‘Randomistas’ who used controlled trials to fight poverty win economics Nobel

Abhijit Banerjee, Esther Duflo and Michael Kremer have been awarded the prize for their experimental approach to alleviating poverty.

Ewen Callaway

doi:10.1038/nindia.2019.137 Published online 14 October 2019

Esther Duflo (left), Michael Kremer and Abhijit Banerjee (right) applied techniques from the medical sciences to research on poverty.

© Eric Fougere/VIP Images/Corbis/Getty, Jon Chase/Harvard University, Saumya Khandelwal/Hindustan Times/Getty

Three economists who used randomized controlled trials to determine how best to lift people out of poverty and improve their health have been awarded this year’s Nobel prize in economic sciences.

The winners are Michael Kremer at Harvard University in Cambridge, Massachusetts, and Abhijit Banerjee and Esther Duflo, both at the Massachusetts Institute of Technology, also in Cambridge. Duflo is only the second woman, after Elinor Ostrom in 2009, to win the economics Nobel. All the winners of this year’s other Nobel prizes were men.

Kremer, Banerjee and Duflo are at the vanguard of the ‘randomista’ movement, which applies the methods of rigorous medical trials — in which large numbers of participants are randomized to receive either a particular intervention or a standard treatment, and followed over time — to social interventions such as improving education1, 2.

“Creating a movement is basically what they did,” says Dean Karlan, an economist at Northwestern University in Evanston, Illinois, who has worked with all three winners. “This is exciting for everybody and exciting for the world.”

The prize — worth 9 million Swedish kronor (US$910,000) — will be shared equally between the three laureates.

Beginning in the 1990s in rural Kenya, Kremer and his colleagues tested whether students perform better when their schools receive extra resources, such as textbooks and free meals, and found that these interventions did not improve learning outcomes. Separate trials in India, led by Banerjee and Duflo, showed that targeted tutoring did help children with special educational needs in the schools that received that intervention.

Global health interventions have been heavily influenced by the work of the randomistas. In one influential study, Kremer’s team found that parents in low-income countries are much less likely to give their children deworming pills when the medication is heavily subsidized than when it is available for free2.

And a 2015 study of more than 10,000 households in six low- and-middle-income countries authored by Banerjee, Duflo and others, found that giving poor people a two-year aid package — including cash, food, health-care services, skills training and advice — improved their livelihoods for at least a year after the support was cut off1.

The winners’ findings have “dramatically improved our ability to fight poverty in practice”, said the Nobel committee in a statement. The committee noted that “as a direct result of one of their studies, more than five million Indian children have benefitted from effective programmes of remedial tutoring in schools”.

Karlan says that randomized trials can offer clarity and objectivity that other forms of evaluation do not – but they’re not easy to do. “They require a lot of work on the ground,“ he says. Nor are they a silver bullet, he adds. “There’s a set of questions for which this is the right tool. It’s not the right tool for all questions.”

This article was first published in Nature.


1. Banerjee, A. et al. Science 348, 1260799 (2015) Doi: 10.1126/science.1260799

2. Kremer, M. & Miguel, E. Q. J. Econ. 122, 1007–1065 (2007) Doi: 10.1162/qjec.122.3.1007