Poor Economics

by Abhijit Banerjee

  Yet neither chlorine nor ORS is used very much. In Zambia, thanks to the efforts of Population Service International (PSI), a large organization that markets it at subsidized prices worldwide, chlorine is cheap and widely available. At the cost of 800 kwachas ($0.18 USD PPP), a family of six can buy enough bleach to purify its water supply, avoiding waterborne diarrhea. But only 10 percent of families use it.2 In India, according to the United Nations Children’s Fund (UNICEF), only one-third of children under five who had diarrhea were given ORS.3 Why are some 1.5 million children dying every year of diarrhea, a disease that could often be avoided in the first place, and could often be treated with boiled water, sugar, and salt?

  Bleach and ORS are not unique examples. There is other relatively “low-hanging fruit” with promise to improve health and save many lives. These are cheap and simple technologies that, if properly utilized, would save much in resources (in terms of extra days worked, less antibiotics used, stronger bodies, and so on). They could pay for themselves, in addition to saving lives. But too many of these fruits are left unpicked. It is not that people don’t care about their health. They do, and they devote considerable resources to it. They just seem to spend money elsewhere: on antibiotics that are not always necessary, on surgery that comes too late to help. Why does it have to be this way?

  THE HEALTH TRAP

  In a village in Indonesia we met Ibu Emptat, the wife of a basket weaver. A few years before our first meeting (in summer 2008), her husband was having trouble with his vision and could no longer work. She had no choice but to borrow money from the local moneylender—100, 000 rupiah ($18.75 USD PPP) to pay for medicine so that her husband could work again, and 300,000 rupiah ($56 USD PPP) for food for the period when her husband was recovering and could not work (three of her seven children were still living with them). They had to pay 10 percent per month in interest on the loan. However, they fell behind on their interest payments and by the time we met, her debt had ballooned to 1 million rupiah ($187 USD PPP); the moneylender was threatening to take everything they had. To make matters worse, one of her younger sons had recently been diagnosed with severe asthma. Because the family was already mired in debt, she couldn’t afford the medicine needed to treat his condition. He sat with us throughout our visit, coughing every few minutes; he was no longer able to attend school regularly. The family seemed to be caught in a classic poverty trap—the father’s illness made them poor, which is why the child stayed sick, and because he was too sick to get a proper education, poverty loomed in his future.

  Health certainly has the potential to be a source of a number of different traps. For example, workers living in an insalubrious environment may miss many workdays; children may be sick often and unable to do well in school; mothers who give birth there may have sickly babies. Each of these channels is potentially a mechanism for current misfortunes to turn into future poverty.

  The good news is that if something like this is what is going on, we may only need one push, one generation that gets to grow up and work in a healthy environment, to set the trap loose. This is Jeffrey Sachs’s view, for example. As he sees it, a large proportion of the world’s poorest people, and indeed entire countries, are stuck in a health-based poverty trap. Malaria is his favorite example: Countries in which a large fraction of the population is exposed to malaria are much poorer (on average, countries like Côte d’Ivoire or Zambia, where 50 percent or more of the population is exposed to malaria, have per capita incomes that are one-third of those in the countries where no one today gets malaria).4 And being so much poorer makes it harder for them to take steps to prevent malaria, which in turns keeps them poor. But this also means, according to Sachs, that public health investments aimed at controlling malaria (such as the distribution of bed nets to keep the mosquitoes at bay during the night) in these countries could have very high returns: People would be sick less often and able to work harder, and the resulting income gains would easily cover the costs of these interventions and more. To put it in terms of the S—shaped curve in Chapter 1, African countries where malaria is endemic are stuck in the left part of the curve, where their malaria-weakened labor force is too unproductive and hence too poor to be able to pay for malaria eradication. But if someone did them the favor of financing malaria eradication, they would end up on the right part of the curve, on the road to prosperity. The same argument could be made about other diseases that are prevalent in poor countries. This is the core of the optimistic message of Sachs’s book The End of Poverty.

