Income is the annual total amount of money an individual earns in a year before taxes, including money from wages and salaries, self-employment, interest, dividends, rent, and government cash transfers. It does not include nonwage compensation such as the value of health insurance. For all but the top 1 percent of households, the majority of income comes from wages.
In the U.S., many jobs are low paid, and many workers earn at or below poverty incomes. Table 3 shows the fifteen largest occupations in the U.S., the number of workers in each occupation, their median hourly wage, and their median annual income.
Table 3. Employment and Median Hourly Wages in the Largest U.S. Occupations, May 2019
||Median Hourly Wage
||Median Annual Income
|Fast food and counter workers
|Home health and personal care aides
|Office clerks, general
|Laborers and freight, stock,
and material movers, hand
|Customer service representatives
|Waiters and waitresses
|General and operations managers
|Janitors and cleaners, except
maids and housekeeping cleaners
|Stockers and order fillers
|Secretaries and administrative assistants,
except legal, medical, exec.
|Heavy and tractor-trailer truck drivers
|Bookkeeping, accounting, and auditing clerks
Many of these very large occupations (shaded) pay less than $15 per hour at the median, which for a full-time (forty hours per week), full-year (fifty-two weeks per year) worker is $31,200. Given that many full-time workers do not have paid vacation or sick leave, $31,200 is a maximum; it is only feasible if the worker does not miss a single hour of work in the year. For that reason, that maximum is often well above what most of these workers take home, as shown in the final column. For reference, the official federal poverty threshold for a family of four in 2019 was $25,750. In total, at least 46.5 million workers are in occupations that pay below $15 per hour at the median. This number represents almost a third of all employees in the U.S.
The large number of workers in jobs with low expected wages is the result of years of weak, or negative, wage growth. Figure 2 shows the change in real (adjusted for inflation) wages at different points of the wage distribution.
Figure 2. Cumulative Change in Real Hourly Wages of Workers, by Wage Percentile, 1979-2019
Notes: Shaded areas denote recessions. the xth-percentile wage is the wage at which x% of wage earners earn less and (100-x)% earn more.
Source: Elise Gould’s analysis of EPI Current Population Survey Extracts, Version 1.0 (2020), https://microdata.epi.org
At the bottom, workers at the 10th percentile did not see a real wage increase for the thirty-seven-year period from 1979 to 2016. Finally, in 2017—after eight years of GDP growth following the Great Recession—the 10th percentile experienced an increase in real wages relative to 1979 of about 3 percent. Over a forty-year period, this group observed a raise of 32 cents—from $9.75 to $10.07. Workers at the median wage experienced higher, but still relatively anemic, growth in wages after 1996, reaching $19.33 in 2019. Researchers point out that weak growth in wages has occurred despite overall increases in labor productivity, but there is debate as to why that is the case.
This stall in wage growth is similarly reflected in income growth. Figure 3 shows real (adjusted for inflation) pre-tax income growth from 1968 to 2019 from households at key points of the distribution.“Income” in Figures 3 through 6 refers to all pre-tax cash income, including sources of unearned income; it excludes the value of noncash transfers such as benefits from SNAP and post-tax cash transfers such as benefits from the EITC.
Figure 3. Pre-Tax Household Income by Percentiles and Average Income in 2020 Dollars, 1968–2019
At the bottom, the 10th percentile household real income in 1968 was $12,444, and it grew to $14,874 in 2019, an increase of about $2,500 over fifty years (a 19.5 percent increase). The 25th percentile had a similarly small gain of $3,300 (an 11.7 percent increase), and incomes at the median had a gain of about $10,400 (a 19.4 percent increase). Both Figures 2 and 3 depict income growth at the bottom and median as slow or even negligible in recent history.
These figures are not comparing a single person over time, but rather, people of the same relative wage or same relative income at different points in time. Slow growth at the bottom of the wage or income distribution would be less problematic if most workers did not stay at that wage or income for long. For instance, most people earn their lowest wage in their first job because they are young and do not have any experience and earn more as they accrue more experience. A person may start at the 35th percentile but retire at the 85th. Studies of lifetime earnings, however, are pessimistic. Even when looking at the total a person earns over their career, the growth in income and wages of workers in the bottom half of the distribution is small, especially in comparison to the top end of the distribution.
