Divergent Paths: A New Perspective on Earnings Differences Between Black and White Men Since 1940
Based on BFI Working Paper No. 2018-45, “Divergent Paths: A New Perspective on Earning Differences Between Black and White Men Since 1940,” by Patrick Bayer, Duke University professor of economics, and Kerwin Kofi Charles, professor, Harris School of Public Policy
- In 1940, the median black man’s earnings would have placed him at the 24th percentile of the earnings distribution for white men
- Seventy years later, with decades of anti-discrimination rulings in place, that number has barely improved, as the median black man’s earnings would place him at the 27th percentile of white men
- Race-neutral policies like minimum wage laws are more effective than race-specific policies in narrowing the earnings rank gap
- However, there is still a place for race-specific policies, as the gains from education show
What that means is that a median black man’s earnings today, relative to a median white man’s, is roughly the same as his father’s and grandfather’s. This notable finding is just one of a number of insights into the relationship between the earnings of black and white men in the United States revealed in “Divergent Paths: A New Perspective on Earning Differences Between Black and White Men Since 1940,” by Patrick Bayer, Duke University professor of economics, and Kerwin Kofi Charles, professor, Harris School of Public Policy.
Bayer and Charles present new evidence on the evolution of black-white earnings differences among all men at different points in the distribution—not just the median or mean—and their findings challenge conventional views about the efficacy of antidiscrimination policies, as well as pose challenges for future policy considerations.
Explaining the stubborn persistence of earnings inequality
Economists have long studied inequality among black and white workers (and here the focus is on the earnings of men), with their attention fixed on mean or median earnings. This has led to the conclusion that the racial earnings gap converged until the 1970s, benefiting from Civil Rights Era legislation, and that this convergence has stagnated ever since. Bayer and Charles challenge this view and show that rather than plateauing, the earnings gap has actually been widening since the 1970s. Rather than just median earnings, the authors measure differences among those at the bottom, middle, and top of the earnings distribution. In addition, they take account of the recent rise in nonworking white and black men, including the increasing rates of incarceration among black men, and also incorporate them into their analysis.
Bayer and Charles make their case in three sections: the revelation of new facts stemming, in part, from the analysis of earnings rank gaps described above (wherein black workers’ earnings are considered within the earnings distribution of white workers); a quantitative assessment of the two primary forces that affect earnings; and the role of educational attainment in explaining gaps in earnings.
Previous research on the levels gap in earnings has revealed that median black and white earnings converged until 1970 but then plateaued until the present, giving rise to questions about why the convergence stalled and how to reignite it. However, focusing only on median wages, and not including wages from top to bottom, including those of nonearners, shrouds a more troubling story: the widening median gap in annual earnings among all black and white men since 1970, which increased from 52.3 in
1970 to 68.4 in 2014. These trends also hold at the 90th percentile, where the median gap converged until 1970, plateaued and then worsened slightly. (See Figure 1.)
Also, as described above, when considering the rank gap in earnings—that is, where a median black man’s earnings would place him in the white male earnings distribution—very little has been gained over the past 70 years or so: a median black man in 2014 would earn at the 27th percentile of white male earnings, just 3 percentage points more than a black man in 1940. Also, as with the levels gap, an analysis of the rank gap shows that black earners at the higher end are doing relatively better: black earners at the 90th percentile of black earnings are at the 75th percentile within the white wage distribution, far above the median level.
What explains these facts?
Bayer and Charles describe two primary forces that explain these phenomena: positional and distributional convergence. Positional convergence shifts the relative positions of blacks and whites within the earnings distribution, and is the result of attempts to address, say, discrimination in education and the workplace. Such laws or policies—like desegregation, anti-discriminatory laws, or education programs aimed at blacks—are meant to expressly benefit black outcomes and do not directly affect white wage earners.
Note: Figure displays earnings rank gap, measured in percentiles, for the median and 90th quantile in the population of all non-Hispanic black and white men aged 25-54, including non-workers. Source: Census, 1940-2000; American Community Survey, 2005-2014. The sample year labeled ‘2007’ combines ACS samples from 2005-07 and ‘2014’ combines those from 2013-14.
