Poor economics – Esther Duflo, Abhijit Banerjee – 2011



The field of antipoverty policy is littered with the detritus of instant miracles that proved less than miraculous. To progress we have to abandon the habit of reducing the poor to cartoon characters and take the time to really understand their life.

The poor are no less rational than anyone else. Precisely because they have so little, we often find them putting much careful thought into their choices: they have to be sophisticated economists just to survive.

Living on 99 cents a day means you have limited access to information and just don’t know certain facts that the rest of the world takes as given, like, for example, that vaccines can stop your child from getting measles….It means not having anywhere safe to keep your money, because what the bank manager can make from your little savings won’t cover his cost of handling it.

It is not easy  to escape poverty but a little bit of well-targeted help (a piece of information, a little nudge) can sometimes have surprisingly large effects. Similarly minor hurdles can be devastating.

About an experiment : students, it seems, were willing to take some responsibility for helping a 7 year old from Mali, but when faced with the scale of the global problem (3 millions Zambian face hunger) they felt discouraged.

Jeffrey Sacks, the poverty trap: countries are poor because they are hot, infertile, malaria infested, often landlocked. This makes it hard to be productive without initial large investment to help them deal with these endemic problem. But they cannot pay for these investments precisely because they are poor. Until something is done, neither free market nor democracy will do very much for them. This way foreign aid is key.

William Easterly has become one of the most influencial anti-aid public figures (joined by Bambisa Moyo) and argues that aid does more bad than good.: it prevents people from searching their own solutions, undermine local institutions and create a self perpetuating lobby of aid agencies. The best bet is to rely on free markets. The poverty trap does not exist. Some rich country becomes poor, poor countries become rich: if the condition of poverty is not permanent then the idea of a poverty trap is bogus (P10)

The data on a couple of hundred countries show that those who receive aid more aid did not grow faster than the rest. This is often interpreted as evidence that aid does not work. But perhaps the aid help them avoid a major disaster, and things would have been much worse without it. We simply do not know. This book will say whether particular instance of aid did some good or not.

It is not clear that answering some of the big question is important. Although aid looms large in London, Paris or Washington, it is only a very small part of the money that is spent on the poor every day. India received virtually no aid in 2004-2005, yet it spend US$31bn PPP on primary education for the poor. In Africa, aid represent 5.7% of total budgets in 2003 (12% if we exclude Nigeria and South Africa, two countries that receive little aid).

More important : where the money goes. This a matter of choosing the right kind of project to fund – should it be food for the indigent, pensions, clinics – and then figuring out how best to run it.

Amartya Sen: poverty is not just a lack of money; it is not having the capability to realize one’s full potential as a human being.


Millenium Village project: 500,000 per village a year, to transition from self subsistence farming to self sustaining commercial activities.

The lack of grand universal answer might sound vaguely disappointing but it is what policy want to know  – that there are few key factors that create trap, and that alleviating those particular problems could set them free and point toward a virtuous cycle of increasing wealth and investment.

First there are now high-quality data from a number of poor countries and second we have powerful tool: randomized controlled trials (large scale experiment – individuals or communities are randomly assigned different treatments. Since the individuals/communities are comparable, any different between them is the result of the treatment.

2003 the poverty action lab (Abdul Latif Jameel PAL) to encourage researchers and diffuse knowledge among policy makers. By 2010, 240 experiments in 40 countries.


MDG : poor people were essentially defined at having not enough to eat.

It is assumed that poor people need food and without food, they are less productive and remain poor. Yet this is not what we see. Most people living with less than 99 cents a day do not seem to act as if they are starving. Even with the very poor, food expenditure increase much less than one for one with the budget. Equally remarkable, when poor people put some money into more food, they do not maximize calories or nutrients. They buy better tasting, more expensive calories. Remarkably in an experiment in India, the caloric intake of those who received the subsidies did not increase. Probably because the subsidies made people richer, they consumed less of the staple food (the food of the poor).

