This is the first brief in a series exploring comprehensive federal policy recommendations around AI and workers.

Introduction

Amid the seemingly endless predictions about AI, one point is essential to remember: How this plays out is not predetermined. Policy choices determine whether workers are helped or harmed by technological change. Even without AI, many workers already face harmful disruptions. Every year, 20 million workers are laid off in the U.S. economy1—due to employer mismanagement, once-profitable products becoming less so, offshoring, automation, and other factors. For individual workers, losing a job for any reason and for any period of time is traumatic and damaging in large part because we lack adequate social insurance programs to smooth those transitions.

As lawmakers debate how to best regulate AI deployment in our workplaces, today’s priority should be to strengthen the safety net for workers in any scenario and to acknowledge how AI is already affecting workers’ rights, protections, and livelihoods. Decades of underinvestment, coupled with attacks on unions and other labor supports, have left workers without an adequate safety net.

We can improve our current systems to better provide livable benefits for workers who lose their jobs because of large-scale economic shocks that are not their fault and outside their control. We can also learn from other periods of workforce disruption—the COVID-19 pandemic, trade-related shocks, and the 2008 recession—about which policy choices were effective and which fell short.

When workers lose their jobs due to factors outside their control, they are considered “displaced” or “dislocated.” If they meet certain requirements, these workers are eligible for unemployment insurance (UI). To strengthen outcomes for workers, this brief will provide recommendations to:

• increase UI benefit amount and duration
• expand and standardize eligibility to receive benefits
• align employer incentives with worker access to benefits
• center worker needs and power
• strengthen data infrastructure to support workers

These recommendations should be considered amid conversations focused on creating a stand-alone unemployment program geared toward AI-driven job loss. These discussions should weigh considerations and potential pitfalls, including:

• a blank slate for program design
• relevant roles for key stakeholders
• eligibility criteria challenges
• the allocation of resources, time, and awareness to be effective
• reauthorization vulnerability

If policymakers are serious about easing the impact of AI-related job transitions for workers, the priority solution should be strengthening already existing social insurance programs to reduce the economic consequences of losing any job and to provide adequate support to workers while they seek other employment.

Changes to the current system or establishing a new program will require robust resources, including time, money, and expertise. This investment can help all workers, regardless of their risk of AI-based displacement. With a strong safety net, workers who lose their jobs can get the assistance they need, including the opportunity to find work that is well suited to their skills and experiences, and businesses are more inclined to improve their practices because they know workers aren’t trapped in their current jobs by economic necessity.2 This moment presents an opportunity to strengthen the labor market for everyone and to ensure workers have the safety net they deserve.

How to strengthen our current system

Any effective approach to fixing the UI system must be comprehensive and should apply regardless of whether a worker is displaced due to AI. Numerous experts have weighed in on the best approach—including the National Academy of Social Insurance (NASI) Unemployment Insurance Taskforce,3 the National Employment Law Project (NELP),4,5 and The Century Foundation,6 among others. This brief draws from those recommendations to strengthen outcomes for workers. Other important issues, such as modernizing UI system technology and infrastructure, fraud detection and prevention, and the use of AI in benefit determinations and delivery, also shape worker outcomes but fall outside this brief’s scope.

Increase benefit amount and duration

First and foremost, UI should be designed to ensure that workers can afford housing, food, and other essentials during their period of joblessness. The percentage of wages that UI replaces for workers has declined over time and has not kept pace with inflation.7 In 2025, national average UI benefits replaced well under half of national average wages.8 This gap leaves workers across the income spectrum unable to maintain their standard of living and can push those whose earnings were close to the minimum wage into poverty.

Ensure benefit amounts are consistent across states

Benefit amounts vary across states. In Arizona and Florida, UI benefits replace less than 15% of average weekly wages.9 In some states, the weekly benefit amount is just $222 per week.10 While state variation in benefit amounts hurts all workers in stingy states, it weighs especially heavily on Black and Hispanic workers.

One recent study finds that differences in state rules cause the wage replacement rate for Black workers to be meaningfully lower than the rate for white workers, and that the discrepancies harm overall well-being.11 These disparities compound. Even among those likely eligible, Black workers are less likely than white workers to receive UI benefits—a gap explained primarily by structural factors, not a difference in eligibility, that has held steady for three decades.12

Wage replacement rates should be consistent across states. A recent federal bill has proposed changes to increase the replacement rate to 75% and provide additional weekly funding for dependents.13 Given growing wage inequality, the rate also should vary by income level. In one proposal, the lowest earners would receive benefits that represent 85% of their wages.14

Extend the duration of benefits across states

The duration of benefits also varies greatly from state to state. Until 15 years ago, eligible workers across the country had access to 26 weeks of benefits, a generally agreed-upon standard.15 After the Great Recession however, federal law barred states from cutting benefit amounts, so many states instead cut the duration of unemployment benefits to stabilize their trust funds and lower tax rates for employers.16 Some states now only provide 12 weeks of benefits, a short bridge to find a new job well-matched to workers’ skills and experiences.17 A minimum of at least 26 weeks of benefits should be a uniform requirement across states.

Workers cannot currently depend on the Extended Benefits program, which was intended to provide support during economic downturns by providing additional weeks of benefits during periods of high and rising unemployment. The triggers for Extended Benefits have not been effective, leaving workers to wonder if they are available. In their absence, workers must rely on temporary programs funded by Congress, which depend on political priorities.18 More reliable triggers, higher wage replacement levels, and a longer duration of benefits would make the Extended Benefits program more effective.19

Expand and standardize eligibility for benefits

Only certain workers are eligible for UI, leaving others stranded without benefits if they lose their jobs. Eligibility criteria and the base period used to calculate benefits are set by state law, rather than federal statute. This state-by-state variation can make it difficult for workers to know whether they qualify for assistance and plan accordingly.

Generally, the group of workers ineligible for benefits includes independent contractors, job seekers who have not been recently employed, and those who have exhausted their benefits. Some employers have incorrectly classified many workers as independent contractors, and this misclassification prevents those workers from requesting access to the UI benefits to which they would otherwise be legally entitled.20

UI should cover a broader range of workers, including part-time workers, independent contractors, job seekers who have not been recently employed, and seasonal and temporary workers. During the COVID-19 pandemic, Congress tasked states with implementing new UI programs for gig and platform workers who fell outside of the usual verification system, creating significant challenges.21 Barring changes to account for the unique situation of these workers, these same issues will arise again.22 By incorporating these workers into the UI system now, instead of waiting for a crisis, future challenges can be mitigated.

Policymakers should also consider stronger presumptive eligibility standards, under which workers are presumed eligible for benefits and paid promptly, with the burden placed on employers or the state to prove otherwise. Lowering the barriers for workers to receive benefits could meaningfully increase uptake among workers who currently go without benefits.

Align employer incentives with worker access to benefits

All eligible workers should be able to access UI benefits if they want them. State UI programs are generally financed through employer experience rating: Employers pay more taxes when they lay off more workers who are eligible for UI.23 Therefore, the system’s setup incentivizes employers to limit their tax liability by restricting eligibility for the program in different ways. This can include not hiring certain workers as employees, not proactively informing workers about their eligibility, attempting to reclassify workers as independent contractors, and other forms of suppressing UI program access.

A Bureau of Labor Statistics survey found that only about a quarter of workers who were unemployed in 2022 applied for benefits.24 Further, a recent survey found that 19% of recently unemployed workers encountered actions from an employer that were meant to deter them from applying, in addition to a quarter of all UI filers who said that their employer contested their benefits.25 Low-income workers more frequently reported employer contestations , and workers who reported contestations of their claims by employers were less likely to receive benefits and to experience greater material and psychological burdens during the filing process.26 There is a real opportunity to better align employer incentives to support workers in accessing UI benefits.

Presumptive eligibility, discussed above, would address some of these access barriers by shifting the burden of proof away from workers, but it would not resolve the underlying incentive employers have to minimize their tax liability. There have been many proposals for how to address this issue: broadening the tax base, payroll contributions from employees to UI, or increasing state taxes to fund programs.27 Most basically, employers could pay taxes based on the total number of workers they lay off, not just those eligible for UI.28 These questions are tied to broader considerations about how to finance a more effective program.

Center on worker needs and power

The UI system should be structured around ensuring that workers have the support they need as they seek reemployment. This requires a real investment in active labor market policy, including communicating with workers and unions about their needs in order to align training and placement. Congress should allocate funds to revitalize the Employment Service in the U.S. Department of Labor, which once played a key role in supporting reemployment.29 For example, in a 1986 New Jersey pilot program, job search assistance measurably helped dislocated workers return to work sooner and increased their earnings.30 A well-resourced and staffed U.S. Employment Service can leverage state labor market information—frequently used by employers—to help job seekers find opportunities aligned with their skill sets.31

Worker organizations can also play an important role in the delivery of benefits, including facilitating training and connections to job search assistance.32 In 2022, unemployed people who were members of a union were twice as likely to apply for UI benefits, and among applicants, 30 percentage points more likely to receive those benefits.33 This sizable gap suggests that when workers have organizational support navigating the UI system, they are far more likely to receive the benefits they are entitled to, reinforcing the case for involving worker organizations directly in UI outreach, filing assistance, and perhaps even delivery. Furthermore, unions routinely negotiate with employers, both in collective bargaining agreements and when bargaining before layoffs, to secure agreements that employers will not contest workers’ applications for UI benefits.

In some Northern European countries, unions are responsible for administering unemployment benefits, and scholars have found that this system leads to stronger unions and worker voice in the economy.34 Policymakers in the United States should consider how unions and other worker organizations could deliver unemployment benefits to workers.

Strengthen data infrastructure to improve outcomes

State workforce agencies collect quarterly wage records from employers to facilitate UI administration to gauge eligibility for benefits as well as benefit levels. States share wage records with the federal Census Bureau, and these data are included in longitudinal data sets that can follow workers over time. These data are used for both research and the administration of other policies (such as facilitating child support payments or tracking the performance of job training programs). Wage records provide an important but limited set of information that varies by state. A growing number of states have expanded their wage records to include additional information such as job titles, total hours worked, and job location.35

State-level data are a valuable source of information about what is happening to workers, but the lack of standardization across states makes it difficult to gain a clearer picture of employment. To remedy this, lawmakers should create standardized requirements for enhanced wage records to analyze who is losing their job and where workers are finding new opportunities.

The enhanced information coming from states would be especially useful with the advent of AI since no single federal data source can reliably track how AI adoption is reshaping demand for specific occupations in real time or identify when firms are reducing jobs through attrition. Further, workers would benefit from better information about opportunities that align with their skill set.36 Better integration with job posting data—including the National Labor Exchange—and reemployment systems would help workers navigate the job-search process.

Considerations for a stand-alone program

The UI system should be improved regardless of how AI affects employment and job opportunities. Additionally, some policymakers have suggested creating a stand-alone program geared toward AI-driven job loss. There is precedent for this approach. Congress has acted nine times since 1958 to create temporary, additional UI benefits, most recently during the COVID-19 pandemic.37 There is much to be learned from successes and challenges of these previous programs.

Some of these programs also included benefits or services beyond unemployment insurance for workers. For example, Trade Adjustment Assistance has, at various points, included relocation assistance, training programs, a health coverage tax credit, and wage insurance for older workers who took a lower-paying job.38 A new stand-alone program could similarly go beyond cash benefits, offering workers support to develop new skills, cover the costs of a transition, or find new opportunities.

Blank slate for program design

A new program can address the key structural issues with the current system discussed in the previous section. Any new program should be uniform across states so that benefit levels, duration, and eligibility are consistent for every worker regardless of where they live. A new program also presents the opportunity to build in the other improvements outlined above, including benefit levels set to a livable amount, expanded eligibility, and stronger data infrastructure. A new program could also adopt more fundamental changes to the claims process, such as the presumptive eligibility standards that Michigan moved toward during the COVID-19 pandemic.39

As discussed in the worker power section, a new program should be explicitly structured around building worker power, for example, by involving worker organizations in program administration or benefit delivery. Further, a new program can focus on effective job placement by leveraging relationships between employers, unions, and other worker organizations.

