In March, Abhijit Banerjee and Angus Deaton, two of the most brilliant development economists of our time, squared off at DRI’s Debates in Development conference.

Here’s the four-minute version:

For Banerjee, co-author with Esther Duflo of Poor Economics, development experiments are valuable because they force researchers to think rigorously about causality, and help create an agenda for learning.

For Princeton development economist Angus Deaton, identifying a causal effect is not so useful because it inevitably depends upon the interaction with other “helping factors.”  More generally, blind trust in randomized controlled trials (RCTs) leads to overconfidence and lack of sufficient scrutiny of  potentially bad evidence.

So that was the short, short version, but if you’ve got a little more time, we have got a longer version. Continue Reading