Economic and Social Inequality Chapter 1

Principles of Economics in Context, Second Edition ECONOMIC AND SOCIAL INEQUALITY As the United States economy began recovering from the Great Recession of 2007– 2009, economic data indicated that the vast majority of all income growth was going to the richest Americans. From 2009–2012, over 90 percent of new income accrued to just the top 1 percent of income earners. As the economy recovered further, new income distribution was less lopsided, but still uneven. The top 1 percent captured over half of all income growth in the United States over the period 2009–2015.1 The trend toward higher economic inequality is not limited to the United States. Over the last few decades, inequality has been increasing in most industrialized nations, as well as most of Asia, including China and India. And while inequality has generally been decreasing in Latin American and sub-Saharan African countries, these regions still have the highest overall levels of inequality.2 Analysis of inequality, like most economic issues, involves both positive and normative questions. Positive analysis can help us measure inequality, determine whether it is increasing or decreasing, and explore the causes and consequences of inequality. But whether current levels of inequality are acceptable, and what policies, if any, should be implemented to counter inequality are normative questions. While our discussion of inequality in this chapter focuses mainly on positive analysis, we will also consider the ethical and policy debates that are often driven by strongly held values. 1. DEFINING AND MEASURING INEQUALITY One of the final economic goals we discussed in Chapter 1 was “fairness.” Note that this goal is subtly, but fundamentally, different from “equality.” Income differences within a society may be considered fair even if they are somewhat unequal. Few desire a society in which everyone earns exactly the same income. But what does it mean to have a society that is neither “too equal” nor “too unequal”? In order to discuss how to achieve a good balance of income and wealth distribution, we first need some objective measures of inequality, which allow us to draw comparisons across time and across societies. We will first consider what we are measuring, and then how we measure it. 1.1 INEQUALITY OF WHAT? When the subject of inequality is raised, most people think of income or wealth inequality. These are indeed central to any economic analysis of the topic. But it is also important to recognize that inequality is a broader concept that extends beyond the realm of money. Let us consider a few examples. Vast inequality exists in the quality of health care across the world. Preventable or treatable diseases in numerous tropical countries (such as malaria, measles, and tuberculosis) cause average life expectancy to be significantly shorter than in the United States or in other rich countries. There is also significant health inequality within many countries. According to a 2017 analysis, average life expectancy in the United States is 10–15 years longer for the wealthiest Americans than for the poorest.3 There is also a considerable imbalance in education, both nationally and internationally. Children in Australia can expect to receive, on average, about 20 years of schooling—the most years of any country. Meanwhile, the average for children in the sub-Saharan countries of Niger, Chad, and the Central African Republic is less than eight years of education.4 Inequalities arise not only due to income differences, but also due to race and gender. In the United States, the difference in academic achievement between white and black students has decreased significantly in recent decades but still remains evident. However, the achievement gap between students from low- and highincome families in the United States has dramatically increased.5 There are mixed results for gender-based educational inequality. By 2016, 24 countries had fully closed the educational gap by gender, while in 17 countries women still had less than 90 percent of the educational outcomes that men have.6 Related to both health and education is what Nobel laureate Amartya Sen has famously referred to as “capabilities.” By his reckoning, money is only one dimension— albeit an important one—of an individual’s “capability” to function in his or her economic environment. To Sen, what matters most is that people possess the necessary tools— including money, health, education, friends, and social connections—to provide them with realistic economic choices. As Sen has pointed out, there is considerable inequality of capabilities in the world, not just in the poor countries. Inequality is also manifest in certain environmental outcomes. Proponents of “environmental justice” point out that polluting industries and toxic waste disposal sites in the United States tend to be located disproportionately near poor and minority communities. This effect is even more pronounced in some developing countries. Oil and gas development in Nigeria by international corporations has resulted in thousands of oil spills that have impoverished local residents due to reduced agricultural production, lower fish harvests, and polluted drinking water.7 In many developed countries, there are stronger regulations on industrial pollution, but major impacts from oil and chemical spills and other