Finally, a new program should also be designed to avoid UI’s employer-financing structure, which—as discussed—in most states, disincentivizes employers from making sure workers get the benefits they are due. Left unaddressed, that same structure would allow employers to game or underreport AI-related job loss in a new program.

Relevant roles for key stakeholders

A successful program requires effective coordination among stakeholders. As with UI and previous stand-alone programs, states would likely retain a central role in delivering benefits and services, given their existing infrastructure and more direct knowledge of local labor market conditions. Local agencies can provide labor market information to workers, worker organizations, and employers to facilitate job placement. Any program should carefully consider how to balance flexibility for states to tailor the program to meet local needs, while maintaining strong, uniform standards across states.

As discussed, worker organizations and unions can play a central role in program administration. Worker organizations can support workers in accessing relevant benefits and training opportunities, ensuring that worker needs and outcomes are central to programmatic strategy and implementation.

Employers bear significant responsibility for the program’s success. They are often the first to know a layoff is coming and should proactively connect workers to available services, for example, by supporting on-site services when requested by state or local Rapid Response teams. Employers can proactively share specific information about their talent and skills needs with labor agencies and worker organizations, and hire workers during or after relevant training.

Eligibility criteria challenges

Creating clear and consistent eligibility criteria will be a significant challenge in establishing a stand-alone program for workers affected by AI-related job displacement. For instance, a program to compensate Department of Energy workers harmed by nuclear materials has faced significant challenges determining eligibility, including the complexity of reconstructing historical exposure to nuclear materials to establish causation.40 Challenges in establishing causation have also been an enormous barrier to worker relief through the Trade Adjustment Assistance program.41

In another example, some pandemic unemployment assistance benefits are still hung up over eligibility disputes years later.42 And from the start of the pandemic, there have been hotly contested debates about who should receive benefits.43 Moreover, concerns about fraud can also complicate eligibility determinations. In the wake of fraud during the COVID-19 pandemic, many states adopted stricter verification requirements to confirm eligibility.44 A new program focused on AI-driven job loss would need to balance similar integrity concerns against the risk of preventing eligible workers from receiving the benefits they are due.45

It is virtually impossible to systematically link job loss with AI adoption. In many instances, AI’s effects on a worker or industry may be indirect. For example, workers may lose their jobs in part due to AI adoption months or years earlier or because of broader economic disruption caused by AI’s impact on different industries. Even setting aside these implementation challenges, restricting eligibility to AI-driven job loss raises a separate concern: fairness. AI is not the only source of economic disruption. Global trade (including offshoring and outsourcing) and the green energy transition continue to displace workers across industries.

Other workers should be entitled to the same benefits as those directly displaced by AI. Drawing arbitrary boundaries around a single cause of job loss will create inequities, a risk compounded if states must independently interpret what counts as AI-driven job loss, therefore creating inconsistent standards nationwide.

Resources, time, and awareness to be effective

Standing up a new program will take significant time and resources. In addition to providing benefits, this includes making meaningful investment in technological systems and program administration, a particular challenge in light of years of underinvestment in modernizing UI systems.46 Federal and state governments need adequate resources for user experience testing, efficient application determinations, coordination with relevant partners, data collection and analysis, and programmatic evaluation. The challenges of implementing relief programs during COVID-19 are instructive. Slow initial payments, overwhelming caseloads that strained public servants, technological breakdowns, and significant confusion about eligibility requirements demonstrated the need to proactively develop systems that can respond more nimbly in times of crisis.47

Starting a new program also requires sustained public outreach, as new and stand-alone programs have historically struggled because of a lack of public familiarity. For example, over half of workers did not know they had access to emergency sick leave during COVID-19.48 In another example, when the Reagan administration made changes to the Trade Adjustment Assistance program, applications plummeted, perhaps because many workers believed the program had been abolished.49 Further, most workers did not know that other services including training, job counseling, and relocation assistance were available through the program.50

A stand-alone program could also come at the expense of the existing UI system, which the vast majority of states say is underfunded.51 From 1999 to 2019, federal funding for UI administration declined by 30% on an inflation-adjusted basis.52 If policymakers decide to divert resources that would otherwise go to the existing UI system, the system’s further deterioration could harm workers served by that system in both the short and long term.

Reauthorization vulnerability in Congress

If an AI-specific UI program requires congressional reauthorization, it will be vulnerable to the difficulty of coordinating bipartisan action in Congress. Several previous programs have been subject to a sunset date, requiring periodic reauthorization to continue accepting new claims. Sunsets and other reauthorization requirements leave workers vulnerable to congressional inaction, especially if the impacts of AI on the workforce play out over an extended time period.

For an example of a program caught up in this type of reauthorization limbo, Trade Adjustment Assistance lapsed multiple times, including most recently in 2022, leaving additional workers without access to the program. Another example is The Emergency Unemployment Compensation program, which was created in 2008 and lapsed five times between 2010 and 2013, before expiring for good at the end of that year.53 This immediately cut off benefits for 1.3 million workers and for more than 3.6 million workers by the following September.54

Financing and the path forward

The current worker safety net has failed to expand enough to cover the basic realities of today’s labor market, and inadequate benefit levels, incomplete coverage, and diminished worker representation are persistent problems. Deployment of AI and related technologies in the workplace adds urgency to addressing these issues.

Implementing the recommendations in this issue brief requires a significant investment that should be considered, along with program parameters and design. As AI drives up stock prices and generates wealth for a select few, policymakers are considering how the public should share in these gains. Ideas include increasing taxes for wealthy individuals, closing the gap between labor and capital tax rates, taxing AI use, and taxing data center development, with the proceeds from these new taxes financing programs that redistribute income more broadly across the economy. Considerations for potential taxation approaches will be discussed in a forthcoming brief.

This moment requires big thinking and corresponding action. While AI drives changes to the economy, improvements to the worker safety net are long overdue. We can learn from the past to improve the labor market for everyone.

Acknowledgments

Angela Hanks, Alex Hertel-Fernandez, Ben Zipperer, Gillian Gutierrez, Jacob Leibenluft, Janelle Jones, Josh Bivens, Michele Evermore, Patrick Oakford, Raj Nayak, and Stephen Nuñez.

Notes

1. Bureau of Labor Statistics, “Layoffs and Discharges: Total Nonfarm” [JTSLDL]. Accessed from FRED, Federal Reserve Bank of St. Louis, October 2026.

2. Patrick Oakford, Josh Bivens, and Celine McNicholas, Federal AI Legislation: An Evaluation of Existing Proposals and a Road Map Forward, Economic Policy Institute, September 2024.

3. National Academy of Social Insurance, Unemployment Insurance Task Force, Unemployment Insurance Task Force Final Report, July 2024.

4. Amy Traub, Boosting Economic Resilience: Seven Ways States Can Support More Workers with Unemployment Insurance, National Employment Law Project, July 2025.

5. Michele Evermore, AI Is New. The Challenge of Mass Unemployment Isn’t. National Employment Law Project, September 2026.

6. Michele Evermore and David Balducchi, A Roadmap for Unemployment Insurance Reform, The Century Foundation, October 2023.

7. National Employment Law Project, Policy Advocacy Brief: Benefit Amounts, November 2023.

8. U.S. Department of Labor, Employment and Training Administration, “UI Replacement Rates Report,” Benefit Accuracy Measurement data. Public data prepared by the OUI Division of Performance Management, Accessed August 7, 2026.

9. Megan Rivera, Why Unemployment Insurance Alone Is Not Enough in the AI Era, Washington Center for Equitable Growth, July 2026.

10. Julie Su, Andrew Stettner, and Michele Evermore, Our Unemployment System Needs Modernizing. Trump Is Doing the Opposite. The Century Foundation, September 2025.

11. Daphné Skandalis, Ioana Marinescu, and Maxim Massenkoff, Racial Inequality in the U.S. Unemployment Insurance System, Washington Center for Equitable Growth Working Paper, July 2023.

12. Elira Kuka and Bryan A. Stuart, Racial Inequality in Unemployment Insurance Receipt and Take-Up, National Bureau of Economic Research Working Paper no. 29595, December 2021.

13. Office of U.S. Senator Michael Bennet, “Bennet, Wyden, Beyer Reintroduce Bicameral Bill to Overhaul Unemployment Insurance” (press release), July 16, 2025.

14. Josh Bivens, Melissa Boteach, Rachel Deutsch, Francisco Díez, Rebecca Dixon, Brian Galle, Alix Gould-Werth, Nicole Marquez, Lily Roberts, Heidi Shierholz, William Spriggs, and Andrew Stettner, Reforming Unemployment Insurance: Stabilizing a System in Crisis and Laying the Foundation for Equity, a joint report of the Center for American Progress, Center for Popular Democracy, Economic Policy Institute, Groundwork Collaborative, National Employment Law Project, National Women’s Law Center, and Washington Center for Equitable Growth, June 2021.

15. National Academy of Social Insurance, Unemployment Insurance Task Force, Unemployment Insurance Task Force Final Report, July 2024.

16. National Academy of Social Insurance, Unemployment Insurance Task Force, Unemployment Insurance Task Force Final Report, July 2024.

17. Julie Su, Andrew Stettner, and Michele Evermore, Our Unemployment System Needs Modernizing. Trump Is Doing the Opposite. The Century Foundation, September 2025.

18. Michele Evermore, “There’s a Growing Bipartisan Consensus to Fix Unemployment Insurance” (commentary), The Century Foundation website, July 17, 2024.

19. Michele Evermore, “There’s a Growing Bipartisan Consensus to Fix Unemployment Insurance” (commentary), The Century Foundation website, July 17, 2024.

20. Ismael Cid-Martinez, Nina Mast, Margaret Poydock, and Valerie Wilson, Misclassifying Workers as Independent Contractors Is Costly for Workers and Social Insurance Systems, Economic Policy Institute, April 2026.

21. Julie Su, Andrew Stettner, and Michele Evermore, Our Unemployment System Needs Modernizing. Trump Is Doing the Opposite. The Century Foundation, September 2025.

22. Adam Grundy and Lucie Alleyne, “Nonemployer Businesses Are Key Contributors to Gig Activities and Economic Growth Nationally” (blog post), U.S. Census Bureau website, July 1, 2025.

23. Michele Evermore, “There’s a Growing Bipartisan Consensus to Fix Unemployment Insurance” (commentary), The Century Foundation website, July 17, 2024

24. Bureau of Labor Statistics, U.S. Department of Labor, “Most Unemployed People in 2022 Did Not Apply for Unemployment Insurance Benefits” (blog post), The Economics Daily (TED), April 14, 2023.

25. Amy Traub, Alexander Hertel-Fernandez, and Sanjay Pinto, The Unemployed Worker Study, National Employment Law Project, April 2025.

26. Alexander Hertel-Fernandez and Alix Gould-Werth, “Employer Contestations and Access to Unemployment Insurance: Implications for Benefit Receipt, Material Hardship, and Administrative Burden,” Social Service Review 100, no. 1 (March 2026): 38–68.

27. Michele Evermore and David Balducchi, A Roadmap for Unemployment Insurance Reform, The Century Foundation, October 2023.

28. Josh Bivens, Melissa Boteach, Rachel Deutsch, Francisco Díez, Rebecca Dixon, Brian Galle, Alix Gould-Werth, Nicole Marquez, Lily Roberts, Heidi Shierholz, William Spriggs, and Andrew Stettner, Reforming Unemployment Insurance: Stabilizing a System in Crisis and Laying the Foundation for Equity, a joint report of the Center for American Progress, Center for Popular Democracy, Economic Policy Institute, Groundwork Collaborative, National Employment Law Project, National Women’s Law Center, and Washington Center for Equitable Growth, June 2021.

