Workers of color are far more likely to be paid poverty-level wages than white workers
Marking the 50th anniversary of the 1968 Poor People’s Campaign, EPI has described the potential to reduce poverty through work, provided there are jobs with decent wages and adequate hours available to everyone who can and wants to work. Unfortunately, even when jobs are available, workers are often paid so little that they can still be left in poverty. Today, one in nine U.S. workers are paid wages that would leave them in poverty for their family size if they are the sole earner in their family—even with a full-time, year-round schedule.
Although the share of workers earning poverty wages has declined over the past three decades, there are still large racial and ethnic differences in the shares of workers being paid at adequate wage levels. As shown in the first figure below, workers of color are far more likely to be paid poverty-level wages than white workers. In 2017, 8.6 percent of white workers were paid poverty wages—i.e., hourly wages that would leave them below the federal poverty guideline for their family size if they are the sole earner in the family, even if they work full-time, year-round. In contrast, 19.2 percent—nearly one in five—Hispanic workers were paid poverty wages, and 14.3 percent—roughly one in seven—black workers were paid poverty wages. Asian or Pacific Islander workers also had higher poverty-wage rates than white workers, at 10.9 percent.
Workers of color are far more likely to be paid poverty-level wages than white workers: Share of workers earning poverty-level wages, by race/ethnicity, 1986–2017
| Year | White | Black | Hispanic | Asian/P.I. |
|---|---|---|---|---|
| 1986 | 15.4% | 23.5% | 28.3% | |
| 1987 | 14.6% | 22.5% | 27.7% | |
| 1988 | 14.4% | 21.8% | 26.9% | |
| 1989 | 13.5% | 21.0% | 26.6% | 16.8% |
| 1990 | 14.5% | 22.7% | 30.2% | 18.3% |
| 1991 | 14.4% | 22.7% | 30.7% | 18.7% |
| 1992 | 14.6% | 22.4% | 30.9% | 19.2% |
| 1993 | 14.4% | 22.1% | 31.2% | 18.6% |
| 1994 | 14.8% | 22.3% | 33.1% | 19.3% |
| 1995 | 14.8% | 21.8% | 34.2% | 20.0% |
| 1996 | 13.9% | 22.0% | 32.2% | 19.9% |
| 1997 | 13.9% | 21.8% | 33.5% | 20.4% |
| 1998 | 12.6% | 18.7% | 30.8% | 18.3% |
| 1999 | 12.0% | 18.2% | 29.6% | 17.1% |
| 2000 | 10.8% | 16.2% | 28.1% | 15.4% |
| 2001 | 10.4% | 16.0% | 27.1% | 13.4% |
| 2002 | 10.3% | 15.3% | 26.5% | 14.2% |
| 2003 | 10.0% | 14.6% | 24.7% | 12.8% |
| 2004 | 10.7% | 15.0% | 26.0% | 14.1% |
| 2005 | 10.2% | 15.5% | 25.6% | 13.7% |
| 2006 | 9.8% | 14.1% | 24.3% | 12.2% |
| 2007 | 9.9% | 14.7% | 23.9% | 12.5% |
| 2008 | 10.1% | 15.1% | 23.0% | 13.0% |
| 2009 | 11.1% | 15.2% | 25.6% | 12.5% |
| 2010 | 10.5% | 14.7% | 26.0% | 13.1% |
| 2011 | 10.6% | 14.9% | 25.3% | 13.1% |
| 2012 | 11.2% | 16.6% | 26.6% | 14.4% |
| 2013 | 11.0% | 16.4% | 26.1% | 14.0% |
| 2014 | 11.1% | 17.8% | 24.6% | 13.1% |
| 2015 | 10.7% | 16.8% | 24.9% | 14.0% |
| 2016 | 9.6% | 15.4% | 21.0% | 11.8% |
| 2017 | 8.6% | 14.3% | 19.2% | 10.9% |
Notes: A “poverty-level wage” is a wage that would leave a full-time, year-round worker below the federal poverty guideline for their family size if they are the sole earner in the family. Poverty wage thresholds are specific to each family size, and family sizes are calculated using the total number of people in each family or subfamily within the CPS data. The "Asian or Pacific Islander" race category is only available beginning in 1989.
Source: EPI analysis of Current Population Survey Outgoing Rotation Group microdata
Among workers of all races and ethnicities, the shares being paid poverty wages have declined from highs reached in either the mid-1980s or the mid-1990s. However, the share of black workers earning poverty wages in 2017 (14.3 percent) was still slightly above where it was in 2006 (14.1 percent.) Black workers are the only group for whom the share receiving poverty wages is not at its lowest level on record.
