The Nobel Prize in Economics for 2019 has been awarded to Abhijit Banerjee, Esther Duflo and Michael Kremer for “their experimental approach to alleviating global poverty”. The approach, popularly known as Randomised Control Trial (RCT), has been the buzzword among development economists for almost two decades. Banerjee, Duflo and Kremer have used this technique (inspired by the use of RCTs in medical science) to test the effect of small interventions on individual behaviour.
Most of these interventions carried out under the aegis of Abdul Lateef Jameel Poverty Action Lab (J-PAL), co-founded by Banerjee and Duflo, in Africa and Asia, have produced evidence on the response to a particular intervention by the poor using these randomised trials. The approach basically examines the impact of these micro interventions by treating one set of individuals/households and comparing the outcome with another set of individuals/households, which are similar in all other respects but have not been treated with the intervention. India has been among the biggest laboratories of these experiments with several experiments on diverse themes such as literacy, nutrition, health, micro-finance and so on.
The RCT approach has its share of supporters as well as critics. While it has enamoured a large number of development economists for its simplicity, where inferences on what works or not are drawn from field experiments, it has also been criticised for reducing the study of poverty to small interventions unconnected to the lived experiences of the poor. The discomfort among many established scholars is that this fashionable trend has made the historical, institutional and social structures of the persistence of poverty less relevant to understanding why the poor continue to remain poor. Others have picked holes in the methodology. However, it has not deterred development economists from using this approach for designing experiments and conducting them to understand how the lives of poor people change as a result of these micro interventions. There have been questions about whether the results can be replicated in different societies, as well as on the ethics of some of the experiments, which have been conducted in collaboration with participating governments. It is also worth pointing out that the method is as good as the range of interventions that can be undertaken.
While critics may have been unfair to RCTs in some respects while correctly pointing out the pitfalls in an RCT-based approach, there is no denying that all the three scholars have contributed a great deal to putting poverty and development economics back on the agenda of economics. Newer methods and approaches are necessary for the discipline struggling to find relevance in an increasingly complex world, which is as much defined by the microeconomics of small interventions as well as the macroeconomics of development such as government policy and structures of production. As Angus Deaton (Nobel Prize winner of 2015) says: “RCTs can play a role in building scientific knowledge and useful predictions but they can only do so as part of a cumulative programme, combining with other methods, including conceptual and theoretical development, to discover not ‘what works’, but ‘why things work’”.
RCT has become almost like a movement, encouraging many young economists (sometimes called “randomistas”) to visit rural areas and observe the lives of the poor. It may not have had any credible and long lasting impact on the lives of researchers and the population studied, but the fact that so many young economists are immersing themselves in the lives of the poor and trying to understand poverty is itself an achievement. More so at a time when economics has often been criticised for being far removed from reality.
The other achievement, although not necessarily for the better, has been the attempt to give scientific colour to the discipline of economics through the use of evidence generated from these experiments. It certainly has convinced many governments to use facts and evidence in policy prescriptions and induced a degree of caution while introducing new interventions. Even in India, there is evidence of RCTs contributing to improvements in financial management and flow of funds for various government programmes including in the field of education.
While it would have been good if RCTs could predict the effects of demonetisation on the lives of the poor, it is also a reality that most such decisions are not contingent on evidence based on hard facts but on the whims and fancies of the government of the day. Despite the tentative nature of much of this evidence, there is no denying that policy interventions do require better facts and evidence for efficient outcomes. This is true not only for evidence generated by RCTs, but also data generated by our statistical systems including the National Sample Survey (NSS).
Incidentally, both Kremer and Banerjee did their PhD work at Harvard University. Banerjee had completed his MA in economics from the Centre for Economic Studies and Planning (CESP), JNU, before proceeding to Harvard for doctoral studies. Banerjee supervised Duflo’s doctoral work at MIT. While both Banerjee and Duflo remain engaged with research in India, Kremer was one of the first to use these experimental methods and look at micro-interventions to examine their impact on poverty. The Nobel recognition will hopefully encourage more rigorous work on some of the long-standing problems of development economics, including on poverty and social mobility. Hopefully, it will spur our own government to take data and evidence more seriously.
This article first appeared in the print edition on October 15, 2019 under the title ‘An economics for the poor’. The writer teaches economics at JNU.