On average, extreme inequality is costing the median income full-time worker about $42,000 a year. Adjusted for inflation using the CPI, the numbers are even worse: half of all full-time workers (those at or below the median income of $50,000 a year) now earn less than half what they would have had incomes across the distribution continued to keep pace with economic growth. And that’s per worker, not per household. At both the 25th and 50th percentiles, households comprised of a married couple with one full-time worker earned thousands of dollars less in 2018 dollars than a comparable household in 1975—and $50,000 and $66,000 less respectively than if inequality had held constant—a predicament compounded by the rising costs of maintaining a dignified middle-class life. According to Oren Cass, executive director of the conservative think tank American Compass, the median male worker needed 30 weeks of income in 1985 to pay for housing, healthcare, transportation, and education for his family. By 2018, that “Cost of Thriving Index” had increased to 53 weeks (more weeks than in an actual year). But the counterfactual reveals an even starker picture: In 2018, the combined income of married households with two full-time workers was barely more than what the income of a single-earner household would have earned had inequality held constant. Two-income families are now working twice the hours to maintain a shrinking share of the pie, while struggling to pay housing, healthcare, education, childcare, and transportations costs that have grown at two to three times the rate of inflation.
This dramatic redistribution of income from the majority of workers to those at the very top is so complete that even at the 95th percentile, most workers are still earning less than they would have had inequality held constant. It is only at the 99th percentile that we see incomes growing faster than economic growth: at 171 percent of the rate of per capita GDP. But even this understates the disparity. “The average income growth for the top one percent was substantially higher,” write Price and Edwards, “at more than 300 percent of the real per capita GDP rate.” The higher your income, the larger your percentage gains. As a result, the top 1 percent’s share of total taxable income has more than doubled, from 9 percent in 1975, to 22 percent in 2018, while the bottom 90 percent have seen their income share fall, from 67 percent to 50 percent. This represents a direct transfer of income—and over time, wealth—from the vast majority of working Americans to a handful at the very top.
But given the changing demographic composition of the U.S. workforce, these topline numbers can only tell part of the story. The U.S. workforce is now better educated and more urban than it was in 1975. It is also far less white and male—with white men falling from over 60 percent of the prime-aged workforce in 1974 to less than 45 percent by 2018. These changes are important, because while there was far more equality between the income distributions in 1975, there was also more inequality within them—notably in regard to gender and race.
For example, in 1975, the median income of white women was only 31 percent of that of white men; by 2018 white women were earning 68 percent as much. Likewise, the median income of Black men as a share of their white counterparts’ earnings rose from 74 percent in 1975, to 80 percent in 2018. Clearly, income disparities between races, and especially between men and women, have narrowed since 1975, and that is a good thing. But unfortunately, much of the narrowing we see is more an artifact of four decades of flat or declining wages for low- and middle-income white men than it is of substantial gains for women and nonwhites.
Much has been made about white male grievance in the age of Trump, and given their falling or stagnant real incomes, one can understand why some white men might feel aggrieved. White, non-urban, non-college educated men have the slowest wage growth in every demographic category. But to blame their woes on compe ion from women or minorities would be to completely miss the target. In fact, white men still earn more than white women at all income distributions, and substantially more than most non-white men and women. Only Asian-American men earn higher. Yet there is no moral or practical justification for the persistence of any income disparity based on race or gender.
The counterfactuals in the table above appear vastly unequal because they extrapolate from the indefensible 1975-levels of race and gender inequality; they assume that inequality remained constant both between income distributions and within them—that women and nonwhites had not narrowed the income gap with white men. But surely, this cannot be our goal. In an economy freed from race and gender bias, and that shares the fruits of growth broadly across all income distributions, the most appropriate counterfactual for all the groups in this table would be the aggregate counterfactual for “All Groups”: a median income of $57,000 a year for all adults with positive earnings ($92,000 for full-time prime-age workers). That would be the income for all workers at the 50th percentile, regardless of race or gender, had race and gender inequality within distributions been eliminated, and inequality between distributions not grown. By this measure we can see that in real dollars, women and nonwhites have actually lost more income to rising inequality than white men, because starting from their disadvantaged positions in 1975, they had far more to potentially gain. Per capita GDP grew by 118 percent over the following four decades, so there was plenty of new income to spread around. That the majority of white men have benefited from almost none of this growth isn’t because they have lost income to women or minorities; it’s because they’ve lost it to their largely white male counterparts in the top 1 percent who have captured nearly all of the income growth for themselves. According to economist Thomas Piketty, men accounted for 85 percent of the top income centile in the mid-2010s—and while he doesn’t specify, these men are overwhelmingly white.
Thus, by far the single largest driver of rising inequality these past forty years has been the dramatic rise in inequality between white men.
The data on income distribution by educational attainment is equally revealing, in that it calls the lie on the notion of a “skills gap”—a dominant narrative that has argued that rising inequality is largely a consequence of a majority of American workers failing to acquire the higher skills necessary to compete in our modern global economy. If workers were better educated, this narrative argues, they would earn more money. Problem solved.
Indeed, at every income distribution, the education premium has increased since 1975, with the income of college graduates rising faster than their less educated counterparts. But this growing gap is more a consequence of falling incomes for workers without a college degree than it is of rising real incomes for most workers with one—for not only have workers without a degree secured none of the gains from four decades of economic growth, below the 50th percentile they’ve actually seen their real incomes decline. College educated workers are doing better. The median real income for full-time workers with a four-year degree has grown from $55,000 a year in 1975 to $72,000 in 2018. But that still falls far short of the $120,000 they’d be earning had incomes grown with per capita GDP. Even at the 90th percentile, a college educated full-time worker making $191,000 a year is earning less than 78 percent what they would have had inequality held constant.