by Dean Karlan
You could raise the commonsense objection that, even if they wanted to, poor people simply don’t have money to cut themselves checks for good behavior, or to penalize themselves for making bad choices—precisely because they’re poor. But we’ve seen this kind of thing happen before. In chapter 7 we met Sunny, who tied her money up in a SEED commitment savings account so she could make improvements on her house. It turned out that Sunny was not alone. SEED ultimately helped hundreds of people resist the temptation to spend and instead save up for much-needed expenses.
Encouraged by the success of SEED, I went back to the Philippines in 2006, this time with Xavier Giné and Jonathan Zinman, to see if a similar product could work for health too. We worked again with Green Bank, our partner from the SEED study, to develop CARES (for “Committed Action to Reduce and End Smoking”) and test it with an RCT.
CARES was a simple, no-nonsense commitment savings account. Clients started by making an initial deposit of fifty pesos (about a dollar) or more. Once a week for the next six months, each client received a visit from a Green Bank deposit collector and had a chance to add to his account. At the end of six months there was a one-time urine test for nicotine. If the client passed, he got all his money back (with no interest); if any nicotine was found, the entire balance went to a local orphanage.
From the perspective of traditional economics, CARES should have been even less attractive than SEED. The worst that would happen with a SEED account was that you’d have to wait to spend your money until you reached the balance or time goal you had set. With CARES, the best-case scenario was the same—you just cashed out the money you had put in—but you could actually lose your savings if you slipped up!
We usually cannot get the free market to increase the price of things that we’d rather avoid, like cigarettes or fatty foods. But, all other things being equal, it’s much easier to resist our temptations when giving in to them is expensive. The simplest way to explain CARES, and stickK.com, is that they provide people a tool to change the relative prices of good and bad behavior—making their vices more expensive, or their virtues cheaper.
Our study left no doubt that at least some of the poor were ready to lay down such a gauntlet for themselves. Of 640 smokers approached on the street by Green Bank staff, 75 (about 12 percent) opened accounts when given the chance. That’s a pretty impressive figure, I think: More than one in ten smokers not only wanted to quit, but wanted it bad enough to put money at risk on the spot.
People were willing to sign up, but would CARES really help them kick the habit? Maybe smokers who were serious enough about quitting to wager their hard-earned money in a CARES account would have been successful even without the aid of a commitment device. To find out, we had to compare. So in addition to the 640 CARES offers, we monitored about 600 more smokers (without offering them accounts) as a control group.
The upshot was that CARES worked. Six months after the marketing, those who were offered CARES—including those who turned it down!—were about 45 percent more likely to pass the nicotine test than the control group. Now, two things are worth noting. First, quitting wasn’t easy for anybody. Only 8 percent of the control group smokers passed the six-month test; even more striking, just a third of those ambitious enough to put money on the line with a CARES account wound up recovering their savings. Second, one could argue that the six-month test was an unfair comparison anyway. Unlike those in the control group, people who opened CARES accounts had a lot riding on that nicotine test. Maybe they were just staying on their best behavior to recover their savings, and would relapse when the pressure was off.
So we waited another six months to see whether the results would stick. A full year after we had begun, we approached all the smokers again and asked them to take another nicotine test. This one was on-the-spot and equally unexpected for everyone. And all the CARES accounts were closed, so there was no incentive for anyone to stay smoke-free just to game the system. Here again, the clients who had been offered CARES—including those who turned it down—did markedly better. It looked like they truly had managed to quit.
Of course, CARES didn’t work for everybody. The 88 percent of people who passed on the offer to open an account probably didn’t benefit much from it, and even those who did sign up were hardly guaranteed success in quitting. But there are two important lessons here. First, CARES was a powerful motivator for those who opted in, and, compared with health incentive programs like Progresa, it’s cheap to offer. All CARES really provided was a lockbox, a deposit collector, and a nicotine test. Those cost something, but they’re much less dear than the essential ingredients—willpower and financial incentives to harness it—which participants brought to the table themselves.
And that is the second lesson: The developing world is ready for these kinds of solutions. There are, among the poor, countless people who stand on the cusp, who genuinely want to improve their lives and will go to great lengths to do it. All they need is a vehicle, a tool. No single tool is a universal cure-all. But if we can make one that’s proven to work for some—and if we can get it to those people—then it is a step in the right direction.
Malaria
Four days out of five, Davis P. Charway came to work in a three-piece suit; and even when he didn’t, his shirts were always impeccable. They were that kind of thick, soft, textured cotton you see on the front table at Nordstrom’s, woven with a tiny herringbone design. As a rule, he wore a French cuff. His cufflinks, like the exotic beetles at the Smithsonian, were shiny and came in every color. They always matched his necktie. Davis looked sharp sitting at his desk and he also looked sharp sitting behind the wheel of his car, a black Mercedes E-Class sedan. When it came up in conversation, he never failed to poke fun at himself. “It is only an E-Class,” he would say. “When I am really a big man, then I will upgrade to the S.”
