What it will take to rebuild trust in our economy: A conversation with Nobel prize-winning economist Esther Duflo

MIT’s Esther Duflo likens her work as an economist to that of a plumber. Instead of relying on models and theories alone, she prefers to get her hands dirty, go out into the field and, through a combination of tinkering, educated guesswork and trial and error, stress test potential solutions to some of the world’s most stubborn challenges, like global poverty. 

Why not first test out how very small loans to low-income borrowers affects their earning potential in a few cities, for example, rather than commit millions to a project based mainly on good intentions? 

This approach to economics — which she has championed alongside her research partner and husband Abhijit Banerjee — sparked a sea change in the field, culminating in a 2019 Nobel Prize win for Duflo, Banerjee and Harvard’s Michael Kremer. Duflo, 47, is the second woman and the youngest person to receive an economics Nobel. 

LinkedIn recently spoke with Duflo about what the win means for her career, how she approaches her research and the thinking behind her latest book, “Good Economics for Hard Times,” which she co-wrote with Banerjee. Below is an edited transcript of the conversation. 

Congrats on the Nobel Prize. Did you have any idea this was coming? 

Esther Duflo: It was completely out of the blue, since in economics it tends to be older people. I was not even in contention, in my opinion. They keep the cards very close to their chests. There are murmurings, but they are nonsense. 

When you win something like this, does it mean you can breathe a sigh of relief, or does it mean the real work begins?

Neither. For most people, it comes much, much later in life, so this is kind of like when people have retired and it's kind of a crowning achievement, and then you can look forward to a life on the lecture circuit.

So, in your case, what comes next? 

So, the question is more, how do we leverage this to further not just our work, but the work of the movement? The way we think about the prize is that it's not just for the three of us. However scintillating our individual contributions might have been, the reason we can get a Nobel Prize is because of the field of development economics and also the policy towards poor people around the world have really changed. And that's not an effect of our work, singularly, but of a huge network of people who are doing this work.

How is the approach you and your colleagues take different from what came before? 

You take a big problem like poverty and what can be done about it, and you cut it into smaller, manageable pieces. For example, there are many things about being poor, one of them is you don't have access to a good education. So, then you start being interested in education. There could be any number of ways to address this problem. Remedial education, grouping people by ability, using computers so that people can follow their rhythm. And then once you have the question defined like that, then you can set up an experiment to actually learn about the impact of that proposed intervention.

So, it's taking economics and putting it into a laboratory-like setting?

The setting is the real world. You're not putting it in a lab setting, but it's applying the tools that you might apply in the lab. Knowing that it's much more complicated than in a lab, because you cannot fully control everything. In [one] experiment, one year there were massive riots. And one year there was an earthquake. Stuff happens.

What was development economics like before? 

Almost nobody was doing it. Maybe a lot of it was about trying to understand why some countries grow and some countries don't. But it's a very difficult question to answer, because countries grow, they stop growing, they start growing in ways nobody really understands. In 1989, The Wall Street Journal predicted that China was not going to go anywhere, unlike Zimbabwe.

Oops.

We can see how that particular prediction played out. 

When news that you won the Nobel came out, critics of your approach pounced. What do you make of the criticism? 

I don't find them particularly illuminating. One that I find particularly unhelpful is [the message that], "Oh, it doesn't get at the big questions." I, too, would like to have the cure for eternal youth, but in the meantime, what do we do?

Fair enough. 

Another criticism that has more teeth to it is the question of whether, if you run one experiment somewhere, you're learning something that you can apply elsewhere. Here it's been very helpful that it's [become] a big movement. So, you can see whether the lesson that you get from one context applies in another context.

For example, one of the big findings early on was that microcredit, making small loans to people, had a pretty disappointing effect on poverty. The first experiment was in urban India. At this point, we kind of kept it quiet, because we thought maybe it wouldn't apply elsewhere. But then, in the intervening years, there were about seven studies of microcredit in very different contexts, from rural Ethiopia to Bosnia-Herzegovina, Mexico, etc. All of them found the same thing, which is, for the average person, it really does nothing to take them out of poverty. 

People always want a magic bullet. But that gives you an example of why I don't have any patience for the big question business. Is it really worthwhile putting so much effort, so much resources, and so much money into a program that actually doesn't work?

At the same time, wouldn’t you say your new book tackles a ton of big questions?

