Scarcity

Before you criticize someone, you should walk a mile in their shoes. That way, when you criticize them, you’re a mile away and you have their shoes.

Poverty is surely the most widespread and important example of scarcity. The breadth and depth of poverty in the modern world is striking. UNICEF estimates that 22,000 children die each day due to poverty. Nearly one billion people are so illiterate that they cannot even sign their names. Half the children in the world live below the global poverty line. Roughly 1.6 billion people live without electricity. Even in a country like the United States, poverty is stark. Nearly 50 percent of all children in the United States will at some point be on food stamps. About 15 percent of American households had trouble finding food for the family at some point during the year.

We have thus far treated the varieties of scarcity as if they were interchangeable. We have bounced from dieting to deep poverty to time pressure with little concern for the differences. This, after all, is our thesis. If scarcity evokes a unique psychology irrespective of its source, then we are free to treat the varieties of scarcity all the same. If there is a common psychology of scarcity, shouldn’t everything we observe about the poor also hold for the busy or for dieters?

Just because the different forms of scarcity share common ingredients does not mean they will have similar outcomes. In chemistry, the same basic elements can produce different compounds, depending on the proportions. Carbon and oxygen can form carbon dioxide—an essential ingredient for the cycle of life—or they can form carbon monoxide, a deadly pollutant. Same ingredients, very different outcomes. Our analysis of scarcity follows a similar logic. There are the common ingredients: tunneling, borrowing, a lack of slack, the bandwidth tax. But these play themselves out in different ways, depending on the context. In the case of money scarcity, borrowing is an obvious feature. In the case of loneliness, however, it is unclear what borrowing even means. That particular ingredient, like that additional atom of oxygen, is simply missing in the case of the lonely. The ingredients of poverty create circumstances that are particularly hostile to the scarcity mindset.

A well-to-do professional who is very busy is in that situation because he has taken on many projects. He would be less busy if he simply took on fewer. He could, in effect, choose to have less scarcity. The extent of his scarcity is, to some extent, discretionary.

This discretion provides a critical safety valve that can limit scarcity’s stress and damage. The tourist frantically trying to see Italy in a week can only get so worked up about her scarcity of time. At some point, she may simply say, “Forget this, I’ll just see the Colosseum on another visit,” or, “I’ll stay another day in Rome and see less of the south.” This safety valve limits the damage and depth of the scarcity trap. For those who have some discretion, the scarcity trap threatens but only so much. The overcommitted can miss a few deadlines. Dieters can take a break from their diet. The busy can take vacations.

One cannot take a vacation from poverty. Simply deciding not to be poor—even for a bit—is never an option. There is no equivalent in the world of poverty to the dieter deciding to live with being overweight or the busy person giving up on some of his ambitions. It would be silly to suggest that the rural poor in India should cope with money scarcity by simply moderating their desires. Basic desires, for clothing, freedom from disease, even modest toys to bring joy to one’s children, are significantly harder to cast off. The poor are not alone in having mandated scarcity. The dieter who faces a serious medical condition, the profoundly lonely, and those who are busy because they must work two jobs to pay the rent all have little choice. A lack of discretion makes for a particularly extreme form of scarcity.

This discussion clarifies what we mean by poverty. We mean cases of economic scarcity where changing what you want, or think you need, is simply not viable. Some of these hard-to-change needs are biological, such as hunger for the subsistence farmer, and some are socially constructed. What we feel we need depends on what others have and on what we’ve gotten used to. Indoor plumbing, for example, would hardly make anyone in the developed world feel terribly lucky these days, yet it was pretty much inconceivable until the last quarter of the nineteenth century, and it is still a dream in many places today. To the subsistence farmer, it is a luxury; to someone living in New Jersey, it is a necessity. Driving a car was a status symbol in the fifties and remains so in many parts of the world. In other parts of the world today, it is a necessity. A deep and complicated question is: How exactly do these needs compare? Does the poor American who cannot afford adequate plumbing really feel a lot like the poor farmer who cannot afford a shirt or the poor European who cannot afford a car? There is too little evidence to know how comparable these two forms of poverty—absolute versus relative—are psychologically. For our purposes, they are all examples of poverty.