Notably, racial gaps have been remarkably consistent—if not gotten worse—over time. The share of black workers earning poverty-level wages has consistently been 1.5 times that of white workers for the entirety of the series. The ratio of the Hispanic poverty-wage rate to the white poverty-wage rate has actually grown since the 1980s. In 1986, the share of Hispanic workers earning poverty-level wages was 1.8 times that of white workers; in 2017, it was 2.2 times the share of white workers.Read more
Nonemployer establishments grew in 2016 but their real revenues were stable: This confirms other data on self-employment showing more activity, but little economic impact
Last week’s release of the new Bureau of Labor Statistics (BLS) Contingent Worker Survey (CWS) has renewed public interest in tracking the number of self-employed or independent contractors. Today’s release of the U.S. Census nonemployer establishment data for 2016 provides another useful data point about the trends in self-employment.
Our analysis of these new nonemployer establishment data affirm the pattern seen through 2015. As we wrote yesterday: these data confirm previous general findings regarding Uber drivers, all online demand platform work, and independent contractors more generally. These previous findings are:
- There is an increase in the amount of people involved in self-employment or online platform types of work activity;
- The increase is primarily among people who are doing so to earn supplementary income and for a short amount of hours. The increase in the various self-employment activities has not occurred in people’s “main job” or as their main source of income; and
- The economic scale of these activities has not increased greatly when measured as a share of hours worked or compensation earned. While “headcount” measures do show a large increase, the overall economic impact is relatively small.
As we show, there has been a significant increase in the headcount of nonemployer establishments. However, the economic impact, as measured by the share of nonemployer establishment revenues in total revenues, has not increased in roughly 20 years. Nonemployer establishments grew by 2.0 percent from 2015 to 2016 but their real revenues grew by just 0.2 percent, evidence of a very small economic impact. In fact, over the medium term there has been a slight decline in the nonemployer establishment share of all revenue between 2007 and 2016, from 3.3 to 3.1 percent of all revenue. This indicates that the growth of nonemployer establishments seems to reflect the growth of self-employed individuals operating unincorporated businesses that generate very little revenue, including the period since 2011 when Uber, Lyft, and other online platform work expanded rapidly.Read more
Has self-employment surged? Data on nonemployer establishments confirm other data showing more activity, but not much economic impact
Update: Read the analysis of 2016 nonemployer establishment data.
Last week’s release of the new Bureau of Labor Statistics (BLS) Contingent Worker Survey (CWS) has renewed public interest in tracking the number of self-employed or independent contractors. The upcoming release of the U.S. Census nonemployer establishment data for 2016 this Thursday will provide another useful data point about the trends in self-employment.
Our analysis of nonemployer establishment data through 2015 indicates that these data confirm previous general findings regarding Uber drivers, all online demand platform work, and independent contractors more generally. These previous findings are:
- There is an increase in the amount of people involved in these types of work activity;
- The increase is primarily among people who are doing so to earn supplementary income and for a short amount of hours. The increase in the various self-employment activities has not occurred in people’s “main job” or as their main source of income; and
- The economic scale of these activities has not increased greatly when measured as a share of hours worked or compensation earned. While “headcount” measures do show a large increase, the overall economic impact is relatively small.
As we will show, there has been a significant increase in the headcount of nonemployer establishments. However, the economic impact, as measured by the share of nonemployer establishment revenues in total revenues, has not increased in roughly 20 years. This indicates that the growth of nonemployer establishments seems to reflect the growth of self-employed individuals operating unincorporated businesses that generate very little revenue.
Background on CWS and nonemployer data
The CWS is based on a household survey and showed that in 2017, the share of independent contractors in total employment was 6.9 percent, comparable to what was reported in all of the other CWS surveys back to 1995. This has elicited surprise by many observers, especially those that have touted the explosion of gig work or freelancing.
The release of new Census data on nonemployer establishments for 2016 this month provides another opportunity to track these trends. These Census data capture the phenomenon of self-employment by examining business establishment/firm data rather than asking individual workers or households about employment. Census describes the data as:
“data for businesses that have no paid employees and are subject to federal income tax. Most nonemployers are self-employed individuals operating unincorporated businesses (known as sole proprietorships), which may or may not be the owner’s principal source of income.”
