The United States is a rich country. In 2007, before the recession hit, the average U.S. household had an income (in 2012 prices) of $74,869, far exceeding the poverty threshold. Even after a devastating recession followed by a sluggish recovery, average household income in 2013 was $72,641. How is it possible, then, that so many Americans still live in poverty? The answer is that income is unequally distributed, with many households earning much less than the average and others earning much more.

Table 18-2 shows the distribution of pre-tax income—income before federal income taxes are paid—among U.S. families in 2013, as estimated by the Census Bureau. Households are grouped into quintiles, each containing 20%, or one-fifth, of the population. The first, or bottom, quintile contains households whose income put them below the 20th percentile in income, the second quintile contains households whose income put them between the 20th and 40th percentiles, and so on.

For each group, Table 18-2 shows three numbers. The second column shows the income ranges that define the group. For example, in 2013, the bottom quintile consisted of households with annual incomes of less than $20,900, the next quintile of households had incomes between $20,900 and $40,187, and so on. The third column shows the average income in each group, ranging from $11,657 for the bottom fifth to $322,674 for the top 5%. The fourth column shows the percentage of total U.S. income received by each group.

Mean versus Median Household Income At the bottom of Table 18-2 are two useful numbers for thinking about the incomes of American households. Mean household income, also called average household income, is the total income of all U.S. households divided by the number of households. Median household income is the income of a household in the exact middle of the income distribution—the level of income at which half of all households have lower income and half have higher income. It’s very important to realize that these two numbers do not measure the same thing.

Economists often illustrate the difference by asking people first to imagine a room containing several dozen more or less ordinary wage-earners, then to think about what happens to the mean and median incomes of the people in the room if a Wall Street tycoon, some of whom earn more than a billion dollars a year, walks in. The mean income soars, because the tycoon’s income pulls up the average, but median income hardly rises at all.

This example helps explain why economists generally regard median income as a better guide to the economic status of typical American families than mean income: mean income is strongly affected by the incomes of a relatively small number of very-high-income Americans, who are not representative of the population as a whole; median income is not.

What we learn from Table 18-2 is that income in the United States is quite unequally distributed. The average income of the poorest fifth of families is less than a quarter of the average income of families in the middle, and the richest fifth have an average income more than three times that of families in the middle. The incomes of the richest fifth of the population are, on average, about 15 times as high as those of the poorest fifth. In fact, the distribution of income in America has become more unequal since 1980, rising to a level that has made it a significant political issue. The upcoming Economics in Action discusses long-term trends in U.S. income inequality, which declined in the 1930s and 1940s, was stable for more than 30 years after World War II, but began rising again in the late 1970s.

The Gini Coefficient It’s often convenient to have a single number that summarizes a country’s level of income inequality. The Gini coefficient, the most widely used measure of inequality, is based on how disparately income is distributed across the quintiles (as we learned in the preceding Global Comparison). A country with a perfectly equal distribution of income—that is, one in which the bottom 20% of the population received 20% of the income, the bottom 40% of the population received 40% of the income, and so on—would have a Gini coefficient of 0. At the other extreme, the highest possible value for the Gini coefficient is 1—the level it would attain if all a country’s income went to just one person.

One way to get a sense of what Gini coefficients mean in practice is to look at international comparisons. Figure 18-2 shows recent estimates of the Gini coefficient for many of the world’s countries. Aside from a few countries in Africa, the highest levels of income inequality are found in Latin America, especially Colombia; countries with a high degree of inequality have Gini coefficients close to 0.6. The most equal distributions of income are in Europe, especially in Scandinavia; countries with very equal income distributions, such as Sweden, have Gini coefficients around 0.25. Compared to other wealthy countries, as of 2012 the United States, with a Gini coefficient of 0.47, has unusually high inequality, though it isn’t as unequal as in Latin America.

How serious an issue is income inequality? In a direct sense, high income inequality means that some people don’t share in a nation’s overall prosperity. As we’ve seen, rising inequality explains how it’s possible that the U.S. poverty rate has failed to fall for the past 40 years even though the country as a whole has become considerably richer. Also, extreme inequality, as found in Latin America, is often associated with political instability because of tension between a wealthy minority and the rest of the population.

It’s important to realize, however, that the data shown in Table 18-2 overstate the true degree of inequality in America, for several reasons. One is that the data represent a snapshot for a single year, whereas the incomes of many individual families fluctuate over time. That is, many of those near the bottom in any given year are having an unusually bad year and many of those at the top are having an unusually good one. Over time, their incomes will revert to a more normal level. So a table showing average incomes within quintiles over a longer period, such as a decade, would not show as much inequality.

Furthermore, a family’s income tends to vary over its life cycle: most people earn considerably less in their early working years than they will later in life, then experience a considerable drop in income when they retire. Consequently, the numbers in Table 18-2, which combine young workers, mature workers, and retirees, show more inequality than would a table that compares families of similar ages.

Despite these qualifications, there is a considerable amount of genuine inequality in the United States. In fact, inequality not only persists for long periods of time for individuals, it extends across generations. The children of poor parents are much more likely to be poor than the children of affluent parents, and vice versa—a correlation that is even stronger in the United States than in other rich countries. Moreover, the fact that families’ incomes fluctuate from year to year isn’t entirely good news. Measures of inequality in a given year do overstate true inequality. But those year-to-year fluctuations are part of a problem that worries even affluent families—economic insecurity.

Economic Insecurity As we stated earlier, although the rationale for the welfare state rests in part on the social benefits of reducing poverty and inequality, it also rests in part on the benefits of reducing economic insecurity, which afflicts even relatively well-off families.

One form economic insecurity takes is the risk of a sudden loss of income, which usually happens when a family member loses a job and either spends an extended period without work or is forced to take a new job that pays considerably less. In a given year, according to recent estimates, about one in six American families will see their income cut in half from the previous year. Related estimates show that the percentage of people who find themselves below the poverty threshold for at least one year over the course of a decade is several times higher than the percentage of people below the poverty threshold in any given year.

Even if a family doesn’t face a loss in income, it can face a surge in expenses. Until implementation of the Affordable Care Act in 2014, the most common reason for such surges was a medical problem that required expensive treatment, such as heart disease or cancer. In fact, in 2013 it was estimated that 60 percent of the personal bankruptcies of Americans were due to medical expenses. The rise in medical-bill bankruptcies—up nearly 50 percent from 46 percent in 2001, and on up to 62 percent in 2007—was a major source of support for the passage of the ACA.