by Atanu Sengupta and Sanjoy De
Introduction
The conferment of the 2019 Nobel Prize in Economics to Abhijit Banerjee, Esther Duflo and Michael Kremer is an accolade to the randomised control trial (RCT)1 method used by these economists to understand and alleviate global poverty. The Prize Committee noted that “in just two decades, the pioneering work by this year's Laureates has turned development economics—the field that studies what causes global poverty and how best to combat it—into a blossoming, largely experimental field.”2 The RCT approach of rethinking of development economics has provided outcomes of policy- intervention in the areas of health, schooling and credit. For instance, researches by this new crew of development economists have found that the expenditure made by the poor people in preventive health care is highly sensitive to the costs of the health products and services. This amplifies the argument for munificent subsidisation of preventive healthcare invest-ment. (Miguel and Kremer 2004; Kremer and Miguel 2007; Banerjee, Duflo, Glennerster and Kothari 2010) Again, this refutes the long-standing idea that the micro-credit programme has strong developmental connections (Banerjee and Duflo 2014).3 In education, Duflo, Dupas and Kremer (2015) showed that the students' learning can be improved by hiring contract teachers. The increasingly higher application of this experimental method amply testifies the transformation phase that the discipline of development economics is passing through.
Basics of RCT
Despite considerable efforts in the past few decades to alleviate it, global poverty, in various forms and magnitudes, still remains a serious problem. Presently, over 700 million people live their lives on abysmally low incomes. From the development perspective, measurement of poverty is very important, if not more important than the efforts towards poverty eradication. The main focus of their research is poverty eradication and development through the experi-mental method. According to them, poor people are not identical. Their perceptions towards well-being and its prerequisites are different from each other. The first and foremost criterion to solve their problems is to directly reach out to them. Poor people, whose problems are to be dealt with, have to be given the utmost importance. The RCT method is different from the theoretical or empirical research in this respect. Paying heed to those who are directly problem-stricken is the main tenet of democracy. This resembles John Stuart Mill's concept of classical democracy, which is built around people's larger participation and parleys.
The application of RCT as a research tool was particularly limited to the fields of bio-medical sciences. In the clinical trials, the efficacy of various drugs is tested using this method. Since the beginning of the 1990s, Banerjee, Duflo and Kremer have been applying this technique in the field of economics. Kremer (2003) applied this technique in Kenya to find the effectiveness of free meals and textbooks on children's learning at school. Similar types of experiments were undertaken by Banerjee and Duflo (2007, 2009) in various parts of India, and they popularised the concept of RCT through their book, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty, published in 2011.
In RCT, in order to judge the efficacy of a particular policy, two sets of individuals—‘treatment group' and ‘control group'—are chosen in a statistically unbiased manner. Then the policy is applied to the treatment group whereas no such intervention is initiated on the control group. In other words, the control group continues to remain as it was earlier. After a certain interval, it is checked whether the condition of the treatment group has changed and to what extent. For example, in order to judge the effectiveness of the vaccination programme, Banerjee, et al. (2009) carried out RCT in some villages of Udaipur, Rajasthan. The programme offered one kilogram of pulse to the parents for single vaccination of their child and the gift of a stainless steel plate for the completion of the entire vaccination course. It has been found that there has been a substantial increase in the vaccination programme through this incentive method. Again, to find out the factors responsible for educational outcomes, Kremer (2003) conducted a series of field experiments in tandem with a Non-Government Organisation (NGO) in western Kenya in the mid-1990s. These experiments were instru-mental in separating the different components in the educational production function.
The brilliance of the RCT experiments lies in the fact that it helps the researchers to isolate the individual impact of a factor on the total outcome. It is believed by many economists (Kremer and Miguel 2007; Ghatak 2019) that RCT can help assist governments and policy- makers to determine, in a very scientific manner, the most appropriate policy-measures that could help alleviate poverty in a rapid way.
