You're listening to a podcast from the BBC World Service.

This is Hard Talk with me, Sean Lay.

Thanks for downloading this edition of the programme, and I hope you enjoy it.

Welcome to Hard Talk on the BBC World Service with me, Sean Lay.

We test the efficacy of drugs using randomised trials.

Why not test treatments for poverty the same way?

That's what Esté DuFlo and two of our colleagues

have been doing. As a result, they've demonstrated what's wrong with food aid and why the poorest kids

need more time than books and won the Nobel Prize for Economics to boot. But in focusing on what

works locally, are we in danger of missing the causes of inequality and of poverty?

Este de Flou, welcome to our hard talk.

In awarding the prize to you and your fellow economist, Abidette Banerji and Michael Kramer,

the Nobel Committee praised your experimental approach to alleviating global poverty.

How would you explain what you do?

First of all, you take a big problem, like, how are we going to eradicate poverty?

And you break it into manageable pieces, pieces that, you know, smaller questions, but questions

that admit rigorous answers. And then once you have one of those questions, you deploy something

which is very much like a randomized control trial

in medicine to test one approach against the other. So to give you an example, suppose that

you want to know how to motivate parents to take their kids to be immunized, you could set

up a randomized control trial where in some villages randomly chosen,

you work with a member of the communities that are going to mobilize parents to get their kids

...