Fighting Poverty With Field Experiments: the Nobel Laureates’ Revolution

As a student, Harvard Business School professor Vincent Pons decided to become an economist after a project in Morocco with Esther Duflo. The co-recipient of the 2019 Nobel Prize in Economics popularized a new research method based on scientific rigor but also human contact and listening.


Children in Morocco. Photo by Monica Volpin


Nobel Prizes in Economics generally reward theorists in their sixties or seventies, for research they conducted several decades earlier. That’s the time required to assess the influence of a new model of economic behavior. This year’s prize is different. It rewards a new empirical method and, at 46, Esther Duflo is the youngest economist to ever receive the Nobel. This recognizes the unprecedented rapidity with which she and her co-laureates, Abhijit Banerjee and Michael Kremer, have transformed economic research and policy-making.


Michael Kremer’s first field experiments are less than twenty years old. Together with co-authors, he measured the effects of offering free meals, uniforms, and textbooks to students enrolled in randomly selected Kenyan schools by contrasting their school attendance with comparison schools. Since then, the three awardees have turned this experimental method into an industrial powerhouse, with a simple goal: ending global poverty.


I first learned about field experiments twelve years ago. A French student in exchange at the University of Chicago, I was considering an academic career and wrote to Esther Duflo for advice. By chance, my email arrived at a time in which she was recruiting research assistants for the Poverty Action Lab, a research lab she had created just three years before. She offered me a one year position as the field outpost of one of her evaluation teams.


The study I joined sought to estimate the impact of a program connecting poor households’ homes to the water network in Tangiers, Morocco. 845 eligible households were included in the experiment. One half, randomly chosen, was offered an in-home water connection on credit. The other half would be informed about the program only one year later.


My first job was to build a salesforce to contact all households in the first group door-to-door, inform them about the program, and help them apply for it. A few months after the connections to the water main were built, I recruited surveyors to administer a one-hour long questionnaire to all households in the study. Because present beneficiaries had been selected randomly, comparing their answers to those of the future beneficiaries would provide a direct estimate of the program’s impact.


This comparison unearthed surprising findings. In-home access to water did not decrease the incidence of water-borne diseases, contrary to our expectation, but it freed up time households had previously spent collecting water from public fountains and increased their leisure time. In addition, it decreased water-related conflicts with neighbors and increased beneficiaries’ social integration and overall well-being.


This early experience convinced me to apply for a PhD in Economics. Two dimensions of field experiments were particularly seductive. First, field experiments transpose hard science laboratory methods to the real-world study of society. This makes it possible to estimate the impact of a development program with the same rigor as a physicist’s lab experiment. Second, field experiments put people at the center.


The co-recipients of the 2019 Nobel Prize in Economic Sciences


The new approach to economic research that Duflo, Banerjee and Kremer’s work has popularized is first and foremost based on human contact and listening. The quality of a field experiment depends as much on the time spent by the researcher “on the field,” discussing with people and learning from them, as on fidelity to the method. I encountered countless numbers of Moroccan people during my time in Tangiers, and I remember vividly the long hours that Esther Duflo herself devoted to talking to surveyors and households when she visited the project. Pragmatism, modesty, and humility are at the core of the method. The three awardees have demonstrated these qualities both in their research and in their professional demeanor.


Yet, the end goal of Duflo, Banerjee, and Kremer’s scientific project has been as ambitious as their method is modest. Like the Moroccan study, their field experiments aim to identify programs that will improve poor households’ access to water, education, health, and jobs, with the hope of lifting people living on one or two dollars a day out of poverty.


Their studies have successfully identified market inefficiencies and other factors preventing households from escaping poverty traps and discovered effective ways to address these challenges. In addition, the transparency of the experimental method, and the fact that most experiments are conducted in partnership with governments, NGOs, and private companies, has facilitated scaling up the most effective interventions. As a result, local governments are equipped with sounder evidence on programs that work, and foreign aid money is spent more effectively. Today, it would be unthinkable for the World Bank and other international institutions to fund large-scale policies without first estimating their effects, if possible through a randomized experiment. In fifteen years, the three laureates and their colleagues have simply transformed both research and policymaking in economic development.


My Moroccan story is, of course, far from unique. Dozens of students have enrolled in PhDs in economics because they were enthralled by the field experience they acquired in their early twenties while working with Banerjee, Duflo, and Kremer. Many of them have since then become professors and have moved on to run their own experiments and train their own students to the method. In a word, the three laureates have done no less than create a new school. This is not the least of their contributions. In their endeavor, they have revealed entrepreneurial skills as impressive as their academic brilliance. Over the years, they have built a research infrastructure with ramifications throughout the world. The Poverty Action Lab, a key piece of the project, counts 176 affiliated researchers in 56 universities, and offices in seven countries. In addition to offices in Chile, South Africa, India, and Indonesia, two offices are located in Paris and Boston.


This is not by chance. Beyond the developing world, Duflo, Banerjee, and Kremer’s experimental approach may very well prove the most effective way to better understand and successfully alleviate poverty in our own societies. For all the laureates’ accomplishments, the task remains enormous. Today is a day for celebration. Tomorrow, let’s go back on the field!


Vincent Pons is an assistant professor of business administration in the Business, Government, and International Economy Unit at Harvard Business School. He is currently visiting the University of Chicago Booth School of Business.


The ProMarket blog is dedicated to discussing how competition tends to be subverted by special interests. The posts represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty. For more information, please visit ProMarket Blog Policy.