The author thanks Karen Conner, Sarah Rawlins, and Matt Sedlar for editorial assistance and Dean Baker for editorial review. This research was funded by the Annie E. Casey Foundation. We thank them for their support but acknowledge that the findings and conclusions presented in this report are those of the author(s) alone, and do not necessarily reflect the opinions of the Foundation.
The problem of joblessness for Black men is on average three times worse than what is generally assumed. We typically assess joblessness based on the unemployment rate, but prime-age (ages 25 to 54) Black men’s employment-to-population ratio (EPOP) lags the EPOPs of prime-age white, Latinx, and Asian men by over ten percentage points. Among prime-age men, Black men’s EPOP is an outlier.
The white-Black EPOP jobs gap is about three times the unemployment rate jobs gap during a period of moderately high unemployment. When we use the unemployment rate to understand joblessness for Black men, we grossly underestimate the problem, the harm it causes to Black communities, and the need for bold policy interventions. If we could close the white-Black EPOP jobs gap, we could add about $30 billion annually to Black communities and make a significant reduction in Black poverty.
This report calculates the white-Black unemployment rate and EPOP jobs gaps during periods of “low,” “moderate,” and “high” Black unemployment. Using 2014 as the year for “moderate” unemployment, the analysis finds that for Black men to have a similar EPOP to white, Latinx, and Asian men would have required 947,000 jobs, 2.8 times the number to close the unemployment rate gap.
A problem with the official labor market statistics is that they do not include the Black men who are incarcerated or allow us to evaluate the economic impact of the higher mortality rate of Black men. Prime-age Black men who are incarcerated or deceased still have children, family members, and partners who, under different circumstances, could benefit from their financial support. When one takes into account the incarceration and mortality rates of Black men, the EPOP jobs gap jumps to four times the unemployment rate jobs gap, and the income deficit approaches $50 billion a year.
Addressing the prime-age men’s white-Black EPOP jobs gap is one important step in building up the economic health of Black communities. Among the other steps are reducing the high incarceration and mortality rates of Black men.
In an average year, the jobs crisis for Black men is about three times worse than what is generally assumed. Prime-age (ages 25 to 54) white, Latinx, and Asian men all have about the same rate of employment relative to their populations. Prime-age Black men’s rate of employment lags these other groups of men by over ten percentage points. This employment deficit, based on the employment-to-population ratio, translates to a jobs gap three times the size of the jobs gap based on the unemployment rate, the measure that receives the most popular attention. Because we typically assess joblessness among Black men based on the unemployment rate, we grossly underestimate the problem of joblessness for Black men, the harm it causes to Black communities, and the need for bold policy interventions.
This report shows that the jobs gap for prime-age Black men is typically around 900,000 jobs when we think it is around 300,000. It also shows that even the 900,000 jobs figure can be argued to be a significant underestimate of the problem. If we could eliminate this jobs gap for Black men, we could add about $30 billion annually to Black communities and make a significant reduction in Black poverty.
All groups experiencing high unemployment and labor market discrimination (e.g., Black women, Latinx men, Latinx women, etc.) deserve attention from policymakers. As this report shows, there are issues specific to the labor market situation of Black men that are not usually addressed when Black men are included as part of a larger group. It is beneficial to examine and acknowledge the similarities Black men share with other groups as well as the differences. This report takes a close look at how prime-age Black men differ from other groups without attempting to downplay the importance of other types of analyses.
We will be comparing the number of jobs needed for prime-age Black men to have the same unemployment rate and the same employment-to-population ratio as prime-age white men. We will look at these jobs gaps during years when the overall Black unemployment rate is “low,” “moderate,” and “high.” These adjectives are in quotation marks because the Black unemployment rate is never truly low.
Figure 1 shows the overall annual average Black unemployment rate from 1972, when the Bureau of Labor Statistics’ series for Black unemployment begins, to 2020. The Black unemployment rate reached its lowest point on record in 2019. While that unemployment rate (6.1 percent) was the lowest on record for Black people, it was still above the median white unemployment rate (5.2 percent) for that time period. An unemployment rate above the white median is not a low unemployment rate.
One goal in selecting representative years is that it be as close to the current year as possible. The more recent years are more relevant to the current state of the US economy. For the calculations below, the year 2019 will represent the period of “low” unemployment. We should note that 2019 represents a period of unusually “low” Black unemployment. This rate is lower than the prior “lows” for Black people. But we can hope that this rate of “low” unemployment will not be rare going forward.