It is not that the lack of food could not be a problem or isn’t a problem from time to tim, but the world we live in today is for the most part too rich for hunger to be a big part of the story of the persistence of poverty. Sen has shown that most recent famine have been caused not by lack of food availability  but by institutional failures that led to poor distribution of available food.

No large country has less medals in the Olympics than India. India has 0.92 medal per Olympic, just below Trinidad and Tobago. Bengladesh, Pakistan are also doing very badly. It seems plausible that there is a connection between wasted children and Olympians.

In rich and poor country, taller people earn more. In UK, US studies showed that good nutrition in early childhood lead to taller and smarter people. And because they are smarter, they earn more.

P35 A family (from South Africa) that had just lost one of its potential earners might have spent some 40% of the household annual income for the funeral party. In Africa social norms for funeral were set at a time when most deaths occurred in old age or infancy. As a result of HIV/AIDS epidemic many prime age adult died without having accumulated burial savings. Poor people in developing countries spend large amounts on social events (wedding, christening…) probably in part as a result of a compulsion not too loose face.

Generally, things that makes life less boring are priority for the poor. This may be television, something special to eat, just a cup of sugary tea. In Udaipur, poor spend 15% of there budget on festivals.

The poor  may well be more skeptical about supposed opportunities and the possibility of radical change in their lifes. They often behave as if they think that any change that I significant enough to be worth sacrificing for will simply take too long. This could explain their focus on the here and now, on living their lives as pleasantly as possible, celebrating when occasion demands it.

Child who gets proper nutrients in utero or during early childhood wil earn more money every year of his or her life. This suggests that we need to rethink food policy (based on supply of more grains, more money – even increase income). In contrast investing in pregnant women and children is tremendous (fortified food in school, treating children for worms…)

Sacks’ poverty trap : health is a source of different traps: workers many workdays, children unable to do well in school, mother may have sickly babies…’The end of poverty’ optimistic message is about eradicating disease that affect people in poor country (malaria etc.). Studies shows that eradicating malaria does results in a reduction of long term poverty, although the effects are not as nearly as large as those suggested by Sacks (yet they can be significant 50% increase in income for the entire adult life – no enough to get people out of poverty in a generation, as Sacks argues).


P45 In 2008, 13% of population lacked access to improved water sources and 25% access to water that is safe to drink. It is estimated that by piping uncontaminated, chlorinated water to households, it is possible to reduce diarrhea by 95%. Lorna Fewtrell, HNP discussion paper 2004 “water, sanitation and hygiene kinks to health”. Nevertheless, today at $20 per household per month, providing piped water and sanitation is too expensive for the budget of most developing countries. Experience in India show that it is possible to do it much more cheaply: $4 per household.

We argued that buying bed nets to reduce the risk of getting malaria has the potential to raise annual incomes by a substantial 15%. However, the demand for bed nets, though very sensitive to price, is not very sensitive to income: people who are 15% richer are only 5% more likely to buy a net.

There is a “psychological sunk cost” effect: people are more likely to make use of something they have paid a lot for. In addition people may judge quality by price. In the “white man’s burden” Easterly seems to suggest that this is what is going on, based on anecdotes (bed nets used as wedding veil etc.). However there are a number of careful experiments that such anecdotes are oversold. Studies that have tested whether people use things less because they go them for free found no evidence of such behavior.

Immunization just prevents some diseases – there are many others – an uneducated parents do not necessarily understand what their child is supposed to be protected against. So when the child gets sick despite immunization, they are likely not to go through it again. They may not understand why all the different shots are needed. And they generally do not trust their doctors (which may not even be doctors at all, as in India).

The poor might appear to believe in all kind of things, but there is no much conviction behind many of those belief.

It is easy to see why small incentives encourage the use of life saving device and why small cost discourage it. A we may want to postpone the purchase of a bottle of chlorine until later because we have a better use for the money now. Vice versa a small incentive received today compensates for the little extra cost (in time or money or side effect) she bears today (by getting her child immunized for instance).

