The Authors of ‘Poor Economics’ on Ending Poverty

The winners of the 2019 Nobel Prize in Economic Sciences have for years been conducting innovative economic experiments on ways to alleviate poverty that have produced real-world — rather than theoretical — results. The winners are Abhijit Banerjee, Esther Duflo and Michael Kremer.

Noted the Royal Swedish Academy of Sciences, “In just two decades, their new, experiment-based approach has transformed development economics, which is now a flourishing field.” Banerjee and Duflo, who are husband and wife, teach at MIT and Kremer is a professor at Harvard.

Some of their experiments have showed, for example, that giving impoverished children textbooks in school did not actually make them better students. Instead, the economists found that achievement improved by providing tutors for low-performing students in India. In other cases, they determined that providing free mosquito nets to reduce malaria or treating children for intestinal worms made them more likely to attend school in the first place and to then eventually work in higher-paying jobs.

Knowledge@Wharton interviewed Banerjee and Duflo about their landmark book, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty, at a conference in Goa, India in 2011. (That interview appears below.) The book sought to answer such questions as: “Why would a man in Morocco who doesn’t have enough to eat buy a television? Why is it hard for children in poor areas to learn, even when they attend school? Why do the poorest people in the Indian state of Maharashtra spend 7% of their food budget on sugar?” 

Unlike economists who focus on macro issues such as aid, Banerjee and Duflo approach poverty much as medical researchers might set about finding the treatment for a disease — by conducting clinical trials. Duflo, who delivered a TED talk on this theme in 2010, explained that the effects of aid are often hard to measure, but it is “possible to know which development efforts help and which hurt — by testing solutions with randomized trials.” Their method involves putting social initiatives to the same rigorous scientific tests that medical researchers use for drugs. This takes the guesswork out of policy-making, according to Duflo, by showing “what works, what doesn’t work and why.” 

Banerjee and Duflo never expected Poor Economics to make for good business. “We wonder how our book is a business book,” Duflo told Knowledge@Wharton. But the attention that their book received included being named the 2011 Financial Times Goldman Sachs Business Book of the Year award winner. 

An edited transcript of the conversation follows.

Knowledge@Wharton: How does it feel to win the book award?

Esther Duflo: We are certainly happy but at the same time, baffled. We wonder how our book [can be considered] a business book since it is about finding and thinking of ways to end global poverty. Nonetheless, we are happy that some businesses and businessmen are interested in our work.

Knowledge@Wharton: What, in your view, is the best way to tackle poverty?

Abhijit Banerjee: The central point of our book is that there isn’t a single answer, that the question itself is wrong. There is no single action that is going to solve the problem of poverty. There are perhaps a few hundred steps that we need to take, each of which will do something, as long as we take the right steps. There is no evidence that we could adopt one step that is far more important than the others. I think that the one-size-fits-all recipe for tackling poverty is an illusion. It is a convenient illusion, so that you can believe that you can solve the problem with a single step. But this does not seem to be supported by data.

“There is no single action that is going to solve the problem of poverty. There are perhaps a few hundred steps that we need to take, each of which will do something, as long as we take the right steps.”

Knowledge@Wharton: But surely of those hundreds of steps, there must be some crucial ones that come to mind when you talk of eliminating poverty?

Duflo: Yes, there are some crucial steps. I can’t say they are the most important but these, as we know today, are very effective. However, that is not to say that in the future there will not be other steps which would be even more effective.

According to our current state of knowledge, there is a zone of shadow where we are not sure exactly what to do. But there are some things we do know that work across sectors. Educating children, for example, is one of them — imparting quality education to them right from a young age. Similarly, there could be positive social and political impact of health care for the poor, which includes steps like better access to preventive health [and] finding ways to put iron, vitamins, etc., in the food that poor people consume, which we know will be good from a medical point of view. Giving an asset — like a cow — to extremely poor people, and then some help in taking care of that asset, also works. These, we feel, are some of the effective steps that can be taken in the initiative towards ending poverty.

Knowledge@Wharton: In your presentation at the Goa conference, you emphasized the quality of education, quality of curriculum, etc. In the case of the poor, where educating children is often seen as the only passport to escape poverty, does the quality of education matter?