29. Michele Evermore, AI Is New. The Challenge of Mass Unemployment Isn’t. National Employment Law Project, September 2026.

30. Stephen A. Wandner, Solving the Reemployment Puzzle: From Research to Policy, W.E. Upjohn Institute for Employment Research, January 2010.

31. Michele Evermore, AI Is New. The Challenge of Mass Unemployment Isn’t. National Employment Law Project, September 2026.

32. Alexander Hertel-Fernandez, “Solidarity Insurance: Reimagining Unemployment Insurance as a Platform for Worker Power.” (policy brief), Political Economy Research Institute, University of Massachusetts Amherst, July 2026.

33. Bureau of Labor Statistics, U.S. Department of Labor, “Characteristics of Unemployment Insurance Applicants and Benefit Recipients—2022” (economic news release), March 29, 2023.

34. Bruce Western, Between Class and Market: Postwar Unionization in the Capitalist Democracies (Princeton, N.J.: Princeton Univ. Press, 1997). 

35. Michele Evermore and Will Raderman, The Case for Enhanced Wage Records: Lessons from States, National Academy of Social Insurance and The Niskanen Center, January 2026.

36. Michele Evermore, “There’s a Growing Bipartisan Consensus to Fix Unemployment Insurance” (commentary), The Century Foundation website, July 17, 2024.

37. Katelin P. Isaacs and Julie M. Whittaker, “The Fundamentals of Unemployment Compensation,” Congressional Research Service, IF10336, updated January 12, 2026.

38. Jacob Leibenluft, Lessons from Past Trade Adjustment Policies to Support Displaced Workers in the Era of Artificial Intelligence, Washington Center for Equitable Growth, March 2026.

39. Michele Evermore, “Michigan Unemployment Claims Processing,” testimony on behalf of National Employment Law Project to a hearing before the Michigan Legislature, Joint Select Committee on the COVID-19 Pandemic, May 21, 2020.

40. U.S. Government Accountability Office, Energy Employees Compensation: DOL Generally Followed Its Procedures to Process Claims but Could Strengthen Some Internal Controls, GAO-16-74, March 2016.

41. Jacob Leibenluft, Lessons from Past Trade Adjustment Policies to Support Displaced Workers in the Era of Artificial Intelligence, Washington Center for Equitable Growth, March 2026.

42. Julie Kay, “‘No Accountability’: Pandemic-Era Unemployment Benefit Delays Persist, 5 Years Later,” WCNC Charlotte, April 3, 2025.

43. Annelies Goger, Tracy Hadden Loh, and Nicole Bateman, Debunking Myths About COVID-19 Relief’s ‘Unemployment Insurance on Steroids,’ The Brookings Institution, May 2020.

44. National Employment Law Project, ID Verification, November 14, 2023.

45. National Academy of Social Insurance, “Testimony of Michele Evermore, Senior Fellow, National Academy of Social Insurance,” testimony before the U.S. House of Representatives, Committee on Ways and Means, Work and Welfare Subcommittee, March 5, 2026. 

46. Julie Su, Andrew Stettner, and Michele Evermore, Our Unemployment System Needs Modernizing. Trump Is Doing the Opposite. The Century Foundation, September 2025.

47. Michele Evermore, AI Is New. The Challenge of Mass Unemployment Isn’t. National Employment Law Project, September 2026.

48. Emma Jelliffe, Paul Pangburn, Stefan Pichler, and Nicolas R. Ziebarth, “Awareness and Use of (Emergency) Sick Leave: US Employees’ Unaddressed Sick Leave Needs in a Global Pandemic,” Proceedings of the National Academy of Sciences 118, no. 29 (July 20, 2021): e2107670118.

49. U.S. Congress, Office of Technology Assessment, Trade Adjustment Assistance: New Ideas for an Old Program, 1987.

50. U.S. Congress, Office of Technology Assessment, Trade Adjustment Assistance: New Ideas for an Old Program, 1987.

51. Bipartisan Policy Center, “Administrative Failures Plague State Unemployment Insurance Programs,” July 1, 2020.

52. Bipartisan Policy Center, “Administrative Failures Plague State Unemployment Insurance Programs,” July 1, 2020.

53. Katelin P. Isaacs and Julie M. Whittaker, Emergency Unemployment Compensation (EUC08): Status of Benefits Prior to Expiration, Library of Congress, Congressional Research Service, CRS Product Number R42444, August 11, 2014.

54. U.S. House of Representatives, Ways and Means Committee Democrats, Unemployment Insurance (fact sheet), 2014.

The numbers bear this out. According to new Bureau of Labor Statistics data:

• Union workers have greater access to employer retirement plans. 94% of unionized workers have access to employer-sponsored retirement plans compared with just 72% of nonunion workers.
  • 85% of unionized workers participate in employer-sponsored retirement plans compared with 52% of nonunionized workers.
• Union workers are more likely to have pensions. 62% of union workers have access to pensions compared with only 10% of nonunion workers.

Not only do unions provide workers with secure and dignified retirements, they also create healthier and safer workplaces and communities.

• Union workers have greater access to health insurance. 93% of unionized workers have access to employer-sponsored health benefits compared with 71% of nonunion workers. Union employers also contribute more to their employees’ health care benefits.
• Union workers have greater access to paid sick leave. 93% of unionized workers have access to paid sick leave compared with 81% of nonunion workers.
• Unions create safer workplaces. Unions empower workers to bring up unsafe working conditions without fear of retaliation.

When unions are strong, they help raise standards for all workers. States with higher union density are more likely to have higher incomes, greater access to retirement and health insurance, and enacted paid sick leave legislation.

Read more on the benefits of unions and the roadmap to rebuilding union power.

EPI’s Director of Immigration Law and Policy Research Daniel Costa and Senior Economist Ben Zipperer join National Domestic Workers Alliance Vice President of Policy and Advocacy Haeyoung Yoon in a conversation moderated by Deputy Director of EPI’s Economic Analysis and Research Network (EARN) Jennifer Sherer.

Webinar links, notes and discussion

Impact of DHS Agent Surge on Minneapolis-Saint Paul Metro Area Labor Outcomes by Aaron Sojourner & Aaron Rosenthal

The ICE Effect: How Immigration Enforcement Affects the Economy and Jobs by Chloe East & Elizabeth Cox

Economic Impacts of Trump Administration Immigration Policy, October 2026 by Robert Lynch & Michael Ettlinger

Losing Status: An Estimated 3.5 Million Previously Protected Immigrants in the U.S. Are Projected to Lose Status by 2028 by FWD.us

EPI’s Cost of Deportations Calculator

The cost of DHS attacks on immigrants: Watch the first webinar in EPI’s immigration series.

Listen on The State of Working America Podcast

If you are an academic, student, non-profit researcher or advocate, or a journalist, you may view and use the content of this webinar and its related materials without requesting any further permission.

This is permitted under a non-commercial use Creative Commons license CC BY-NC-SA 4.0.

If you are a commercial enterprise looking to this information or data in any product that will be sold or as part of services and data you provide to paying customers, request commercial use by contacting EPI.

Find out about upcoming webinars first! Subscribe to EPI newsletters.

Key takeaways

• The Child Care and Development Fund (CCDF) was created to provide child care for low-income families while parents work or obtain education and training. But the program is so underfunded that it provides subsidies to just 16% of eligible children. 
• Child care is so costly that a full-time minimum-wage worker in Alabama would need to work 29 weeks, from January to July, and use every penny of their earnings to cover the cost of unsubsidized child care for one infant. 
• The Trump administration is proposing a new plan to divert CCDF funds from current recipients and give the resources to married couples with a stay-at-home parent. This will result in low-income working families losing CCDF funds, leading to increased child care costs for everyone and possibly resulting in the closure of some child care providers. 

Reporting in The New York Times describes an upcoming proposal from the Trump administration to use funds from the Child Care and Development Fund to provide “parent-based child care” in which a married parent could receive financial assistance to stay home to care for their children, while the other parent works. Providing additional income to support economically struggling families, including support for family members who want to be full-time caregivers for children, is a great idea. Pulling the funds to do so from the Child Care and Development Fund is not. 

The CCDF was created in the 1990s to provide child care subsidies for low-income families, so parents, primarily mothers, could work or go to school. The program is severely underfunded, however. It serves just 16% of all eligible children and just 17% of eligible children, age 5 and younger. That leaves hundreds of thousands of eligible children sitting on waitlists. Worse, some states don’t maintain a waitlist or have simply implemented an “enrollment freeze,” turning eligible applicants away rather than adding them to a waitlist. 

Who are the families that rely on subsidized childcare? 

Not adequately funding the CCDF hits low-income families, families with infants and toddlers, and families headed by single parents the hardest. To even be eligible for the program, federal criteria require families to earn less than 85% of the state median income. And states can make the income requirement even more stringent. For example, in seven states, a family with an income above 150% of the poverty line—just 47% of the state median income in Florida or 41% in Ohio—would not qualify for child care assistance. In 20 states, a family with an income above 200% of the poverty line would not qualify. This is just 55% of the state median wage in Michigan. 

The most recent data on the characteristics of families served by CCDF show that in 2023, 82% were headed by a single parent. In many states, an even larger share of families participating in the CCDF program are headed by single parents: 95% in Alabama, 93% in Illinois, North Carolina, and Ohio, and 92% in Pennsylvania. These are families with little choice but for the head of household—mothers in 75% of cases—to engage in formal employment if the family is to survive.

Then there are parents of infants and toddlers, pre-school-age children for whom child care is the most costly. The program provides subsidies for children up to age 13, but younger children are more likely than older children to receive subsidies. This reflects the fact that infants and toddlers have the greatest need for care, but parents often face a shortage of affordable qualified providers. This is in part due to younger children requiring more qualified providers than older children require and smaller child-to-staff ratios. In a Pulse Household Survey, 15% of all parents reported not working because they were caring for children, but for parents of the youngest children aged 0 to 4, it was 35%. 

Consequences of moving forward with the Trump–Vance plan

If married couple families are added to the list of eligible families without substantially increasing funding to provide subsidies to all eligible families, as is reportedly proposed, it will do significant harm to families of all types, whether they are married or not and whether they rely on subsidies or not.

With the new strain on available funds, families currently relying on the CCDF could lose the subsidies that allow them to provide even a modest living for themselves and their children. Without the subsidized care provided by the CCDF, these families could work full-time for most of the year just to pay for child care. In Alabama, for example, a minimum-wage worker working full-time would have to work 29 weeks—the equivalent of working from January to July—and use every cent of those wages to be able to cover unsubsidized child care for one infant. In North Carolina and Ohio, these workers would need to work from January to October. 

These are parents who want to work to support their families but cannot afford unsubsidized child care. They also cannot afford not to work.

Reducing the availability of child care funding for working mothers and fathers will not only hurt families who rely on child care subsidies, it will hurt all families who rely on child care. Siphoning these funds away from working families—disproportionately Black single mothers—will also deprive child care providers of a key source of revenue, potentially leading them to charge other families more for care, be forced to shed staff, or even close down. With families in so many communities across the nation already experiencing a child care crisis, this would exacerbate the shortage issue. 

We should fully fund the CCDF and provide needed support to all families

So many American families are already struggling to pay for the basic necessities, including rent, food, and gas. In an economy that requires dual-earner households where possible, child care subsidies provide struggling families with some basic support at a time when prices continue to rise, and safety nets, including health care subsidies, food assistance, and even heating assistance, continue to be cut. It’s notable that instead of providing real solutions to the affordability pressures many Americans face, Vice President Vance—who has championed this policy—and Trump’s Department of Health and Human Services are proposing what is effectively an attack on single parents, working parents, and working women in an effort to center married households and to push a political and cultural agenda about supporting supposedly “traditional” values. We must invest in families, regardless of their composition—married couples, single mothers, or grandparents caring for their grandchildren. We should fully fund the CCDF to ensure all eligible families are supported. 