Well, we are trying to apply the same approach to these big questions, like immigration and trade and growth and climate change and polarization and inequality. For example, on migration, a key question on migration is whether, when you have a lot of low-skilled migrants who are coming into a country, does it hurt the low-skill people in that country or not?

People have a strong intuition that it would, because more people, lower wages. That makes sense, it's kind of economics 101. But a lot of intuitions turn out to be wrong. And this one, in particular, is wrong. And that's not something we discovered, that's something that there is a huge literature on.

David Card did a very famous paper on the Mariel Boatlift, a wave of Cuban immigrants coming to Florida. And it was a big shock when he found no effect whatsoever on the wage of the low-skilled people. And people attacked his study, there was criticism, and a response to the criticism, and a response to the response. And then this type of study was replicated again, and again, and again, and again. 

So, now we have learned one thing. It's just one thing, but it's enormously important for policy, obviously. Because if people understood that, then I think their view on migration might be changed. 

You are presenting all of these facts to support an argument that that runs counter to many people’s feelings. But you also admit in the book that it's unclear how much facts matter to people in these kinds of cases. 

Yeah, and one of the reasons we wrote this book is to try and play a small part in changing that. It's kind of a human predisposition. We don't like to change our minds. People like to be consistent and it feels like admitting defeat somehow. So, one of the things that is very important to do, we feel, is to try and explain how we get to the conclusions we get at. It is not easy, because it takes a little longer than just asserting them. But if we manage to do that, of going through the argument, then we could start having conversations on this issue that are a little bit more reasonable.

You would need to change the political climate to get that to happen, wouldn't you?

I don't think we have any reason to be fatalistic about it. Sometimes, very strongly held positions that people have, you can actually nudge them out of it without very much.

How does that work? 

I'll give you an example that's from India. In India, people tend to vote for a party that's aligned to their caste. Which is not so different than in the US, people will vote Republican because they've always voted Republican, or vice-versa. If you always vote for someone from your caste, then there is no particular reason for a candidate to make an effort. You could be a criminal and people are still going to vote for you. So, that's terrible for politics. And, of course, it doesn't really encourage conversation, because you just root for your guy, you don't need to think of anything else. 

I was involved in an experiment where they ran puppet shows in villages and the message of the puppet show was, "Vote on issues, not on caste." They didn't tell them which issues they can be concerned about. They gave them no information. They just told them, vote on issue, not on caste. And that's sufficient to move about 10 percentage points of people from voting aligned with their caste to voting to another party.

A puppet show?

A puppet show. 

Alright. 

I mean, it's a nice puppet show. It’s a tradition. Here you wouldn't do puppet shows, we would do something else. 

One reason that they vote on caste is that they don't trust politicians to deliver on anything else. And in the US there is a bit of that, I think. In particular, the bottom 50 percent of the population in terms of income, their wages have remained stagnant, even as GDP was growing and inequality was increasing.

A lot of people feel that there is nothing in it for them in the political discourse, and so they are not really willing to engage in the political discourse. Therefore, very naturally, they will go and vote with whatever their group is. And I think that explains the tribalization of politics we are experiencing.

How do you change this dynamic? 

In the 2018 [midterm] elections [in the U.S.], in very red states, a lot of people voted for Obamacare extension, which is their economic self-interest against their tribal interests. Which means that people are perfectly capable of thinking when there are economic interests.

For, really, decades, since the 1980s, nobody has been concerned [about these people], telling them, "Oh, eventually it's going to go back to you. It's going to trickle down." And, actually, as it turns out, it's not trickling down.

I think it's very hard to have a conversation with someone who is persuaded, with some good reasons, that you despise them. 

This sounds like a much bigger project than just writing a book and telling people, "Here's new information."

Yeah, exactly. You can give them new information till they are blue in the face. That's not going to work. You need to reestablish legitimacy, for which I think you need to do something real, economically, for people who are displaced.

What would that involve?

Veterans have a GI bill. And we consider someone who is a veteran of war, not as someone who is lazy or whatever, but someone who is a hero. I think we need to take that image onboard and say, "Well, suppose we took the GI bill and modeled a program for people whose jobs got lost because of some disruption. Trade or automation." And suppose there was a GI bill for veteran of those disruptions. It would give people time by paying them extended unemployment insurance on off time. Second, it would give them tuition credit for very good places, like the GI bill. It would cost a lot of money, so it does require to raise the money somewhere. But people would immediately get something out of it. Then, I think you could talk. Once there is a little bit of trust reestablished.