Given high sensitivity to prices, delivering preventive services for free, rewarding people for getting them or making them the default option when possible. There should be public investment in water and sanitation infrastructure in densely population areas.

Aren’t we , those who live in the rich world, the constant beneficiairies of a paternalism now so thogroughly embedded in the system that we hardly notice it? It not only ensures that we take care of ourselves better than if we had to be on top of every decision, but also,  by freeing us from having to think about these decisions, it give us the mental space to focus on the rest of our lives.

Enrollment rates have increased in all developing countries from 1999 to 2006, the number of children of school age who were out of school went from 103M in 1999 to 73M in 2006. The implicite assumption was that learning would follow from enrollment. Unfortunately, things are no so simple.

Conditional cash tranfers: PROGRESA in Brazil, parents gets more money if their children are in secondary than primary, more money if it was a girl who went to school rather than a boy. And payment were presented as compensation to the family for wages lost when children went to school. Such programme does substantially increase school enrolment rates.

Comparing conditional and unconditional transfers, a study in Malawi showed that the effects were large on drop out between those who received transfers and those who did not, but effect were the same for those who receive conditional and unconditional transfers: parents did not need to be forced to send children to school, they needed to be helped financially.

If parental income plays a vital role in the educational investment, rich children will get more education even if they are not particularly talented. And talented poor children may be deprived of education. Leaving it the market will not allow every child to be educated according to his/her ability.

P82 Unless parents are unwilling to treat their children differently from one another, it makes sense for them to put all their educational eggs in the basket of the child they perceive to be the most promising….rather than spreading the investment evenly across all the children.

Lower caste teachers were actually more likely to assign worse grade to lower-caste students (compared with higher-caste teacher). They must have been convinced that these children could not do well.

I should now be clear why private schools do not do better at educating the average child – their entire point is to prepare the best performing children for some difficult public exam that is the stepping-stone toward greater things….

In 1991, 15,000 more or less indigent Ethiopian Jews, including many children, were airlifted out of Addis Ababa in a single and dispersed into communities all over Israel. These children, whose parents completed on average between one and 2 years of schooling, reached graduation level with as much success as kids from other Israeli children, whose parent had on average 11.5 years of schooling. Even the most severe disadvantage in terms of family background can largely be compensated for.

It takes relatively little training to be an effective remedial teacher, at least in the lower grades.

By the end of 1976, 21% of Indian couples were sterilized (large financial incentives for those who agreed to be sterilized; compulsory sterilization laws). But the violations of civil liberties that were integral part of the implementation of the program were widely resented. It is widely believed that Indira Gandhi’ defeat in 1977 elections was in part driven by popular hatred for this program.

“The gift of dying” in a article Alwyn Young argued that epidemic (of HIV/AIDS) would make future generation better off by reducing fertility. Reduction occurs first because of the reluctance to engage in unprotected safe and because the resulting labor scarcity  makes it more attractive for women to work. South Africa could be 5.6% richer in perpetuity as a direct consequence of HIV/AIDs. “one cannot endlessly lament the scourge of high population growth in the developing world and then conclude that a reversal of such processes is an equal economic disaster.”

However the claim that HIV/AIDS will lead to reduce fertility has been refuted and the fact that reduced fertilityy will make everyone richer is less obvious than it sounds (we are richer today – and more numerous – than at the time of Malthus….)

Today there is no argument that large family are bad for children, contrary to what Sacks argues in “Common Wealth”.  That family does not adversely affect children seems counterintuitive, however: if the same resources have to be shared among more people, some of them should end up with less. If children do not suffer, who does? One possible answer is the mother.

Contraceptive access may make people happy by giving them a much more convenient way to control fertility. But it appears to do, in itself, little to reduce fertility. But better access to contraceptives help teenagers postpone pregnancies.