“Moderate” unemployment is represented by the year 2014. The Black unemployment rate in 2014 (11.3 percent) is very close to the median Black unemployment rate from 1972 to 2020 (11.4 percent). The year 2013, with a 13.1 percent Black unemployment rate, is too high above the median, and the year 2015, with a 9.6 percent unemployment rate, is too low. The 11.4 percent rate of 1990 does not meet the goal of being relatively recent. The year 2020 was not selected because the unemployment dynamics during the COVID-19 pandemic is atypical.
“High” Black unemployment is represented by the year 2010 when the Black unemployment rate reached 16 percent. This is the most recent Black unemployment rate peak, but it is not the highest Black unemployment rate on record.
Black people never experience low unemployment. It would be more accurate to say that they experience periods of high (referred to in this report as “low”), very high (“moderate”), and extremely high (“high”) unemployment. The persistence of high unemployment for Black people should command more attention and policy intervention than it currently does. As bad as the picture for Black joblessness is when looking at the unemployment rate, it is worse when looking at the employment-to-population ratio for Black men.
The unemployment rate varies by age group. Because age distributions differ by race, it is helpful to compare groups within the same age range. The prime-age range of 25 to 54 years old is important because it is an age range when people are more likely to be working. It is also important because it encompasses most of the ages when individuals are likely to be raising young children, if they have children. Thus, joblessness for prime-age workers can lead to children experiencing poverty and can have negative repercussions for the next generation.
For the past two decades, prime-age Black men have had the highest unemployment rate by race and gender for the prime-age major groups. Figure 2 shows the average of the annual prime-age unemployment rates by gender, race, and Latinx ethnicity from 2000 to 2020. Latinx men, Latinx women, and Black women have had unemployment rates significantly above the white rates. Black men have had a higher average than all of these other groups.
While the Black male unemployment rate has been consistently the highest, it is important to note that the rate fluctuates with the business cycle (Figure 3). Because of this fluctuation, the gap between the prime-age white and Black unemployment rate ranged from a low of 2.7 percentage points in 2019 to a high of 8.7 percentage points in 2011. Tighter labor markets are beneficial for reducing Black joblessness, but they have failed to eliminate joblessness gaps.1
Comparing the graph of prime-age men’s unemployment rates (Figure 3) to the employment-to-population ratios (EPOPs) (Figure 4), one sees important differences. While each racial or ethnic group has a different and largely separate unemployment rate trend line, white, Latinx, and Asian men all have very similar and sometimes overlapping EPOP trend lines. Black men stand out with an EPOP trend line that is much lower than those of the other men.
There are also important differences in the dynamics of the men’s white-Black gap in the unemployment rate versus the white-Black gap in the EPOP. The unemployment rate gap averages 4.8 percentage points from 2000 to 2020. The EPOP gap averages 12.1 percentage points. There is also more relative variation in the size of the unemployment rate gap. The size of the unemployment rate gap triples from its smallest point (2.7 percentage points) to its largest point (8.7 percentage points). The smallest EPOP gap is 9.6 percentage points in 2018 and 2019, and the largest is 15.5 percentage points in 2011. Thus, the EPOP gap only increases about 60 percent from its smallest point to its largest. The white-Black EPOP gap for prime-age men is relatively large and relatively unchanging over the business cycle.
We can also note the differences in the prime-age EPOP for Black men compared with Black women (Figure 5). While prime-age Black women’s EPOP falls more than prime-age white women’s when the labor market is slack, when it is tight, Black women’s EPOP becomes very similar and sometimes overlaps with white women’s. Among women, the group that is most similar to Black men is Latinx women. Although the EPOP gap for white and Latinx women is not quite as large as that for white and Black men, they are both relatively large and relatively stable in size from 2000 to 2020. Asian women’s EPOP gap with white women also has a similar pattern to Latinx women and to the white-Black gap for men but with a smaller size gap. Thus, the EPOP story for prime-age Black men and Black women are quite different.
