Nobel Prize in Economics 2019: Field Experiments and the Fight Against Poverty

This note covers the Nobel Prize in Economics 2019: who won it, what "experimental approach to alleviating global poverty" means, how randomised field experiments work in development economics, the laureates' main findings on education, health and credit, how the field changed over two decades, why it matters for real-world policy, and quick facts for exams.

What was the Nobel Prize in Economics 2019 awarded for?

The official citation reads: "for their experimental approach to alleviating global poverty". The prize was shared equally by three economists who changed how researchers study poverty.

In plain words, the laureates showed that instead of asking a huge, vague question like "how do we end poverty?", researchers should break it into small, testable questions, such as "does giving children more textbooks improve their test scores?" They then answered such questions using field experiments, a method where real people in their everyday lives are randomly split into groups, some receiving a specific help or treatment and others not, so that researchers can fairly compare the results.

This turned development economics, the branch of economics that studies why some countries and people remain poor, into a far more evidence-based field.

The prize's official name is the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, widely called the Nobel Prize in Economics. It was announced on 14 October 2019 and is awarded by the Royal Swedish Academy of Sciences.

Who are the laureates?

Abhijit Banerjee

Abhijit Banerjee was born on 21 February 1961 in Mumbai, India. At the time of the award he was affiliated with the Massachusetts Institute of Technology (MIT), Cambridge, MA, USA, and held one third of the prize.

According to the nobelprize.org facts page, both his parents were professors of economics, and he studied at the University of Calcutta and Jawaharlal Nehru University in Delhi before earning his doctorate at Harvard University.

He later taught at Harvard and Princeton before joining MIT. He is married to fellow laureate Esther Duflo.

Banerjee, often working with Duflo, extended field-experiment methods from education into many other areas of poverty research, and helped build the conceptual case for using small-scale, carefully designed studies to understand why some individuals and firms in poor countries fail to use the best available technology or opportunities.

Esther Duflo

Esther Duflo was born on 25 October 1972 in Paris, France. At the time of the award she was affiliated with MIT and held one third of the prize.

Her mother was a paediatrician and her father a professor of mathematics. She studied history and economics at the École Normale Supérieure before completing her doctorate in economics at MIT in 1999, and has worked there ever since (apart from a period at Princeton).

Duflo co-founded, with Banerjee and Sendhil Mullainathan, the Abdul Latif Jameel Poverty Action Lab (J-PAL) at MIT, a research network that promotes randomised evaluation of anti-poverty programmes worldwide. Her studies, often with Banerjee and Kremer, covered schooling, teacher incentives, health subsidies and farmers' adoption of fertiliser.

Michael Kremer

Michael Kremer was born on 12 November 1964 in New York, NY, USA. At the time of the award he was affiliated with Harvard University, Cambridge, MA, USA, and held one third of the prize.

He is described as having Jewish ancestry from Poland and Austria through immigrant forebears, and his mother worked as a professor of literature.

He earned his doctorate in economics at Harvard in 1992, then worked at MIT before returning to Harvard as a professor in 1999.

In the mid-1990s, Kremer and colleagues launched a landmark series of field experiments on schooling in rural western Kenya, working with a local non-governmental organisation. This work is credited as the starting point that showed how powerful the field-experiment approach could be for development questions.

What problem were the laureates trying to solve?

Despite decades of economic progress, the press release notes that more than 700 million people still subsisted on extremely low incomes at the time of the award.

Every year, close to five million children under five died from illnesses that could often have been prevented or treated with inexpensive care, and half of the world's children still left school without basic literacy and numeracy skills.

The scientific background explains that economists had long debated how to reduce global poverty, but economic theory alone could not say how large an effect a particular policy would have in practice.

For example, theory could not tell researchers whether hiring extra contract teachers on short-term deals was more cost-effective than simply reducing class sizes, nor whether microcredit programmes genuinely boosted small businesses.

Before the 1990s, much development research relied on comparing countries or regions that differed in many uncontrolled ways, so it was hard to know whether a measured difference in outcomes was really caused by a specific policy or by some other unseen factor, a problem economists call selection bias.

The laureates' contribution was to find a way of answering specific poverty-related questions with much greater certainty, by borrowing an old scientific method, the randomised trial, and applying it to real people in real village and school settings.

How does the experimental approach work?

At the heart of the laureates' method is the randomised controlled trial (RCT), adapted into what the committee called a "field experiment": a trial in which participants make genuine choices in their normal day-to-day environment, rather than in a laboratory.