Various analysts have used these data to track the rise of self-employment with many arguing that the rapid rise of nonemployer establishments reflect a surge of self-employment and gig work. Abraham et al. (2017) note that the growth of nonemployer establishments was faster than self-employment in the Current Population Survey (CPS). They cite this evidence (along with tax records) as being consistent with the argument that the CPS is missing some of the growth of in self-employment, pointing out that nonemployer establishments “have trended upwards as a percent of the number of earners.” Steve King of Emergent Research, a leading analyst of gig and “independent worker” trends, has also said that nonemployer data are “a useful general indicator of U.S. self-employment.”
All of the analyses we have seen of these data solely focus on headcount measures, tracking the number of nonemployer establishments, and do not examine the economic impact, as we do in our analysis by examining the revenues of nonemployer establishments.
This may be surprising since the Census used to (but now does not) warn data users of the following:
“The majority of all business establishments in the United States are nonemployers, yet these firms average less than 4 percent of all sales and receipts nationally. Due to their small economic impact, these firms are excluded from most other Census Bureau business statistics.”Read more
How big is AI-related employment? Not that big at all—despite what Stanford’s AI Index Annual Report tries to claim
There is much discussion about the potential for artificial intelligence (AI) to transform our workplaces, possibly leading to fewer overall jobs and greater economic inequality in the future. This discussion has always made me wonder to what extent AI is already being used to produce goods and services. So, I was excited to learn about the AI Index from reading the recent Furman and Seamans paper on “AI and the Economy.” Furman and Seamans write: “The AI index, a non-profit project designed to track activity and progress in AI, provides a number of interesting facts designed to track the scientific progress in and impact of artificial intelligence and robotics.” I was even more excited to see the AI index reporting on “the growth of the share of US jobs requiring AI skills.” However, my excitement soon ended when I realized that the information they offered completely obscured the (lack of) importance of the underlying data, and I have to reluctantly conclude that the AI index offers extremely misleading hype on AI jobs rather than useful information. I am getting tired of overhyped phenomena, so this blog post is to alert the public.
The AI Index highlights the fact that “the share of jobs requiring AI skills in the US has grown 4.5x since 2013” (see the graph below). That sounds like a huge explosion of jobs related to AI, right? But exactly how many jobs are they talking about being AI-related now compared with in the past? One cannot tell from their information, because while it shows the growth in the share of AI jobs, it provides no information of the level of these shares (either now or in the past). I am always careful to not present data showing the percent changes of percentages, because this is often misleading. For example, when a share rises from 1 percentage point to 2 percentage points, it has risen 100 percent. If a share rises from 20 percentage points to 22 percentage points, it has only risen by 10 percent. Yet in the latter example the change in the level of the share is twice as large (2 percentage points, compared to 1). So what does it tell us about the share of AI jobs in the economy today to tell us that this share has grown 350 percent, without telling us the initial level of the share? Absolutely nothing.
CEO pay: Still not related to performance
This post originally appeared on the CEPR blog.
Earlier this year we did an analysis of CEO compensation in the health insurance industry to see if it was affected by the cap on deductibility imposed by the Affordable Care Act (ACA). One of the provisions of the ACA limited the amount of CEO pay that health insurers could deduct on their taxes to $500,000, beginning in 2013.
This provision effectively raised the cost of CEO pay to insurers by more than 50 percent. Prior to 2013, the deduction in effect meant that the government was picking up 35 cents of every dollar of CEO pay, while the companies were paying just 65 cents. 1 With the new provision in place, insurers are now paying 100 cents of every dollar of CEO pay in excess of $500,000.
If the pay reflects the value of the CEO to the company, we should expect this change to reduce the pay of CEOs in the insurance industry. For example, if a CEO gets paid $20 million a year, this should mean that she delivers roughly $20 million in additional value to shareholders.
When the CEO’s pay was fully deductible, the $20 million paid to the CEO actually only cost the company $13 million. This would presumably be the number that matters to shareholders since they care about how much money comes out of their pockets, not the number on the CEO’s paycheck.
Why is wealthy Westport trying to gut police pensions?
Police in Westport, Connecticut are resisting an attempt to slash their pension and partially replace it with a 401(k)-style plan. Such a move would mess with a tried-and-true system that promotes secure and orderly retirement, in favor of an inefficient one that harms workers with no benefit to taxpayers.
Westport is a latecomer to a trend that appears to have nearly run its course. A few years ago, the Wall Street Journal praised San Jose Mayor Chuck Reed for taking an axe to police and other public employee pensions. Along with Rhode Island Treasurer (later Governor) Gina Raimondo, Reed had become a public face of the pension gutting movement, leading a statewide initiative backed by hedge fund billionaires. Reed and Raimondo, both Democrats, were lauded by conservative think tanks and the Journal for taking on public-sector unions, ignoring the fact that they were courting more powerful interests—the financial industry and wealthy donors.
