last question. What did you learn from your father?

BANERJEE: To listen carefully to other people’s arguments and try to see what they’re saying. And before you dismiss them, be mindful that they have thought about it. Be respectful of what their argument is and kind of unwind it carefully. He was ruthless about unwinding bad arguments, but he would do it in a way that’s funny and nonconfrontational. I try, but I’m not as good as he is.

COWEN: Abhijit, thank you very much.

BANERJEE: Thank you. That was wonderful.