Esther Duflo


Esther DufloI was born in 1972 at the Port Royal hospital, the “baby factory” of Paris.

I grew up with my parents, Violaine Duflo (a pediatrician) and Michel Duflo (a math professor) in Asnières, a western suburb of Paris. I went to the local public schools until grade 11, and transferred to Henri IV, a magnet high school in the center of Paris, for the last year of high school. I was not stellar at anything in particular, but very organized, quite solid in every school subject, and generally keen to please. As a result, I was a good, uncomplicated student, well-liked by teachers, and with a great set of friends. I have two siblings whom I adore, my brother Colas who is four years older than me, and my sister Annie who is 6 years younger. My parents were busy, but our house was always filled with Au Pairs, cousins and friends. We grew up with a lot of freedom and joy.

I was very tiny as a child. My mother says that this is how I became so confident: I only learnt to read at six, but since I looked like a four-year old, adults thought I was really smart. I was also what you then called a tomboy, and would I guess now be called gender-fluid. I wanted to be a boy, I dressed like a boy, and cut my hair short; all my friends were boys. At adolescence, both things changed: I grew enough to be counted at the short end of “normal-sized”, and I became comfortable being a girl. But I retained both my self-confidence and the desire to never let my gender define what I would or could do.

I was fortunate to be made aware at a very early age of the diversity of life circumstances in the world. My mother volunteered across the world, helping children who were victims of war. In school, in the magazines I subscribed to, in the various youth groups I belonged to, we were always exposed to the lives of others. I was amazed and somewhat awed by my luck: how come I, Esther, get to be born in this middle-class, intellectual family, with loving parents, decent schools, and all the food and books I need, while some other kids are born in Congo, in the middle of a war, and are forced to carry a Kalashnikov to fight?

I was also disturbed by the French education system’s propensity to rank everyone according to their grades and treat harshly the students who did not conform. My best friend throughout primary school was just as small as me, but – in my opinion – much smarter. Yet he struggled in school and was constantly humiliated by teachers and other children. He eventually had to repeat a grade. I theorized early that everyone must be excellent at least ONE thing, but do not necessarily get to find their spe­cial talent. This made my luck even more flagrant: since my one talent was to be a jack of all forms of schoolwork, I did not have to look very hard to find it: I was rewarded for it every day! In contrast, the boy with the Kalashnikov may never live to discover his talent. Closer to home, perhaps my friend would have been an amazing programmer or cook or artist or tennis player, but the school system was not allowing him to dis­cover what he was good at.

I felt that the only way I could ever repay this huge cosmic debt to the world was first, to nourish and exploit my own unremarkable talent, and second, to play some role in helping others get the opportunity to find and nurture their talents. The meant I resolved very early that I would be an academic, and also that I needed to pick a course that allowed me not to specialize early (since, once again, I was not particu­larly stellar at any one thing, but was able to keep many balls in the air). The French system allows good students the chance to stay in high school for their first two years of college. This gives them access to small classes and time to prepare for competitive examinations with the goal of entering the Grandes Écoles. This is what I did after finishing high school in 1990. As a field, I chose the social sciences, because it allowed me to combine math with philosophy and history (and sociol­ogy and economics, but I considered those as prices to pay to keep math and history in the mix). It served me well and I got admission to the École Normale Supérieure.

There I decided to study at least two subjects, to avoid specializing for as long as possible. I was determined to study history, but the choice of economics as a second subject was a (fortuitous) accident. On the day of admissions to the École Normale, in June 1992, Daniel Cohen, the profes­sor in charge of economics, was out recruiting students. Daniel is one of the most charming economists I know. He persuaded me that economics was the perfect discipline for a jack-of-all-trades like me.

To accomplish the second goal of helping others, I did what I could on the side: as a child, I helped my classmates with homework, I helped out in soup kitchens; in college, I volunteered in a prison for a year. But I found all of this rather unsatisfactory – it was hardly making a dent. Meanwhile, I started to find history really too remote from anyone’s life to be useful. And economics was just dreadfully boring.

All in all, at twenty, I was ever more fortunate and privileged, being a member of the elite of French students, but my project of improving the world was not going well. I felt I was not living up to my own luck. A little despondent, I seized an opportunity the school offered me to spend a year as a French teaching assistant in the new Social Science university in Moscow. In 1993–1994, I happily gave up economics and moved to Mos­cow. I wrote my history masters dissertation there, which at the time was sufficient to get me my degree and it also gave me a year to clarify my ideas on what to do to accomplish my second goal.