emissions still occur, often affecting lower-income communities. One also sees considerable inequality when confronting the issue of climate change. Numerous studies find that climate change will hit poor countries the hardest, exacerbating global inequality. Warmer temperatures and changing precipitation patterns in Africa and other developing regions could reduce the growing season and lower yields, leading to a 20 percent global increase in the number of people at risk of hunger by 2050.8 According to a 2015 analysis in the journal Nature, by the end of the twenty-first century climate change will have a significantly higher proportionate impact on incomes in the world’s poorest countries.9 In addition to these specific effects, a critical fact about climate change, as well as other environmental damage, is that the rich can generally protect themselves much better than the poor can. 1.2 MEASURING INEQUALITY While recognizing these various types of inequality, for the purposes of economic analysis we will focus primarily on inequality of income and wealth. The two most common metrics used to measure income inequality are: 1. Measure the income share (percent of all income) held by various groups ordered by income from poorest to richest, such as the bottom 20 percent, the middle 20 percent, the top 1 percent, etc. 2. Measure the overall distribution of income in a society, using mathematical and graphical techniques. Income Distribution Data Let’s consider the first approach. Table 11.1 presents the distribution of household income in the United States in 2016. The data are arranged in order of income, and the share of the total income “pie” that accrues to each twentieth percentile (or quintile) is in the second column. To understand what this table means, imagine dividing up U.S. households into five equal-sized groups, with the lowest-income households all in one group, the next-lowest in the next group, and so on. Note that the table also breaks out the richest 5 percent as a separate group. Table 11.1 Household Income Distribution in the United States, 2016 Group of Households Bottom 20% Second 20% Third 20% Fourth 20% Top 20% Top 5% Share of Income (Percent) 3.1 8.3 14.2 22.9 51.5 22.6 Annual Income Range Below $24,002 $24,003 - $45,600 $45,601 - $74,869 $74,870 - $121,018 Above $121,018 Above $225,250 Source: U.S. Census Bureau, Historical Income Tables: Households, Tables H-1 and H-2. The lowest-income quintile, with household incomes below $24,002, received only 3.1 percent of all the household income in the country. The richest quintile, those with incomes of $121,019 or more, received 51.5 percent—in other words, more than half—of all the income received in the United States. The top 5 percent of households receive nearly as much income as the bottom 60 percent. (Note that the graph in Chapter 0 presents income inequality slightly differently, looking at average incomes in each group rather than income shares.) Using these data, we can now construct several measures of inequality based on the ratios of the income share of one group compared to another group. One common measure is the ratio of the income share of the richest fifth to that of the poorest fifth of the population; in this case, we obtain 51.5/3.1 = 16.6—that is, households in the richest quintile have over 16 times the income, on average, of households in the poorest quintile. We can then see how this ratio has changed over time to track changes in inequality. For example, in 1980 this ratio was only about 10, indicating an increase in the spread between the richest and poorest fifth of the population. The U.S. Census Bureau publishes various ratios based on the incomes at different percentiles of the distribution, such as the 90th/10th ratio, the 95th/20th ratio, and the 80th/50th ratio. Again, these can be tracked over time to determine how inequality has changed. The Lorenz Curve and Gini Coefficients However, a simple ratio is somewhat arbitrary, focusing on some parts of the income distribution while ignoring others. Economists frequently prefer to use a more comprehensive measure that reflects the shape of the entire income distribution. This measure first involves creating a graph of the income distribution, referred to as a Lorenz curve—named after Max Lorenz, the statistician who first developed the technique. A Lorenz curve for household income in the United States is shown in Figure 11.1. In this graph, the horizontal axis represents the cumulative percent of households, lined up from left to right in order of increasing income. The vertical axis measures the cumulative percentage of all income received by different groups of households (the lowest 20 percent, the lowest 40 percent, etc.). Lorenz curve: a line used to portray an income distribution, drawn on a graph with percentiles of households on the horizontal axis and the cumulative percentage of income on the vertical axis Figure 11.1 Lorenz Curve for the United States, 2016 F 100 Cumulative Percent of Income 80 E 60 D G 40 C 20 A B 0 0 20 40 60 80 Cumulative Percent of Households (Ordered) 100 Source: U.S. Census Bureau, Historical Income Tables: Households, Tables H-1 and H-2 . We use the data in Table 11.1 to draw the Lorenz curve in Figure 11.1. Point A represents the fact that the poorest 20 percent of households received 3.1 percent of all income. To obtain point B, we need