The percentage point difference in the EPOPs for prime-age white and Black men is consistently larger than the difference in the unemployment rate. What does this mean for estimates of joblessness for Black men? The civilian noninstitutionalized population, which is at the base of the Black EPOP, is larger than the Black labor force, which is the base for the unemployment rate. In other words, the denominator in the calculation of the EPOP is larger than the denominator in the calculation of the unemployment rate. This means that one percentage point in the EPOP represents more Black men than one percentage point in the unemployment rate. Thus, if the men’s white-Black unemployment rate gap were equal to the EPOP gap, the estimate of joblessness would still be larger for the EPOP gap. However, the EPOP gap is also larger in percentage points. The EPOP gap, therefore, must be significantly larger than the unemployment rate gap.
How much larger?
Figures 6 and 7 show that the prime-age men’s white-Black EPOP jobs gap can be about four times the gap measured by the unemployment rate during periods of “low” Black unemployment. During the historically “low” unemployment rate period of 2019, the prime-age men’s white-Black unemployment rate jobs gap was 179,000 jobs. For Black men to be employed at a rate similar to white, Latinx, and Asian men would have required Black men to gain 770,000 jobs or 4.3 times as many as with the white-Black unemployment rate jobs gap.
The prime-age men’s white-Black EPOP jobs gap is about three times the unemployment rate gap during periods of “moderate” unemployment. In 2014, closing the unemployment rate gap would have required Black men to gain 341,000 jobs. For Black men to have a similar EPOP to white, Latinx, and Asian men would have required 947,000 jobs, 2.8 times the number to close the unemployment rate gap.
The EPOP jobs gap is about twice the unemployment rate gap during periods of “high” unemployment. In 2010, the unemployment rate jobs gap was 461,000 jobs. The EPOP jobs gap was 1,005,000 or 2.2 times the number required to close the unemployment rate gap. The unemployment rate therefore significantly underestimates the problem of joblessness for Black men, in the best of times and in the worst of times.2
If one averages the ratios of the EPOP jobs gap to the unemployment rate jobs gap for each year from 2000 to 2020, one finds an average ratio of 3.3. The median of these ratios is also 3.3. Thus, it is fair to say that the problem of joblessness for Black men is on average three times worse than most people would think it is.
As large as the prime-age white-Black EPOP jobs gap is, it still underestimates the degree of jobs deprivation in Black communities caused by Black male joblessness. To gain a fuller picture, we need to factor in the high incarceration rate and the high mortality rate of Black men.
The jobs gap estimates are based on the noninstitutional population which excludes the incarcerated population.3 The incarceration rate for Black men is about six times the rate for white men.4 This means that there is a significant and disproportionate share of the Black male population being excluded from earning an income to support themselves, their families, and their communities. Additionally, Black people suffer from a higher rate of mortality than white people at most ages,5 and Black men have a higher mortality rate than Black women.6 As a result, the male-to-female sex ratio for prime-age Black people is lower than the ratio for white people. In a more racially equitable society, there would be many more Black men alive and available to work. When we adjust the prime-age men’s white-Black EPOP jobs gap estimates to include the incarcerated and to equalize the white and Black male-female ratios, we end up with significantly larger jobs gaps.
It is impossible to separate incarceration rates from the racial and class inequality in society. Violent crime can be understood as a product of mainly males growing up in the most socially disadvantaged communities.7 Of course, most males in these communities are not involved in criminal violence, but the rates of violence tend to be highest in the communities suffering the most from a society’s socioeconomic inequality. Black communities are grossly overrepresented among the most socioeconomically disadvantaged communities. For example, recent research finds that Black neighborhoods are 77 times as likely as white neighborhoods to be both socioeconomically poor and socially isolated from better-resourced neighborhoods.8
Further, the criminal justice system has racial and class biases that also lead to higher rates of incarceration for Black males. For example, illicit drug use and sales on college campuses are largely invisible to policing, but the same activity on city streets is highly policed in comparison.9 White youth are more likely to be involved with illicit drugs on college campuses; Black youth are more likely to be involved with drugs on city streets. These structural inequities lead to high rates of incarceration for Black men.
Although Black men in prison are not counted as part of the population when the Bureau of Labor Statistics calculates the employment-to-population ratio, they are clearly part of the population. In federal prison, about half of them have children under 20 years old that they cannot financially support while they are incarcerated. It is reasonable to expect that a substantial share in state prisons also have minor children.10 Even those without children may have other family members or partners who could benefit if they were working.
