The economist couple broke the mould to carry out field experiments to answer questions about global poverty.
Abhijit Banerjee, Esther Duflo and Michael Kremer won the Nobel Prize in Economics for 2019.
The press release accompanying the announcement of their win states: “This year’s Laureates have introduced a new approach to obtaining reliable answers about the best ways to fight global poverty. In brief, it involves dividing this issue into smaller, more manageable, questions – for example, the most effective interventions for improving educational outcomes or child health. They have shown that these smaller, more precise, questions are often best answered via carefully designed experiments among the people who are most affected.”
It further pointed out: “In the mid-1990s, Michael Kremer and his colleagues demonstrated how powerful this approach can be, using field experiments to test a range of interventions that could improve school results in western Kenya. Abhijit Banerjee and Esther Duflo, often with Michael Kremer, soon performed similar studies of other issues and in other countries.”
In the aftermath of the Second World War, the subject of economics became fairly mathematical. Despite this, economists had a huge physics envy. Jonathan Aldred in his book Licence to be Bad: How Economics Corrupted Us, defines physics envy as “the desire of economists to emulate sciences such as physics”.
Scientists are scientists because they carry out controlled experiments. The trouble is that suitable conditions for controlled experiments rarely exist in real life. In this scenario, most experiments in economics happen in a lab or through students who “play games or answer hypothetical questions about contrived situations, such as how they would respond to a financial incentive”.
Banerjee, Duflo and Kremer have broken this mould and carried out real-life experiments trying to answer, as the Nobel Prize press release put it, the smaller and the more manageable questions, which would help in fighting global poverty.
A particularly difficult problem that Banerjee and Duflo tackled was the bad learning outcomes in many Indian schools. Students, even in higher classes, could not and still cannot, read and write properly, or do basic mathematics.
At a literature festival in Mumbai, in the winter of 2014, Banerjee gave an extremely detailed talk around his work. I happened to be there and recorded the entire talk on tape. Banerjee talked about an experiment involving government school teachers in Bihar, sometime in the summer of 2012.
As he put it: “The teachers were asked that instead of teaching like you usually teach, your job for the next six weeks is to get the children to learn some basic skills. If they can’t read, teach them to read. If they can’t do math, teach them to do math. At the end of six weeks, these teachers were given a small stipend. They had also been given a couple of days of training. At the end of six weeks, the children had closed half the gap between the best performing children and the worst performing children. They had really improved enormously.”
So, what happened here? Why did the school teachers do much better than they had in the past? As Banerjee explained it: “The reason was they were asked to do a job that actually made sense. They were asked to teach the children what they don’t know. The usual jobs teachers are asked to do is teach the syllabus — which is very different. Under the Right to Education Act, every year you are supposed to cover the syllabus. It doesn’t matter whether the children understand anything. Think of all the class 4 children who can’t read. They are learning social studies and all kinds of other wonderful things — except they can’t read. They are learning nothing. They are sitting in a class watching some movie in some foreign language without subtitles. Hence, the dropout rates are high. And I am shocked why anybody comes to school at all.”
As Duflo told The Indian Express in 2015: “Learning is not about enrolment, teacher-student ratio [or] having latrines in schools; it’s about if we are serious about learning.”
The question is: are we serious about learning? It turns out that we are not. A basic problem with the RTE remains that: it expects children to be ready for elementary education when they enter class 1. In fact, as Duflo told the Financial Times in 2015: “By the end of Grade 1, they are supposed to be done with reading…It’s a complete fantasy.”
Through this small experiment, the answer to a very big problem was found. In the first few years, instead of trying to complete the syllabus it is important that the teachers and the education system teach basic skills to children. This basically means to teach them to read, write and do some basic mathematics. If a student has problems reading, you teach them to read. If the student does not recognise numbers, you help them with that. And so on. In fact, Banerjee and Duflo have carried out such experiments in different states and seen the learning levels of children improve.
As the Nobel Prize press release put it: “As a direct result of one of their studies, more than five million Indian children have benefitted from effective programmes of remedial tutoring in schools.”
Apart from education, Banerjee and Duflo have carried out experiments in the area of healthcare as well. Different countries around the world have implemented the right to food with the idea that if poor people are given subsidised food, their nutrition will improve.
In an interview I did with Banerjee in 2011, he talked about a very interesting example he had encountered in Morocco. As he said: “We met a person in Morocco and we were talking to him and he seemed to represent the classic starving poor. We asked him what will you do if you got some more money and he said I will buy more food. And when we asked him what will you do if you got even more money? And he said he will buy some more food. We felt very bad for him. Till we walked into his house. And then we found his TV, his antenna and his DVD player. And we asked him why do you have this television when there is not enough to eat? And his answer was television is more important than food.”
This was a very interesting insight. In his talk in Mumbai in 2014 and also in the 2011 interview, Banerjee talked about an experiment around the example discussed above that they had carried out in China.
As he said: “We carried out a nice experiment in China. We gave some people a voucher to buy cheap rice. Instead of buying rice, let’s say for Rs 10, they could buy it for Rs 2, using the vouchers. The presumption was that this would improve nutrition. This was done as an experiment and hence some people were randomly given vouchers and others were not [technically referred to as a randomised controlled trial].”
The result of the experiment was very different from what had been expected. As Banerjee put it: “People with vouchers were worse off in nutrition. They felt that now that they have the vouchers, they are rich and no longer need to eat rice. They could eat pork, shrimp etc. They went and bought pork and shrimp and as a result their net calories went down. This is perfectly rational. These people were waiting for pleasure.”
And how was this perfectly rational? “Pleasure is something very important not just for us to live but also in terms of being able to control our destiny. You think the rest of my life will be drab, and it becomes very difficult to live. In that particular sense they had little opportunity and they knew that this wouldn’t last forever. They could improve their nutrition or for the next ten days they could also eat a little bit better. Fun is something that we forget about.”
What Banerjee is saying here is that when food security policies are designed, they don’t take pleasure into account. It is assumed that if food is sold at subsidised rates, people will automatically think of nutrition. But that doesn’t turn out to be the case.
As Banerjee put it: “You have to think of a life where pleasure has its space. If we don’t take this idea where pleasure needs space seriously, we end up with an idea that if you dump food on people they are going to eat it. In fact, suppose people eat 30 kg of rice per month and we give them 20 kg for free then that makes them richer that doesn’t make them buy any more rice.”
This is a very simple but profound point which could have only come to the fore from an experiment. Most of the research carried out by Banerjee, Duflo and Kremer is along these lines.