to calculate the cumulative percent of income received by the bottom 40 percent of households. So we add the income received by the bottom 20 percent to the income received by the next 20 percent. Thus the cumulative percent of income received by the bottom 40 percent is 3.1 + 8.3 = 11.4 percent of total income. For point C, we need to calculate the cumulative percent of income received by the bottom 60 percent of households, which is 3.1 + 8.3 + 14.2 = 25.6 percent of total income. Similarly, point D shows that the income share of the bottom 80 percent is 48.5 percent of all income. Finally, point E shows that the bottom 95 percent received 77.4 percent of all income (everyone except the top 5 percent). The Lorenz curve must start at the origin, at the lower left corner of the graph (because 0 percent of households have 0 percent of the total income) and must end at point F in the upper right corner (because 100 percent of households must have 100 percent of the total income). The Lorenz curve provides information about the degree of income inequality in a country. Note that the 45-degree line in Figure 11.1 represents a situation of absolute equality. If every household had exactly the same income, then, for example, the “bottom” 40 percent of households would receive 40 percent of all income. This is shown by point G in Figure 11.1. Imagine the other extreme—a situation in which one household received all the income in a country. In this case, the Lorenz curve would be a flat line along the horizontal axis at a value of zero until the very end, where it would suddenly shoot up to 100 percent of income (at point F). Of course these two extremes do not occur in reality, but they indicate that the closer a country’s Lorenz curve is to the 45-degree line, the more equal its income distribution is. This is illustrated in Figure 11.2, which shows the Lorenz curve for four countries: Sweden, South Africa, India, and the United States. Income is distributed relatively equally in Sweden; its Lorenz curve is closest to the 45-degree line of absolute equality. South Africa has one of the most unequal income distributions—we see its Lorenz curve bows far from the line of equality. The lower portion of India’s Lorenz curve is similar to Sweden, but the upper portion is similar to the U.S. We would conclude that income inequality in India is more unequal than Sweden, but more equal than in the United States. Figure 11.2 Lorenz Curves for Sweden, South Africa, India, and the United States Cumulative Percent of Income 100 80 Absolute Equality 60 40 20 0 0 20 40 60 80 Cumulative Percent of Households (Ordered) Sweden South Africa India 100 United States Source: World Bank, World Development Indicators database. Year of data varies from 2011 to 2013. Thus the more the Lorenz curve bows away from the line of absolute equality, the greater is the extent of inequality in the income distribution. This observation led a statistician by the name of Corrado Gini to introduce a numerical measure of inequality that came to be known as the Gini ratio (or “Gini coefficient”), which is defined as the ratio of the area between the Lorenz curve and the diagonal line of equality to the total area under the diagonal line. Gini ratio (or Gini coefficient): a measure of inequality, based on the Lorenz curve, that goes from 0 (absolute equality) up to 1 (absolute inequality). Greater inequality shows up as a larger area between the Lorenz curve and the diagonal line of absolute equality Referring to areas A and B in Figure 11.3, the Gini ratio is A/(A+B). Clearly, the Gini ratio can vary from 0 for absolute equality (since in such a case area A would equal zero as the Lorenz curve overlaps the line of absolute equality) to 1 for absolute inequality (where area B would equal zero). According to U.S. Census Bureau calculations, the Gini ratio for U.S. household income in 2016 was 0.481. We will present international comparisons of inequality, along with data trends, later in the chapter. Figure 11.3 The Gini Coefficient: A/(A+B) 100 Cumulative Percent of Income 80 60 A 40 B 20 0 0 20 40 60 80 100 Cumulative Percent of Households (Ordered) You might be wondering about some details of the measure of income we are using. The definition of income used for the data in Table 11.1 is pre-tax income excluding the value of noncash government benefits such as food assistance and Medicare, and also excluding the value of employer-provided benefits such as health care. How might the Gini coefficient change if we defined income differently? Higherincome people, after all, pay more in taxes, so perhaps we should look instead at disposable income after taxes. Meanwhile, poor people may qualify for noncash programs such as food assistance, or for subsidized housing and medical care, and arguably the value of these programs should be included as part of income. On the basis of considerations like these, the U.S. Census Bureau has experimented with at least 15 different definitions of income. In addition to the definition used in Table 11.1, another definition is meant to approximate what the distribution of income would be if—hypothetically—the impact of government activity were excluded. For this definition, the Census Bureau starts with pretax income and subtracts government cash transfers (such as welfare payments). Then it adds the value of employer-provided health insurance