  Skeptics have been quick to point out that it is not clear whether malaria-infested countries are poor because of malaria, as Sachs assumes, or perhaps their inability to eradicate malaria is an indicator of the fact that they are poorly governed. If it is the latter, then the mere eradication of malaria may achieve very little, as long as governance remains weak.

  Whose story—the activists’ or the skeptics’—does the evidence support? Successful campaigns to eradicate malaria have been studied in a number of different countries. Each of these studies compares high-malaria-prevalence regions in the country with low-prevalence regions and checks what happens to children born in these areas before and after the campaign. They all find that life outcomes (such as education or earnings) of children born after the campaign in areas where malaria was once prevalent catch up with those of children born in low-incidence areas. This strongly suggests that eradicating malaria indeed results in a reduction in long-term poverty, although the effects are not nearly as large as those suggested by Jeffrey Sachs: One study on malaria eradication in the U.S. South (which had malaria until 1951)5 and several countries in Latin America6 suggests that a child who grew up malaria-free earns 50 percent more per year, for his entire adult life, compared to a child who got the disease. Qualitatively similar results were found in India,7 Paraguay, and Sri Lanka, although the magnitude of the gain varies from country to country.8

  This result suggests that the financial return to investing in malaria prevention can be fantastically high. A long-lasting insecticide-treated bed net costs at most $14 USD PPP in Kenya, and lasts about five years. Assume conservatively that a child in Kenya sleeping under a treated net has 30 percent less risk of being infected with malaria between birth and age two, compared to a child who doesn’t. In Kenya, an adult makes on average $590 USD PPP a year. Thus, if malaria indeed reduces earnings in Kenya by 50 percent, a $14 investment will increase incomes by $295 for the 30 percent of the population that would have gotten malaria without the net. The average return is $88 every year over the child’s entire adult work life—enough for a parent to buy a lifetime supply of bed nets for all his or her children, with a chunk of change left over.

  There are other examples of highly effective health investments. Access to clean water and sanitation is one of them. Overall, in 2008, according to estimates by WHO and UNICEF, approximately 13 percent of the world’s population lacked access to improved water sources (typically meaning a tap or a well) and about one-fourth did not have access to water that is safe to drink.9 And many of these people are the very poor. In our eighteen-country data set, access to tap water at home among the rural extremely poor varied from less than 1 percent (in rural Rajasthan and Uttar Pradesh in India) to 36.8 percent (in Guatemala). The numbers tend to be much better for richer households, though they vary a lot from country to country (from less than 3.2 percent in Papua New Guinea to 80 percent in Brazil, for the rural middle class). They are higher in urban areas, both for the poor and the middle class. Decent sanitation facilities are even rarer among the poor—42 percent of the world’s population live without a toilet at home.

  Most experts agree that access to piped water and sanitation can have a dramatic impact on health. A study concluded that the introduction of piped water, better sanitation, and chlorination of water sources was responsible for something like three-fourths of the decline in infant mortality between 1900 and 1946 and nearly half the overall reduction in mortality over the same per
iod.10 Moreover, repeated bouts of diarrhea during childhood permanently impair both physical and cognitive development. It is estimated that by piping uncontaminated, chlorinated water to households, it is possible to reduce diarrhea by up to 95 percent.11 Poor water quality and pools of stagnant water are also a cause of other major illnesses, including malaria, schistosomiasis, and trachoma,12 any of which can kill children or make them less productive adults.