Outside of a crisis or recession, Americans’ perceptions of how the country and economy are being managed have never been so negative. Many have attributed this voter unhappiness to a crisis of “affordability.”

It is objectively true that it is too hard for most American families to afford a secure and dignified life. But the word “affordability” leads too many people—including policymakers—to fixate on prices. Affordability is not just about prices; instead, it’s the outcome of a race between incomes and prices.

This is not just economists quibbling. Focusing on prices will lead policymakers to ignore far too much of the useful playing field when thinking about what changes could make life better for working families.

In this article, we make the following arguments:

• Far too many families are unable to afford a decent economic life.
• The primary cause is a large increase in income and wage inequality, with incomes and wages for the vast majority of families lagging far behind what they could and should be.
• This rise in inequality was caused by increasingly unequal “market” incomes (e.g., wages and salaries, returns on investments), while changes in taxes or transfers (e.g., Social Security, Medicare, unemployment insurance) slightly dampened the rise of income inequality.
• The large rise in income inequality was driven by intentional policy changes that affected typical workers’ leverage and bargaining power in the labor market—and that means they can be reversed.
• In capitalist economies (like ours), labor markets are inherently tilted toward employers—but historically and globally, broadly shared prosperity has only been achieved when policies that intentionally support workers (like strong unions, adequate minimum wages, and full employment mandates) have provided a countervailing force against employers’ power in labor markets. 
• Much of the post-1979 period in the United States saw an assault on worker-friendly policies, and this led directly to the rise in inequality and to weak income growth for working families.

Americans’ economic dissatisfaction has real roots

The U.S. economy is the richest in the world, yet the gap between what it could deliver to working families versus what it actually delivers is maddening. This gap can be measured with some precision. Figure A shows inflation-adjusted household income for the middle-fifth of U.S. families between 1979 and 2022, as well as what this growth could have been had it simply grown as fast as average incomes did in this period. 

This gap is driven by inequality. Average incomes can only rise faster than incomes at the middle if some groups—the ultra-rich in this case—see strongly above-average growth. This gap between average growth and growth experienced by the middle reached staggering levels by 2022 (the most recent data from the Congressional Budget Office). In that year, inequality’s rise since 1979 deprived middle-income families of an average of $28,100. Life for these families would be far more affordable today if they had this money coming in each year. And that’s well within our grasp. Average income growth is by definition attainable. All that’s needed are policies that ensure income growth is broadly shared, instead of policies that cause staggeringly fast income growth among the top 1% and much slower income growth for working people. 

Figure B shows this inequality another way—charting average annual growth rates for the 1979–2022 period for a number of groups ranked by their position in the income distribution. The strikingly bad news from this figure is that only household groups above the 90th percentile saw income growth that matched or exceeded average income growth. How can more than 90% of households be below average when it comes to income growth? This is possible because the top 5%—and especially the top 1%—saw astoundingly fast growth over this period. 

For any given average growth rate, faster growth at the top of the scale must be matched by slower growth at the middle and/or bottom. It is this zero-sum dynamic of inequality, not anything to do with prices, that has been the crushing drag on regular Americans trying to afford a better life over time.

Staggering inequality is a choice

This growth in inequality has been driven by the rules governing markets—rules our elected leaders determine—not by taxes or transfers. Figure A showed the staggering $28,100 gap in market (pre-tax and transfer) income between what families in the middle-fifth actually made in 2022 versus what they could have made had inequality not risen. Figure C shows how large this gap is after the federal government gets involved on the tax and transfer side of the equation. 

Taxes obviously reduce incomes, but transfers (social insurance like Social Security and income support payments like unemployment insurance) raise incomes. For the middle-fifth of US households, this effect is largely a wash—their current income levels in Figures A and C are very similar. But because the United States still has a progressive federal tax system (though not as progressive as we would like), the rise of inequality in this post-tax and transfer data is slightly muted—i.e., federal taxes and transfers shrink the gap somewhat. By 2022, the annual gap after accounting for federal taxes and transfers is $17,698—still a sum of money that would be transformative for American families.

Essentially, federal taxes and transfers undid roughly one-third of the rise in income inequality, allowing rich households to pocket roughly two-thirds of their gains.*

This rise in inequality was overwhelmingly driven by an intentional, multipronged policy campaign to suppress wages that was undertaken by shareholders, other capital owners, and corporate executives, with policymakers greasing the skids along the way.4 The primacy of wage suppression can be seen in Figure D, which compares the economy’s potential to pay higher wages and incomes with the actual hourly pay of typical workers in the United States. 

We measure the economy’s potential to pay higher wages by productivity, which is the output and income generated in the economy in an hour of work on average. And we define typical workers’ pay as the wages and benefits of workers in production and nonsupervisory positions, a group that constitutes over 80% of the economy’s private-sector workforce, excluding higher-wage managers and executives. As you see in the figure, in the three decades after World War II, productivity and pay mostly moved roughly in tandem, with typical workers’ pay rising 83% as fast as productivity. After 1979, these lines diverge sharply, with workers’ pay rising only about 43% as fast as productivity. 

If typical workers’ pay had risen in line with productivity growth in the years since 1979, their hourly pay would be 43% higher today. For a full-time, full-year worker making the median wage, this would constitute annual wages that are almost $23,000 higher.5 Where did that $23,000 go? Instead of paying workers more as their productivity rose, corporate executives, other already highly paid professionals, and shareholders captured those gains for themselves.

This growing gap between what shows up in typical workers’ paychecks and benefits versus the overall income being generated in the economy is the root story of American inequality and of today’s affordability crisis.

This pay-productivity gap can be decomposed into two parts: the portion driven by rising inequality in “labor” income (income earned from work), and the portion driven by a shift from labor income to “capital” income (income from investments, like when a stock increases in value). Growing inequality within labor incomes—earnings growing much faster among high-paying jobs than among middle- and low-paying jobs—accounts for almost 80% of the gap. The remaining 20% is accounted for by a shift from labor income to capital income.6 Below we say a bit more about each of these.

Rising inequality of labor incomes

The larger factor in the rise of overall income inequality is the growing inequality within labor incomes. This often surprises people, who assume the story of rising inequality is mostly one of the profits of rich corporations rising while most of their workers are left behind. It’s true that most workers in these corporations do not benefit, but the powerful employees who do prosper—CEOs and other executives—receive astronomical salaries that are classified as labor income in economic data, and these inflated executive salaries do cut into corporate profits. 

In addition, below the stratospheric level of corporate managers at large companies, there is a stratum of workers in medicine, legal services, and finance who command huge salaries. It’s not a large group of people, but the rise in their pay has been extreme. Figure E highlights this radical inequality within labor incomes, showing annual earnings of various wage groupings. (To keep the figure legible, pre-1979 data are not shown.) Prior to 1979, wage growth among very high wage workers—those in the top 10%, the top 1%, and the top 0.1%—was roughly in line with wage growth for the vast majority (i.e., for the bottom 90%). But between 1979 and 2023, cumulative growth in average annual earnings for the bottom 90% of workers was 44%, compared with 133% for the top 1%. For the top 0.1%, this growth was 354%—so high it doesn’t fit in the figure.

Given that labor income remains the large majority of all income generated in the economy, this huge rise in inequality within labor incomes is a key driver of the economy-wide march to greater inequality. Workers at the top of the wage scale were largely able to insulate themselves from the campaign of wage suppression launched by corporate owners. Of course, some of them were active participants in this campaign and got a significant cut of its benefits (think CEOs and lawyers for union-busting law firms). But the vast majority of workers (roughly 90%, as we see in Figure B) were on the losing side of this wage suppression campaign and found their wages falling far behind the economy’s potential to deliver strong and sustained wage growth. 

In some ways, the influence of rising inequality within labor incomes might be underestimated. For decades, U.S. tax policy has levied lower tax rates on capital income than labor income, and a great deal of capital gains escapes taxation entirely due to loopholes. 

Many of the same people who have been privileged enough to insulate themselves from wage suppression are also privileged enough to have excellent accountants who can make their incomes appear in whatever form results in the lowest taxes.7 For example, CEOs are overwhelmingly paid with “performance-based” measures, which means measures tied to the value of their companies’ stock prices. Twenty years ago, the large majority of this stock-based pay for CEOs came in the form of stock options, which are contracts that give the CEO the right (but not the obligation) to buy shares of stock at a set price. If the market price went above this set price, CEOs could exercise these options and pocket the difference as pay. The gains from exercised stock options are recognized by the IRS as labor income, taxed accordingly, and classified in data as labor income. But over the past two decades, there has been a pronounced shift in the stock-based pay of CEOs away from stock options and toward the outright granting of stock. In this case, CEOs are not given a right to buy shares at a preferential price; they are simply given shares.8 What’s important for the split between capital and labor incomes is that these non-option forms of stock-based compensation are far less likely to be captured in measures of wage incomes. So this tax evasion strategy artificially depresses estimates of labor income in the economy.

The shift from labor to capital incomes

While most of the pay-productivity gap stemmed from the rising inequality within labor earnings discussed above, a nontrivial portion of this gap stemmed from a shift in overall income from labor to capital. This goes far beyond the tax avoidance trick described above for CEO compensation—including paying regular working people less so that shareholders get more. If, for example, a corporation were able to suppress its workers’ pay while raising customers’ prices and/or cutting what it paid suppliers (which generally means those suppliers paying their workers less), it would earn higher profits. By successfully suppressing wages to boost profits, American corporations have made their stock more valuable. Imagine an investor buys $100 in company stock with the expectation of an annual return of $5. If the company undertakes a successful campaign of wage suppression that boosts annual returns to $10, many other investors will buy company stock—bidding up share prices.

So even though this labor-to-capital shift in overall income is the smaller player in generating overall income inequality, it still had profound effects on American economic life. Estimates indicate that anywhere from 40% to nearly 100% of the entire nominal gains in the U.S. stock market since 1989 can be attributed to this shift of income from workers to capital owners.9

The rise in US stock prices in recent decades is a key driver of another kind of economic inequality: inequality of wealth.^† One of the primary sources of wealth, the ownership of corporate equities (e.g., shares of stocks), is incredibly concentrated; the top 10% of households own about 85% of all corporate equities, and the top 1% own nearly 40%.10

The concentration of corporate equities combined with the role of wage suppression in making these equities far more valuable leads to a clear implication: The wage suppression of recent decades is not just by far the biggest driver of the rise in income inequality; it is also by far the biggest driver of the rise in wealth inequality. Most of the rise in wealth inequality in recent decades has been the outcome of an intentional transfer away from workers to the top. 

Labor markets are not fair

This rise in inequality in recent decades has attracted much attention from researchers—along with everybody else struggling to pay for groceries and keep the lights on. For a long time, economists’ role in the debate over inequality was to look for reasons why well-functioning, competitive markets could generate lots of inequality. This often led to explanations that essentially blamed workers for the outcomes. The argument was that the fair and competitive labor market had spoken and that these workers were falling behind because their skills and efforts had been found wanting. The precise failure identified was often workers’ alleged inability to adapt to the quickening pace of technological change in the economy. 

But the evidence supporting this view of inequality driven by apolitical forces working through fair and competitive markets was incredibly thin.11 That led many researchers to examine whether labor markets are by their very nature tilted against workers, making it very difficult to secure regular raises that match overall economic growth. 