Poverty is extreme in another way. Consider the parents of a newborn, who are suddenly time scarce. They also do not have the option to “want less”; the baby needs to be taken to the doctor, and fed, and changed, and cuddled, and bathed, and rocked (forever) to sleep. There are just so many nondiscretionary activities to juggle. But if you are a parent with money, your time scarcity can be alleviated in another way. You can hire a nanny or a maid, order in food rather than cook, use an accountant, employ a gardener, all of which will free up time. Similarly, if you are on a diet, with plenty of money, you can buy tasty but healthy food. Money, because it is fungible, can be used to compensate for other forms of scarcity.

The reverse—trying to alleviate the scarcity of money—is much harder. Sure, you can try to work a few more hours, but in most cases you don’t have much to give, and it will bring limited extra wealth and leave you even busier and more exhausted. Less money means less time. Less money means it is harder to socialize. Less money means lower quality and less healthy food. Poverty means scarcity in the very commodity that underpins almost all other aspects of life.

We have used the psychology of scarcity to create an empathy bridge. We have used experience with one form of scarcity (say, time) to connect to another form (money). Having known what it’s like to badly need a little more time, we might start to imagine what it’s like to desperately need a little more money or even more friends. We used this bridge to draw a connection between a busy manager fretting about insufficient time before a deadline and a person short on cash fretting about insufficient funds to pay rent.

This empathy bridge, though, only goes so far. After all, the manager can say, “Forget it. I’ll just strive less and alter my work–life balance,” whereas the person stressed for cash can’t simply say, “Forget it. I don’t need the apartment after all.” So while both time and money can tax bandwidth, the magnitude of those taxes—their gravity—can be very different.

THE ELEPHANT IN THE ROOM

Most conversations about poverty feature an elephant in the room.

Take the case of diabetes, which affects 285 million people worldwide. It is a serious disease with consequences that include coma, blindness, limb amputations, and death. Luckily, it is now a manageable disease. Drugs taken regularly—sometimes in pill form, sometimes with an injection—can prevent diabetes from doing too much damage. Yet diabetes remains a major problem. Part of the problem is pharmacology: medicine has not fully cured the disease. But a bigger part of the problem is psychology. For any medicine to work, people must take it. Yet diabetics take their medication only 50 to 75 percent of the time, greatly reducing its efficacy.

Think of how striking this is. Decades of medical research transform a debilitating, deadly disease into a manageable one. But we trip up on the last mile, on the most trivial step: taking a pill or shot. This last mile plagues much of medicine. Twenty years ago, we would have been ecstatic to have the antiretroviral drugs for the treatment of HIV that we have today. Yet millions have died because they did not take the medications consistently. For tuberculosis, the problem is so large that the standard delivery protocol in developing countries, DOTS (directly observed therapy), is designed just to address this problem: someone comes every day to watch you take the pill. In some countries, we cannot afford to provide tuberculosis medication. Not because the pills are expensive—they are cheap antibiotics—but because the cost of DOTS is too high. One remarkable medical achievement after another stumbles on nonadherence, this vagary of human behavior.

Nonadherence affects many people, but it is particularly concentrated in one group: the poor. While people at every income level may fail to take their medications, the poor do so most often. Disease after disease—HIV, diabetes, tuberculosis—the same pattern repeats itself. No matter the location, the kind of medication, or the side effects, one thing stays the same: the poor take their medication least consistently.

Moving to a very different context, consider the case of agricultural yields. The amount of crop that can be grown on a plot of land affects all of society. It determines food prices, world trade, environmental impacts, and even the feasible population of the planet. It matters perhaps most of all to the farmer: his entire income depends on his yield. As with medicine, technology has made terrific strides in improving yields and sustainability: better seeds, farming techniques, and organic farming methods. Yet like the doctors above, agricultural scientists who work on these issues are continuously vexed by one thing: farmer behavior.

For thousands of years, farmers have known that weeding dramatically improves crop yields. Weeds suck away nutrients and water from the main crop. Weeding requires little skill or machinery, merely some tedious work. Yet farmers in the poorest parts of the world fail to weed. Some estimate that losses from not weeding in parts of Africa are more than 28 percent of total yield. In Asia, uncontrolled weed growth has been estimated to cost up to 50 percent of total rice output. It’s possible that these estimates are too large. But even a 10 percent increase in yield would be a fantastic return for a couple of days of tedious work. Besides, since weeding increases output without using more money or land, a 10 percent increase in yield means a 20 to 30 percent increase in earnings, a pretty hefty sum. Nonetheless, many farmers leave this money on the table by failing to weed regularly or enough. And even within these areas, the biggest offenders are again the poorest farmers.