Banerjee: Yes, that is absolutely crucial. If you don’t learn to read or acquire basic math skills by the time you are 13 or 14 — and there is a significant portion of students with this deficiency — then the entire effort is worthless. It is a crime that children should be subjected to such education. No one is talking to them, the whole class is going on and they do not understand a thing. There is nothing worse than this. I don’t believe we are giving children a decent chance to learn.

Duflo: [While] it is not good to say that no education is better than some education, an education that is so blatantly ignorant and irrelevant is a torture.

Knowledge@Wharton: How do you define poverty?

Banerjee: There is no one way to define poverty that is going to satisfy everyone. You have to make an arbitrary call as to what you think poverty is, and when you make that call, you should decide what you are trying to capture. If you are trying to capture a set of people who are so desperate that they need immediate help, that is one thing. If you want to say what is an unacceptable level of standard of living, and the country needs to somehow fix that problem, then that is a different thing. There is no definition that is independent of the question you have posed. If the policy question is, ‘How do I target emergency help?’, then that is one definition. It would be very different if you were to decide to set the target for, say, the next 15 years. Our poverty line for the desperately poor should be much lower than our goal that you will provide everyone at least ‘X’ after 15 years. It depends upon what policy question we are trying to answer.

Knowledge@Wharton: In India, poverty has always been a top-of-mind issue. Amartya Sen won the Nobel Prize for his work on poverty. C.K. Prahalad became famous for The Fortune at the Bottom of the Pyramid. Would you agree that poverty in India is a lucrative opportunity area?

Duflo: At a basic level, there are lots of poor people, and so they are a market. There are some forms of social businesses that have done well in this regard, and they are there for us to see. A lot of people inspired by C.K. Prahalad say that you can make money while helping the poor. However, you have to be a little more careful. I am not saying that these opportunities don’t exist. Sometimes it is true that there is a market and someone can come and creatively tap it. But there are also lots of things that the poor need and the market is not able to provide them. It is a big mistake to think that markets will be able to do everything. This notion is misguided.

Knowledge@Wharton: The Indian Planning Commission has set the poverty line at 65 cents a day. You have different views on this and on what parameters poverty should be measured. Would you briefly explain that?

Banerjee: Let’s not use [the conversion to U.S. cents]; it is very misleading. Let’s say Rs. 32, and let’s be very clear about what that means. The problem with that conversion is that it doesn’t allow for differences in prices. The latest conversion between the rupee and dollar that allows for differences in prices is roughly Rs. 19 to a U.S. dollar. That is the right conversion, not Rs. 48 or Rs. 49 to a dollar [the current exchange rate]. This is what the World Bank uses in PPP [Purchasing Power Parity] estimates. So, under this conversion, Rs. 32 is $1.70. That’s the right number, not 65 cents a day. For any of these questions [about poverty], the market dollar rate is not relevant. You take a bus ride here, it costs Rs. 3. In the U.S., it would cost $2. You have to adjust for the fact that things are cheaper in India than in the U.S. If you don’t do that, you get a completely misleading number.

Duflo: We are not deeply interested in measuring poverty. Others do it. People in the World Bank do it, and they put in lots of effort at measuring poverty levels. India itself has a long tradition of doing that very well. At the end of the day, what is important is not to measure poverty as much as try to understand what to do about it. In some sense, it is a democratic debate in which all people should have something to say.

“At the end of the day, what is important is not to measure poverty as much as try to understand what to do about it.”

Knowledge@Wharton: The poor, you say, are more discerning customers than the rich because they have to make a little go a much longer way. But do the government, aid workers, companies, etc., make any effort to provide choice?

Banerjee: It’s a good question. I think lots of aid policy and, in general, the social policy ignores the free will of the poor. They are seen as sort of desperate — you give them this and they are going to take it. The poor, on the other hand, are trying to have a good life within the constraints they face. And if you tell them to have a particular kind of food every day just because it is healthy food — for instance, eat chickpeas everyday — they are not going to do it. So you have to dig out their reality as to how they want to live their lives. We often tell them to boil water for 20 minutes but we don’t take the reality of their lives [into consideration]. Twenty minutes is a lot of time for a woman who has to do so many things in the house. You have to think of what the priorities are. You tell them to do things and then wonder why they are not doing them. They are not doing them because you have not understood their lives very well. You have to give them choices and understand their choice behavior. If you really want to do something for them, you have to make it attractive enough for them to do it. You can’t think of the poor as machines.