By any reasonable standard, Davis already was a big man. From a typical childhood in the Ashanti region of central Ghana, he had made it to the Ivy League and beyond, ascending finally to a vice president’s office high above lower Manhattan in the headquarters of a major credit card company. After a few years he made good on a growing desire to return home and became an executive manager of a Ghanaian microfinance bank based in Accra, the capital city.
That was the capacity in which Jake knew him. IPA had partnered with his organization to study its microloans, and Jake was a research assistant on the project. Once, after only a few weeks in the country, he went to Davis’s office for a meeting and found the place locked up. His secretary said he was out sick. Jake asked whether everything was all right, and she said, “Oh, don’t worry about Davis. He’s just having some small malaria. I’m sure he will be back by Friday, or else next week.”
At the time, Jake’s knowledge of malaria was limited to what he had read in travel advisories and in news items that invariably characterized the disease as a great scourge of the developing world—none of which had led him to believe that it was something for bank executives to worry about. It conjured up images of dense tropical jungles, rural villages, thatch-roofed huts with unscreened windows, and squalid slums full of standing water. It was hard to see how a malarial mosquito could even have bitten Davis. In his office, his car, and his house, he always kept the windows shut so he could run the air-conditioning.
Over the following twenty months, as he watched dozens of bank employees lose weeks of work to similar flare-ups, Jake got to know better. Malaria respects neither rank nor social standing. It afflicts businessmen, farmers, and beggars alike. (And my eight-year-old daughter, Gabriela, got it this past summer in Ghana. Her case illustrates the mixed blessing of expensive drugs: Even for more than a dollar a day, the prophylactics do not always prevent getting malaria—but they do soften the symptoms significantly. She was quite a trouper.)
The standard depictions that informed Jake’s initial impression were incomplete, but they were not inaccurate. The poor and defenseless do suffer the most, and the most acutely,
when malaria strikes. In the summer of 2009 he visited Port Victoria, Kenya, on the shores of the great lake from which it takes its name. Approaching the town, he came over a rise and saw the place spread out below, a cluster of buildings beside the glittering lake, and big rectangular fields of sorghum and finger millet marching off into the distance.
For residents, living where they do is both a blessing and a curse. Being so close to the lake, fish like tilapia and Nile perch are plentiful, but large areas of the low-lying town flood frequently and for weeks at a time, creating a perfect environment for parasites and mosquitoes to breed. When this happens, the population writhes quietly in the grip of a bitter pestilence. People contract malaria and other parasitic diseases, and they fall sick in droves. Some make it through and some succumb; scores and scores die, leaving behind orphaned children and empty houses. Of the eight hundred students at Port Victoria’s Lunyofu Primary School, three hundred had lost both parents. Both.
But Jake didn’t know any of that when he first arrived. He had actually come to Lunyofu on behalf of IPA, hoping to learn about the school’s experience with the tremendously successful deworming program we saw in chapter 9. The headmaster, Michael, was happy to oblige. He registered his satisfaction with the program in no uncertain terms, and even invited a group of students from the eighth grade class to talk about their impressions of deworming day. Most of the students had vivid stories to tell, but a few girls stood quietly at the back. Once the others had finished, Michael called them forward. The six of them formed a line by the side of his desk and cast their eyes down. They seemed embarrassed.
Michael looked them over, turned his attention to Jake, and began: “I want to make a compassionate plea.” The girls, he went on to explain, came from small, remote islands in the lake, far from shore. Their parents had sent them to Port Victoria for schooling, but without relatives on the mainland, they had no place to stay. Michael had done what he could, which was to offer them the schoolhouse. It wasn’t much, but it was their only option. At night, hours after the other students had returned home, the girls pushed the long wooden desks to the edge of their classroom and slept on the floor. They had little more than the roof over their heads. With their ill-fitting doors and without screens on their windows, the rudimentary buildings were hardly an obstacle for the mosquitoes.
Every morning the girls woke up to find new bites, and they were often stricken with malaria. Bed nets would have provided the protection they needed, but the school couldn’t afford them. There was, after all, no budget for housing. Had they been expecting mothers, or under five years old, the girls could have acquired the nets for free at any government health clinic. (In Kenya those groups are eligible for fully subsidized bed nets.) Being neither, they would have to pay. And since they couldn’t pay, they would continue to come into the headmaster’s office when they were called; they would stand bashfully, averting their eyes; and they would wait for Michael’s compassionate plea to fall on sympathetic ears.
Fighting Malaria: Selling Versus Giving
Biologically, malaria is the same everywhere—the same five protozoan parasites, carried from person to person by the female Anopheles mosquito. A mosquito becomes a carrier when she bites an infected person. Then she has to stay alive for two weeks (which is hardly a given; most Anopheles mosquitos’ entire lives last about two weeks) while the parasite gestates in her abdomen. Only then can she pass it to other humans, which she does when she bites them to feed.
Which particular people she bites makes all the difference. For Davis P. Charway, malaria is an inconvenience; for the people of Port Victoria, it’s a plague. While the particulars vary widely, all malaria stories are sewn together with the thread of needless suffering, which is reason enough to work toward eradication. So what’s the best way to fight?
