Readers & Thinkers: The 2019 Nobel Prize in Economics. Flipping Coins for Poverty

Dear All:

Well, who won?

The Nobel prize for economic sciences for 2019 was announced today, October 14, 2019. It went to Abhijit Banerjee, Esther Duflo, and Michael Kremer

Banerjee is 58 years old and was born in Kolkata, India.   He obtained his B.S. from the University of Calcutta, his M.A. from Jawahalal University, and his Ph.D. from Harvard University.  The majority of his work was done while a Professor at the Massachusetts Institute of Technology.  He is currently a Professor of Economics at MIT.

Duflo is 46 years old and was born in Paris, France.  She obtained her B.A. from Ecole Normale Superieure, her M.A. from DELTA, and her Ph.D. from the Massachusetts Institute of Technology.  All of her work in developmental economics was done while she was at the Massachusetts Institute of Technology.  She is currently a Professor of Developmental Economics at MIT.

Kremer is 54 years old and was born in the United States.  He obtained his B.A. and Ph.D. from Harvard University.  He did his work at both the Massachusetts Institute of Technology and Harvard University.  He is currently a Professor of Economics at Harvard.

In my 2019 prediction of the Nobel Prize, I mentioned the field welfare economics as likely to win and in fact, it did. 

So what did these guys do to make them so Noble?

The Nobel committee wrote that they received the prize “for their experimental approach to alleviating global poverty.”

Although at times it doesn’t seem so, the study of economics is ultimately the study of how to make society work better to improve people’s lives.  Some economists spend their time deriving elegant theories and some spend their time analyzing data to see what’s actually happening and some do both.  Analyzing data for causality is much more difficult in economics than it is in physics, chemistry, or biology because we can’t really conduct controlled experiments.  A controlled experiment is where you analyze the effect of an independent variable by keeping all other elements the same.  In the hard sciences, this is relatively easy to do.  You create a petri dish of the exact same bacteria and blast it with some chemical and see what happens and do it again and again under the same conditions.

In economics, it is difficult to create experiments under the same conditions because we are dealing with many human beings in a complex system, so we study actual events.  For example, if we want to determine whether more education leads to more economic growth, we collect data from different countries with different education levels and different rates of growth and use statistics to determine whether more education leads to more economic growth.  Economists have developed sophisticated techniques to do this, but causality is always a challenge.  You’ve heard the line, “Correlation doesn’t prove causality.”  The reason is obvious.  There are so many factors, other than education, that are affecting the growth of a nation, including interest rates, culture, type of economy, and so on and so forth.

Economists spend a lot of time trying to resolve the causality issue, sometimes successfully and other times not.  In recent years, economists have used clever techniques to move closer to a controlled experiment with what are called “natural experiments” and “field experiments.”  Field experiments, in some sense, are relatively simple.  In a field experiment, economists randomly select similar groups of people or institutions to be part of the treatment group or the control group.  This randomization, if done correctly, can more clearly identify causality.  It is still not as good as controlled experiments in the hard sciences, but it can be useful.

Kremer, Duflo, and Banerjee applied this very simple and known concept of randomized, field studies in poor and developing countries like India and Kenya.  The hope was to determine what policies might improve the plight of people in these countries.  In this sense, they were flipping coins for poverty.

Kremer really started this push as it was a passion that he had as an undergrad and after teaching in Kenya.  The trio worked on various topics either separately or together in the area of field experiments in two main areas:  education and health.

One of the benefits of applying theory and or testing in the field is that you learn about a multitude of things that you might never have thought of in your office.  This is true for all good empirical research. 

For example, they learned that oftentimes in developing countries, teachers simply do not show up to teach.  They tested a system whereby teachers were given a bonus for showing up to class and installed cameras to monitor them.  This had the obvious effect of increasing teacher attendance and improving student learning

In other experiments, they found that giving more books to a school didn’t raise average test scores (although they did help the best students), but that providing incentives for teachers did.  They also found that it is more effective to tutor the weaker students or teach at their level, rather than have them be taught at a higher level in classrooms with higher level students.  In addition, they found that civil-servant teachers, with little incentives and job security, weren’t as effective as contract-based teachers.

In terms of healthcare, they found that there are externalities to improving the health of even some of the population.  Another perhaps obvious result that the researchers found was that the percentage of people taking medicine is much higher when it’s free than when it costs even just a little bit of money (75% versus 18%).

In terms of giving immunizations, they found that many people in third-world countries don’t get immunizations because the nurses simply aren’t at work.  Thus, they found that a more effective method to get more people vaccinated was to provide mobile vaccination clinics where staff was always present.  They also found that offering a bag of lentils along with a vaccination in India (an incentive) increased the percentage of children that were vaccinated.

All of these field experiments are interesting and these particular economists took the plunge that others might not have wanted to.  However, all of these small, field experiments have a larger issue.  Will they be valid on a larger scale or are they artifacts of a particular village at a particular time?  These Nobel winners have tried to address some of these issues, but they are always a real issue. 

So what about the Nobel choice?

During my time at MIT, Michael Kremer had arrived to do a post-doc and one of my good friends was his research assistant.  He was turning one of his thesis chapters into the published paper “The O-Ring Theory of Economic Development.” 

Around the same time, Abhijit arrived at MIT.  MIT was trying to build their developmental economics program.  Soon after, Duflo arrived and Abhijit was one of her Ph.D. thesis advisors.  They later got married.

As I mentioned at the beginning, the study of economics is ultimately the study of how to make society work better to improve people’s lives.  These three researchers focused on the lives of people in poor, developing countries in order to understand how to improve their lives.

The focus on poor countries and poor people makes this a popular Nobel pick, but the contrarian might ask whether or not the results of many of these studies wasn’t obvious ex-ante.  For example, wouldn’t it be obvious that free pills would lead to a much larger percentage of pill adoption by people or that installing cameras in a building would lead to higher teacher attendance?  And finally, how much have these experiments really changed people’s lives?  And let’s not forget, we also have lots of problems in first-world countries to look at, including poverty.

So is that it?

In order to fulfill their mission, Duflo and Banerjee co-founded MIT’s Abdul Latif Jameel Poverty Action Lab (J-PAL)  in 2003, a global network of antipoverty researchers that conducts field experiments.

Dr. Robert Barro, Kremer’s Ph.D. thesis advisor, told me today that “Michael Kremer combines a humanitarian concern for economic development with keen economic skills and imagination.  Thereby, he has made important practical advances for the provision of improved education and health in developing countries.”

Dr.  James Poterba, Professor of Economics at MIT, told me today “I am thrilled by this prize, which showcases how economic research can tackle incredibly important problems and come up with innovative insights and solutions.”  

It should always be remembered that even though these three people won the prize, there are many others who have contributed to the area of natural experiments and field experiments.  There are also many other researchers who have worked on poverty.

For more information on the winners:

From MIT:

From Harvard:

Congratulations to Abhijit Banerjee, Esther Duflo, and Michael Kremer and their pursuit of knowledge, and the other armies of researchers who walked with them!



P.S. Feel free to email me your thoughts and indicate whether you wish them to remain anonymous or not.

October 14, 2019