Indeed, I could not have anticipated how much the year in Moscow would clarify my ideas. Daniel Cohen, who showed great forbearance, given how undisciplined a student of economics I had been, hired me for one of his projects, and also got me a job in the team that Jeff Sachs had put together to advise the Ministry of Finance. I knew no economics, but I spoke Russian well, and I was ready to be enterprising, so I made myself useful (one of my early tasks was to locate a directory of plants and draw a stratified random sample, which I did by making up an algorithm to do it mostly manually). This gave me a unique ringside view of the process of economic reforms in Russia. 1993-1994 was a very difficult year. The political process was bitter and angry, the economy was tanking, people were literally starving while others were starting to build empires by buy­ing back their shares in the national industries. I did not think any of the competing teams of economists working in Moscow at time was neces­sarily doing the right thing. But I was incredibly impressed by their influ­ence (especially perhaps given how little they knew). I knew that I wanted their job, in pursuit of my second goal.

In Russia, I also met Thomas Piketty, who was at the time teaching at MIT. He told me that MIT would teach me the economics I needed to be useful in the world. He also ensured, when I was applying the second year, that I could get in, by convincing his colleague that I was worth gambling on despite my unusual background (No one had heard of my multidisciplinary program at the time, and an admissions officer called Abhijit Banerjee had initially put my file in the reject pile, I later came to know …).

I came to MIT in 1995 as a graduate student (one year after completing my masters at the ENS, at “Delta” – which would eventually become the Paris School of Economics). One of my very first classes was development economics. Abhijit Banerjee and Michael Kremer were teaching it.

The class was very small: it was taught jointly with Harvard students (in the spring we took Jonathan Morduch’s class at Harvard). Eliana La Ferrara, Asim Khwaja and Jishnu Das from Harvard, and Chris Spohr, Stuti Khemani and me from MIT. Every single one of us is a development economist today.

For me, it took precisely one lecture to be hooked. This class was noth­ing like I had ever experienced. They were not just teaching. They were searching. Searching for what made this field different. Building on the work of – among others – Dasgupta, Ray, Stiglitz and his own research with Andy Newman, Abhijit was busy building the theoretical framework for how to think about development and how to teach it. An essential part of this was the idea of poverty traps: being poor changes people’s oppor­tunity set, and sometimes this means that they stay poor. This theory res­onated with my childhood theory of undiscovered talent. It also offered a way out. Perhaps it is indeed possible to find the right lever to unleash people’s opportunities?

And then, of course, there were Michael’s early forays into randomized controlled trials (RCTs), with the famous textbook projects. We students saw the start of this movement unfold in real life. The 7 schools, the unex­pected results, the larger sample of schools, the results still unexpected, the new tests, and results still were not falling in line!! The most impor­tant lesson was perhaps that this research was hard. The second was that it is worth experimenting. Quite often, you’d be surprised.

Most of his colleagues thought Michael was a little crazy to do this and indulged his experimental forays as a cute way to waste his time before returning to his real talent (macroeconomics). But Abhijit immediately understood the power of RCTs, not just as a tool to evaluate programs, but as a way of turning development economics on its head by giving us the freedom to put any theory to the test. He started running RCTs as well, and even more importantly, creating the partnerships that would allow them to be broader than just a few researchers.

What Michael and Abhijit were trying to do was to bring their pas­sion, their intellect and their hard work to the project of transforming development economics, not just for the sake of economics, but for the sake of poor people. I had found my role models. Now I needed to emu­late them.

But they were also realizing that they could not do it alone, not if they were going to make a real difference. So, their project was intimately tied with teaching and advising. Michael and Abhijit were doing their bit to discover and nurture the talent of others (us, the students), with the greater goal of making sure that, worldwide, the talents of the poor don’t get wasted. The MIT economics department was a wonderful place to do that, for it places graduate education at its very core. I had a wonderful experience there. Abhijit became my advisor, along with Joshua Angrist.

Angrist is one of the architects and main troublemakers of the “credi­bility revolution”. He taught me to think about data in the real world through the lens of an experiment: “What would be the ideal experi­ment?” He would ask. “And how does your set-up differ from that?”