Similar initiatives followed in Dallas, Memphis, and Palm Beach, among other places. But as the Journal reported last year, in an apparent change of heart, cities that slashed police pensions were later forced to restore benefits or spend millions on retention bonuses in efforts to stem outflows of experienced officers. In Palm Beach, for example, 24 mid-career police officers left in the four-year period after the city cut police and firefighter pensions and partly replaced them with 401(k)-style plans, compared with just one mid-career officer in the previous four years. The exodus of police and firefighters to neighboring jurisdictions caused the city to incur millions in additional training and overtime costs. Five years later, the city reversed course, restoring pension benefits and dropping the 401(k)-style plans.
Social Security trustees report shows why we should expand the program—not look for excuses to cut it
There’s no real news in the Social Security trustees report released this afternoon. We’re now a year closer to the date the trust fund will be exhausted, 2034 (same as last year’s projection), at which point current revenues will still be sufficient to cover 77 percent of benefits even if nothing is done to shore up the system’s finances. Each year, the release of the trustees report provides an occasion for Social Security scaremongering by those wanting to shrink our social insurance system. But not only can we afford current benefits, we can afford to expand them.
The average retired worker beneficiary receives an annual benefit of $16,933. Disability and survivor benefits are even more modest. Social Security benefits are replacing a declining share of income at retirement, thanks to a rising normal retirement age and increased taxation of benefits—cuts passed in 1983 that are gradually taking effect (an increase in the retirement age is really just an across-the-board benefit cut that can be offset by retiring later). Rising Medicare premiums, which are deducted from Social Security, also reduce net benefits.
Despite these trends, recent Census research has found that retirees are somewhat better off than previously thought, thanks to income from traditional defined benefit pensions that is underreported in household surveys. Nevertheless, Social Security remains by far the most important source of income for most seniors. It constitutes most of the income of seniors in the bottom half of the income distribution and is the single biggest source of income for all but those in the ninth and tenth income deciles, for whom defined benefit pensions and earned income loom larger.
Seven facts about tipped workers and the tipped minimum wage
As debate continues on a referendum to raise the tipped minimum wage in Washington, D.C., to the minimum wage for nearly all other workers, we wanted to take a few minutes to set the record straight on the facts about tipped worker wages and incomes. Currently, eight states do not have differential treatments of the tipped workforce in terms of the minimum wage.1 Throughout this post, these will be referred to as “equal treatment” states. To be clear, tipped workers in these equal treatment states receive the full, regular state minimum wage plus tips.
Over the last several years, there has been a great deal of research about the minimum wage and tipped restaurant workers, in particular, and we are going to draw on some of that research to make several key points: 1. In the District of Columbia, women, African American, and Hispanic workers are disproportionately minimum wage workers, including tipped minimum wage workers; 2. Maintaining a separate, lower minimum wage for tipped workers perpetuates racial and gender inequities; 3. In states that have a lower tipped minimum wage, tipped workers have worse economic outcomes and higher poverty rates than their counterparts in equal treatment states; 4. Tipped work is overwhelmingly low-wage work, even in D.C.; 5. Wage theft is particularly acute in food and drink service, and restaurants across the country have been found to be in violation of wage and hour laws; 6. Waitstaff have higher take-home pay in equal treatment states than in D.C.; and 7. The restaurant industry thrives in equal treatment states.
Here, we take a closer look at each point:
1. Women, African Americans, and Hispanic workers have disproportionately benefited from minimum wage increases in Washington, D.C. Furthermore, contrary to popular opinion, the vast majority of minimum-wage earners are not teenagers or college students working part-time jobs.
2. Research indicates that having a separate, lower minimum wage for tipped workers perpetuates racial and gender inequities, and results in worse economic outcomes for tipped workers. Forcing service workers to rely on tips for their wages creates tremendous instability in income flows, making it more difficult to budget or absorb financial shocks. Furthermore, research has also shown that the practice of tipping is often discriminatory, with white service workers receiving larger tips than black service workers for the same quality of service.