The first step is to keep the enemy at bay. During waking hours, a wave (or slap) of the hand is usually sufficient defense, but at night—which happens to be peak biting hours for the Anopheles mosquito—people need help. The most effective preventative measure developed thus far has been the bed net, a simple fabric screen, impregnated with insecticide, which hangs over a bed and tucks under the mattress.
Beyond protecting those who sleep under them, bed nets (much like the deworming pills we saw in chapter 9) also confer indirect benefits on the broader community by breaking the chain of transmission. As more people are directly protected, fewer opportunities exist for mosquitoes to become carriers, and consequently the risk of infection decreases for everybody. You could argue for providing free bed nets to the poor on a purely humanitarian basis; but these cascading social benefits make the case even stronger. A number of influential economists, most notably Jeffrey Sachs, have recommended just that. Their arguments have found traction in the aid community, which has distributed millions of free nets in the developing world over the past decade.
But others contend that it’s not as simple as setting up a card table in an impoverished community and doling out nets to all the passersby. Getting nets into the hands (and bedrooms) of the poor is, after all, only the first step. To be effective, the nets must be used properly; and this is where some economists raise a concern. They argue that blind giving is wasteful, that it ignores the preferences of those we intend to help. Think, for instance, of the coupons slid indiscriminately under windshield wiper blades in the parking lot of the grocery store. How many of them end up littering the ground? For whatever reason, some people might have no interest in sleeping under an insecticide-soaked screen; surely any bed net that fell into the hands of such a person would only collect dust. In his book The White Man’s Burden, William Easterly, NYU economist and former senior adviser at the World Bank, reports, “a study of a program to hand out free nets in Zambia to people, whether they wanted them or not . . . found that 70 percent of the recipients didn’t use the nets.”
This camp proposes to address the issue of waste with a market solution—by selling nets for a nominal price instead of giving them away. Requiring people to pay something, they argue, achieves two goals. First, it filters out those who don’t want, or don’t need, the product; second, it creates a sense of investment for those who do choose to purchase. Having sunk hard-earned cash into the nets, buyers should want to get their money’s worth.
Ever buy tickets to a show, and then when the night comes you don’t really want to go—but you still feel obligated? Behavioral economists call that the “sunk cost effect.” It is the phenomenon that makes you feel like you should go to the event, just because you paid for it, or that you really should finish that lobster dinner you ordered, even if you’re already full.
Now, an Econ would stop eating his lobster: He knows he’s already bought the meal, and that he’s going to pay for the whole thing, no matter how of it much he eats. So he’ll just eat the amount that makes him happiest. We Humans don’t always think that way. We force ourselves to get out of the house and go to the show. We finish our lobster dinners.
The idea of the market solution camp is to keep protection within reach for everyone—hence the big discounts—while using insights from behavioral economics to ensure that the nets go to people who will use them. Among practitioners, the market-based approach also has its adherents. Population Services International, a leading international health NGO that sells discounted nets in dozens of countries (but doesn’t give them away), claims that its programs prevented nineteen million episodes of malaria worldwide in 2007 alone.
What can we make of these conflicting stories, each with its cadre of supporters and its battery of compelling statistics? Amid the assertions and invective, there are a few pieces of hard evidence to guide our thinking. One comes from Jessica Cohen of the Brookings Institute and Harvard School of Public Health and Pascaline Dupas of UCLA and IPA, who worked together on an RCT to learn how the method of distributing bed nets impacts the way they are used. They snatched the debate between free and cost-sharing out of the realm of abstraction and put it d
own on the dusty ground in western Kenya. In their study, pregnant women who had come to public health clinics for prenatal care were offered bed nets at randomly assigned prices. Some women were offered nets for free, and others for a nominal price between fifteen and sixty cents apiece.
What they found was simple supply and demand, the bread and butter of classical economics, at work: Fewer people bought nets when they were more expensive. While this discovery lacked the power of revelation, the intensity of the women’s responses to price was remarkable. Cohen and Dupas calculated that increasing the price from zero to seventy-five cents (which was the going rate for a discounted net from Population Services International) would drive off three-quarters of customers!
Of course the drop-off in demand might have been acceptable, maybe even desirable, if the people being filtered out were less likely to use, or to need, the nets (that is, after all, the advocates’ reason for charging something in the first place). But it was not so. When the nets were distributed, clinic workers measured each recipient’s hemoglobin level, a strong indicator of malaria, and found that those who paid more were no more likely to be sick than those who paid less or nothing. So the invisible hand of the free market was not doling out protection to those who needed it most. Higher prices also failed to screen out people who wouldn’t use the nets effectively. Surveyors visited the homes of recipients a few weeks after the nets had been distributed to see whether they had been installed properly and to ask how they were being used. If, like those in the market solution camp, they were expecting to find most of the free nets still in their wrappers, they would have been disappointed. Roughly the same portion of nets—just over half—were found hanging in all the homes, regardless of the price that had been paid.
