A prize for evidence-based policy

Econ Nobelists transformed policy evaluation

For the second time in five years, the Royal Swedish Academy of Sciences has awarded the Nobel Memorial Prize in Economic Sciences to researchers in the field of development economics. Angus Deaton, who won in 2015, has been joined this year by Abhijit Banerjee, Esther Duflo, and Michael Kremer. But the three economists this year have a very different approach to Prof. Deaton’s, although they look at similar questions of poverty, inequality, and welfare. The latter won for work that uses large survey data sets to make useful but arguable inferences. This year’s laureates, on the other hand, have reduced the same questions to smaller, bite-sized pieces about which something certain can be said. The hope is that coming out with more rigorous results will allow for evidence-based policy to take hold in developing economies.

The basic approach is similar to the clinical trials performed in medicine: A randomised group is exposed to the “treatment” of the policy being investigated, and another group is not — they are, in other words, served a policy placebo. Once the policy has been administered, then the effects on the two groups are compared. If it is statistically clear that there is a greater difference in the desired outcome for the experimental group than the control group, then a clear inference can be drawn that it is due to the effects of the policy. Since the two groups were randomly constituted, no other factor is likely to be responsible for a difference in outcomes. This approach is extremely important in contested and ideological fields like development economics, where policies could be attacked for merely enhancing inherent advantages of a particular group, or contrariwise pushed forward in spite of not having any real merit. Randomised control trials (RCT) have, over time, become dominant in the field of development economics — much to the annoyance of “big picture” econometricians and theorists like Prof. Deaton.

The impact of such thinking on the Indian state must be acknowledged. Some influential recent policy innovations in this country, from Aadhaar to reform of the public distribution system, have been pioneered following RCT studies. This Nobel would not, indeed, have been won without the co-operation with the three laureates and others in their field of multiple individuals and agencies of the Indian government at both the Union and state levels. It is easy to see why RCT-based analysis is so appealing in the Indian context. Bureaucrats are inundated with more proposals than they have the time, energy, or resources to implement. Proposals that are accompanied with a clear proof of their worth are, therefore, likely to be prioritised. But it is also true that evidence-based policy-making has not spread as much as it should have. In fact, one common criticism of RCTs is that governments have been reluctant to scale them up. In the education sector in particular, only one or two interventions emerging from RCT studies have been adopted at scale by the Indian bureaucracy. It is to be hoped that the Nobel awards will cause many more in the policy hierarchy to sit up and take note of what the academics have been producing.