benefits, generally received by workers with higher incomes. Under this definition, the Gini ratio, not surprisingly, rises, showing greater inequality. The share of the bottom fifth drops considerably, while the share of the top fifth rises. Adjusting income for the effects of the tax system mainly lowers incomes at the top, though as we will see in the next chapter all households pay taxes to some extent. When the Census Bureau further adds in the effects of noncash government transfer programs such as food assistance and Medicare, the distribution becomes somewhat less unequal. Income Inequality and Well-Being How much importance should we place on income inequality and the Gini index? Many important goods and services are, after all, obtained without the use of cash income. Many families produce at least some services (such as child care and cooking) for themselves. In addition, many of the things that we enjoy—such as pleasant parks, safe roads, or clean air—add to our well-being without requiring payments (although some of these things are financed through taxes). If we were to look at the distribution of wellbeing rather than just the distribution of income, we would need to take account of these other sources of important goods and services. Some of these goods may contribute to lessening inequality—for example, everyone, rich or poor, can enjoy a public park or use a public library. Evidence suggests, however, that at least in some cases the distribution of such nonpurchased goods may accentuate, rather than lessen, inequality. For example, as noted earlier, proponents of “environmental justice” point out that polluting industries and toxic waste disposal sites tend to be located disproportionately near poor and minority communities. Another interesting issue is the relationship between income and leisure time. Data for the United States indicate that higher education, and thus higher income, is associated with less leisure time. But this does not mean that poor people simply enjoy lives of greater leisure and well-being. Instead, unemployment rates are much higher for people with less education, suggesting that some leisure time is involuntary. Meanwhile, job satisfaction increases with education, which also contributes to well-being.10 As we’ve seen before, well-being is multidimensional and we should be wary about drawing conclusions about well-being based on any single variable. Discussion Questions 1. 2. 2. What are some of the differences between inequality of income and inequality of “capabilities” or well-being? How are these three concepts related? Which one do you think deserves the most attention from policymakers? What do you think is the minimal amount of annual income that an individual, or a small family, would need to live in your community? (Think about the rent or mortgage on a one- or two-bedroom residence, etc.) What does this probably mean about where the average level of income in your community fits into the U.S. income distribution shown in Table 11.1? INEQUALITY TRENDS AND FURTHER CONSIDERATIONS We now can use inequality data to track how inequality changes over time. In this section we first explore income inequality trends in the United States and then discuss some additional perspectives on inequality, including inequality of wealth and how inequality is related to race, age, education, and other factors. 2.1 INCOME INEQUALITY OVER TIME IN THE UNITED STATES No one disputes that income inequality in the United States has increased in recent decades. We can see this in Figure 11.4, which shows the Gini coefficient in the United States from 1967 to 2016, based on data from the U.S. Census Bureau. The Gini coefficient reached a record low of 0.386 in 1968. After that, the Gini coefficient increased in 39 of the next 48 years. Figure 11.4 Gini Coefficient in the United States, 1967-2016 0.50 Gini Coefficient 0.48 0.46 0.44 0.42 0.40 0.38 1965 1975 1985 1995 2005 2015 Year Source: U.S. Census Bureau, Historical Income Tables: Households, Table H-4. While comparable government data are not available for the years prior to 1967, academic researchers have estimated longer trends in income inequality by focusing on the share of total income going to the top income groups. Figure 11.5 shows how the income share of three high-income groups in the United States—the top 10 percent, the top 1 percent, and the top 0.1 percent—has changed since the early twentieth century. After the Great Depression, the share of income going to the top income groups generally declined, suggesting that income inequality was decreasing. The share of income going to the top 10 percent remained low at around 32 percent from 1950 until the early 1970s. The share of income going to the top 0.1 percent reached a low of less than 2 percent in the early 1970s. Since the early 1970s, the income shares going to these top groups have increased, generally surpassing the high levels that occurred prior to the Great Depression. We will consider some the explanations for the recent trend toward higher inequality later in this chapter. Figure 11.5 Income Shares of Top-Income Groups, United States, 1917-2016 Share of Income (Percent) 50 40 30 20 10 0 1915 1935 1955 Top 10% Top 1% 1975 1995 2015 Top 0.1% Source: Saez, 2016. Note: Data exclude capital gains. 