How these 2 economists are using randomized trials to solve global poverty

Read the Full Transcript

• Judy Woodruff:

  This year, a trio of economists were awarded the Nobel Prize for their work to alleviate global poverty. Their research helped more than five million children in India benefit from remedial tutoring in schools.

  Tonight, Paul Solman talks with two of those winners, a husband-and-wife duo.

  It's part of our series Making Sense.

• Paul Solman:

  MIT's Abhijit Banerjee and Esther Duflo, the first married couple to win the Nobel Prize in economics.

  Duflo, 47, is also the youngest economics laureate ever and only the second woman to receive the prize. They met in the mid-'90s, when Duflo, then a grad student, took Banerjee's course on economics and poverty.

  And she says:

• Esther Duflo:

  I was going to study development, no matter what happened.

• Paul Solman:

  Development to help poor people with data?

• Esther Duflo:

  Exactly, and link sort of careful thinking, not go with your intuitions, because our intuitions are often wrong.

• Paul Solman:

  Or they have been taught to you in economics classes.

• Abhijit Banerjee:

  By authority figures.

• Paul Solman:

  By authority figures.

• Abhijit Banerjee:

  The issues are fundamental.

• Paul Solman:

  Banerjee was such a figure. But economic theory, his forte, was totally divorced from the Mumbai neighborhood in which he grew up.

• Abhijit Banerjee:

  I played with the kids from the slum all the time. And I think, over time, I started to realize that what I was doing could be connected with my previous life.

• Paul Solman:

  Previous life as an economist, you mean?

• Abhijit Banerjee:

  No, previous life as a little boy playing with other little boys who didn't go to school. And, mostly, I think I was always a little bit conscious of the fact that the economics that I practiced were not necessarily always deeply connected.

• Esther Duflo:

  Well, here's a thought experiment for you.

• Paul Solman:

  Thus began their rigorous approach to combating poverty, testing policy solutions through randomized controlled experiments, the way new treatments are tested in medicine.

• Esther Duflo:

  It's not the Middle Ages anymore. It's the 21st century. Randomized controlled trials have revolutionized medicine by allowing us to distinguish between drugs that work and drugs that don't work. And you can do the same randomized control trial for social policy.

• Paul Solman:

  How to improve the dreadful state of schooling in India, for example, at the lowest cost.

• Esther Duflo:

  So you could think of any number of solutions to address this problem, giving more textbooks, cutting class size, giving incentives to teachers.

• Abhijit Banerjee:

  Once you think in the scale of the Indian school system, these are massive resource implications, 600,000 schools, so that's — it's not cheap.

  So you want to figure out what exactly you need to do. Can it be done within the school system, with normal teachers, in the normal teaching hours?

• Paul Solman:

  And what emerged from the experiments was a simple cost-benefit conclusion: Teach students what they don't know in dedicated classes, rather than one size fits none.

  But the team isn't known only for finding out what works, but what doesn't, in a word, debunking.

• Abhijit Banerjee:

  One of the places where we partially debunked was microcredit. Microcredit was kind of the flavor of 2000.

• Paul Solman:

  Was it ever, and well before 2000.

• Robert MacNeil:

  Microlending, small loans to small entrepreneurs.

• Paul Solman:

  I did a "NewsHour" story in 1994 in Pine Bluff, Arkansas, where entrepreneurs like hair weaver Jessie Pearl Jackson were having trouble getting bank loans.

• Jessie Pearl Jackson:

  Here, you have to have money to get money. Then you don't need the money anyway. So I don't understand the banking process. If you don't have it, you don't get it.

• Paul Solman:

  Instead, she applied for, and got, a $7,000 loan and business advice from a nonprofit based on the idea of Nobel Peace Prize winner Muhammad Yunus.

• Muhammad Yunus:

  To create a job, I need money. And banks will not lend me money. Once you have a micro-enterprise coming up, you are allowing a person to show his work and her work.

• Paul Solman:

  It sounded great, looked great, but this was anecdata, based on a tiny sample. The randomized controlled trials in India were anything but.

• Abhijit Banerjee:

  So, these were 104 neighborhoods in the city of Hyderabad; 52 were going to get microcredit now, 52 will get in two years.

  And we compare the places which got microcredit with ones which didn't. And we found that, on average, it did nothing for the earnings of the people who lived there. Didn't get richer.

• Paul Solman:

  You realize, of course, that, for me and for our audience, this is an extremely depressing result.

• Abhijit Banerjee:

  Yes. It was extremely depressing for us too.

• Esther Duflo:

  But for the few people who already had a business before, there is actually a positive effect.

• Abhijit Banerjee:

  I think this presumption that it's going to win, win, win, win, win is what the problem was. It was oversold.

• Paul Solman:

  Also controversial are the economist couple's randomized trials on local politics.

• Abhijit Banerjee:

  We focus on the question of, how do we get voters to be responsive to the fact that this politician isn't doing his job?

• Paul Solman:

  All they actually did, publicize politicians' voting records in local newspapers.

• Abhijit Banerjee:

  When you ask poor people in poor neighborhoods, 2 percent say, we want roads, just 2 percent; 57 percent say, we want drains, sewers, et cetera.

• Paul Solman:

  This politician voted for drains and sewers. This one voted for roads. And, in fact, people begin to vote for the drain person.

• Abhijit Banerjee:

  Over the roads person.

• Paul Solman:

  Once they see in a newspaper which way they voted.

• Abhijit Banerjee:

  Exactly.

• Paul Solman:

  Another intervention was just to let the locals know that what politicians do matters, informational theater.

• Esther Duflo:

  Street plays, street actors. And we see people deciding, hey, I'm going to try this out.

• Paul Solman:

  More candidates for office…

• Esther Duflo:

  Yes.

• Paul Solman:

  … in the places where the…

• Esther Duflo:

  This was done.

• Paul Solman:

  … play was staged…

• Esther Duflo:

  Yes, exactly.

• Paul Solman:

  … as opposed to the places where it wasn't.

• Esther Duflo:

  And the second consequence is that the incumbent got fewer votes.

• Paul Solman:

  The incumbent got fewer votes?

• Esther Duflo:

  And, even more importantly, the worst incumbents are the ones that get completely clobbered, completely clobbered.

• Paul Solman:

  Clobbered.

• Abhijit Banerjee:

  They get zero.

  So the next intervention we did is, we told people, two years from now, we're going to put out a report card on you, OK? And, indeed, when you do that, they start building those drains or whatever they want to do.

• Paul Solman:

  Who would have guessed?

  But that's what the Duflo/Banerjee research is all about, trying to reduce the guesswork of economic development policy by seeing what seems to work, and what doesn't, at least in its current form.

  For the "PBS NewsHour," Paul Solman in Boston, Massachusetts.