To move to yet another example, take parenting. Researchers have now spent a great deal of time studying how people raise their children. Do parents raise their voices needlessly? Do they show love and support in times of need? Are they consistent in their application of rules, or do they make demands that are haphazard and arbitrary? Do they give positive feedback when the child does well? How much do they engage with the child as opposed to plopping her in front of the TV? Do they help with homework?

One broad theme emerges from decades of this research: the poor are worse parents. They are harsher with their kids, they are less consistent, more disconnected, and thus appear less loving. They are more likely to take out their own anger on the child; one day they will admonish the child for one thing and the next day they will admonish her for the opposite; they fail to engage with their children in substantive ways; they assist less often with the homework; they will have the kid watch television rather than read to her. We now know more about what makes for a good home environment, and poor parents are less likely to provide it.

The poor fall short in many ways. The poor in the United States are more obese. In most of the developing world, the poor are less likely to send their children to school. The poor do not save enough. The poor are less likely to get their children vaccinated. The poorest in a village are the ones least likely to wash their hands or treat their water before drinking it. When they are pregnant, poor women are less likely to eat properly or engage in prenatal care. We could go on. And on.

These facts follow like a tedious argument of insidious intent (to butcher T. S. Eliot). The overwhelming question in this case is an old, almost tired one. Why do the poor fail so badly and in so many ways?

This is the elephant in the room.

CONFRONTING THE ELEPHANT

When we do confront the disturbing facts, it is natural first to question their interpretation. Perhaps the poor are not “failing” to take their medication; perhaps these pills are simply too expensive. Why do they not weed? Because they are too busy. Why do they not parent better? Because they grew up in similar circumstances and have not been taught other parenting skills. Surely, all these issues of access and cost and skills play some role. But time after time, when you look at the data, these factors alone cannot explain the failures. For example, the poor in the United States who are on Medicaid pay nothing for their medications, yet they fail to take them regularly. The poor in rural areas report that their time is abundant between harvests, yet they do not weed. These failings cannot be dismissed as merely circumstantial: at the core there is a problem of behavior.

Another instinctive response is to question the facts themselves. Whether the poor fail or not is really in the eye of the beholder. Perhaps they are not failing. Perhaps those who created the data are biased. There is plenty of compelling psychology to back up such assertions. In one study, for example, subjects watch a video of a young girl, Hannah, taking a test. Her performance is ambiguous: she gets some hard questions right and some easy ones wrong. One group of subjects sees Hannah against a background that suggests she comes from a poor family; another sees a background suggestive of an upper-middle-class family. Both groups watch her take the test and then gauge her performance and abilities. Those who observed “poor” Hannah saw more errors, judged that she did worse, and guessed she was at a lower grade level than those who observed “rich” Hannah.

It seems easy to be biased in our interpretation of the data about the poor. Given that we hold highly negative stereotypes about the poor, essentially defined by a failure (they are poor!), it is natural to attribute personal failure to them. Is it a surprise then that researchers “see” the disadvantaged failing? Unfortunately, when you look closer, the elephant cannot be taken out of the room so easily. Most of these data are genuine correlations, not just biased perceptions.

Nor can the data be dismissed as the result of researchers’ political bias. The data are often collected by researchers without an agenda, and when they have one, it is often contrary to what they find. Other times, the findings are incidental to their research, not something they were looking for. Agronomists and medical researchers collect large data sets where income is but one variable; they report this among many other correlations. They neither went looking for findings about the poor nor do they trumpet them. Moreover, when researchers finally do focus on poverty, they often come to the subject with a pro-poor bias. Scholars working on families or obesity or any number of other domains that focus on poverty tend to have a natural affinity for their subjects, and they report discomfort with what they find. Perhaps most compelling is the sheer breadth and depth of this evidence. It comes not from an isolated study or polemic piece of research. Many efforts have accumulated quite a bit of data. And together they present quite a large elephant.