About 60 percent of the incarcerated work while in prison.11 But this work does not count as real employment because it does not provide an income of any significance. A 2017 investigation by the Prison Policy Initiative found that the highest average hourly wage for prisoners was $1.41.12 This highest average is less than 20 percent of the already low federal minimum wage. In Mississippi, the highest a prisoner could expect to earn was $1.30 per hour.13 In Louisiana, the highest pay rate was $1.00 per hour.14 The Prison Policy Initiative clarifies that “[i]n many states, most regular prison jobs pay well below the highest rates.”15 Prisoners can also be required to work for no pay. The Prison Policy Initiative found that on average wages had declined from when they checked in 2001.16 Wages could be even lower today.
Prisoners who work in prison earn very low wages and many earn no wages at all. Further, if they have earnings, the earnings may be garnished to pay criminal justice-related fees and charges that can eat up half of their meager earnings.17 If prisoners earn income from work in prison, it is negligible.
|Age Category||White||Black||US-Born Latinx||US-Born Asian|
|55 and older||0.86||0.75||0.84||0.89|
Source: Author’s analysis of American Community Survey data from IPUMS USA, University of Minnesota, www.ipums.org.
As a matter of biology, more males are born than females. There should be about 103 to 107 males born for each 100 females born in a population or a male-to-female sex ratio of 1.03 to 1.07.18 After birth, however, males tend to have a higher mortality rate than females, which leads to there being more females than males at older ages.19
We see this dynamic in Table 1. For the 0 to 24-years-old group, there are more males than females. Black people have the lowest sex ratio of 1.03, but there are still 103 males to every 100 females. The white ratio is 1.05. (The analysis of the Latinx and Asian populations are of the US-born populations to avoid distortions from gender biases in terms of who immigrates and from sex-selective abortions favoring males, which are more common in Asia.20) For all races, the sex ratio declines as one moves to older age categories.
But notice that the flip to more females than males occurs fastest for Black people. In the prime-age category of 25- to 54-years-old, there are 101 white men for every 100 white women. For the US-born Latinx and Asian populations, there is an equal number of males to females. For the Black population, however, there are only 90 men for every 100 women.
Figure 8 shows the 2014 mortality rates for white and Black males and females under 55-years-old. The males have higher mortality rates than the females. Although the mortality rate for Black females is higher than the rate for white females, the difference between the male and female mortality rates is still larger for Blacks than for whites. The Black male mortality rate is 103.4 points above the Black female rate, while the white male rate is only 73.9 points above the white female rate. Black males have a higher mortality rate relative to Black females than white males relative to white females.
As with incarceration, society-wide systems of racial discrimination led to the high mortality rate of Black men. The Centers for Disease Control and Prevention states,
. . . centuries of racism in this country has had a profound and negative impact on communities of color. The impact is pervasive and deeply embedded in our society—affecting where one lives, learns, works, worships and plays and creating inequities in access to a range of social and economic benefits—such as housing, education, wealth, and employment. These conditions—often referred to as social determinants of health—are key drivers of health inequities within communities of color, placing those within these populations at greater risk for poor health outcomes.21
It seems likely that for Black men, joblessness and racial discrimination in the labor market are among the social factors contributing to their higher mortality rate. This higher rate of mortality for Black men than for Black women leads to the low Black sex ratio. In a more equitable society, the prime-age Black male-to-female ratio would be more like the white ratio, which means that there would be more Black men alive and available for work.
When Black men are incarcerated or deceased that does not mean that they do not still have children, family members, and partners who, under different circumstances, could benefit from their financial support. It is important that the jobs gap be understood not just for its impact on the jobless Black men but also for its impact on Black families and communities.
In 2014, the year of “moderate” unemployment, the prime-age men’s white-Black jobs gap was 341,000 jobs as measured by the unemployment rate. It was 947,000 jobs, 2.8 times as large when one looked at the gap in the official employment-to-population ratios (EPOPs). But the official EPOP calculation excluded the many white and Black men in state and federal prisons. Including them in the calculations increases the jobs gap to 1.2 million, making it 3.6 times the unemployment rate jobs gap (Figure 9). (See Figures A2 and A3 in the Appendix for the calculations for periods of “low” and “high” unemployment.)