2.2 WEALTH INEQUALITY Gini coefficients may also be calculated for the distribution of wealth rather than income. This distribution, which depends on what people own in assets, tends to be much more unequal than income distribution. Many lower-income people have almost no net wealth, and even people with middle-class income levels often have only a relatively small amount of wealth. It is even possible to have negative net wealth. This happens when the value of a person’s debts (e.g., for a car, house, or credit cards) is higher than the value of her assets. For people in the middle class, the equity that they have in their house is often their most significant asset. By contrast, those who do own substantial wealth are generally in a position to put much of it into assets that increase in value over time or yield a flow of income and dividends—which can in turn be invested in the acquisition of still more assets. The distribution of wealth is, however, less frequently and less systematically recorded than the distribution of income—in part because wealth can be hard to measure. Much wealth is held in the form of unrealized capital gains. A household realizes—turns into actual dollars—capital gains if it sells an appreciated asset, such as shares in a company, land, or antiques, for more than the price at which it purchased the asset. An asset can appreciate in value for a long time before it is actually sold. No one, however, will know exactly how much such an asset has really gained or lost in value until the owner actually does sell it, thus “realizing” the capital gain. Another reason that it is harder to get information on wealth is that although governments normally require people to report their annual income from wages and many investments for tax purposes, most governments do not require everyone to regularly report their asset holdings. Finally, wealth consists not only of financial assets but also commodities, paintings, real estate, and the like. Such disparate forms of wealth make it difficult to obtain reliable estimates of aggregate wealth statistics. capital gains: increase in the value of an asset at the time it is sold compared to the price at which it was originally purchased by the same owner These caveats notwithstanding, reasonable estimates of the U.S. Gini coefficient for wealth have been made. They are in the neighborhood of 0.8, significantly higher than the income Gini coefficient of 0.48.11 While the top 10 percent of U.S. households by income receive about 30 percent of all income, as shown in Figure 11.6 the top 10 percent by wealth own 77 percent of all wealth. The top 1 percent (those with more than $4 million in assets) own 42 percent of all wealth, much more than the bottom 90 percent combined. And the top 0.01 percent (about 16,000 families with at least $111 million in assets each) own 11 percent of U.S. wealth.12 For an interesting study of Americans’ perceptions of current wealth inequality, see Box 11.1. Figure 11.6 The Distribution of Wealth in the United States, 2012 Top 0.01% (11% of wealth) Top 0.1%, except the top 0.01% (11% of wealth) Top 1%, except the top 0.1% (20% of wealth) Bottom 90% (23% of wealth) Top 10%, except the top 1% (35% of wealth) Source: Saez and Zucman, 2016. Just as income inequality has been increasing in recent decades, so has wealth inequality. A plot of the wealth shares owned by the top groups in the United States over time looks much like the income shares in Figure 11.5. The share of national wealth owned by the top 1 percent was over 50 percent prior to the Great Depression, declined to less than 25 percent by the late 1970s, but then steadily increased to around 45 percent today.13 Contemplating such vast wealth inequality brings us back to the question of opportunity. Do those with little or even negative wealth have the opportunity to achieve an adequate level of well-being? In addition, great wealth often confers upon its owners both economic and political power. When the ownership of wealth is highly uneven, the ability to direct the operations of businesses and to influence government policy through campaign contributions and the like may become concentrated in the hands of relatively few. They may then use this power to maintain or exacerbate existing inequalities. We return to this point again later in the chapter. BOX 11.1 WEALTH INEQUALITY IN THE UNITED STATES Figure 11.6 presents data on the actual distribution of wealth in the United States. However, political debates about inequality are often based upon perceptions rather than facts. A 2011 study surveyed people regarding their perceptions of wealth inequality in the U.S.14 Specifically, respondents were asked to estimate what percentage of total wealth was actually owned by each wealth quintile. Further, people were also asked to construct their ideal distribution of wealth, again assigning a percentage of total wealth to each quintile. The results are presented in Figure 11.7, along with the actual distribution of wealth in the U.S. We see, for example, that the top quintile actually owns 84% of all wealth in the U.S. according to the paper. (Note that the “actual” distribution of wealth in Figure 11.7 differs somewhat from the distribution given in Figure 11.6—the two figures rely upon different data sources and apply to different years.) However, respondents estimated that the top quintile only owned 59% of all wealth. But most respondents thought that even