There is ample evidence for this view—that excess employer-side power makes labor markets generally unfair and inefficient, and that truly fair, competitive labor markets are the exception, not the rule. For example, many employers, and particularly those of low- and moderate-wage workers, rarely if ever negotiate pay; instead, they post take-it-or-leave-it wage offers.12 And when a given employer lets its wages lag behind those of potential competitors, workers’ exit from the lower-wage firm is far less common than would be predicted under truly competitive labor markets (where employers robustly compete for workers).13

This employer-side power is rooted in many obvious factors in real-world labor markets that make it hard for workers to effectively search for better jobs and, therefore, force employers to compete over them. These include things like lack of information about wages and benefits offered by other employers, transportation restrictions that require workers to look for jobs only in places near their homes or public transit nodes, and child care considerations that require a job’s location be compatible with picking up kids at a regular time, along with many other factors. 

Another barrier to competition is the obvious fact that in the short run, most employers need the income a new worker would generate for their business far less than most workers need the income from a job. In a jobsite of 100 workers, having a month go by understaffed by a single worker reduces business income by roughly 1%. In a household with a single worker, having a month go by without a job reduces income by essentially 100%. 

Employers exploit these barriers to employees finding better options by “marking down” wages below what would be necessary for employers to attract and retain workers in competitive labor markets. These markdowns can be large enough to push workers’ pay well below the value they produce for the employer (i.e., below the “market clearing” wage). This makes them not just unfair, but inefficient—a drag on economic growth. 

Key policy choices that led to rising inequality 

Having realized the role of employers’ power in determining labor market outcomes, the importance of specific policies is magnified. For example, before the 1990s, many economists were extremely skeptical that minimum wages could do much good in raising wages without steep downsides like job loss. Why? Because they erroneously used models of competitive labor markets. 

But more accurate models that include employers’ power reveal significant room to raise minimum wages without generating job losses. The evidence over the past 30 years has been highly persuasive that minimum wages could be much higher than they were in the 1980s and 1990s without causing job losses and that the benefits for low-wage workers would be large.14

The period of rising inequality since 1979 was one of profound institutional change in labor markets. The federal minimum wage, for example, lost 35% of its value between 1979 and 2025 as legislative inaction (i.e., not raising it) allowed it to be battered into irrelevance by inflation.15 This was also a period that saw a pronounced acceleration in the decline of unionization rates of American workers.16 It was a time when high levels of unemployment were tolerated by policymakers for extended periods in the name of fighting inflation.17 And it was a time when increasing integration between the rich United States and a poorer global economy was done on terms that were written by and for corporate interests.^‡

Several years ago, researchers at our organization, the Economic Policy Institute, reviewed the research on how much specific policy choices likely contributed to growing inequality. Adding together the impacts of the changes like those listed above could easily explain the lion’s share of the rise in inequality since 1979.18

For example, one key policy change was the practical abandonment of the Federal Reserve’s full employment mandate. By law, the Fed is supposed to pursue both stable inflation and full employment (which means trying to keep unemployment as low as is consistent with stable inflation). But between 1979 and 2007 (right before the Great Recession), the Fed largely acted as if it did have a mandate to pursue stable inflation but did not have one to pursue full employment. 

After the Great Recession, the Fed admirably reversed course and tried to push the economy back to full employment, but its tools proved too weak given the magnitude of the shock. In such situations, fiscal policy—taxes and spending—should be used aggressively to restore full employment. But in the 2010s, political gridlock and excess caution kept policymakers from doing this, and much of that decade was plagued by excess unemployment. 

Excess unemployment does not just leave willing workers locked out of jobs. It also saps the ability of still-employed workers to demand raises. For nonunion workers, their chief leverage for getting wage increases is threatening to quit. This threat is only credible when unemployment is low. Consequently, the too-high unemployment rates for most of the period from 1979 to 2019 were a drag on wage growth. In our estimates, too-high unemployment may well have explained nearly a third of the entire pay-productivity gap over that period19—not to mention the devastation it wrought for millions of families.

Another key policy change was the failure to keep the playing field level between workers looking to organize and join unions and the employers who wanted to stop them. The National Labor Relations Board is supposed to safeguard this right, but its tools have proved too weak in the face of fierce employer opposition to unions, and policy changes to strengthen these tools have consistently been blocked. The results of throttling the growth of new unions have been profound; our estimates are that declining unionization likely explains a quarter of the pay-productivity gap since 1979.20 Crucially, the decline in unionization did not just hurt workers who otherwise would have been unionized. By far the biggest of the wage-suppressing effects of deunionization has been on the broad pool of nonunion workers. As unions lose strength, they stop being able to set industry-wide pay standards that even nonunion employers feel like they have to meet to avoid hemorrhaging employees. 

The wage-depressing effect of trade flows from poorer nations—flows encouraged by the corporate-led trade agreements the United States has signed in recent decades—can likely explain another 10 to 15% of the pay-productivity divergence since 1979.21

The wrong incentives

Tolerating excess unemployment, throttling workers’ ability to join unions, failing to update the minimum wage as costs rise, and signing corporate-friendly trade agreements were some of the many instruments of wage suppression undertaken and abetted by policymakers in recent decades. At the same time, choices legislators made on tax policy boosted the incentive for capital owners and corporate managers to aggressively use these instruments to increase their wealth. 

When ultra-high incomes and corporate profits are taxed at high (i.e., appropriate) rates, the incentives for powerful individuals to rig the rules of markets to suppress regular workers’ wages are much smaller. Key research shows that this incentive effect is real and powerful. For example, across countries, the larger the tax cuts on the rich enacted in recent decades, the greater the increase in pre-tax inequality.22 And, the lower the top tax rates for individuals, the higher the levels of pre-tax CEO pay.23 High taxes reduce the benefits of rule-rigging, so cutting taxes increases rule-rigging. This means that raising taxes on the richest households and corporations results in new revenue and more equal pre-tax incomes. But from the mid-1970s, tax rates for high-income households and corporations have been cut steadily and deeply in the United States, reducing both tax revenues and wages for working people.

Income inequality, not high prices, is behind the affordability crisis

We opened this article with a claim that affordability is the outcome of a race between income and prices. Our long walk through the economics and history of recent American inequality highlights that intentional policy choices have deprived typical households of income they could have otherwise claimed. Without this inequality, a middle-income household today would have tens of thousands of dollars more per year—and this would obviously make affording a decent life much easier. 

But some might wonder if we have still given prices short shrift in how much they contribute to affordability challenges. We don’t think so, for a number of reasons. We sketch three of them here.

First, all of the income, wage, and productivity statistics we have included in this article have been real (i.e., they have been adjusted for the impact of inflation). And it is unambiguously true that real (inflation-adjusted) incomes are the proper way to measure living standards and economic possibilities for households. 

Getting distracted by price growth while missing what’s happening with income will lead to wrong conclusions about economic performance over even relatively recent periods of time. Figure F compares two periods, both starting one year before a deep recession struck and then running five years: 2007–2012 and 2019–2024. In the first period, inflation averaged 1.8%, while in the second it ran more than twice as fast at 4.2%. Yet real (inflation-adjusted) wage growth for low- and middle-wage workers was far faster in the second period. For the lowest-wage workers, real wages fell by 2.1% in the first period but rose by 15.3% in the second. For workers in the middle of the wage scale, real wages fell by 1.5% in the first period but rose by 5.8% in the second. 

Over very short periods of time (one to two years), it is true that a rapid spike in prices tends to drive down real incomes and wages. But over any longer period (even as short as three to five years), assessing how the economy is doing for typical families rarely bears much relationship to price growth.

Second, even when researchers adjust for inflation differently at different parts of the wage and income distribution, the impact is modest. Such measures account for things like lower-income families spending a higher share of their income on rent and groceries and a lower share on vacations. But there are surprisingly small differences in overall price growth faced by families at different income levels. For example, from 2019 to 2025, when the price of housing and groceries was on peoples’ minds for good reasons, the inflation rate faced by the bottom 40% of households was just 0.2% higher than for the top 20% of households.24 In short, the growth in prices faced by different groups varies far less than the growth of their incomes and wages. 

Third, a key insight in assessing affordability debates is that one person’s cost is another person’s income. If the cost of a pound of coffee doubles from $10 to $20, this constitutes $10 of additional income that somebody is getting. Perhaps the coffee grower or the shipper or the grocery store shareholders or the CEO or the cashiers or some other link in the supply chain is getting an extra $10 (or several of them are getting some slice of it). This means that rapidly rising prices cannot result in less income overall, so they are highly unlikely to actually make an entire economy poorer. Instead, the groups that face only the price increase lose out while groups receiving the extra income win. 

This fact that every price is an amalgamation of various income streams also means policymakers can more usefully target wage and income policies rather than price policies. Again, my bill at the grocery store pays for the wages of cashiers, the pay of the company CEO, the dividends to shareholders, the payments to suppliers, and more. Even if we’re unhappy about this grocery bill, we likely don’t want all of those price components to get squeezed. We probably want the wages of cashiers to rise while hoping to rein in CEO pay and shareholder dividends. Policies that only look to restrain prices—price controls, for example—make no such distinction, so we don’t know who in the grocery supply chain will bear their burden (though we can guess it’s more likely to be the cashiers than the CEO). But if we raise minimum wages, change labor law to allow more widespread unionization, and raise taxes on the ultra-rich and on corporate profits, we have a very good idea of which incomes will be boosted and which will get squeezed.

Creating a fairer economy

It is deeply depressing that intentional policy acts led to the enormous rise in inequality that is making life so much harder for so many people. If tens of millions of American households had tens of thousands of extra dollars in their bank accounts each year while billionaires had significantly less money, the country would be a much better and happier place.

What brings us hope is the knowledge that because the rise in inequality was not the inevitable outcome of a modern economy, it can be halted and reversed. Today’s workers have the skills and abilities needed to support much higher incomes with no loss in efficiency or employment—if we change policy to give them these higher wages. This is excellent news. Of course, many of today’s elected leaders—and their donors—have little interest in reducing inequality, so the road ahead is long. But the foundational ingredients for a fairer and more efficient economy are clear: 

• Keep unemployment rates low for long periods of time and fight recessions fiercely when they inevitably occur. 
• Restore the right to organize new unions and bargain collectively. 
• Raise minimum wages, including the federal minimum wage. 
• Enact rules for the global economy that support healthy wage growth, not just healthy corporate profits. 
• Crush the incentive to rig the rules of the economy by raising taxes significantly on ultra-rich households and corporations. 

The details on how we create a fairer economy are more complex—and they do matter! But understanding that the affordability crisis facing American families is overwhelmingly an inequality crisis is a necessary and useful place to start.

*Since this analysis goes through 2022, it does not include the tax or benefits cuts (including to Medicaid and the Supplemental Nutrition Assistance Program) in the One Big Beautiful Bill Act that President Trump signed into law in July 2025. These will further increase inequality.

^†Wealth is the value of a person’s assets (e.g., the equity in their home, stocks and bonds in their retirement accounts, or their baseball card collections) minus the value of their debts.

^‡For details, see “A Trade Policy That Puts Working Families First.” 

Footnotes

1. Congressional Budget Office, The Distribution of Household Income, 2022 (January 2026), cbo.gov/publication/61911.

2. Congressional Budget Office, The Distribution.

3. Congressional Budget Office, The Distribution.

4. For a much deeper dive into the specifics of this policy campaign of wage suppression, along with empirical assessments of how much it cost American families, see L. Mishel and J. Bivens, “Identifying the Policy Levers Generating Wage Suppression and Wage Inequality,” Economic Policy Institute, May 13, 2021, epi.org/unequalpower/publications/wage-suppression-inequality.

5. The median wage for U.S. workers in 2025 was $25.67. This (and a lot more) can be found at data.epi.org. Multiplying this median wage by 0.43 and then by 2,080 (hours worked by a full-time/full-year worker) yields the $23,000 figure.