Black men have a higher mortality rate than white men, and this lowers the Black sex ratio below the white sex ratio. If we make the prime-age Black sex ratio equal to the white sex ratio, we can see the potential jobs Black communities have lost due to the higher mortality rate for Black men. Adding these additional jobs to our prior jobs gap calculation increases the gap to nearly 1.4 million or 4.1 times the unemployment rate jobs gap.
In 2014, even if we were able to employ the 947,000 Black men needed to close the EPOP jobs gap, Black families and communities would still be functioning with a deficit of the income from (1,388,000 – 947,000 =) 441,000 jobs because of the Black men who were incarcerated or who had died as a result of a higher mortality rate than white men.
Addressing the prime-age men’s white-Black EPOP jobs gap is one important step in building up the economic health of Black communities, but it is only one of several steps that are necessary. Among the other steps are reducing the high incarceration and mortality rates of Black men. The good news is that increasing employment will probably lead to future reductions in incarceration and mortality. Reducing joblessness lessens the socioeconomic marginalization of Black neighborhoods which should reduce street crime. Reducing joblessness improves one factor in the social determinants causing worse health outcomes for Black men.
. . . the gaps, the disparities are understated in the standard data because of the problem of missing Black men. Nearly 20 percent of Black men are frequently missing from Census-, [Current-Population-Survey]-type datasets. And, if they were included—they are almost all drawn from the lower end of the earnings distribution—the disparities would all be much worse.22
–Economist Harry J. Holzer
Researchers have found that the official labor market statistics from the Bureau of Labor Statistics (BLS) undercounts joblessness, particularly for Black men. Individuals who fail to respond to the survey from BLS are more likely to be jobless than those who respond. This nonresponse bias is stronger for Black men than for white men, and it is growing over time.23 The calculations presented do not make any adjustment for this problem. Therefore, we should consider them conservative, lower-bound estimates.
|Unemployment Rate Gap||Employment-to-Population Ratio (EPOP) Gap||EPOP Gap Adjusted for Incarceration||EPOP Gap Adjusted for Incarceration and Higher Mortality|
|$ 11,734,000,000||$ 32,638,000,000||$ 42,480,000,000||$ 47,805,000,000|
Source: Author’s calculations based on 2020 “Current Population Survey Tables for Personal Income: Personal Income: PINC-03,” US Census Bureau.
Applying the median annual earnings for a Black man with a high school diploma24 to the jobs gap numbers, we can see the collective loss of income Black men, Black families, and Black communities face from different jobs gap estimates (Table 2). If we were to address the typical unemployment rate gap, we would add about $12 billion to Black communities annually. This increase in income could lift hundreds of thousands of families out of poverty.
Of course, the gains would be greater if we are able to close the EPOP jobs gap. The value to Black communities of closing the EPOP jobs gap is about $33 billion a year. With this amount, we could lift hundreds of thousands more families out of poverty than if we just addressed the unemployment rate gap. But even with the $33 billion a year from addressing the EPOP jobs gap, Black communities would still potentially be missing roughly an additional ($48 billion – $33 billion =) $15 billion a year due to the high rate of incarceration and the high rate of mortality for Black men. In other words, the full cost of Black men’s joblessness is about $50 billion a year, especially when one considers that the official statistics undercount Black men’s joblessness.
Even if the unemployment rate provided an accurate gauge of the jobless crisis facing Black men, it would be deserving of more attention and more policy action than it currently receives. After all, Black men have had the highest annual unemployment rate consistently for the last 20 years. Now that we know the crisis is three times worse than we think it is—or four times worse, if we factor in incarceration and mortality—maybe policymakers will be spurred to act.
The Bureau of Labor Statistics (BLS) data series are primarily available with the Latinx population included in racial groups (e.g., Latinx men who identify racially as white are included in the data for white men). All the data include the Latinx population within racial groups or are adjusted based on estimates from the American Community Survey to estimate what the racial group would be with the Latinx population included.
The prison population data from the Bureau of Justice Statistics (BJS) exclude the Latinx population from racial groups. The institutional group-quarters data from the American Community Survey was used to adjust the BJS data for estimates of the prime-age Black and white incarcerated population to include the incarcerated Latinx members of those racial groups.
