Finding the pulse of the poor
Armed with data, an MIT lab offers fresh insight on some of the world’s most vexing problems
CAMBRIDGE - It’s no one’s idea of an MIT laboratory: not a beaker or an oscilloscope in sight. But in a wood-paneled suite, on the third floor of a bland, concrete building, researchers are tackling problems as complex and vexing as any in technology, science, or medicine.
This is the Abdul Latif Jameel Poverty Action Lab, or J-PAL, where economists through precise, detailed studies are trying to find ways to alleviate poverty. For nearly a decade, MIT economics professors Esther Duflo, and Abhijit Banerjee, have worked with a global network of researchers to conduct experiments in the world’s poorest places - where families live on less than $1 day - and reached conclusions that are changing the way economists and policy makers think about development in impoverished areas.
The findings are contained in their new book, “Poor Economics,’’ which earlier this month won the Financial Times and Goldman Sachs Business Book of the Year Award for “the most compelling insight . . . into modern business issues.’’ In the book, the MIT professors argue that antipoverty policies must be built on evidence from careful, controlled tests that detail how the poor live, act, and react.
For example, Duflo and Banerjee set a goal of increasing immunizations of children, so they divided the people of several villages into two groups. One group was offered free monthly immunizations at a mobile camp; the other was offered the same immunizations, but parents were given one bag of lentils per immunization.
The study concluded that small, nonfinancial incentives not only dramatically increased immunization rates, but also made such programs cost-effective by lowering the cost per child immunized - even considering the price of lentils.
“The prominence and enthusiasm of Duflo and Banerjee have helped make development economics one of the most compelling areas in economics today,’’ said Christopher Udry, an economics professor at Yale University. “They’ve advanced the idea that it’s possible to evaluate small interventions at the community level and they’ve set up a tool kit [of techniques] to make that approach more accessible to economists around the world.’’
Duflo and Banerjee focused their studies on the poorest of the poor when many other economists were concentrating on Wall Street and financial markets. The MIT professors and their network of 59 affiliated professors have conducted about 275 narrowly focused studies in 49 countries to measure how the poor respond to different aid strategies, subjecting antipoverty policies to the same rigor that pharmaceutical researchers use to test new drugs.
Their findings have reached beyond broad generalizations to show that the poor do not always act in ways that conform to the views and programs of well-meaning relief agencies, revealing a portrait of poverty as complex as the movements of financial markets. Why do families without enough to eat choose to buy a television instead of food? Why do antimalaria nets go unused, even when distributed free? Why are bags of healthy grain often not the best response for regions starving for food?
“We don’t think poverty is one problem, it’s 300 separate problems,’’ said Banerjee, in a recent interview with Duflo. “The reason many poverty problems don’t get solved is because everybody wants them to be one problem.’’
Duflo and Banerjee came to the study of global poverty from very different starting points. To Duflo, who grew up in a comfortable, academic family in France, global poverty was a remote abstraction. For Banerjee, abject poverty was as close as the ramshackle houses behind his childhood home in Calcutta. They met at MIT when Duflo, as a student, took a course in development economics from Banerjee. After Duflo earned her doctorate in 1999 and joined the MIT faculty, the two founded the Poverty Action Lab in 2003.
Since then Duflo, 39, and Banerjee, 50, have received multiple honors and prizes. In 2010, Duflo was awarded the prestigious John Bates Clark Medal for the best American economist under 40. The year before, she received a MacArthur Fellowship, the so-called genius award.
What inspires their interest in poverty, Banerjee explained, is the idea that the poorest people must deal with different conditions than the rest of the world’s population.
Duflo said the gap in the quality of life between the US middle class and the poor in a country such as Kenya, where annual per capita income is $770, is so startling that it is a moral imperative to try to close it.
“How you can bring that person in Kenya anywhere closer to the quality of life of us, for example, seems like a primordial question,’’ said Duflo. “It’s hard to imagine that there is a more obvious thing to study. Why study anything else?’’
Today, the poverty lab is a global operation. The staff at MIT consists of 29 employees in research, training, policy, and administration. Affiliated professors are scattered around the world, at universities in India, France, Chile, and South Africa.
The budget of the lab was more than $10 million in 2010. Major benefactors include the Bill and Melinda Gates Foundation, William and Flora Hewlett Foundation, Nike, and the John D. and Catherine T. MacArthur Foundation.
As the number of lab studies approaches 300, creating a body of data to help identify effective antipoverty programs,Duflo and Banerjee hope this work will exert a larger influence on policy makers around the world.
It already has. For example, research by lab affiliates showed that school-based deworming, which removes parasites that can harm the health and learning abilities of children, is one of the most cost-effective methods of improving school participation. The evidence from that study helped convince governments and nonprofit development groups in 26 countries to adopt school-based programs that have dewormed - and helped educate - 23 million children.
“Because of the Poverty Lab, we now have a much richer empirical understanding of the effects of antipoverty policies,’’ said Yale’s Udry. “Duflo and Banerjee have advanced the idea that economists can say very specific things about very specific programs to policy makers.’’
Duflo said she believes the lab’s influence will expand as it generates more definitive studies of poverty programs and more hard data on their effectiveness.
“Five years ago, we didn’t have many results to share, but now we have more lessons to discuss,’’ she said. “Come back in two years, and we’ll have even more to say.’’