6. Earlier estimates of how much inequality within wages contributed to the pay-productivity gap can be found here: L. Mishel, “Growing Inequalities, Reflecting Growing Employer Power, Have Generated a Productivity–Pay Gap Since 1979,” Working Economics Blog, September 2, 2021, epi.org/blog/growing-inequalities-reflecting-growing-employer-power-have-generated-a-productivity-pay-gap-since-1979-productivity-has-grown-3-5-times-as-much-as-pay-for-the-typical-worker. The easy way to update this (which we did for this report) is to compare productivity with growth in overall average compensation of American workers since 1979. This overall average compensation rose by roughly 76% since 1979. Given typical workers’ pay growth of just under 30%, this means that 46% (76% minus 30%) of the divergence between typical workers’ pay and productivity is a difference between typical workers’ pay and average pay.

7. See here for an estimate of how much of today’s reported capital incomes would be more properly classified as the returns to work (i.e., labor incomes): A. Eisfeldt, A. Falato, and M. Xiaolan, “Human Capitalists,” NBER Working Paper no. 28815, National Bureau of Economic Research, April 2022, nber.org/papers/w28815.

8. For more on CEO pay levels and their composition, see J. Bivens, E. Gould, and J. Kandra, “CEO Pay Has Skyrocketed Since 1978,” Economic Policy Institute, September 25, 2025, epi.org/publication/ceo-pay.

9. For this estimate, see D. Greenwald, M. Lettau, and S. Ludvigson, “How the Wealth Was Won: Factor Shares as Market Fundamentals,” Journal of Political Economy 133, no. 4 (April 2025): 1083–1132; and A. Atkeson, J. Heathcote, and F. Perri, A Macroeconomic Perspective on Stock Market Valuation Ratios (Federal Reserve Bank of Minneapolis, Research Division, January 2026), minneapolisfed.org/research/sr/sr682.pdf.

10. See Table 10 in: E. Wolff, “Household Wealth Trends in the United States, 1962 to 2019: Median Wealth Rebounds… but Not Enough,” NBER Working Paper no. 28383, National Bureau of Economic Research, January 2021, nber.org/system/files/working_papers/w28383/w28383.pdf.

11. For a much deeper dive into the weakness of claims that inequality was driven by technology rewarding skilled workers and penalizing less-skilled workers, see J. Schmitt, H. Shierholz, and L. Mishel, Don’t Blame the Robots: Assessing the Job Polarization Explanation of Growing Wage Inequality (Economic Policy Institute, November 19, 2013), epi.org/publication/technology-inequality-dont-blame-the-robots.

12. One study found that roughly 75% of low-wage jobs were ones where employers made take-it-or-leave-it posted offers: R. Hall and A. Krueger, “Evidence on the Incidence of Wage Posting, Wage Bargaining, and On-the-Job Search,” American Economic Journal: Macroeconomics 4, no. 4 (October 2012): 56–67; and R. Hall and A. Krueger, “Evidence on the Determinants of the Choice Between Wage Posting and Wage Bargaining,” NBER Working Paper no. 16033, National Bureau of Economic Research, May 2010, nber.org/system/files/working_papers/w16033/w16033.pdf.

13. For evidence on how nonresponsive worker quits are to wage cuts relative to predictions of competitive markets, see A. Dube, L. Giuliano, and J. Leonard, “Fairness and Frictions: The Impact of Unequal Raises on Quit Behavior,” American Economic Review 109, no. 2 (February 2019): 620–63.

14. For a comprehensive review of this evidence, see D. Cengiz et al., “The Effect of Minimum Wages on Low-Wage Jobs,” Quarterly Journal of Economics 134, no. 3 (August 2019): 1405–54.

15. Economic Policy Institute, “Minimum Wages: Real Minimum Wage (2025$),” 2026, data.epi.org/minimum_wage/minimum_wage_levels/line/year/national/real_minimum_wage_2025/overall?timeStart=1938-01-01&timeEnd=2025-01-01&dateString=1979-01-01&highlightedLines=overall.

16. P. Romero and J. Whittaker, A Brief Examination of Union Membership Data (Library of Congress, June 16, 2023), congress.gov/crs-product/R47596; and H. Meyerson, “Economic Inequality Is Undermining America: Worker Solidarity Will Build a Better Future,” AFT Health Care 3, no. 2 (Fall 2022): 33–36.

17. S. Galan, “Monthly Federal Funds Effective Rate, Unemployment Rate and Inflation Rate in the U.S. During Paul Volcker’s Terms as Federal Reserve Chairperson from 1979 to 1987,” Statista, October 2022, statista.com/statistics/1338105/volcker-shock-interest-rates-unemployment-inflation/?srsltid=AfmBOoqoXQ4yTpqiSJMevwNFbnNGEETrwhlltEeVrJuA1rThyYCBBkFl.

18. Mishel and Bivens, “Identifying the Policy Levers.”

19. J. Bivens, “Focus on the Boom, Not the Slump—the Fed’s New Policy Framework Needs to Stop Cutting Recoveries Short,” Working Economics Blog, Economic Policy Institute, June 18, 2019, epi.org/blog/focus-on-the-boom-not-the-slump-the-feds-new-policy-framework-needs-to-stop-cutting-recoveries-short-epi-macroeconomics-newsletter.

20. Mishel and Bivens, “Identifying the Policy Levers.”

21. Mishel and Bivens, “Identifying the Policy Levers.”

22. A. Fieldhouse, Rising Income Inequality and the Role of Shifting Market-Income Distribution, Tax Burdens, and Tax Rates (Economic Policy Institute, June 14, 2013), epi.org/publication/rising-income-inequality-role-shifting-market.

23. J. Bivens, “Using Tax Policy to Restrain CEO Pay: Best Practices and Smart Alternatives,” Economic Policy Institute, December 13, 2023, epi.org/publication/using-tax-policy-to-restrain-ceo-pay-best-practices-and-smart-alternatives.

24. Authors’ analysis of data obtained from: Federal Reserve Bank of New York, “Economic Heterogeneity Indicators (EHIs),” 2026, newyorkfed.org/research/economic-heterogeneity-indicators.

The cost of this wage suppression is real. Because wages make up a large share of income for most U.S. working families, lost wage growth meant sluggish income growth. However, there was one group where wage growth, and corresponding income growth, was substantial: the top 1%, whose wages skyrocketed by 182% since 1979. This uneven growth implies that today’s labor market, and the income derived from it, is deeply unbalanced. Rather than workers reaping the rewards of their rising productivity, a growing share of their labor has gone to the top 1%, leading to rising wage and income inequality.  

Decades of sluggish wage growth for typical workers—paired with the message that it was workers’ own fault for not obtaining the necessary skills for today’s economy—have left many pessimistic about whether policy changes can help them earn more. But if we want to achieve broad-based economic security in the U.S. economy, there is no alternative to raising pay for workers up and down the wage distribution. The U.S. economy generates enough income in aggregate to provide decent rates of pay growth for all workers—the challenge is a political one.

1

1

The cost of inequality to middle-class households

The stakes of rising inequality for middle-income households in the United States are enormous. By 2022, steadily rising inequality was costing these households roughly $30,000 per year.

Figure 1 compares actual income growth of the middle fifth of nonelderly households since 1979 with what this income growth could have been had inequality not increased. The graph highlights market-based incomes because rising inequality was driven entirely by this income category—earnings from the labor market and business income like dividends and interest payments. We focus on incomes of nonelderly households because inequality largely stemmed from unbalanced labor market power, and nonelderly income is dominated by labor earnings.

In 2022, the average market income of the middle fifth of households was $96,335. But their income would have been $127,011—roughly 32% (over $30,000) higher—had inequality not increased after 1979. Put another way, if the middle fifth of households had seen market income grow at the overall average rate—an overall average pulled up by stratospheric growth at the very top—their income could have been roughly $30,000 higher by 2022.

Share this chart:

2

2

Income inequality is driven by unbalanced power in labor markets

The inequality in market-based incomes highlighted in Figure 1 was overwhelmingly driven by unbalanced power in labor markets that prevented typical workers from seeing their paychecks keep up with overall economic growth. Figure 2 shows hourly pay (including benefits) for the roughly 80% of the private-sector workforce who are not managers or supervisors and a measure of economy-wide productivity—defined as the income generated in an average hour of work in the U.S. economy. The rising gap between these lines is how we define wage suppression—keeping typical workers’ wage growth slower than growth in incomes in the rest of the economy.

In the three decades following World War II, hourly compensation for the vast majority of workers rose 2.1% on an annualized basis, roughly in line with productivity growth of 2.5%. But for most of the last five decades (except for a brief period in the late 1990s), pay for the vast majority lagged further behind overall productivity. Over the entire 1979 to 2025 period, hourly pay for a typical worker rose just 0.6% while productivity increased 1.4% on an annualized basis. This means that workers have been producing far more than they receive in their paychecks and benefit packages from their employers, and this wedge has grown substantially over time. Some of those gains in productivity went to greater profits for corporations but an even larger portion went to highly paid managers and professionals, resulting in a vast increase in wage inequality over the last five decades.

Had we pursued policies to support broad-based wage growth—instead of wage suppression—typical workers would have much higher pay today. In fact, if annual pay for production and nonsupervisory workers had instead tracked productivity growth since 1979, their pay would be 45% higher, or $30,000 more than it is today.

Share this chart:

3

3

Wage suppression boosted profits

The declining labor share of corporate-sector income—the share of corporate income received by workers in the form of wages and benefits—explains a portion of the wedge between productivity and pay. The labor share of corporate-sector income has not recovered to its pre-pandemic level, and remains well below its 1979 to 2008 average. Figure 3 illustrates the labor share of corporate-sector income since 1979. Typically, the labor share rises during recessions because profits fall much faster than wages during downturns. Then, in the early stages of recovery, the labor share falls significantly as profits increase much more rapidly than wages. Usually, the labor share then rises again late in an expansion as labor markets tighten and workers regain the bargaining power necessary to secure wage increases. Since 2000, no business cycle expansion has gone on long enough with unemployment low enough to put serious upward pressure on the labor share of income. Small upticks were seen sporadically over the last 15 years, but this progress, but this progress has stalled out.

Two things about the shift from labor compensation to profits in recent decades deserve some context. First, even as measured in official data, this shift explains well under half of the overall wedge between productivity and pay—the biggest piece of this gap remains inequality within labor compensation (we return to this point further below). Second, some of the changes in the official data might be more the outcome of accounting decisions than of fundamental changes in the economy.

Since the early 2000s, capital income (dividends, capital gains, and business income) has been taxed at lower rates than labor income. Highly privileged economic actors who have the ability to engage in tax evasion by reclassifying their income have likely done so in this time—relabeling income once classified as wages and salaries as capital income instead to pay lower tax rates. Recent research has estimated that roughly a third of the decline in the labor share of income might be driven by this kind of tax evasion rather than by a genuine change in which factors of production are seeing income gains.

Share this chart:

4

4

Highly paid corporate managers and professionals benefited the most from wage suppression

The largest portion of the growing wedge between productivity and pay can be explained by the astronomical wage growth for those at the highest end of the wage distribution. The ability of those at the very top to claim an ever-larger share of overall wages is evident in Figure 4. Annual earnings for the top 1% have grown 182% since 1979, while wages of the bottom 90% have grown just 44%. If the wages of the bottom 90% had grown at the average pace over this period—meaning that wages grew equally across the board—then wages for the bottom 90% would have grown by 65%, far higher and roughly 50% faster than the actual growth experienced. Earnings are so concentrated at the top that average growth is well above the 90th percentile of the earnings distribution. That means workers need to be among the highest 10% of wage earners to even experience average earnings growth. And everyone else—more than 90% of the workforce—saw less than average growth over the last four and a half decades.

Share this chart:

5

5

Since 1979, middle-wage workers only saw short-lived real wage growth during tightening labor markets

Between 1979 and 2025, the hourly wages of middle-wage workers (workers earning the median wage) cumulatively grew just 30%—about 0.6% per year. Yet even this disappointing cumulative wage growth occurred only because wages grew in the late 1990s and the late 2010s–2020s. Apart from these two periods, the wages of middle-wage workers were totally flat or in decline since 1979. Figure 5 shows real median wages since 1979. The red portions of the line highlight periods when median wage growth was genuinely stagnant, while the green portions highlight when growth occurred.