  Nevertheless, the conventional wisdom is that today, at $20 per household per month, providing piped water and sanitation is too expensive for the budget of most developing countries.13 The experience of Gram Vikas, an NGO that works in Orissa, India, shows, however, that it is possible to do it much more cheaply. Its CEO, Joe Madiath, a man with a self-deprecating sense of humor who attends the annual meeting of the world’s rich and powerful at the World Economic Forum in Davos, Switzerland, in outfits made from homespun cotton, is used to doing things differently. Madiath’s career as an activist started early: He was twelve when he first got into trouble—for organizing the labor on the plantation that his father owned. He came to Orissa in the early 1970s with a group of left-wing students to help out after a devastating cyclone. After the immediate relief work was over, he decided to stay and see if he could find some more permanent ways to help the poor Oriya villagers. He eventually settled on water and sanitation. What attracted him to the issue was that it was simultaneously a daily challenge and an opportunity to initiate long-term social change. He explained to us that in Orissa, water and sanitation are social issues. Madiath insists that every single household in the villages where Gram Vikas operates should be connected to the same water mains: Water is piped to each house, which contains a toilet, a tap, and a bathing room, all connected to the same system. For the high-caste households, this means sharing water with low-caste households, which, for many in Orissa, was unacceptable when first proposed. It takes the NGO a while to get the agreement of the whole village and some villages eventually refuse, but it has always stuck to the principle that it would not start its work in a village until everyone there agreed to participate. When agreement is finally reached, it is often the first time that some of the upper-caste households participate in a project that involves the rest of the community.

  Once a village agrees to work with Gram Vikas, the building work starts and continues for one to two years. Only after every single house has received its tap and toilet is the system turned on. In the meantime, Gram Vikas collects data every month on who has gone to the health center to get treated for malaria or diarrhea. We can thus directly observe what happens in a village as soon as the water starts flowing. The effects are remarkable: Almost overnight, and for years into the future, the number of severe diarrhea cases falls by one-half, and the number of malaria cases falls by one-third. The monthly cost of the system for each household, including maintenance, is 190 rupees, or $4 per household (in current USD), only 20 percent of what is conventionally assumed to be the cost of such a system.

  There are even cheaper ways to avert diarrhea, such as adding chlorine to water. Other very inexpensive medical or public health technologies with proven effectiveness include ORS, getting children immunized, deworming drugs, exclusive breast-feeding until six months, and some routine antenatal procedures such as a tetanus shot for the would-be mother. Vitamin B against night blindness, iron pills and iron-fortified flour against anemia, and so on are other examples of low-hanging fruit.

  The existence of these technologies is the source of both Jeffrey Sachs’s optimism and his impatience. As he sees it, there are health-based poverty traps, but there are also ladders we can give to the poor to help them escape from these traps. If the poor cannot afford these ladders, the rest of the world should help them out. This is what Gram Vikas does in Orissa, by helping to organize the villages, and by subsidizing the cost of the water systems. A few years ago, Joe Madiath told us he felt he had to turn down funding from the Bill & Melinda Gates Foundation when the grant officer insisted that the villagers should pay the full cost of what they were getting (fortunately, the foundation subsequently changed its view on this question). He argued that villagers simply cannot afford 190 rupees per month, even though it is true that the health benefits are potentially worth far more—Gram Vikas only asks villagers to pay enough money into a village fund to be able to keep the system in good repair and be able to add new households as the village grows. The rest the NGO raises from donors all over the world. In Sachs’s view, this is how things should be.

  WHY AREN’T THESE TECHNOLOGIES USED MORE?

  Underutilized Miracles

  There is one wrinkle with Sachs’s theory that poor people are stuck in a health-based poverty trap and that money can get them out of it. Some of these technologies are so cheap that everyone, even the very poor, should be able to afford them. Breast-feeding, for example, costs nothing at all. And yet fewer than 40 percent of the world’s infants are breast-fed exclusively for six months, as the WHO recommends.14 Another good example is water: Piping water to homes (combined with sewerage) costs 190 rupees per month, or 2,280 rupees per year, as we saw, which in terms of purchasing power is equivalent to about 300,000 Zambian kwachas. It is likely that poor villagers in Zambia cannot afford that much. But for less than 2 percent of that, a Zambian family of six can buy enough chlorine bleach to purify their entire drinking water intake for the year: A bottle of Chlorin (a brand of chlorine distributed by PSI) costs 800 kwachas ($0.18 USD PPP) and lasts a month. This can reduce diarrhea in young children by up to 48 percent.15 People in Zambia know about the benefits of chlorine. Indeed, when asked to name something that cleans drinking water, 98 percent mention Chlorin. Although Zambia is a very poor country, 800 kwachas for a bottle that lasts a month is really not a lot of money—the average family spends 4,800 kwachas ($1.10 USD PPP) per week just on cooking oil. Yet only 10 percent of the population actually uses bleach to treat their water. When, as part of an experiment, some households were offered a discount voucher that would entitle them to a bottle of Chlorin for 700 kwachas ($0.16 USD PPP), only about 50 percent wanted to buy it.16 This fraction went up sharply when the price was lowered to 300 kwachas ($0.07 USD PPP), but remarkably, even at this reduced price one-fourth of the people did not buy the product.