The scientific background notes this method has a long history, having been pioneered in crop studies roughly a century earlier and widely used in medicine.

The basic logic set out in the sources can be summarised as a process:

1. Pick a specific, narrow question about poverty, for example whether free textbooks improve test scores.
2. Identify a group of eligible units, such as schools, villages or households, large enough for reliable comparison.
3. Randomly assign these units into a treatment group, which receives the intervention, and a control group, which does not, so that chance alone decides who gets what.
4. Deliver the intervention and measure outcomes, such as test scores or vaccination rates, in both groups after a set period.
5. Compare the average outcomes between the two groups; because random assignment removes systematic differences between them at the start, any later difference can be credibly attributed to the intervention.

Because assignment is random, the two groups should on average be similar in income, ability and other characteristics before the study starts, so the researchers can be confident that any difference afterwards is caused by the treatment itself, not by pre-existing differences.

The scientific background calls this problem of comparing unlike groups "the fundamental problem of causal inference" and explains that randomisation is the main tool economists use to get around it.

Draw two boxes labelled "Eligible schools or villages", with an arrow splitting into "Treatment group (receives intervention)" and "Control group (no intervention)", then both arrows converging on a box "Compare outcomes after the study period" to show how random assignment isolates the effect of the intervention.

What did the laureates discover about education, health and behaviour?

Education: targeted help beats extra resources

Kremer's early Kenyan trials tested whether simply adding more textbooks or free school meals improved learning. The press material states these had only small effects, while focused support for struggling pupils clearly raised their results.

In India, Banerjee and Duflo tested remedial tutoring for weak pupils in Mumbai and Vadodara, finding that teaching assistants who focused on struggling children produced clear, measurable gains.

As a result of this line of research, the press release states that more than five million Indian children have benefited from programmes of remedial tutoring in schools.

Studies also compared teachers employed on short-term, renewable contracts against permanently employed teachers with smaller classes; pupils taught by contract teachers showed significantly better test results, while reducing the pupil-teacher ratio alone showed no significant effect.

Health: price sensitivity and service quality

A Kremer-led study on deworming pills found that 75 per cent of parents gave children the pills when free, compared with only 18 per cent once they were priced at under a US dollar, a price that was still heavily subsidised.

This showed that poor households are extremely sensitive to even small prices for preventive healthcare.

A separate study by Banerjee and Duflo on vaccination found that mobile clinics, staffed reliably, tripled vaccination rates in randomly selected villages, from 6 per cent to 18 per cent, and adding a bag of lentils as a small bonus raised this further to 39 per cent.

Notably, the overall cost of each vaccination fell by half even with the added cost of the lentils, because the clinic's fixed costs were spread over many more vaccinations.

Behaviour and credit: why people delay good decisions

Duflo and Kremer studied why farmers in sub-Saharan Africa often failed to use fertiliser even when it offered strong returns. They found evidence of present bias, a tendency to keep delaying an investment decision because today's concerns dominate attention.

Consistent with this, temporary fertiliser subsidies boosted adoption far more than permanent ones, because a limited-time offer removed the temptation to postpone action.

On microcredit, Banerjee and Duflo's study in Hyderabad found only small positive effects on existing small businesses and no measurable effect on household consumption at 18 or 36 months, a result later echoed in similar studies in other countries.

How did the discovery unfold?

The experimental approach did not appear all at once. It grew over roughly two decades, starting with a single set of field trials and gradually spreading across topics and countries as the method proved its worth. The table below traces the main steps in this history, from Michael Kremer's earliest Kenyan experiments to the award itself.

Each row in this timeline represents more than a single study: it marks a point where the method was extended to a new question, a new country or a new group of researchers, gradually turning development economics into the "flourishing field" the committee described.

Why does it matter?

The committee said the laureates' research has "considerably improved our ability to fight global poverty" and that development economics is "now a flourishing field of research".

Their findings have fed directly into policy: besides the five million Indian children reached by remedial tutoring programmes, the scientific background notes that the World Health Organization recommends free distribution of deworming medicine in areas where more than 20 per cent of schoolchildren carry certain parasitic infections, a recommendation informed by the price-sensitivity findings.

The research network J-PAL, co-founded by Banerjee and Duflo, has helped scale up evaluated programmes that have reached more than 400 million people.

The scientific background stresses this likely understates the full impact, since many development economists outside J-PAL also use these methods.

The sources also note an important open question: whether results from one setting, such as Kenyan schools, can be generalised to very different places, since administering bodies, scale and local conditions can change how an intervention performs.