The mortality adjustment was based on the sex ratios from the full population (i.e., including the institutional and other group-quarters populations) of 25-to-54-year-olds sex ratios derived from the American Community Survey. The Current Population Survey data from the BLS excludes the incarcerated population and would therefore be inaccurate for the population sex ratios. The prime-age Black male-to-female ratio is made to equal the prime-age white male-to-female ratio based on the size of the Black female population. This adjustment does not adjust for the high Black female mortality.
Baker, Dean, and Jared Bernstein. 2013. Getting Back to Full Employment: A Better Bargain for Working People. Washington, D.C.: Center for Economic and Policy Research.
Bivens, Josh. 2021. The Promise and Limits of High-Pressure Labor Markets for Narrowing Racial Gaps. Economic Policy Institute, August 24.
Bond, M. Jermane, and Allen A. Herman. 2016. “Lagging Life Expectancy for Black Men: A Public Health Imperative.” American Journal of Public Health 106, no. 7: 1167–1169.
Cai, Yixia and Dean Baker. 2021. “Masking Real Unemployment: The Overall and Racial Impact of Survey Non-Response on Measured Labor Market Outcomes,” Working Paper No. 150. Institute for New Economic Thinking, February 21.
Carson, E. Ann. 2020a. Data Collected Under the First Step Act, 2019. Bureau of Justice Statistics.
Carson, E. Ann. 2020b. Prisoners in 2019. Bureau of Justice Statistics.
Chao, Fengqing, Patrick Gerland, Alex R. Cook, and Leontine Alkemad. 2019. “Systematic assessment of the sex ratio at birth for all countries and estimation of national imbalances and regional reference levels.” Proceedings of the National Academy of Science U S A 116, vol. 19: 9303–9311.
Flagg, Anna. 2021. “The Black Mortality Gap, and a Document Written in 1910,” The Upshot. New York Times, September 1.
Glaze, Lauren E. and Laura M. Maruschak. 2010. Parents in Prison and Their Minor Children. Bureau of Justice Statistics.
Holzer, Harry J. 2021. “Understanding Current Racial Disparities and Identifying Ways to Improve Conditions for Students, Workers, and Families of Color—Finding a Path Forward in Divided Times,” Racial Disparities in Today’s Economy: 64th Economic Conference. Federal Reserve Bank of Boston.
Levy, Brian L., Nolan E. Phillips, and Robert J. Sampson. 2020. “Triple Disadvantage: Neighborhood Networks of Everyday Urban Mobility and Violence in U.S. Cities.” American Sociological Review 85, no. 6: 925–956.
Mohamed, A. Rafik, and Erik D. Fritsvold. 2011. Dorm Room Dealers: Drugs and the Privileges of Race and Class. Boulder, Color.: Lynne Rienner Publishers.
Newburn, Tim. 2018. Criminology: A Very Short Introduction. New York: Oxford University Press.
Office of Minority Health and Health Equity. 2021. “Racism and Health.” Centers for Disease Control and Prevention.
Rampey, Bobby D., Shelley Keiper, Leyla Mohadjer, Tom Krenzke, Jianzhu Li, Nina Thornton, Jacquie Hogan, Holly Xie, and Stephen Provasnik. 2016. Highlights from the U.S. PIAAC Survey of Incarcerated Adults: Their Skills, Work Experience, Education, and Training: Program for the International Assessment of Adult Competencies: 2014. National Center for Educational Statistics.
Ritchie, Hannah and Max Roser. 2019. “Gender Ratio.” Our World in Data.
Ruggles, Steven, Sarah Flood, Sophia Foster, Ronald Goeken, Jose Pacas, Megan Schouweiler and Matthew Sobek. IPUMS USA: Version 11.0 [dataset]. Minneapolis, MN: IPUMS, 2021. https://doi.org/10.18128/D010.V11.0.
Sawyer, Wendy. 2017. “How much do incarcerated people earn in each state?” Prison Policy Initiative, April 10.
US Bureau of Labor Statistics. 2015. How the Government Measures Unemployment. US Bureau of Labor Statistics.
US Census Bureau. 2021. “PINC-03. Educational Attainment-People 25 Years Old and Over, by Total Money Earnings, Work Experience, Age, Race, Hispanic Origin, and Sex,” Current Population Survey Tables for Personal Income. US Department of Commerce.
Wolchover, Natalie. 2011. “Why Are More Boys Born Than Girls?” Live Science, September 09.