Slow growth at the bottom of the wage or income distribution would also be less problematic if income at the top of the distribution only grew apace. Instead, growth at the top was much faster than growth at the bottom. At the 90th percentile (or the bottom end of the highest earning 10 percent), wages grew by 44.3 percent between 1979 and 2019 compared to 3.3 percent for the 10th percentile. Over the same period income, at the 90th percentile increased by $54,000, or 40.7 percent, compared to a 0.4 percent decline at the 10th percentile (see figure 3).
Income inequality is not a constant in the U.S. economy, and evidence suggests that it has worsened over the past forty years. Figure 4 shows the income shares of the top 10 percent of households since 1917, or how much of all income in the U.S. was taken home by the top 10 percent of households in the income distribution.
Figure 4. Share of Total Income Taken Home by Top 10 Percent of Households Prior to Taxes and Transfers, 1917–2018
The ten years preceding the Great Depression saw growing income concentration. During the Depression, the top 10 percent income share held steady at 45 percent and then fell beginning in 1940. For the next four decades, the top ten percent of the distribution took home a third of all income. Starting in the late 1970s, the share going to the bottom 90 percent steadily eroded until, in 2012, more than half of all income went to the top ten percent of households. Put another way, the total income of the bottom 90 percent was less than the total income of the top 10 percent. Not only does this show the increase in income accruing to the highest earners but also that this trend is recent and not a permanent or necessary feature of the U.S. economy.
It is not clear what the relationship is between economic inequality and economic insecurity. Slow income growth for the bottom half of households does not necessarily mean that they are all economically insecure. And importantly, inequality is a result, not a cause. It is a summary of the income distribution. At the very least, income inequality greatly curtails the gains in average economic status among the population and demonstrates that not all households share in those gains.
To explore that disparity, we examine two traditional channels of attaining economic security: buying a home and going to college, the former being an asset that provides security and the latter a means of attaining higher income. Figure 5 shows the growth in the median sale price of a new home and median income in nominal dollars—that is, the actual dollar amount not adjusted for the average change in prices over time. The dollars are not adjusted because Figures 5 and 6 examine two goods (home and, separately, tuition) have increased in cost much faster than average prices and, as the figures show, much faster than income.
In 1975, nominal median (the 50th percentile) household income in the U.S. was $11,800 and the nominal median sale price of a new home was $39,300. By 2019, median household income in the U.S. was $68,700 and the median sale price of a new home was $321,500. The price of a home jumped from about three times annual income to nearly five times annual income.
Figure 5. Median Sale Price of a New Home vs. Pre-Tax Median Household Income in Nominal Dollars, 1963–2019
Figure 6 provides the same illustration but compares nominal median income with the nominal cost of tuition, fees, room, and board at a four-year private nonprofit college and a four-year public college. In 1975, median income was $11,800 compared to $3,680 for the cost of a year of private college and $1,780 for a year of public college. In 2019, median income was $68,700 while a year of private college cost $49,870 and a year of public college cost $21,950. Thus, the cost of a year at a public college went from one-sixth of the median family’s income to almost one-third.
Figure 6. Cost of Tuition, Fees, Room, and Board of Four-Year Colleges vs. Pre-Tax Median Household Income in Nominal Dollars, 1971/72–2019/20
The growth in the price of homes and tuition, both hallmarks of economic security, is greatly outpacing ability to pay. In 1975, if prospective homebuyers at median income saved 10 percent a year, they could afford a 20 percent down payment for a home in six and a half years. In 2019, it would take ten and a half years. Similarly, in 1975, if they used that 10 percent instead for college, they could fully finance a four-year private education in twelve years and a public one in six. In 2019, savings at a rate of 10 percent of median income annually would take thirty-three years to finance a four-year private education fully and fourteen to finance a public one.
Comparing rates of growth in housing prices and tuition rates (see Figures 5 and 6), most households have little hope to afford such items without undertaking enormous debt.
Indeed, there is evidence that household debt is increasing. Figure 7 shows that total household debt has more than doubled since 2003. Over the same period, median income increased by less than 14 percent. Like home prices and college tuition, household debt is rising much faster than income.
Figure 7. Household Debt by Type of Debt, Trillions of Nominal Dollars, 2003–2020
Debt itself is not a bad thing; financing an investment that will lead to higher income or economic security in the future is considered a sound practice. Accumulating debt payments, however, may increase economic insecurity.