Distributional convergence arises when earnings distributions change due to broader issues that impact black wage earners. An increase in the minimum wage, for example, is directed at all low-wage earners, but because black wage earners are disproportionately represented among low wage earners, they would benefit more relative to white wage earners. Conversely, distributional divergence may occur when, for example, low-wage jobs are transferred overseas. Distributional forces are race neutral in that they largely affect all workers in the lower to middle income ranges, regardless of race.
Given these two forces, the question arises: How much of the change over time is attributable to each? Again, the authors find something surprising: at the median, the overwhelmingly important force for both black relative gains and the worsening of black positions were distributional forces, and not those designed to directly improve the earnings of black workers. Minimum wage laws, the decline of unions, the churning associated with low-wage jobs, and other economic shocks, had a greater impact on black male earnings than programs and policies designed to have a direct effect.
However, this was not the case at the top of the earnings distribution, where the driving forces were positional, or those that were designed to directly raise the earnings of black workers. That a 90th percentile black man did better relative to a 90th percentile white man between 1950 and 2000 was owing more to the desegregation of good schools, affirmative action, and making all professions open to all races, than to distributional forces.
A median black man’s earnings today, relative to a median white man’s, is roughly the same as his father’s and grandfather’s.
A highly skilled black man had very few prospects ahead of him in 1950 compared with 2000, and those improved prospects were largely the result of positional forces that expanded his opportunities.
The role of education
Learning of the gains made from race-neutral distributional forces, one might wonder how it is possible that educational policies designed to benefit blacks—like desegregation and, at least in part, Head Start—had relatively little effect. The assumption would be that letting black kids get more education would have improved the median earnings of black men over time relative to white men, but it didn’t happen. How could that be?
The answer is that there were indeed positional gains attained by black men, but those gains were overtaken by a sudden increase in the education premium. In other words, just when black students were making significant gains in their rates of high school graduation relative to white students, the value of a high school degree plummeted. In 1960, 57 percent of black men had eight years of schooling or less. Today, the number of black men with a high school diploma or GED is 88 percent—a huge success by almost any measure. However, those gains were almost immediately overwhelmed by the demand for college-educated workers.
A large gap in the relative earnings of black and white men has been a stubbornly persistent feature of the US labor market since the end of slavery. A conventional
view in economics has been that, over the last 75 years, the racial earnings gap initially converged quite sharply through the mid-1970s, due in large part to racial convergence in educational attainment and the Civil Rights legislation of the 1960s and has stagnated ever since. In this paper, the authors argue that this perspective misses several key aspects of the dynamics of the earnings gap.
Though their paper reveals new facts about the relative decline of the earnings of black men relative to white men over the past 70 years, the authors do reveal some good news to policymakers and others hoping to improve earnings of black men: race-neutral distributional policies work. Policies that benefit lower- and middle-income workers will have an unintended benefit for black men because they are disproportionately represented in those demographics.
Minimum wage laws, the decline of unions, the churning associated with low-wage jobs, and other economic shocks, had a greater impact on black male earnings than programs and policies designed to have a direct effect.
However, that doesn’t mean that positional or race-specific programs should be abandoned. Distributional forces can sometimes shield the benefits of positional policies. For example, as described above, does the impact of distributional forces on the benefits of education mean that such positional policies as desegregation and Head Start failed? To answer that, according to Charles, one only has to imagine a world where those programs and others were never implemented, and then imagine how much further behind black men would be today on the earnings distribution with their white counterparts. Race-neutral distributional forces, then, may have a greater effect over time, but that does not mean that the benefits from positional forces are negligible. The sum of those forces matter.
Still, this is not a time to feel sanguine. By some measures, the level of earnings inequality between black and white men in the US is similar to that of the most unequal countries in the world. This paper focuses attention on the divergence in the labor market prospects of black men over the past several decades and offers new insights into interpreting that divergence. While the entire economy has experienced a marked increase in earnings inequality, this increase is even more dramatic for black men, with those at the top making gains within the earnings distribution, and those at the bottom badly affected by mass incarceration and declining labor market options for the less-skilled.