What to Watch on Jobs Day: Signs of stronger wage growth that will eventually improve Americans’ livings standards
Although in last month’s jobs report we saw a fall in the unemployment rate accompanied by a drop in labor force participation—which showed the unemployment rate dropping for the wrong reasons—the longer-term trends suggest that displaced workers continue to return to the labor market. This is to be expected as the labor market improves, and what we’ve been expecting for years. The unemployment rate of 3.9 percent seems to be overstating the strength of the labor market given how many sidelined workers appear to want jobs. Furthermore, upwards of 70 percent of the newly employed are coming from out of the labor force as opposed to those “actively” looking for work, that is, among those officially counted in the U3 unemployment rate. We only need to look as far as nominal wage growth to know that we are not yet unambiguously at full employment. Employers and workers alike seem to recognize the slack out there and workers still do not have sufficient leverage to bid up their wages. Year-over-year nominal wage growth has averaged 2.6 percent over the last couple of years, consistently below target levels.
Unfortunately, nominal wage growth for private-sector workers found in the monthly jobs report’s payroll survey offers only a limited view on wage growth in the economy today. One of the major benefits of a full employment economy is that wage growth isn’t simply strong for workers at the top of the wage distribution or for workers with more educational attainment, but that it allows low-wage workers to make gains as well. To get finer-grained estimates of what’s happening to wage growth for particular groups of workers, we have to turn to the Current Population Survey Outgoing Rotation Group (ORG). The ORG is a household-based survey, not an employer-based one like the payroll survey, which each month provides widely reported estimates of job growth and wage growth for private-sector workers. This is important because this means the ORG can not only ask questions of wages, but also make comparisons of wages across the wage distribution. Adding to that information gleaned from the Current Population Survey Annual Social and Economic Supplement (CPS ASEC) allows for comparisons about incomes, notably information on poverty rates.
Over the last couple of weeks, EPI has been highlighting some important statistics in commemoration of the 50th anniversary of the Poor People’s Campaign. We’ve pointed out the changes in the poverty rate over the last 50 years, where gains have been made, particularly among the elderly, and where disparities remain. We also discussed how poverty reduction can be improved with a stronger safety net and a better labor market. The truth is that full employment that makes more hours available to lower-wage workers and broad-based wage growth are key tools in the fight against poverty, and many people who live below the poverty line rely on a stronger economy to make ends meet.
Ending individual mandatory arbitration alone fails most workers: For real worker power, end the ban on class and collective action lawsuits
Uber made news yesterday when the company announced that it will end mandatory arbitration for sexual harassment and assault complaints. Lyft quickly followed suit and said that it would also do away with mandatory arbitration agreements for sexual misconduct claims. These companies are the latest in a growing number of corporations that have moved to eliminate mandatory arbitration agreements for sexual harassment claims. There is no doubt that these companies are being driven to action by the power of #MeToo and #TimesUp. And, while a move away from mandatory arbitration by firms like Uber and Microsoft should be celebrated as a victory for these movements, it is important to recognize that for women in low-wage jobs, challenging workplace sexual harassment and assault remains largely impossible, unless companies also end bans on class and collective action.
Workers depend on class and collective actions to enforce many workplace rights. Employment class action cases have helped to combat race and sex discrimination and are fundamental to the enforcement of wage and hour standards. Without the ability to aggregate claims, it would be very difficult, if not impossible, for workers to find legal representation in these matters. This is particularly true for low-wage workers, whose cases are unlikely to involve large enough awards to attract attorneys to invest time in the case. That is the power of class and collective action suits: they let workers pool their claims, making it possible for an attorney to earn enough to make the case worth pursuing.
Banning mandatory arbitration in sexual harassment and sexual assault claims but continuing to restrict class and collective action will do little to help women in low-and middle-wage jobs access justice when they face sexual harassment or assault. These women will still face challenges finding legal representation and find the cost of litigation prohibitive. And it will do nothing to help women facing other kinds of workplace violations. This is especially concerning considering the majority of low-wage workers in the United States are women. And research shows that low-wage workers in the United States lose more than $50 billion annually as a result of wage theft by their employers. Workers have the right to a workplace free of sexual harassment as well as their right to be paid fairly, and creating a hierarchy of worker protection laws by privileging certain types of claims over others is fundamentally unjust—particularly for low-wage workers. Companies should not be applauded for such a minimal response to workplace misconduct and advocates should not be fooled into seeing this as a solution.
Our nation’s labor and employment laws need reform. A key element of reform must be to end employers’ ability to require workers to sign away their rights as a condition of employment. The Supreme Court will soon decide National Labor Relations Board v. Murphy Oil USA, which will determine whether employers can lawfully require workers to sign arbitration agreements that include class and collective action waivers. If the Court denies workers this fundamental right, Congress must act to protect it. If policymakers leave the solution to corporations, we will end up with a system that privileges some claims and fails most workers.