ABCD (Abstain, Be faithful, Use condoms or die): tested in 170 schools, no changes in behavior were reported and pregnancy rates after 1, 3 and 5 years after intervention was the same as in school without ABCD teaching.

Fertility is in part a social and religious norm, and deviation from it get punished  (by ostracism, ridicule or religious sanctions). Therefore, it matters what the community deems to be appropriate behavior.

The contraceptive adoption by Hindus had no effect on the adoption by their Muslims neighbors, and vice versa. This pattern must mean that the women were progressively learning about what was acceptable behavior within their communities.

Telenovelas in Brazil (where most female characters under the age of fifty had no child and the rest had one): right after soaps became available in an area, the number of birth would drop sharply (and many children were named after the characters in the soaps)

In India, girl babies stop getting breast fed earlier than boys, which means they start drinking water earlier and have accelerated exposure to waterborne disease diseases. In addition, after the birth of a girl, parents are more likely to want to stop breast feeding earlier in order to increase the wife’s chances of getting pregnant again (breast feeding act as contraceptive).

Economists often ignore the fact that a family is not just one person. There are important policy consequences from ignoring the complicated dynamics within the family. Giving women access to formal property title is important for fertility choices because it makes her view count more.

Perhaps women are expected to do something for the family when they get some windfall cash and men are not. If this is the case not only who get the money but how it is earned, also matters.

Of course the global crisis increased risk for the poor, but it added little to the overall risk they have to deal with daily, even when there is no crisis for the World Bank to worry about. During the Indonesian crisis in 1998….rice farmers, who tend to be among the poorest people, actually gained in terms of purchasing power. Government employees ended up worse off.

Drought may have more negative impact on wages in the villages that are more isolated: workers will compete with each other to find job and, given the limited amount of work available, competition with drive salary down. Working more is not always an effective way of copping with getting paid less.

The insurance company does not want only sick people signing up. To avoid adverse selection, the trick is to start from a large pool of people who came together for some other reason than health (employee of large firm etc.). this is why many microfinance institutions thought of offering health insurance.

Another possibility is that the poor do not understand the concept of insurance very well. It is true that insurance is unlike most of transaction the poor are used to. Something you pay for, hoping you will never need to make use of it….Farmers were able to understand the concept but were simply not interested in buying it. They were however swayed in their decision by relatively small things (home visits). The household must trust the insurer completely: trust, credibility of the insurance is an issue.

There are reasons why people don’t feel comfortable with insurances proposed on the market. On the other hand the poor clearly bear unacceptable risks. There is thus a clear role for government action.

Microfinance, with its 150 to 200 millions clients, has earned its place as one of the most visible anti-poverty policies. But does it works.  Unfortunately, until recently  there was in fact little evidence to prove this.

Many positive case studies, anecdotal evidence was sufficient for supporters of micro credit. Governments however worry that microcredit might be a new usury. Because most of the Microcredit Institution are financially sustainable, evaluating how beneficial they are was not necessary. But in reality they are generally subsidies by donors or by policies (priority sector qualifying for incentives etc.)

Microfinance give the poor a chance to map out the future in a way that was not possible for them before, and this is the first step toward a better life….we found no evidence that women were feeling more empowered, at least along measurable dimensions. They were not exercising greater control over how the household spend the money.

Microcredit has earned its rightful place as one of the key instruments in the fight against poverty. Microcredit is not miraculous, but it is working (though more studies are needed).

Why microcredit is not more popular:  it probably has something to do with precisely what makes it able to lend relatively cheaply and effectively: its rigid rules and the time costs it imposes on clients.

In many ways, their success comes from making repayment an implicit social compact, in which the community ensures that loans will be repaid and MFI continues to provide loans. The situation can quickly unravel is trust disappears (if people believe the MFI will go under then they stop repaying…). Opening the doors to defaults, even as a way to encourage risk taking, may lead to an unraveling of the social contract that allow them to keep repayment rates high and interest rate relatively low. The necessary focus on repayment discipline implies that micro finance is not the natural or best way to finance entrepreneurs who want to go beyond micro-enterprise. Microfinance is not well suited to discover who has a appetite for risk taking.