If we cannot dismiss the elephant, how can we make sense of it? One way is to assume that the causality runs from failure to poverty; that the poor are poor precisely because they are less capable. If your earnings depend on making good choices, then it follows naturally that those who fail end up poor. There are obvious complications to this view. Accidents of birth—such as what continent you are born on—have a large effect on your chance of being poor. Still, one prevailing view explains the strong correlation between poverty and failure by saying failure causes poverty.

Our data suggest causality runs at least as strongly in the other direction: that poverty—the scarcity mindset—causes failure.

PARENTING

One study on parenting focused on air traffic controllers. What made air traffic controllers interesting is that their jobs change daily and can be intense. Some days there are many planes in the air, weather conditions are bad, and there are congestion and delays. On those days the cognitive load—tunneling for long hours on landing all planes safely—is very high. Other days are more relaxed, with not many planes in the air or on the mind. What the researchers found was that the number of planes in the air on a particular day predicted the quality of parenting that night. More planes made for worse parents. Or, if you don’t mind a more vulgar framing, think of it this way. The same air traffic controller acted “middle class” after an easy day at work and acted “poor” after a hard day’s work.

Of course you know this yourself. You come home from work after a long, frustrating day. All you want is some peace and quiet, but your kids are enthusiastically watching cartoons. The TV is not terribly loud but certainly enough to grate on your nerves. You implore your kids to turn the thing off, happy you managed not to be brusque. They respond that this is their television time, that you had explicitly promised that they could watch TV at this hour if they had finished their homework, which they have done. You hesitate for a second but the noise is too much. “Just turn the damn thing off!” you bark. Later you feel bad. It is not how you’d like to be with your wonderful children, but you couldn’t stop yourself.

And you would have good reason to be upset. While research on child rearing is murky, there are a few things that emerge as clearly good, and they are pretty intuitive. Consistency is near the top of the list. It is tough and anxiety-producing for children to learn things—discipline, rules of conduct, a sense of comfort—if parents are inconsistent in their statement and application. Yet this is easier said than done. Being a good parent, even when you know what to do, is hard. Consistency requires constant attention, effort, and steadfastness.

Good parenting generally requires bandwidth. It requires complex decisions and sacrifice. Children need to be motivated to do things they dislike, appointments have to be kept, activities planned, teachers met and their feedback processed, tutoring or extra help provided or procured and then monitored. This is hard for anyone, whatever his resources. It is doubly hard when your bandwidth is reduced. At that moment, you do not have the freedom of mind needed to exercise patience, to do the things you know to be right. A crowded airspace during the day leaves a crowded mind that night. A difficult day as an air traffic controller at work makes for a worse parent at home.

The poor have their own planes in the air. They are juggling rent, loans, late bills, and counting days till the next paycheck. Their bandwidth is used up in managing scarcity. Just as air traffic controllers might have their heads buzzing, so do the poor. An outside observer in their living room who didn’t know about all those planes in the air would indeed conclude that these parents lacked skills.

A recent study showed some evidence of this. As we have seen, poor parents receive food stamps once a month, but by the end of each month they are running short. The end of the month is when their bandwidth is most taxed, the time when parenting is likely to be toughest. The economist Lisa Gennetian and her colleagues showed that these are also the times when children of parents who receive food stamps were most likely to be acting out and end up being disciplined in school.

Being a good parent requires many things. But most of all it requires freedom of mind. That is one luxury the poor do not have.

POOR IN MORE THAN ONE WAY

The poor are not just short on cash. They are also short on bandwidth. This is exactly what we saw in the mall studies and in the harvest studies. The same person when experiencing poverty—or primed to think about his monetary troubles—did significantly worse on several tests. He showed less flexible intelligence. He showed less executive control. With scarcity on his mind, he simply had less mind for everything else.

This is important because so many of our behaviors, not just parenting, rely on bandwidth. For example, an overtaxed bandwidth means a greater propensity to forget. Not so much the things you know (what psychologists call declarative memory), like the make of your first car, but things that fall under what psychologists call prospective memory—memory for things that you had planned to remember, like calling the doctor or paying a bill by the due date. These tasks must be maintained alive in your head, and they get neglected when your bandwidth is reduced. Is it any surprise then that the poor fail to take their medications? Some may find this hard to believe: how can you forget something so important? But memory doesn’t work that way. You don’t remember as a function of long-term value. Certainly no one forgets to take painkillers: the pain is a constant reminder. Diseases such as diabetes, though, are “silent”; their consequences are not immediately felt. There is nothing to remind a person with an overburdened bandwidth to take those medications.