Some Questions
Economics is not a physical or laboratory science. Rather, it is a human science. Again, development economics, which is a key branch of this societal science, deals with the problems of the poor people and their possible solutions in a compassionate, humane and ethical way. Clinical trial-type experiments—which have wide accept- ability in some natural science— may not have similar type of suitability in economics. In pursuits of RCT-based experiments, doling out of gifts or money, to a particular group and offering no such incentive to the other group or withdrawal of money or gift at the end of the experiment, keeps open the possibility of infringement of the ethical standards and this amounts to the assault on human dignity.
Secondly, in empirical research, the casual effect of a variable on the outcome can be statistically ascertained. Like in case of observational science (say, astronomy), in the RCT method, the outcome of a particular intervention can be observed. But, RCT cannot determine the underlying causes of the nature of outcomes due to a certain policy-intervention. Again, RCT cannot say whether a particular intervention, having positive influence, is the best possible policy or not. In other words, the policy, observed in the RCT method, which is in use, may not always be the best policy. So, there is a chance of adoption of a sub-optimal policy with the possibility of loss of welfare, under the RCT experiment.
Secondly, Deaton and Cartwright (2017) argued that picking a group in a random manner for an RCT experiment, does not imply that the group is identical with other possible groups. Though there is a chance of two groups to be identical, there is also a high chance of the two groups to be widely different. This problem of the dissimilarity of the treatment groups puts a question-mark on the outcome of the RCT exercise. Basu (2018) opined that it would be wrong to treat RCT as the ultimate determinant of the efficacy of any intervention. He further added that RCTs can actually be misleading. To determine the efficacy of actual policy-interventions in the future, application of intuition is the key.
Finally, this aid-based RCT experiments may have limited implication for the long-term improvements of the poor. Probably, sustainable poverty alleviation requires a total structural and macroeconomic change in a country. The state has to play a pivotal role in the long-term and permanent poverty eradication. Moreover, this incentive-based research programme is a very costly affair.
Recently, a group of 15 renowned economists, including two Nobel Laureates, in an open letter to the Guardian Online, stated that the aid-based RCT programmes will result in short-run, superficial and misplaced policies. They are concerned about unabated global poverty, deepening hunger as well as environmental and ecological crises. They believe that the RCT-driven solutions are not likely to produce the same outcomes in all the regions.
Basic Issues
So far we have discussed the common perceptions of this year's Nobel Prize. However, it seems to us that the insight may go still further and reveal the metaphysical network embedded in the tapestry which is visible to us all. The Nobel Prize has been awarded to Banerjee, Duflo and Kremer on the initiation and utilisation of RCT. The RCT, on its outside, is merely a new methodology that is developed to answer the old question of poverty. It seems that the prize has been allotted to the develop-ment and fashioning of a new methodology. All the debates—for and against—are based on the application of this new methodology. This type of visualisation misses the core point that is repeatedly cited in the book Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. Banerjee, in many interviews, stated the relevance and implications of poor economics. That is conveyed in his book also. Banerjee and Duflo argued in their book that in the common nomenclature, poor caricatures are a cartoon type character on whom some well-thought-out policy arising out of grand theories of economics are applied. Some of them are ‘pro-Left' (public policy, nationalisation) and some are ‘pro-Right' (liberalisation, operation of the free market). Banerjee and other associates feel that this is not an appropriate approach. The poor have their own lives, culture, beliefs and utility. It is necessary to learn from them and acquire it into the knowledge of development economics. For example, giving more to a poor will not raise his consumption of food that is so much necessary for his ill-nourished life. Similarly, it is not used for increased health or educational expenditure that might enrich his/her life. Rather, the life of a poor is boring. (Banerjee and Duflo, 2011) So, he/she spends the extra money in buying a second-hand TV or devouring liquor or any other item that makes his/her life more interesting. The policy of money-transfer is useless here.
The present authors have seen a similar type of situation in a poor village of Purba Burdwan. The government was giving wheat in the fair price shops. Many commentators were against this because wheat is not a staple food of the Bengalees. However, in that village, an old widow sold wheat in the market to earn her livelihood. This situation is in complete contrast to the commentators. Thus, the perception of grand theories would not catch the lesson that we can gain from learning the lives of the poor. This is the essence of the poor economics that Banerjee and others try to permeate.