In 1999, I got my PhD, and immediately joined MIT as an assistant pro­fessor. It was not common to be hired by your own department, but Michael had unexpectedly decided to move to Harvard and the depart­ment felt that Abhijit needed a colleague to continue building the field of development in the department. I felt this would be a place that would let me do whatever it takes to discover exactly what my talent was (within economics), without giving me too much direction. This freedom, and Joshua Angrist’s rigor and clarity of thinking, gave me the courage to take the plunge and conduct my first “real” experiment as soon as I got my first paycheck and a little bit of research money.

In 2003, Abhijit Banerjee, Sendhil Mullainathan and I were all mulling over various outside opportunities, and the institution decided to do something to humor us. We proposed to the chair of our department, Bengt Holmström, the idea of creating a center that would focus on doing randomized controlled trials and spreading the results to policy makers. From the very beginning we conceived of a poverty action lab (later renamed J-PAL) as a network. There was so much work to do that what we needed to do was to support the work of all those ready to use RCTs to improve the policies affecting poor people in developing countries. Bengt saw the value of this proposition much more clearly than we did. He secured some funding for us from MIT, we hired Rachel Glennerster, and the Poverty Action Lab was created.

Rachel Glennerster (who happened to be Michael Kremer’s wife) was absolutely instrumental in the creation and the development of J-PAL. She gave it a name; she gave it a mission (“Our mission is to reduce pov­erty by ensuring that policy is informed by scientific evidence. We do this through research, policy outreach, and training”) and kept us on it, and she oversaw a massive growth from our humble beginnings (two barely heated offices, three staff members besides Rachel, and eight affiliates) to the moment she left to become the Chief Economist at the Department for International Development (DFID), in the UK.

Even with Rachel at the helm, with only the initial funding provided by MIT, it would have been difficult to go very far. We were fortunate that Susan Hockfield became MIT’s president. She met us very early on and also understood our model, and why it made sense. She decided to make the Poverty Action Lab a priority for fundraising. This is how we were brought to the attention of an alumnus, Mohammed Jameel.

Mohammed, a pragmatic enthusiast who had chosen the fight against poverty as one of his key philanthropic pursuits, thought the project was worthwhile, and in turn challenged us to affect 100 million lives with bet­ter policies in our first ten years. This sounded like a lot, but since my early days as a very short six-year-old, I don’t lack confidence. Rachel, Sendhil, and Abhijit are not the kind of people to say no to a challenge either. We agreed.

Mohammed endowed our lab, which became J-PAL (the Abdul Latif Jameel Poverty Action Lab). Today, J-PAL is a network of 194 affiliated professors and just as many other invited professors from around the world. We have 998 ongoing or completed projects in our database, run by our affiliates or invited researchers in 84 countries. And at least 450 million people have been affected by policies scaled up after a J-PAL eval­uation found them effective.

I have stayed at MIT ever since I got that first job there. I have wonder­ful friends and colleagues there, and this has been a great home for J-PAL Global and J-PAL North America (led by Amy Finkelstein and Larry Katz). The other J-PAL offices are scattered in academic institutions around the world, and I feel that those are also a little bit my homes.

As I mentioned, when I first met Abhijit Banerjee as a first-year stu­dent, I found him unbelievably inspiring. He was kind and a bit aloof, but I knew crossing his path had changed my life. Little did I know that many, many years later, I would not only become a colleague, but also, eventu­ally, a life partner. I continue to find him just as inspiring, and just as kind, although the aloofness is gone. We have two delightful children, Noemie and Milan. I continue not to believe how lucky I have been throughout my life. That my path crossed Abhijit’s was definitely the most amazing part of it all.

From The Nobel Prizes 2019. Published on behalf of The Nobel Foundation by Science History Publications/USA, division Watson Publishing International LLC, Sagamore Beach, 2020

This autobiography/biography was written at the time of the award and later published in the book series Les Prix Nobel/ Nobel Lectures/The Nobel Prizes. The information is sometimes updated with an addendum submitted by the Laureate.

Copyright © The Nobel Foundation 2019

To cite this section
MLA style: Esther Duflo – Biographical. Nobel Prize Outreach AB 2023. Fri. 29 Sep 2023. <>

Back to top Back To Top Takes users back to the top of the page

Coming up

This year’s Nobel Prize announcements will take place 2–9 October. All announcements will be streamed live here on

See the full schedule
Announcement dates

Explore prizes and laureates

Look for popular awards and laureates in different fields, and discover the history of the Nobel Prize.