Another consequence is reduced productivity at work. Nearly every task—from processing drive-thru orders to arranging grocery shelves—requires working memory, the capacity to hold several pieces of information active in our minds, until we use them. By taxing working memory, poverty leads us to perform less well. It makes us less productive because our mental processor is occupied with other concerns. This creates a tragic situation where the poor, who most need the wages of their labor, also have their productivity most heavily taxed.

An overtaxed bandwidth means a reduced ability to process new information. How much of a lecture will you absorb if your mind constantly gets pulled away? Now think of a low-income college student whose mind keeps going back to making rent. How much will she absorb? Our data above suggest that much of the correlation between income and classroom performance may be explained by the bandwidth tax. And learning is impeded not only in the classroom. Many public health programs rely on the poor to absorb new information. Campaigns try to educate the public about the importance of eating healthier, smoking less, obtaining prenatal care, getting screened for HIV, and so on. In poor countries, extension workers reach out to farmers to educate them about the latest crops or the latest pests. It should not come as a surprise that these efforts are less successful with the poor, largely failing to get them to smoke less, eat healthier, or adopt the latest farm practices. Absorbing new information requires working memory.

The bandwidth tax also means that you have fewer mental resources to exert self-control. After a long day hard at work, are you likely to floss? Or will you say, “Never mind, I’ll do it tomorrow.” To make matters worse, we have seen how the constant struggle with poverty (and scarcity generally) further depletes self-control. When you can afford so little, so many more things need to be resisted, and your self-control ends up being run down. Now picture yourself as a farmer preoccupied by thoughts of how you will make ends meet this week. You go to sleep preoccupied by how you’ll afford the dentist for your son who’s been complaining of toothache. You may need to forgo the night out with friends that you’ve been looking forward to. And you need to weed soon. You wake up, tired and still anxious. Like failing to floss, it is all too easy to imagine how you might decide, “I’ll just weed tomorrow.”

We see this in the data on smoking: smokers with financial stress are less likely to follow through on an attempt to quit. The poor will end up fatter too; eating well is a substantial self-control endeavor. One study found that when low-income women were moved to higher-income neighborhoods, rates of extreme obesity and diabetes dropped tremendously; other factors may have played a role, but a reduction in stress is almost certainly part of the story. Being a good parent requires self-control. Showing up at work even when you are sick requires self-control. Not snapping at your boss or at a customer requires self-control. Regularly attending a job-training program requires self-control. When you live in a rural village, ensuring that your kid gets to school every day requires self-control. So many of the “failures” surrounding poverty can be understood through the bandwidth tax.

Finally, think about the following. You have a big presentation tomorrow, for which you have prepared intensively. You know the value of rest, so you make sure to finish work by 5 p.m. You go home, have a good dinner with the family, and turn in early. But your mind is buzzing with thoughts of the presentation. So despite the need to sleep, you do not sleep well. Sleep research shows you are not alone. In one study, thirty-eight good sleepers were instructed to go to sleep as quickly as possible. Some of them were told that after the nap they would be giving a speech. Most people really do not like to give speeches. Indeed, this group had far more trouble falling asleep and slept less well when they did. Other data on insomniacs show that they are more likely to be worriers. Put simply, it is hard to sleep well when you have things on your mind.

This is perhaps the most pernicious, long-term detrimental way in which scarcity may tax bandwidth: thoughts of scarcity erode sleep. Studies of the lonely show that they sleep less well and get fewer hours. These effects are quite strong for the poor: they too have lower-quality sleep. And not sleeping enough can be disastrous. The U.S. Army has shown how lack of sleep can lead soldiers to fire on their own troops. The oil tanker Exxon Valdez crashed in Alaska in 1989 arguably in part because of the crew’s sleep deprivation and sleep debt. These effects cumulate. Studies show that sleeping four to six hours a night for two weeks leads to a decay in performance comparable to going without sleep for two nights in a row. Insufficient sleep further compromises bandwidth.