The question is: how can we learn from the lives of the poor? The answer is simple—to live with them. Kremer has claimed that his experience as a teacher in Kenya has helped him a lot to know about their life. Banerjee and Duflo claim that setting up of a poverty lab has helped them to interact with the lives of the poor. This has become a sizeable experience apart from the classroom and text-bookish knowledge from the echelon of learning.
This concept of addressing the problems of the poor is not new. Chambers (1999), in his article on poverty, claimed that the definition of poverty is highly localised and culture-specific. To put them in a blank envelope, in order to quantify them is a gross dis-service to the poor. Giving an example of the district of Birbhum in West Bengal, Chambers showed how a social network developed by the then ruling party has helped the so-called poor to live a more meaningful life than what the common theory prescribes.
In a way, Chambers thus proposed the path to understand poverty from the local point of view. Later many studies were developed by sociologists (and some economists) in the pathway of understanding the culture of poverty. In fact the famous work by Professor Pranab Bardhan (1989), Conversations Between Economists and Anthropologists: Methodological Issues in Measuring Economic Change in Rural India, discusses this very issue. Professor Banerjee and his co-Laureates used this signal in a more compre-hensive and neater framework.
The RCT was developed by this methodology to understand the poor and his/her world from the perspective of a development economist. However, the Nobel Committee, in its citation, granted this Prize not to ‘poor economics' but to the RCT. This is strange since RCT is a mere tool of understanding the culture of the poor. It is in this strangeness that the dilemma and strength of Banerjee's work has been revealed.
Economic diaspora is still clogged with the panorama of the grand narrative. In their earlier work, both Banerjee and Kremer have contributed to the development of this grand narrative. However, the narrative failed to remove global poverty as the fifteen economists in their letter in the Guardian have stated. Their argument against RCT is basically an argument in favour of the grand narrative and the power of knowledge as Michel Foucault had long ago cited. Banerjee and his associates tried to remove this deficiency of the grand narrative by appealing to the localised poverty culture. In a sense, thus, it is a great breakthrough in the field of economics. The Nobel Committee, in its citation, meticulously mentioned how Banerjee's work deviates from the standard economic theory which won Nobel Prizes earlier. His work challenges the hegemony of economic grand narrative. It forces us to think in a new way.
But, herein lies the problem. If the grand narrative is utterly broken, the power-centre of knowledge-creation will collapse, which will be truly democratic. However, it cannot occur in a society ruled by knowledge alike. The RCT is thus a tool of translating the poor's culture in the language of the powerful.
Thus “the poor are like hedge-fund managers—they live with huge amounts of risk. The only difference is in their levels of income.” (Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty) The authors have compared the most hapless creature of humanity to the most powerful people in the world. There is practically no similarity between the likes of hunger, ill-health, ill-education, social and economic exploitation, harassment and the exclusion that a poor in a Third World economy faces to the rosy, flowery lives of the hedge managers. Yet, this is the only way in which the powerful ‘understands' the powerless. Banerjee and Duflo themselves found evidence of this. Self-help groups can help individuals to rise above the poverty line but cannot make them successful entrepreneurs. The Nobel Laureates even faced life-threats by bringing this truth to light.
Nonetheless, the Laureates made a compro-mise between living in a world of grand narrative and garnering localised knowledge. The Nobel Committee also faced this dilemma. It repeatedly reiterated that the Laureates emphasised microeconomic application to development economics. Any student of economics knows the problem of ‘Sophie's Choice' that a poor man faces. In the classic story of Sophie, she was arrested by the German Gestapo because of her husband being a Communist. Identified as a non-Jew by a camp officer, she was given the choice of selecting one of her two children as the survivor. The other is automatically put to the gas chamber. This dilemma of a prisoner at a German base camp during World War II is the daily dilemma of the poor. He/she has to chooses between clothes and food, both of which is at their bare minimum. He/she choose between empty stomach and full clothes or full appetite and nakedness. This is a cruel choice or no choice at all. This argument is given by Eswaran and Kotwal (1994) in their book, Why Poverty Persists in India: A Framework for Understanding the Indian Economy. They stated that the poor man's preference is lexicographic thereby making the whole consumer choice based on continuous utility as irrelevant. Thus, the Nobel Committee points to the fissures of the grand narrative and simultaneously tries to mend this by the relevance of the current economic theory. On the whole Banerjee's poor economics does not get the prize but that goes to the RCT which translates poor economics to the economics of the rich. Yet, by questioning the grand narrative, Banerjee and his associates have shaken the discipline of economics from its long slumber like that of the legendary Kum-bhakarna. We have to congratulate Banerjee and his associates for this grand achievement.