  Demand is similarly low for bed nets. In Kenya, Jessica Cohen and Pascaline Dupas set up an NGO called TAMTAM (Together Against Malaria), to distribute free mosquito nets in prenatal clinics in Kenya.17 At some point, PSI started distributing subsidized, but not free, nets in the same clinics. Cohen and Dupas wanted to find out whether their organization was still needed. They set up a simple test: They offered nets at various prices in different clinics, chosen at random. The price varied from free in some places to the (still subsidized) price charged by PSI in others. Much as in the case of Chlorin, they found that the purchase of nets was indeed very sensitive to the price. Almost everybody took a free net home. But the demand for nets fell to very close to zero at the PSI price (about $0.75 USD PPP). When Dupas replicated the experiment in different market towns but gave people the time to go home and collect cash (rather than having to buy on the spot), more people bought at the PSI price, but demand still went up by several times when the price was brought down toward zero.18

  Even more troubling is the related fact that the demand for bed nets, though very sensitive to price, is not very sensitive to income. To get on the right part of the S—shaped curve and start a virtuous circle where improved health and increased income reinforce each other, the increase in income coming from one person avoiding malaria should be enough to make it very likely that his or her children buy a net and avoid malaria as well. We argued above that buying bed nets to reduce the risk of getting malaria has the potential to raise annual incomes by a substantial 15 percent on average. However, even though a 15 percent increase in income is far more than the cost of a bed net, people who are 15 percent richer are only 5 percent more likely to buy a net than others.19 In other words, f
ar from virtually ensuring that the next generation sleeps under a net, distributing free bed nets once would only increase the number of children in the next generation sleeping under a net from 47 percent to 52 percent. That is not nearly enough to eradicate malaria.

  What the lack of demand underscores is perhaps the fundamental difficulty of the problem of health: The ladders to get out of the poverty trap exist but are not always in the right place, and people do not seem to know how to step onto them or even want to do so.

  The Desire for Better Health

  Since they do not seem to be willing to sacrifice much money or time to get clean water, bed nets, or for that matter, deworming pills or fortified flour, despite their potentially large health benefits, does that mean the poor do not care about health? The evidence suggests the opposite. When asked whether there was a period of a month in the recent past when they felt “worried, tense, or anxious,” roughly one-fourth of the poor in both rural Udaipur and urban South Africa said yes.20 This is much higher than what we see in the United States. And the most frequent source of such stress (44 percent of the time in Udaipur) is their own health or that of their close relatives. In many of the countries in our eighteen-country data set, the poor spend a considerable amount of their own money on health care. The average extremely poor household spends up to 5 percent of its monthly budget on health in rural India, and 3 percent to 4 percent in Pakistan, Panama, and Nicaragua. In most countries, more than one-fourth of the households had made at least one visit to a health practitioner in the previous month. The poor also spend large amounts of money on single health events: Among the poor families in Udaipur, 8 percent of the households recorded total expenditures on health of more than 5,000 rupees ($228 USD PPP) in the previous month, ten times the monthly budget per capita for the average family, and some households (the top 1 percent spenders) spent up to twenty-six times the average monthly budget per capita. When faced with a serious health issue, poor households cut spending, sell assets, or borrow, like Ibu Emptat, often at very high rates: In Udaipur, every third household we interviewed was currently repaying a loan taken out to pay for health care. A substantial proportion of those loans are from moneylenders, at rates that can be very high: The standard interest rate is 3 percent per month (42 percent per year).