One of the things the poor lack most is bandwidth. The very struggle of making ends meet leaves them with less of this vital resource. This shortfall is not of the standard physiological variety, having to do with a lack of nutrition or stress from early childhood hindering brain development. Nor is bandwidth permanently compromised by poverty. It is the present-day cognitive load of making ends meet: when income rises, so, too, does cognitive capacity. The bandwidth of the farmers was restored as soon as crop payments were received. Poverty at its very core taxes bandwidth and diminishes capacity.

Bandwidth underpins nearly every aspect of our behavior. We use it to calculate our odds of winning in poker, to judge other people’s facial expressions, to control our emotions, to resist our impulses, to read a book, or to think creatively. Nearly every advanced cognitive function relies on bandwidth. Yet a tax on bandwidth is easy to overlook. Perhaps the best analogy is this: Think of talking to someone who is clearly doing something else, say surfing the web, while talking to you. If you did not know what they were doing, how would they seem to you? Daft? Confused? Uninterested? Not all there? A bandwidth tax can create the same perception.

So if you want to understand the poor, imagine yourself with your mind elsewhere. You did not sleep much the night before. You find it hard to think clearly. Self-control feels like a challenge. You are distracted and easily perturbed. And this happens every day. On top of the other material challenges poverty brings, it also brings a mental one.

In this light, the elephant in the room no longer seems so puzzling. The failures of the poor are part and parcel of the misfortune of being poor in the first place. Under these conditions, we all would have (and have!) failed.

IS BANDWIDTH TAX THE CULPRIT?

We began with a small sample of observations all pointing at the elephant in the room. In a great variety of circumstances, poverty appears to correlate with failure. We have given one explanation for these findings: the bandwidth tax. But how do we know that this, in fact, is the explanation? You might wonder, for example, whether the bandwidth tax is large enough to explain everything from failed adherence to forgotten weeding. We think it is. In the mall study from chapter 2, where the low-income group would not even qualify as truly poor, the bandwidth tax was sizable: roughly thirteen to fourteen IQ points, with an equally large effect on executive control. In the harvest study in India, we found an eight-to-nine-point effect on IQ and an even larger effect on executive control. These are, as we have pointed out, very large effects on cognitive function. In terms of the standard IQ classification, they can take you from “normal” to “superior” intelligence, or from “normal” to “dull,” or even “borderline deficient.” Not only is the bandwidth tax large, but the fact that we find it in two very different contexts is powerful confirmation. The poor in rural India are quite different from low-income shoppers at a New Jersey mall, yet they exhibit broadly similar bandwidth taxes. It is therefore not unreasonable to expect that the bandwidth tax plays a similarly large role in the lives of the poor everywhere.

The bandwidth tax is an appealing explanation because it accounts for a diverse set of phenomena. Explanations of the poor’s failure are normally piecemeal. Perhaps farmers do not weed for cultural reasons; perhaps diabetics do not take their medications because of side effects; perhaps poor parents just lack the knowledge. These explanations are scattered because the circumstances of the poor are so very different. What people don’t know in Trenton they might know in Nairobi. And what is a norm in Nairobi might not be so in the rural Philippines. In contrast, a single, fundamental mechanism—bandwidth—can make sense of this diverse set of empirical facts across behaviors, time, and place. Surely the specific circumstances also matter for understanding the lives of the poor, but bandwidth is fundamentally important and applies across all of them.

Understanding the role of bandwidth also helps us to better understand the specific circumstances of the poor. Disease, noise, and malnutrition are no longer simply sources of misery but also additional forms of bandwidth taxation. Take the idea that the poor lack certain basic skills. Rather than viewing this as an established fact, we may consider how a bandwidth tax can be one reason for this skill shortfall. Any form of skill acquisition, whether it be learning social skills or developing good spending habits, requires bandwidth. If the poor lack bandwidth, they will be disadvantaged at acquiring useful skills.

All this proposes a new lens through which to understand poverty. We need to look at data that have already been collected—on drug adherence, weeding, parenting, and other behaviors—with a cognitive lens, informed by scarcity considerations. Rather than isolated behaviors, each requiring its own account, these ought to be viewed as predictable consequences of overtaxed bandwidth. This perspective also suggests a new focus for collecting data. When we study poverty, we tend to focus on material conditions, but we also ought to look at psychological conditions—at bandwidth. In this way, existing puzzles may become less puzzling. To understand the poor, we must recognize that they focus and they tunnel and they make mistakes; that they lack not only money but also bandwidth.