The inverse of debt is savings, and as debt has increased, savings has decreased.
Figure 8. Personal Savings as a Percentage of Disposable Income, 1959–2019
Figure 8 depicts the downward trend in saving rates among individuals in the U.S. over the past sixty years. Between 1959 and 1985—in periods of economic expansion and decline—the average savings rate in a given year rarely fell below 10 percent. Since then, 2020 was the only year in which the average savings rate exceeded 10 percent (individual months may be higher in the figure). This period is underscored by historically low savings rates prior to the Great Recession; they reached an annual average as low as 3.1 percent of disposable income in 2005. Altogether, weak income growth has a parallel trend of declining savings. Consequently, many today have minimal savings to draw on.
An oft-cited study states that about half of U.S. households cannot cover an unexpected $400 expense. The implications of this finding are often exaggerated. The question asks whether the person has cash on hand to cover the expense. A large majority of the half who do not have sufficient cash answered that they would borrow from a friend or family member, sell something, delay other payments, or employ other strategies that would allow payment of the $400 expense. While the finding does not mean that half of the population is $400 away from ruin, it does mean that half have hardly any breathing room in their budgets. Critically, the individuals who answered that they did not have $400 in cash were not all poor. About a third of the respondents had $35,000–$40,000 in income, but also had student loans, installment loans, a mortgage, or a combination of the three.
As we noted, income inequality greatly curtails the gains in average economic status among the population and demonstrates that not all households share in those gains. We showed that inequality by comparing the top half of households to the bottom half. There is also persistent inequality among different racial groups. The median wage or income for Black households and Hispanic households was much less than the wage or income for White households, even when looking within categories of educational attainment (Table 4).
Table 4. Wage and Income by Race and Education Level, 2019
|< High school
A difference in income might reflect benign causes. There is not an identical distribution of Black and White individuals across U.S. states or regions, for example.
But it certainly reflects more malicious causes as well. There is a large and robust literature documenting discrimination against Black workers in the hiring process. This has implications in terms of longer unemployment spells, higher unemployment rates, and lower income.
But the differences do not stop with income. The ability to save, for example, is greatly hindered by not having a bank account. Six percent of the U.S. population is unbanked, meaning that they do not have a checking, savings, or money market account. Virtually all of those unbanked individuals had incomes of less than $40,000 a year. On racial lines, 14 percent of Black individuals were unbanked compared with 10 percent of Hispanic individuals and only 3 percent of White individuals. Those unbanked individuals report instead using alternative financial services that are often associated with high fees or interest rates, such as money orders, check cashing services, payday loans, and pawn shops.
Similarly, Black and Hispanic individuals also hold more student loan debt than White individuals and are more likely to be behind on payments. Being behind on student debt is also correlated with being a first-generation college student. A more troubling form of debt held by Black and Hispanic households is unpaid legal expenses, fines, or court costs. This is debt associated with interaction with the criminal justice system. Only 5 percent of White individuals have this type of debt, compared to 12 percent of Black individuals and 9 percent of Hispanic individuals. A report from the U.S. Commission on Civil Rights found that “municipalities target poor citizens and communities of color for fines and fees.”
It is important to keep in mind, however, that many surveys of income, wealth, savings, and debt do not ask about identification with certain demographic groups. It is typical for a household survey, like the Current Population Survey (which is used to estimate the unemployment rate) to ask about race, gender, age, and education. It is less common for a survey to ask about sexual orientation or religion.
Research has shown that the LGBTQ community also faces discrimination in the labor market. Federal law did not explicitly prohibit from firing or discriminating against a worker for their sexual orientation until June 2020. And many LGBTQ individuals are, or were at some point in time, cut off from their families, including financially. Additionally, cities that are typically welcoming to LGBTQ individuals tend to be relatively high-priced cities. Labor market discrimination, lack of help from family, and a higher likelihood of living in an expensive city are all thought to be contributors to the higher levels of poverty and financial insecurity among LGBTQ households.
Hence, the level of income growth of the past five decades has not been sufficient for many in the U.S. to establish economic security. The growth was weak for the bottom half of households and, indicative of that weak growth, coincided with decreases in savings and increases in debt.