References
Banerjee, Abhijit; Deaton, Angus; Duflo, Esther; Glennerster, Rachel; Kothari, Dhruva, 2009, “Udaipur Health and Immunisation Studies”, Harvard Dataverse, V13.
Banerjee, Abhijit and Esther Duflo, 2007, “The Economic Lives of the Poor,”Journal of Economic Perspectives 21(1): 141-167.
Banerjee, Abhijit and Esther Duflo, 2009, “The Experimental Approach to Development Economics”, Annual Review of Economics 1: 151—178.
Banerjee, Abhijit and Esther Duflo, 2011, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. New York, NY: Public Affairs.
Banerjee, Abhijit and Esther Duflo, 2014, “Do Firms Want to Borrow More? Testing Credit Constraints Using a Directed Lending Programme”, Review of Economic Studies 81(2): 572-607.
Banerjee, Abhijit, Esther Duflo, Rachel Glennerster and Dhruva Kothari, 2010, “Improving Immunisation Coverage in Rural India: A Clustered Randomised Controlled Evaluation of Immunisation Campaigns with and without Incentives”, British Medical Journal 340: C2220.
Bardhan, Pranab, 1989, Conversations Between Economists and Anthropologists: Methodological Issues in Measuring Economic Change in Rural India, Oxford University Press.
Basu, Kaushik, 2018, “The Trials of Randomised Control: Probability, Intuition and the Dinosaur Risk”Social Science and Medicine.
Chambers, Robert, 1999, “Whose Reality Counts? Putting the First last”, Intermediate Technology Publications, London.
Deaton, A. and Cartwright, N, 2017, “Understanding and misunderstanding randomized controlled trials.”Social Science and Medicine.
Duflo, Esther, Pascaline Dupas and Michael Kremer, 2015, “School Governance, Teacher Incentives, and Pupil-Teacher Ratios: Experimental Evidence from Kenyan Primary Schools”, Journal of Public Economics 123: 92-110.
Eswaran, Mukesh and Kotwal, Ashok, 1994, Why Poverty Persists in India: A Framework for Understanding the Indian Economy, Oxford University Press.
Ghatak, M., 2019, “Vital additions to empirical research”, The Hindu. Retrieved at: https://www.thehindu.com/opinion/lead/vital-additions-to-empirical research/article29728657.ece
Kremer, Michael, 2003, “Randomised Evaluations of Educational Programs in Developing Countries: Some Lessons”, American Economic Review 93(2): 102-106.
Kremer, Michael and Edward Miguel, 2007, “The Illusion of Sustainability”, Quarterly Journal of Economics 122(3): 1007-1065.
Miguel, Edward and Michael Kremer, 2004, “Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities.”Econometrica 72(1): 159—217.
Endnotes
1. RCT is an experiment that is intended to segregate the impact that a particular intervention/policy/variable has on an outcome. For example, a social science researcher, who is interested to get an idea about the impact of additional teachers in a school on students' learning, can carry out an RCT.
2. Scientific Background on the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2019, Understanding Development and Poverty Alleviation, The Committee for the Prize in Economic Sciences in Memory of Alfred Nobel, October 14, 2019.
3. This is in contrast with the Bangladeshi Nobel Peace Prize winner Muhammad Yunus's pioneering work on Grameen Bank and efforts through microcredit for championing economic and social development.
Dr Atanu Sengupta is a Professor, Department of Economics, Burdwan University, Burdwan (e-mail: sengupta_atanu[at]yahoo.com). Dr Sanjoy De is an Assistant Professor, Department of Economics, Shyampur Siddheswari Mahavidyalaya, Calcutta University (e-mail: sanjayda2000[at]gmail.com)