Knowledge@Wharton: Is the problem of global poverty too huge to envision and address?

Duflo: We need to cut this huge problem into many problems, problems that can be dealt with one by one. Some people term it as one huge problem and then conclude that there is one big solution to it. Projected this way, people get depressed by the enormity of the problem. The right thing is to say that it is not one giant problem, but a series of issues that need to be addressed in numerous ways. This way, there will be incremental victories and progress towards ending poverty.

Knowledge@Wharton: You say that the three villains of efforts to eradicate poverty are ideology, ignorance and inertia. And the three problem interfaces are the expert, the aid worker and the local policy maker. In rural India, it is the local government and the aid worker who are regarded as the biggest hurdles. What is your experience of other countries?

Duflo: Essentially, there is not much aid coming to India, and hence you cannot blame the aid workers. India gets very little aid, and most of the money on anti-poverty programs is spent by the government. So, there is no large presence of aid workers in India, but there are many good NGOs [non-governmental organizations] in India who are doing excellent work. There is a lot of difference between the places where policies are being designed, say in New Delhi, and people in the field. These field workers are doing good work, and they have a much better sense of what is actually going on at the field level. Their knowledge is not necessarily being harnessed. There are countries — in Africa, for example — which receive lots of aid, but essentially face the same issues.

Knowledge@Wharton: Do anti-poverty campaigns receive enough money? Is there adequate funding for projects such as yours?

Banerjee: No. There is a lot more money that could be spent. And even within what is being spent, little money goes into supporting innovative programs that can be carefully tested, carefully identified and carefully implemented. Too much money and governmental resources are spent on programs that have not been tested. Rather than spending it carefully, too much is spent on [the] wrong programs. There is too little money going into anti-poverty programs that we would like to see.

Knowledge@Wharton: What has your Poverty Action Lab (the Abdul Latif Jamil Poverty Action Lab or J-PAL) accomplished so far?

Banerjee: In 2003, we founded the Poverty Action Lab to encourage and support research on a new way of doing economics, based on what we call randomized control trials. These give researchers, working with local partners, a chance to implement large-scale experiments designed to test their theories. As of 2010, J-PAL researchers had completed or were engaged in 240 experiments across 40 countries. A very large number of organizations, researchers and policy makers have embraced the idea of randomized trials. Many have come to share our basic premise — it is possible to make very significant progress against the biggest problems in the world through a set of small steps, each well thought-out, carefully tested and judiciously implemented.

Knowledge@Wharton: How do you think the Occupy protest movement, that started on Wall Street and has spread globally, will impact the ongoing debate on global poverty?

Duflo: At the moment, the Occupy Wall Street protest is very much in response to domestic issues in the U.S., to the increasing inequalities in the U.S. in the past 10 to 15 years, to inertia and to the inadequate response to the economic crisis in the U.S. Global poverty is not at the forefront of their consideration at this point in time. So I frankly don’t know whether or not this protest is going to have any impact on how to think about solving the problem of global poverty.

“We need talent from the whole world to think about ways of ending poverty.”

Banerjee: There is a general worry that it will lead to an irresponsible, populist backlash in policymaking in the West, like anti-trade and all that. It could turn into something like the Tea Party. And that will be bad. But right now, they are just reacting to what is the real problem in all the OECD [Organisation for Economic Co-operation and Development] countries. In some of these countries, there is ballooning inequality, and there is very little official response to tackle this.

Knowledge@Wharton: Recent studies have shown that poverty has increased dramatically in the U.S. over the past decade, due in part to the economic downturn. In a world of continuing global financial uncertainty, what challenges do we face in fighting global poverty?

Banerjee: Slowing growth in the West is a huge problem for growing countries like India, China, Bangladesh and Pakistan, which rely on servicing these markets. They are all going to face some constraints on that account. It is also clear that there is a certain amount of policy attention or creativity that is now being directed to finding the equilibrium within [Western] economies. After all, we need talent from the whole world to think about ways of ending poverty.

Duflo: The crises in the end have to affect the lives of the very poor. The immediate impact of the global financial crisis was not as harsh on the very poor as on the middle classes of the rich countries. What is more worrisome is the inability to get out of the crisis in the past few years. That eventually will create additional challenges, particularly for the people in the developing world.