this estimated concentration of wealth was excessive. On average, their ideal wealth distribution allocated only 32% of all wealth to the top quintile. Looking at the other end of the wealth spectrum, the bottom quintile actually owns only 0.1% of wealth in the U.S. Respondents estimated that the bottom quintile owns about 3% of wealth. According to their ideal distribution, the bottom quintile should own about 11% of all wealth. The results clearly illustrate the difference between reality, perceptions, and subjective preferences. The study authors draw two primary messages from the results: First, a large nationally representative sample of Americans seems to prefer to live in a country more like Sweden than like the United States. Americans also construct ideal distributions that are far more equal than they estimated the United States to be—estimates which themselves were far more equal than the actual level of inequality. Second, there was much more consensus than disagreement across groups from different sides of the political spectrum about this desire for a more equal distribution of wealth, suggesting that Americans may possess a commonly held ‘‘normative’’ standard for the distribution of wealth despite the many disagreements about policies that affect that distribution, such as taxation and welfare.15 Figure 11.7 Actual, Estimated, and Ideal Distribution of Wealth in the United States Actual Top Quintile Estimated Second Quintile Third Quintile Fourth Quintile Bottom Quintile Ideal 0 20 40 60 80 100 Percent of Wealth Source: Norton and Ariely, 2011. 2.3 FURTHER PERSPECTIVES ON INEQUALITY So far we have documented the extent of income and wealth inequality in the United States. But we need to delve a little further to better understand what drives inequality. For example, income inequality is clearly related to race in the United States, as shown in Figure 11.8. Asian households have the highest median annual income, about $77,000, while black households have the lowest at only $37,000. Median income also changes with age, increasing up to middle-age, and then declining as people retire. Married couples, with the potential for two adult workers, have higher incomes than households with just one adult male or female. Further, whether a family with only one adult is headed by a male or a female can make an income difference of nearly 50 percent. Finally, households in metropolitan areas have median incomes about 33 percent higher than those outside of metropolitan areas. Figure 11.8 Median Household Income in the United States by Select Characteristics, 2015 By Race Asian Black Hispanic White By Age 15 - 24 years 25 - 34 years 35 - 44 years 45 - 54 years 55 - 64 years 65 years and older By Type of Household Family with female householder, no spouse Family with male householder, no spouse Married-couple family By Residence Area Inside metropolitan area Outside metropolitan area 0 20,000 40,000 60,000 80,000 Median Household Annual Income Source: Proctor et al., 2016, Table 1. Economic inequalities based on race, age, and other demographic factors are even more pronounced when we consider household wealth. Figure 11.9 presents data on the median value of household assets for different types of households.16 In some cases, we can see how inequalities arising due to differences in income are magnified when it comes to wealth. While white households’ incomes are 63 percent higher than the incomes of black households, the assets of white households are more than 8 times higher than those of black households. Hispanic households also have little in assets, only about $12,000. The median value of household assets tends to rise with age. So while older households (aged 65 and older) have relatively low income as seen in Figure 11.8, they have comparatively high assets. While married couples have incomes about twice as high as households with just one adult, their assets are more than 6 times larger. We also see that education has a significant impact on household assets. For example, those with a college degree have over four times as much household wealth as those with only a high school diploma. Finally, those owning their own homes (including those still paying a mortgage) have 90 times the assets of renters. This demonstrates the importance of real estate equity in building household wealth. Figure 11.9 Median Value of Household Assets in the United States by Select Characteristics, 2013 By Race Asian Black Hispanic White By Age Less than 35 years 35 - 44 years 45 - 54 years 55 - 64 years 65 years and older By Type of Household Households with female householders Households with male householders Married-couple households By Education No high school degree High school degree Associate's degree Bachelor's degree Graduate/professional degree By Housing Type Renters Owners 0 50 100 150 200 250 300 350 Median Household Assets (thousands) Source: U.S. Census Bureau, 2017. 