In the absence of strong institutions and labor standards like robust unions and high minimum wages, strong wage growth occurs only when labor markets are tight. In tight labor markets, the relative scarcity of workers gives them more power in the workplace to effectively demand decent hourly wage growth across the board.

Share this chart:

6

6

Wage suppression was not driven by increased benefits

The suppressed “pay” portion of the pay–productivity chart includes wages and benefits, suggesting that benefits have not offset slow wage growth. In fact, by many measures benefit generosity and availability have deteriorated alongside wage suppression. We explore trends in two key employment benefits in Figure 6: employer-sponsored health insurance and retirement coverage. We define employer-provided health insurance as insurance for which the employer pays some or all of the premium. We define retiree coverage as coverage a worker receives at their firm that offers it. We isolate workers who are strongly attached to their job, working at least 20 hours per week and 26 weeks per year.

For workers, sluggish pay has been compounded by the fact that health insurance and retirement coverage have declined significantly over time. Health insurance fell from 69.0% to 52.2% between 1979 and 2024. At the same time, retiree coverage nearly fell by half from 50.6% down to 29.0%. Furthermore, plan generosity has suffered with the trend away from defined benefit to defined contribution retirement benefits, as well as the shift to high deductible health plans.

Share this chart:

7

7

CEO pay grabs a larger share of wages

In a nation of increasing inequality, the most extreme disparities are between the heads of large American corporations and typical workers. Figure 7 tracks the ratio of pay of CEOs at the 350 largest public U.S. firms to the pay of typical workers. In 1965, these CEOs made 21 times what typical workers made. As of 2025, they make 325 times typical workers’ pay. This higher pay for CEOs does not reflect any increased contribution to corporate output or growth in the skills or productivity of CEOs. CEO pay gains help explain the growing divergence between pay and productivity. Further, the pay of CEOs and other highly placed corporate managers sets norms and market benchmarks across a range of elite occupations. The pay of specialty physicians, lawyers, consultants, and even sectors like university and nonprofit administration is likely pulled upward by rising pay for corporate executives.

Share this chart:

8

8

Tolerating excess unemployment has been deeply damaging for wage growth

As a rough rule of thumb, real wages for middle-wage workers outright fall when unemployment is greater than 5%.

Prioritizing high-pressure labor markets—those identified by extended periods of low unemployment—is key to providing lower- and middle-wage workers the leverage to bid up their wages. Figure 8 shows that the threshold for achieving any real wage growth at all is around 5% unemployment. In general, real wages grow when unemployment is lower than 5%, but they fall when unemployment is higher. Extended periods of low unemployment can also reduce wage inequality and racial disparities.

Share this chart:

9

9

The federal minimum wage no longer lifts wages

Figure 9 shows the decline in the real (inflation-adjusted) value of the federal minimum wage since its high in 1968. The last increase to the minimum wage was enacted in 2009, 17 years ago, which is the longest period without a minimum wage increase since its inception in 1938.

The minimum wage is essential for establishing the wage levels for the bottom fifth of wage earners. Today’s low-wage workers are far more educated, older, and experienced than low-wage workers in 1968. Yet, despite being more skilled and productive, their wages are 43% lower than wages earned in 1968, and 32% lower than the most recent increase in the minimum wage 17 years ago. The failure to raise the federal minimum wage has especially adverse effects on women and Black and Hispanic workers.

Luckily, many states have refused to follow the federal government’s inaction and have passed higher minimum wages at an increasing rate over the past two decades. Seventeen states and the District of Columbia now have minimum wages of at least $15 per hour, and more workers live in states with minimum wages over $15 than in states only bound by the too low federal minimum wage.

Share this chart:

10

10

The assault on collective bargaining hurts all workers

One of the main causes of wage suppression and rising wage inequality is the decline of collective bargaining, which lowers the wages of both union and nonunion workers. Collective bargaining not only raises wages for organized workers but also leads other employers to raise the wages and benefits of nonunion workers to come closer to union wage standards.

The erosion of collective bargaining can explain from one-fifth to one-third of the growth of wage inequality between 1973 and 2007; it had a greater impact on men than women. This erosion of unionization occurred despite large numbers of workers indicating they would prefer collective bargaining if they had a choice. But the political power of those with the most income, wealth, and power has prevented the adoption of laws to modernize our labor–management system and enable workers to pursue collective bargaining.

Figure 10 shows how wages could have grown if unionization hadn’t declined since 1973. If union density in 2025 had been the same level as union density levels of 1973 (14 percentage points higher than it is today), median wages would rise by 10.1%.

Share this chart:

Click here for the 2015 archive of this page.

Introduction

For Black Americans, public-sector employment has historically provided a pathway to better, more equitable and secure job opportunities compared with available private-sector jobs. The federal government has played an especially vital role in establishing a robust Black middle class in the Washington, D.C. metro area. According to the 2023 American Community Survey, roughly 2 out of 5 Black adults in the D.C. metro area were college graduates, Black median household income was nearly $90,000 and the Black homeownership rate was 52.8%. Postal service jobs have been particularly valuable to Black workers without college degrees because of the uniform wage and benefit structure (all postal employees who have the same job title and job tenure are paid the same nationwide) and higher pay relative to comparable private-sector employment. With a minimum education requirement of a high school diploma, the median hourly wage of a postal worker is 43% higher than the typical high school graduate. While federal employment has opened the door to social and economic mobility for generations of Black Americans, it has often been the battleground and served as a compass in setting higher labor standards and equal employment policies in the United States.

Opportunity. Backlash. Resistance. Change: A brief history of Black federal workers

The history of Black workers employed in the federal government dates to the Civil War when the federal government hired its first Black employee in the Treasury Department in 1863. In time, the federal government quickly became the largest employer of formerly enslaved people, with large concentrations in the military and the U.S. Postal Service (USPS). By 1912, the federal government was the largest employer of Black Americans in the nation, including highly skilled Black workers who were hired in high-ranking white-collar positions.

One of the earliest actions aimed at weakening the position of Black federal workers came shortly after the inauguration of President Woodrow Wilson. In 1913, Wilson racially segregated the USPS and Treasury department—the first federal agencies to employ, and in the case of USPS, promote Black workers to management positions. The administrative practice of segregating the federal workforce extended to the demotion of Black civil servants from white-collar positions, at-will firings, and refusal to fill open jobs with qualified Black candidates. Later that year, a group of Black workers formed the National Alliance of Postal Employees, the first industrial union in the federal service, to resist the administration’s racist tactics.

In the 1940s and 1950s, Presidents Franklin D. Roosevelt, Harry S. Truman, and Dwight D. Eisenhower each issued executive orders that took measured steps to undo the overtly racist and discriminatory federal employment practices put in place by Wilson. Those orders were largely directed at national defense industries, armed forces, and government contractors in response to the demands imposed by World War II. But, throughout the 1950s and 1960s, civil rights activists pushed the federal government to do more to expand its hiring of Black workers. In response, President Eisenhower’s Executive Order 10590 established the President’s Committee on Government Employment Policy (PCGEP) in 1955. The PCGEP involved federal agencies more fully in the government’s anti-discrimination agenda and called for departments to develop regulations in accordance with its mission to stop all discrimination in all federal employment. However, the group lacked the enforcement power necessary to accomplish that mission.

Over the following decades, job prospects for Black federal workers were most improved by a series of executive actions and legislation introduced in the 1960s and 1970s. On March 6, 1961, President John F. Kennedy’s Executive Order 10925 required the federal government and federal government contractors to practice non-discrimination in their hiring practices. Additionally, E.O. 10925 established the President’s Committee on Equal Employment Opportunity (PCEEO) to monitor non-discrimination on government contracts. In a move that distinguished the PCEEO from prior ineffective, enforcement-lacking efforts like Eisenhower’s PCGEP, Kennedy granted policy-making authority to the group led by Vice President Lyndon Johnson and Secretary of Labor Arthur Goldberg.

On January 17, 1962, Kennedy signed Executive Order 10988 which allowed limited collective bargaining for federal employees for the first time and opened the door to federal employee union membership under three different classifications: informal, formal, and exclusive recognition. Public-sector collective bargaining would play a central role in maintaining the quality and accessibility of federal jobs through labor contracts that fostered transparency with clearly defined policies and pay structures. Labor contracts also served to limit discriminatory outcomes while providing critical protections and recourse against other forms of exploitation or mistreatment.

The power of Kennedy’s executive orders was reinforced when Title VII of the historic Civil Rights Act of 1964, signed by President Lyndon Johnson, formally prohibited employment discrimination in the United States and established the Equal Employment Opportunity Commission (EEOC) to enforce the law. The Equal Employment Opportunity Act of 1972 extended Title VII protections to cover more employers and strengthened the enforcement power of EEOC by allowing them to litigate against employers, including federal agencies, who violated Title VII.

Within the span of the 1960s and 1970s, the federal government had established a clear definition of what it meant to be an equal opportunity employer, leveraged its purchasing power to compel private contractors to meet similar standards, extended limited collective bargaining rights to federal workers, and assigned the EEOC a central role in enforcing anti-discrimination law. Black federal employees also continued to support and advocate for one another, establishing the non-profit organization, Blacks in Government (BIG), in 1975. The progress made during 1960s and 1970s would be gradually chipped away in the decades that followed. 

Federal job losses since the 1980s

During the 1980s, the Reagan administration took a swipe at federal employees, unions, and anti-discrimination enforcement, but that record pales in comparison to more recent developments. While Reagan announced plans to make federal job cuts, and infamously fired 11,000 striking air traffic controllers in the early 1980s, federal payrolls actually rose by more than 200,000 during his presidency before dropping by 427,000 during the 1990s and taking another hit of 244,000 between 2010 and 2014. Since the 1980s, the postal service, a major employer of Black workers, has been under sustained assault, including attempts to undercut employee compensation and the agency’s solvency.

In 2025, the Trump administration took steps to implement massive cuts to the federal sector and reverse course in the government’s pursuit of equity by rescinding at least a dozen prior executive orders related to racial and/or gender equality and terminating workers in DEI departments within federal agencies. In a series of legally challenged actions, Trump fired decisionmakers at the EEOC and National Labor Relations Board (NLRB)—rendering two independent agencies responsible for enforcing workers’ rights non-operational for several months—while his newly created Department of Government Efficiency (DOGE) made severe staff reductions and eliminated entire federal agencies. Trump’s attacks on the federal workforce have also included attempts to limit the approval of collective bargaining agreements with federal workers. The actions of Trump and DOGE contributed to the loss of 288,000 federal jobs between January and December of 2025, based on data from the Bureau of Labor Statistics. Ironically, while federal jobs once provided Black workers relatively more job security, early evidence suggests the burden of federal job cuts has fallen disproportionately on Black women. The potential consequences of these actions go beyond job losses and include major implications for Black family incomes and racial and gender pay equity.

An accounting of the significance of federal sector employment for Black workers and families

As detailed in the history presented above, between 1941 and 1981, Black workers gradually improved their employment status in the federal government through collective and individual activism of groups like the National Alliance and Blacks in Government, within a context of official support for their rights through executive orders and landmark civil rights legislation. This improved employment status expanded the ranks of Black federal workers who were able to secure higher incomes. By 1970, the median household income for Black families was just $6,279 compared with a range of $7,178–$10,987 for those earning GS 5–8 salaries in the federal government. In fact, Black federal employees compensated between grades GS 5–8 were either close to or slightly above the national median of $9,867. This remains a factor today as the high concentration of federal employment and related professional job opportunities in the Washington, D.C. metro area helps to make metro D.C.’s Black median household income ($89,912 in 2023) one of the highest in the nation and well above the overall national median of $77,719.