We are far from seeing the equivalent of the microfinance revolution for SME, nobody has figured out  how to do it profitably on a large scale. This is the next big challenge for finance in developing countries. Changes in business environment, such as improvement in the functioning of the courts, may well make that difference.

That view of the poor as essentially different people, whose innate inclination toward shortsighted behavior is what keeps them poor, has persisted over the years. Garry Becker, 1997: the possession of wealth encourages people to invest in becoming more patient (by implication poverty makes people more impatient).

As a result of administrative fees, most of the poor may not want a bank account even when they are entitled to one.

The poor live under considerable stress, and stress induced cortisol makes is choose more impulsive decisions. The poor thus have to do a harder job on fewer resources.

The vast majority of the businesses run by the poor never grow to the point where they start having employees or much in a way of assets. In addition they are not making much money. The very small scale of many of these businesses explains why their overall returns are often so low, despite the high marginal return. This low profitability of business run by the poor also explain why microcredit does not see, to lead to radical transformation in the client’s lives. Small business can’t borrow and what they borrow is very expensive. And even if money was available, developing business may require some management and other skills that they don’t have and can’t afford to buy. Therefore they are stuck staying small.

Given that micro entrepreneur may not be fully committed to make every penny count (knowing that it would not change their life in a meaningful way) may also explain the disappointing effects of the business training programs that MFIs offers to their clients. This lead to improve business knowledge but no changes in profits, sales and assets.

The enterprises of the poor often seem more a way to buy a job when a more conventional employment opportunity is not available. It is not a reflection of a particular entrepreneurial urge.

Stability of employment is what distinguish the poor from the middle class (weekly or monthly pay rather than daily pay). A sense of stability may be necessary for people for people to be able to take the long view.

Easterly has criticized randomized controlled trials (RCTs): they are infeasible for many big questions in development, like economy wide effects of good institutions and led development researchers to lower their ambitions.

Why Nations Fails: these institutions (political, economic institutions) are the prime driver of success or failure of a society. Acemoglu, Robinson showed that  former colonies where the disease environment prevented large-scale settlement of Europeans tended to have worse institutions during colonial times (because they were picked for being exploited from afar) and these bad institutions continued after decolonization.

The Bottom Billion : Paul Collier, it is western world’s duty to get the countries stuck in vicious circle of bad economic and bad political institution out, if necessary through military intervention.

Easterly : freedom (political and economic) is the best advice we can give to poor countries. But it can’t be imposed from outside. All that can be done is to campaign for the ideals of individual equality and rights.

Even a very imperfect local elections can make a substantial difference in how local government are run.

It might therefore be better for the decentralization to be designed by a centralized authority, with the interest of the less advantaged or less powerful in mind.

In Benin, the clientelist speech was a clear winner (comparing political campaigner speech – one about national unity, one more clientelist) with 10% more vote. Being from the right caste of ethnic group is sufficient to be elected.

Brazil, since 2003, 60 municipalities are drawn at random every month in a television lottery, and their accounts are audited. Results are made public through  the internet and local media. Honest incumbent were 13% more likely to be elected at the next elections. Corrupt incumbent were 12% less likely to be re-elected.

First the poor often lack critical pieces of information and believe things are that not true (e.g.in health, educ etc.).  They bear responsibility for too many aspects of their lives (i.e. the richer you are the more right decision are made for you, social insurance etc.).There are good reasons that some markets are missing for the poor, or the poor face unfavorable prices in them (e.g. they get negative interest on their accounts – too little in bank, too large admin fees). Poor countries are not doomed to fail because of their history. Finally, expectation about what people can or can not do  end up turning into self-fulfilling prophecies (role of expectations means that success feeds on itself).