The laureates' work also changed how aid organisations operate, as many now evaluate new measures using field experiments before scaling them up, and some ineffective programmes have been closed after rigorous testing showed they did not work, freeing resources for more effective ones.

How does this connect to what you study?

The idea of a fair, controlled comparison that the laureates used is the same logic behind a controlled experiment in a science laboratory, where one variable is changed while everything else is kept the same. If you have studied experimental design in biology or chemistry, comparing a treatment group against a control group under similar conditions, you have already met the core idea behind this prize, just applied to schools, clinics and farms instead of test tubes.

In economics or civics classes that discuss poverty, education policy or public health, this prize shows how a social-science question such as "does free healthcare work better than paid healthcare?" can be tested with the same rigour as a laboratory hypothesis, rather than settled only by opinion or theory.

The underlying statistical idea, that randomly splitting a group removes bias so that any later difference between groups can be trusted, also connects to basic statistics taught in mathematics: it is the same reasoning behind sampling methods designed to produce a fair, representative comparison rather than a skewed one.

Quick facts for exams

The Nobel Prize in Economics 2019, formally the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, was awarded jointly to Abhijit Banerjee, Esther Duflo and Michael Kremer, each receiving one third of the prize, "for their experimental approach to alleviating global poverty".

It was announced on 14 October 2019 by the Royal Swedish Academy of Sciences. The laureates pioneered the use of randomised field experiments in development economics, showing how breaking poverty into small, testable questions and answering them with carefully designed trials among affected people could reveal which anti-poverty measures actually work, in areas from school tutoring to vaccination to microcredit.

Note: Source. The prize facts in this note are from the Nobel Prize's official site, nobelprize.org.

Glossary

• Development economics — the branch of economics that studies the causes of poverty in low and middle-income countries and how best to reduce it.
• Field experiment — a randomised trial run among people in their real, everyday environment rather than in a laboratory.
• Randomised controlled trial (RCT) — a method of testing a policy by randomly placing subjects into a treatment group and a control group and comparing outcomes.
• Control group — the group in an experiment that does not receive the intervention being tested, used for comparison.
• Treatment group — the group in an experiment that receives the intervention or programme under study.
• Selection bias — a distortion that arises when the groups being compared differ systematically for reasons other than the treatment itself.
• External validity — the extent to which results from one experiment can be trusted to hold true in a different setting or at a larger scale.
• Internal validity — the reliability of a single experiment's own result, showing that the intervention, not something else, caused the measured effect.
• Present bias — a tendency to give excessive weight to immediate concerns, leading people to repeatedly delay beneficial decisions.
• J-PAL — the Abdul Latif Jameel Poverty Action Lab, a research network founded by Banerjee, Duflo and Sendhil Mullainathan to promote randomised evaluation of anti-poverty programmes.
• Microcredit — small loans given to poor households or entrepreneurs, often without requiring traditional collateral.
• Remedial tutoring — extra, targeted teaching support given to students who are falling behind their grade level.

Common errors and misconceptions

• Misconception: The prize was for inventing field experiments. Correct: Randomised trials already existed in medicine and agriculture; the laureates applied and refined this method systematically in development economics.
• Misconception: The laureates proved that more resources never help poor schools. Correct: Their studies found that general resource additions such as textbooks had small effects, while targeted, well-matched interventions had significant benefits.
• Misconception: All three laureates worked at the same university. Correct: Banerjee and Duflo were affiliated with MIT; Kremer was affiliated with Harvard University.
• Misconception: Microcredit was proven to strongly boost household incomes. Correct: Studies found only small effects on business investment and no measurable effect on consumption.
• Misconception: Free healthcare always finds equal demand to paid healthcare. Correct: Demand for deworming pills fell sharply even at a small price, showing strong price sensitivity among poor households.
• Misconception: This is the "Nobel Prize in Economics" in the same sense as the original Nobel prizes. Correct: Its official name is the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, established later and administered alongside the Nobel Prizes.

Exam-style questions with model answers

Q1. State the official citation for the Nobel Prize in Economics 2019. [1 mark]

1. The citation was "for their experimental approach to alleviating global poverty".

Q2. Name the three laureates of the Nobel Prize in Economics 2019 and their affiliations at the time of the award. [2 marks]

1. Abhijit Banerjee and Esther Duflo were affiliated with MIT, while Michael Kremer was affiliated with Harvard University, all in the USA.

