In the mid-1990s, Michael Kremer conducted a series of experiments to improve attendance rates at schools in Kenya. He was probably the first person, in the economics field, to apply randomized controlled trials to obtain precise information about a problem and thus be able to resolve it. Today, Kremer holds is the Gates Professor of Developing Societies, in the Department of Economics at Harvard University.
That approach, completely unheard of at the time, established the bases for a movement of economists who seek innovative solutions in spheres such as education, health, poverty, immigration and the climate crisis. For implementing this kind of experimental approach “to alleviate global poverty”, Kremer received the Nobel Prize for Economics 2019, together with researchers Esther Dufflo and Abhijit Banerjee, of the Abdul Latif Jameel Poverty Action Lab (JPAL), at the Massachusetts Institute of Technology (MIT).
It has been over twenty years since you started researching poverty. Why did you do it?
Because it is an incredibly important, titanic human problem. And it is not unsolvable. Many things can be done at different levels, from governments to NGOs, and even on an individual scale. Research demonstrates that there are small investments that can have an enormous impact.
Diarrhoea is one of the main causes of death among infants in developing countries, because of the faecal contamination of water. Until recently, the proposed solution was to distribute the treatment (a chlorine tablet to disinfect the water) in shops and at a very low price. But one of the great discoveries of behavioural economics was that, in preventive health, the price marks an enormous difference.
And what did you do?
With the collaboration of NGO Evidence Action, we designed a programme of chlorine dispensers which we placed right next to the fountains. We tried to introduce a new habit among the population: the women turned the dispenser handle and received an adequate dose of chlorine for their containers before filling them with water. That way we managed to increase the number of households disinfecting their water from 7% to 50%.
Now, Evidence Action has extended this project, and millions of people in eastern Africa are benefiting from it. Unfortunately, they do not have the necessary funding and will not be able to develop it in other regions. They would need more contributions or for governments to start supporting these kinds of programmes.
The water dispensers project is a good example of new ways that have been developed to study potential solutions for alleviating poverty, solutions that are shared with JPAL. In this case, it is based on the application of the clinical trials method to economic interventions.
Indeed, this is a new tool within the economics toolbox. It is very useful because, among many more contributions, it permits a rigorous estimate to be obtained of the causes and the effects. It requires working, arm in arm, with different social groups on very practical problems, to try to learn about the work that they do and understand their perspectives and the problems that have to be tackled. This way of working also requires larger teams, therefore it is important to have a research infrastructure in other scientific fields. In this sense, the JPAL is fantastic, because it provides this infrastructure.
When and why did you start using randomized controlled trials?
When I finished my university studies I worked as a secondary school teacher for a year in Kenya. After that period, I returned to the United States and, some time later, when I got my first job and money, I returned there on holiday. Then, a good friend who had been a teacher and was by then working for an NGO, commented that they would be starting to work in a region of Kenya to try to tackle high levels of school absenteeism. However, they did not know what the best way would be to approach it. That was towards the end of the 1990s.
Between the two of us, we thought that if we wanted to know what the best way of achieving it or of assessing the impact that the actions would have, the most reasonable thing would be to set up different interventions at different schools, and then analyse which one was working best, and implement it across the entire education network. The NGO thought it was a good idea and told us to go ahead with it.
What solution did you find worked best to increase school attendance rates?
After studying various options, such as donating uniforms or books free of charge, we opted for treating the children against intestinal worms or parasites, helminths. In the West, infections by this parasite in animals are treated systematically. In contrast, in developing countries, although the treatment is very cheap, children do not have access to it and these infections cause anaemia, lethargy and apathy. Of course, ill children are absent from school more frequently than healthy children.
The World Health Organisation advocates the treatment being administered at schools and for it to be the teachers who give it to the pupils. So, we developed a programme to train the teachers to administer the treatment to the children at their schools. And it worked. Afterwards, with the Ministry of Education, we extended the project to the entire country. The programme started up in 2001, therefore we have been able to measure the impact that it had on the pupils that we treated, who today have a better quality of life than those of their peers who did not receive the treatment.
Could having access to better data put an end to poverty?
Using better data and conducting a more rigorous analysis of these data can improve our lives, but I would not go to the extreme of affirming that they can put an end to poverty. Contrary to what we often think, the absolute poverty rate – in other words, living with less than 1.90 dollars per day – is not increasing, but has drastically decreased since 1980. The biggest contribution to that has been the reforms carried out in China, as it was a tremendously poor country, even though we have forgotten this.
The Chinese government probably carried out many experiments, but not in the sense of formal experiments, like those that are carried out in the scientific field with the randomized controlled trials. A part of what the Chinese government learned was thanks to the introduction of reforms in particular areas, and then seeing what impact they had. If they worked, they applied them in other areas of the country. These were political decisions, but certainly they experimented, they obtained data and they acted in consequence.
Talk to us about Precision Agriculture for Development (PAD).
It is an NGO that I helped to found with the aim of improving the lives of farmers in developing countries. We work closely with governments and other NGOs. For example, in Africa, there is a new pest called the fall armyworm. It has just arrived there from America and the farmers know nothing about it, but we can send them messages to their mobile to alert them to this pest and give them advice on how to treat against it. Now the governments of India and of Ethiopia are also implementing this system, which is very cheap and effective.
Another example: farmers from developing countries will have problems due to climate change. It is clear that to combat global warming it is necessary to reduce greenhouse gas emissions, but we also have to provide farmers with information, for example, on weather forecasts or on which are the most drought-resistant crops. That is what we are aiming to do from PAD.
You have just received the Nobel Prize for Economics. It is not very often that the Prize is awarded to researchers in behavioural economics or who work to study poverty. Do you think that this is a change in tendency regarding what is considered important in economics research?
Until quite recently, a large part of the research in economics consisted only of the analysis of impacts of policies implemented by governments. Now, on the other hand, there is a real tendency to innovate, thanks largely to randomized controlled trials. There is a global effort in the economics field to create new techniques and methodologies that will improve the world.
For example, Al Roth, the Nobel Prize in Economics laureate of the year 2012, developed new algorithms within game theory to regulate the design of markets in accordance with supply and demand. And this algorithm had very specific social applications; it was applied to improve kidney exchanges in the United States, matching patients who need a kidney with donors.
In what way does the prize you have received influence your work?
The prize is a recognition for the entire sector, and that includes hundreds of researchers the world over, but also the governments and the NGOs. Let’s hope it is useful for opening more doors. I hope that it narrows the gap between research and policy and raises awareness among legislators, so that they not only learn a lot more about how to make their intervention programmes more effective but start to act in accordance with the results of the evaluations.
Interview by Cristina Sáez