2.4 ECONOMIC MOBILITY Figures 11.8 and 11.9 suggest that some inequality is to be expected in any society, given that people’s incomes and assets tend to increase as they become older and more established in their careers. So at any point in time in a country, we are likely to have younger people with relatively low incomes and few assets, middle-aged people with higher incomes and more assets, and retirees who tend to have relatively low incomes but relatively high assets. Thus we have people moving from lower income groups to higher income groups, and vice versa. This possibility for people or households to change their economic status, for better or worse, is called economic mobility. For a given level of economic inequality, we may be more tolerant if economic mobility is higher because it implies that people have the opportunity to improve their economic condition. economic mobility: the potential for an individual or household to change its economic conditions (for better or worse) over time A common way to measure economic mobility is to track the frequency with which individuals or households move into different income groups, especially in relation to the group in which they were raised. For example, a 2013 U.S. study looks at the income quintiles of people in their late 30s related to their “birth quintile”—the quintile where their parents were, at the same age.17 For people raised in families from the bottom quintile, 44 percent are still in the bottom quintile as adults, 22 percent rise into the second quintile, and about 6 percent rise all the way to the top quintile. Meanwhile, people raised in families from the top quintile are 47 percent likely to also be in the top quintile as adults, with about 25 percent in the fourth quintile and 7 percent falling all the way to the bottom quintile. So while some economic mobility exists, one’s background is clearly an important determinant of one’s adult income. A 2015 study summarized the situation: C]hildren raised in low-income families will probably have very low incomes as adults, while children raised in high-income families can anticipate very high incomes as adults. The differences are extreme: The expected income of children raised in well-off families (90th percentile) is about 200 percent larger than the expected income of children raised in poor families (10th percentile) and about 75 percent larger than that of children raised in middle-class families (50th percentile).18 Other research focuses on how economic mobility in the United States has changed over time. Perhaps the most comprehensive analysis of economic mobility over time in the U.S. found that mobility has remained relatively constant for people born between 1971 and 1993.19 For example, the probability of a child from the bottom quintile reaching the top quintile as an adult was 8.4 percent for those born in 1971 and 9.0 percent for those born in 1986. A 2016 paper took a different approach to studying economic mobility, looking at how one’s income changes throughout a working career.20 This study found that earnings mobility has decreased as inequality has increased since the 1980s. A particularly striking finding was a dramatic decline in upward mobility for those starting their careers in the middle class, even for those with a college degree. Another aspect of economic mobility is whether successive generations are, on average, better off than their parents. With consistent economic growth, each generation can look forward to higher average incomes. However, recent research suggests that this is no longer the case in the United States—see Box 11.2. BOX 11.2 THE FADING AMERICAN DREAM One aspect of the “American Dream” is that each successive generation hopes it will be better off than the previous generation. This continual increase in living standards is referred to as “absolute income mobility.” While this was often taken for granted in the past, is this part of the American Dream still alive? According to a 2017 paper in Science, the answer seems to be mostly “no.”21 Looking at data on children born in the United States from 1940 to 1984, and their parents, the researchers were able to determine the percentage of children who ended up earning more than their parents (after adjusting for inflation). For children born in 1940, over 90 percent of them ended up earning more than their parents. But for children born in the 1980s, this percentage had dropped to 50 percent. Two explanations for the decline in absolute income mobility are proposed: lower GDP growth rates and greater income inequality. Of these two explanations, the paper concludes that: most of the decline in absolute mobility is driven by the more unequal distribution of economic growth in recent decades, rather than by the slowdown in GDP growth rates. In this sense, the rise in inequality and the decline in absolute mobility are closely linked. Growth is an important driver of absolute mobility, but high levels of absolute mobility require broad-based growth across the income distribution. With the current distribution of income, higher GDP growth rates alone are insufficient to restore absolute mobility to the levels experienced by children in the 1940s and 1950s. If one wants to revive the “American dream” of high rates of absolute mobility, then one must have an interest in growth that is spread more broadly across the income distribution.22 Discussion Questions 1. Were your parents better off economically than their parents? Do you believe that you will be better off than your parents? Do you think that this is true of most of your friends? 2. Make a list of the reasons that inequality can be considered desirable, and the ways in which inequality hurts social well-being. Is it possible to limit the negative consequences of inequality while still harnessing the positive aspects?