Q3. Why could economic theory alone not tell policymakers whether contract teachers were more cost-effective than smaller classes? [3 marks]

1. Theory could identify which incentives might matter, such as job security or class size, but it could not say how large their effects would be in practice.
2. Answering such specific, practical questions required an empirical method that could isolate the true causal effect of one policy from other influences.
3. This gap is exactly what the laureates' field-experiment approach was designed to fill, by randomly assigning schools to different policies and comparing outcomes.

Q4. Explain what a field experiment is and why randomisation is important in it. [4 marks]

1. A field experiment is a randomised trial conducted among real people making genuine choices in their everyday environment, rather than in a laboratory.
2. Researchers randomly assign eligible units, such as schools or villages, into a treatment group that receives an intervention and a control group that does not.
3. Randomisation ensures that, on average, the two groups are similar in every respect except for the treatment, removing systematic selection bias.
4. This allows researchers to credibly attribute any later difference in outcomes to the intervention itself, giving reliable, causal evidence for policy.

Q5. Describe the findings of the deworming and vaccination studies associated with this prize. [4 marks]

1. A deworming study found that 75 per cent of parents gave children the pills when free, but only 18 per cent did so when the pills cost less than a US dollar.
2. This showed that poor households are extremely sensitive even to very small prices for preventive healthcare.
3. A separate vaccination study found that staffed mobile clinics tripled vaccination rates in randomly selected villages, from 6 per cent to 18 per cent.
4. Adding a small bag of lentils as a bonus for vaccinating children raised the rate further to 39 per cent, while the cost per vaccination actually fell because fixed costs were spread over more people.

Q6. Discuss how the experimental approach pioneered by this year's laureates changed development economics and influenced real-world policy. [5 marks]

1. Before the 1990s, researchers largely compared countries or regions that differed in many uncontrolled ways, making it hard to identify true causes of poverty.
2. Starting in the mid-1990s, Michael Kremer and colleagues launched field experiments on schooling in Kenya, breaking the broad question of poverty into small, testable pieces.
3. Abhijit Banerjee and Esther Duflo, often working with Kremer, extended this method across education, health, credit and agriculture in many countries.
4. Their studies produced concrete findings, such as the value of targeted remedial tutoring and the price sensitivity of preventive healthcare, that could be directly tested and trusted.
5. These findings fed into real policy: remedial tutoring programmes reached more than five million Indian children, and the World Health Organization adopted free deworming distribution guidance based on the price-sensitivity research.
6. The research network J-PAL, co-founded by Banerjee and Duflo, has helped scale evaluated programmes that have reached more than 400 million people, cementing the experimental approach as the dominant method in development economics.

Key takeaways

• The 2019 prize honoured Banerjee, Duflo and Kremer for an experimental approach to fighting global poverty.
• Their method breaks big poverty questions into small, specific ones tested with randomised field experiments.
• Random assignment of treatment and control groups removes selection bias and allows reliable causal conclusions.
• Targeted remedial tutoring for weak students worked far better than simply adding textbooks or school meals.
• Poor households proved extremely price-sensitive towards preventive healthcare, such as deworming pills and vaccinations.
• Temporary subsidies encouraged farmers to adopt fertiliser more effectively than permanent subsidies, reflecting present bias.
• Microcredit studies found only modest effects on investment and no clear effect on household consumption.
• J-PAL, co-founded by two laureates, has helped scale evaluated anti-poverty programmes worldwide.

Test yourself

Who are the three laureates of the Nobel Prize in Economics 2019?

Abhijit Banerjee, Esther Duflo and Michael Kremer shared the prize equally for their experimental approach to poverty.

What was the official citation for the 2019 prize?

The citation was "for their experimental approach to alleviating global poverty".

Where was Michael Kremer affiliated at the time of the award?

Michael Kremer was affiliated with Harvard University, Cambridge, MA, USA, at the time of the award.

What is a field experiment?

A field experiment is a randomised trial in which real participants make genuine choices in their everyday environment, allowing causal comparison.

What did the Indian remedial tutoring studies find?

They found that targeted help for weak students significantly improved learning outcomes in the short and medium term.

How did price affect demand for deworming pills?

Demand fell sharply, from 75 per cent uptake when free to only 18 per cent when priced at less than a US dollar.

What organisation did Banerjee and Duflo help found?

They co-founded J-PAL, the Abdul Latif Jameel Poverty Action Lab, with Sendhil Mullainathan at MIT.

What effect did temporary fertiliser subsidies have compared with permanent ones?

Temporary subsidies encouraged significantly more fertiliser adoption than permanent subsidies, consistent with present bias.