Experiments at heart of this year’s prize
India-born Abhijit Banerjee, his fellow MIT professor and wife Esther Duflo and Harvard’s Michael Kremer were awarded the 2019 Nobel Memorial Prize in Economic Sciences on Monday “for their experimental approach to alleviating global poverty”. HT explains the significance of their work and its influence on the world.
What makes the Nobel Prize to Abhijit Banerjee, Esther Duflo and Michael Kremer unique?
Economics is considered to be a part of the social sciences, where, unlike in the physical sciences, theories and concepts cannot be tested for validity. Banerjee and his co-awardees winning the Nobel changes that because their work has argued that it is possible to design policies after having tested different versions of them in controlled conditions. They conduct something called Randomised Controlled Trials (RCTs), a technique commonly used in testing medicines on patients, to test whether tweaking an existing policy can deliver significantly better results. If the RCTs show positive results, Banerjee et al argue that they should be implemented on a wider level. They also argue that all policies to fight poverty should be decided on the basis of RCTs. Given the fact that their approach has received the highest recognition available in the discipline, it is likely to significantly boost the acceptability of this approach not just in academia, but also at the policy level. This is asking for the primacy of experiments, earlier associated with the pure sciences, in a social science discipline.
What exactly is an RCT?
An example which Duflo used in her press conference at MIT after the Nobel Prize announcement on Monday can make this clear. They conducted an RCT to find out why immunisation rates were poor in Rajasthan. The perceived reason was that irregular functioning of immunisation centres. The RCT checked this by randomly picking two sets of villages and ensuring regular functioning of the centre in one set and giving a small quantity of lentils at the beginning of the programme and set of steels utensils at the end of the programme to the parents in another set. Their one year long experiment showed that the increase in immunisation was significantly greater in the second set of villages compared to the first. They use these to argue that it is not enough to fix irregular functioning of immunisation centres in order to fight low immunisation rates. Jamal Abdul Lateef Poverty Action Lab, which has been set up by Banerjee and Duflo at MIT, is involved in almost a thousand such experiments across the world.
What’s the philosophy behind this approach?
Banerjee and his co-awardees believe that because anti-poverty programmes are almost always designed by the elite, they have little idea about the actual preferences and constraints faced by the poor. This might lead to a situation in which even well-intentioned policies might not be able to achieve their objectives. Their argument: designing policies on the basis of RCTs can help eradicate such errors born of bias.
Does everybody agree with their approach?
Many economists are extremely critical of the virtue Banerjee and his co-awardees make of RCTs in fighting poverty. This includes Angus Deaton, who was awarded the Nobel Prize in Economics in 2015. In a co-authored 2016 paper, Deaton had very strong views on the usefulness of RCTs in economics. “We suspect that a large fraction of the published results of RCTs in development and health economics are unreliable,” the paper said. The paper listed two main problems with relying on the results from RCTs for policy making: their inability to capture all the differences between the two groups which were subjected to different versions of the policy and the belief that what worked in a particular situation would always work in other places.
Does this mean we should reject RCTs completely?
Not necessarily. Jean Dreze, an eminent development economist, has argued for seeing RCTs as a complementary tool to policy making rather than being the sole driver of policy making. His comments in a 2018 article are reproduced below.
“In short, I feel that economists need to be cautious and modest when it comes to giving policy advice, let alone getting actively involved in ‘policy design’. Their expertise and research can certainly contribute to more informed policy discussions and public debates. But if they give advice, it is best done as concerned citizens rather than plumber-like economists, in collaboration with others from different disciplines and walks of life. In the field of social policy, at least, I see no reason to privilege the advice of economists.”
None of this detracts from the value of RCTs (correctly understood), or from the case for evidence-based policy. If the idea is to bring more evidence to bear on public policy, there is much to be said for it. This endeavour, however, is likely to be all the more useful if we bear in mind that evidence involves more than RCTs, understanding more than evidence, and policy more than understanding.