Analysis of 2024 state-level data from the Office of Personnel Management (OPM) reveals that over 300,000 federal workers (excluding USPS) reside in the D.C. metro area, accounting for 60% of all federal workers in the District of Columbia and surrounding states of Virginia, Maryland, and West Virginia. Black workers are just over one-fourth of the federal workforce in the District of Columbia (28.8%), Maryland (27.9%), and Virginia (26%). While the D.C. metro area is home to the largest concentration of federal workers, over 90% of the federal workforce live and work outside the nation’s capital. Black workers account for at least one-fifth of the state’s federal workforce in 12 states beyond the D.C. metro area.

Implications of massive federal job losses and the unfinished business of equity

To understand the stakes of federal workforce contraction, it is necessary to compare the demographic and wage structure of federal employment with that of the broader labor market. As shown in Table 1, in 2023 and 2024, Black workers were 12.5% of the private-sector workforce, compared with more than a fifth (22.6%) of all workers in the federal sector—a share that also exceeds their representation in the entire public sector (16.4%) which includes state and local governments. Black women’s share of the federal workforce (12.8%) was double their share in the private sector (6.4%).

A national comparison of hourly wages at the median and for low-wage (10th percentile) workers demonstrates the clear monetary benefit of federal over private-sector employment. Figure A shows this is true across race and gender both at the middle and lower end of the wage distribution. The hourly wage of a typical (i.e., median) Black federal worker is more than 40% higher than that of the median Black worker in the private sector. Black federal workers—median and 10th percentile—also have higher wages than same gender white workers in the private sector. It is worth noting that these wage comparisons don’t account for the more generous benefits typically offered to federal and other public-sector workers, which further raises the value of their total compensation. The higher wages earned by federal workers largely reflect the higher share of college and advanced degree holders and higher rates of union coverage relative to private-sector employees. Less than 7% of private-sector workers are in a union or covered by a union contract compared with 35.9% of all public-sector workers and 29.5% of federal workers (see Table 1). While greater union coverage helps to boost wages and benefits for all workers, it is an even more important factor in raising wages of those for whom racial and gender discrimination further restrict individual bargaining power.

Another factor contributing to better pay outcomes in the federal government is the use of the Schedule (GS) pay scale which applies to over 70% of white-collar federal jobs. This helps to mitigate pay discrimination in the federal government by standardizing the qualifications and compensation associated with a specific position and consistent with experience, job performance, and local cost of living. On average, Black federal workers appear to experience only marginally improved pay equity over Black workers in the private sector, while the Black-white wage gap is much smaller in the public sector, overall.

In the federal sector, Black workers earn 12.6% less than white workers with the same levels of education, experience, union coverage status, gender, and state of residence, compared with 14.9% less in the private sector and just 3.8% less in the overall public sector (see Table 2). Although there is a sizable wage gap between Black women and white men across sectors, the federal sector gap (26.1%) is nearly 8 percentage points lower than the gap that exists in the private sector (33.9%). Given enforcement of the GS pay scale, remaining racial and gender pay gaps among federal workers likely reflect disparities in job positions and associated GS levels, a long-documented concern of Black federal worker advocates and activists. These disparities may stem from the underrepresentation of Black workers in higher-level, higher-paying positions, which can reflect differences across agencies in workforce demographic composition, occupational structures, and promotion rates. Notwithstanding the relatively higher economic position of many Black federal workers, these results epitomize the unfinished business of eliminating pay inequity and occupational segregation across all sectors of the labor market.

Conclusion

This brief summarizes the important role federal-sector employment has played in providing better job opportunities for Black Americans than have traditionally been available in the private sector. However, those outcomes have never been a given. A solid history of advocacy and activism by and on behalf of Black federal workers alongside others were critical in securing important wins through executive actions and policy change. Moreover, pushback against some of the most egregious violations of federal worker’s civil and worker rights have at times resulted in stronger, more broadly enforced labor and equal employment standards, improving outcomes to the benefit of all workers.

References

American Community Survey (ACS). 1-Year 2023 microdata, Accessed via IPUMS USA, University of Minnesota, www.ipums.org.

American Community Survey (ACS). 2019–2023 pooled microdata, Accessed via IPUMS USA, University of Minnesota, www.ipums.org.

Center for Migration Studies. 2026. Estimates of Undocumented and Eligible-to-Naturalize Populations by State. Retrieved July 13, 2026.

Davis, Carl, Marco Guzman, and Emma Sifre. 2024. Tax Payments by Undocumented Immigrants, Institute on Taxation and Economic Policy, July 2024.

Economic Policy Institute (EPI). 2024. Current Population Survey Extracts, Version 1.0.60, https://microdata.epi.org.

Escobari, Marcela, Ian Seyal, and Paul Beach. 2026. Shock, Awe, and Economic Fallout: The Employment Effects of ICE Enforcement in U.S. Cities. The Brookings Institution, May 2026.

Peri, Giovanni, and Reem Zaiour. 2021. Citizenship for Undocumented Immigrants Would Boost U.S. Economic Growth, Center for American Progress, June 2021.

U.S. Bureau of Labor Statistics, Consumer Price Index for All Urban Consumers (CPI-U), Accessed via Zipperer, B (2025). realtalk: Price Index Data for the US Economy. R Package Version 0.19.0, https://economic.github.io/realtalk/.

U.S. Citizenship and Immigration Services (USCIS). 2024. Count of Active DACA Recipients by State or Territory, as of September 30, 2024.

Wilson, Jill H. 2025. Temporary Protected Status and Deferred Enforced Departure. Congressional Research Service Report RS20844, August 2025.

Wong, Tom K., Ignacia Rodriguez Kmec, Diana Pliego, Karen Fierro Ruiz, Silva Mathema, Ben Greenho, and Rosa Barrientos-Ferrer. 10th Annual DACA Survey: 2024 Findings Reveal What’s at Stake for Recipients and the United States. Center for American Progress, August 2025.

Zipperer, Ben. 2025. Trump’s Deportation Agenda Will Destroy Millions of Jobs. Economic Policy Institute, July 2025.

Last updated September 2026

At a time when many are struggling to afford basic necessities, it’s easy to understand why. Through unions and collective bargaining, workers have more power to win higher wages, better benefits, safer working conditions, and a fairer share of the wealth they create. Unions help build a strong middle class, reduce inequality, narrow racial economic disparities, and boost participation in our democracy.

Yet only 1 in 10 U.S. workers are in a union today—a sharp decline from the more than 1 in 3 workers who belonged to a union in the 1950s. That drop did not happen because workers stopped wanting or needing unions. It happened due to relentless attacks on unions and collective bargaining, and lawmakers’ failure to fix the broken labor laws that have allowed those attacks to succeed.

The consequences have been enormous. As union power has declined, workers have seen less of the gains from the economic growth they have helped create. Since 1979, productivity (how much average value workers produce in an hour of work) has grown 2.8 times as much as pay for typical workers.

New EPI research makes clear just how much working people stand to gain by rebuilding union power to 1950s’ levels.

Tripling union membership would raise the pay of the median worker by more than $7,700 a year, or nearly $270,000 over a 35-year career. That’s enough to more than cover the cost of sending two children to a four-year public university, for example. Crucially, both union and nonunion workers would see these gains because stronger unions raise standards across the labor market.

Scaled across the workforce, tripling union membership would shift an estimated $1.2 trillion to the pockets of working people every year. This is enough to reverse roughly one-third of the increase in inequality since 1979.

Stronger unions are also good for businesses and the broader economy. When workers earn more, they have more money to spend in their communities, strengthening consumer demand. Businesses would benefit from lower worker turnover, higher productivity, and workers who have a greater stake in the success of their workplaces.

Tripling union membership won’t be easy, but it’s far from a nostalgic pipe dream. It will take continued nationwide organizing and decisive policy action that makes it easier for workers to unionize. Federal lawmakers can start by passing legislation that expands collective bargaining rights, closes loopholes in existing law that allows employers to suppress worker organizing, and holds employers accountable when they violate workers’ labor rights. Further, state lawmakers should provide all public-sector workers with collective bargaining rights and repeal so-called right-to-work laws that weaken workers’ ability to organize and bargain collectively. 

This Labor Day, let’s recommit to putting more power in the hands of working people. That means giving more workers the freedom to organize. And it means setting an ambitious goal worthy of the moment: tripling union membership and building an economy that works for working people.

Key takeaways:

• So-called right-to-work (RTW) laws undermine workers’ ability to form unions. Unionization rates are clearly lower in RTW states—and since unions raise wages, that suppression lowers pay.   

• States with RTW laws are also more likely to have other anti-worker laws and generally weaker labor standards. This is likely because both RTW and other anti-worker policies stem from the same anti-worker political roots, and because RTW’s suppression of unions deprives these states of strong potential political champions (unions) for other pro-worker policies.  

• Non-RTW states are more likely to have pro-worker policies like higher minimum wages and prevailing wage laws, restrictions on noncompete agreements, pay transparency policies, and more protective unemployment insurance benefits. 

• As a result, workers in RTW states are paid 6.7% less on average than workers in non-RTW states—more than double the 3.2% wage penalty we found in 2015. That means the RTW wage penalty has increased in recent years and now translates to over $4,000 less per year for a median full-time worker in a RTW state.

• The RTW wage penalty is steeper for women (7.3%) than for men (6.3%), and for Black (9.7%) and Hispanic workers (10.0%) than for white workers (5.4%).   

• Lawmakers already have the policy tools they need to boost wages and improve other outcomes for workers: 

• • At the state level, rolling back existing RTW laws is the most powerful lever.  

• • At the federal level, the Protecting the Right to Organize Act (which includes a ban on RTW) would be transformational for efforts to organize workers.  

• • Policymakers could also make progress on numerous other fronts, such as making UI systems more protective and raising minimum wages. 

In the United States, so-called right-to-work (RTW) laws are anti-union state policies promoted by big business interests to weaken unions and limit workers’ bargaining power. Despite the name, RTW laws do not provide any sort of job protection or right to a job. Rather, the deceptively named policy starves unions of resources, making it harder for them to negotiate collectively for better wages, benefits, and working conditions.

RTW laws emerged in the 1940s as part of anti-union campaigns to suppress worker organizing and maintain Jim Crow labor relations in Southern states; these campaigns worsened economic inequality and racial disparities (Childers 2024). Today, 26 states have active “right-to-work” statutes.{{1}}

As intended, RTW laws reduce union density (the share of workers who are either members of a union or covered by a union contract). These policies prohibit unions and employers from negotiating over union security—the contract terms under which workers covered by a union contract either join the union or pay an agency fee covering their share of representation costs. This deprives unions of critical resources and suppresses union membership (Sherer et al. 2026). RTW states—specifically those with RTW laws for at least 25 years—have an average union density rate of 6.2%, far less than half the union density rate in non-RTW states (16.0%) (see Appendix Table 1).

RTW laws not only reduce the number of workers with collective bargaining agreements but also limit unions’ political power to advocate for stronger labor standards for workers more broadly. This is, in part, why states without anti-union restrictions are more likely to pass protections such as prevailing wage mandates, restrictions on noncompete agreements, higher minimum wages, and pay transparency laws.

To assess the effects of RTW laws on outcomes for workers, we analyze whether wages are lower in RTW states after controlling for key demographic-, job-, and state-level characteristics (see Appendix Table 1 for full demographic controls as well as median and average wages by RTW status). We find that wages in “right-to-work” states are, on average, 6.7% lower than in non-RTW states. This penalty is worse for Black and Hispanic workers, as well as young workers and those without college degrees. Notably, the penalty is also larger for nonunion workers. This is because strong unions benefit workers across the state, not just those in unionized workplaces (Shierholz et al. 2026). In RTW states with weaker unions, nonunionized workers do not experience these positive “spillover” effects.

Pro-worker policies are more common in states without RTW laws

The strong correlation between wages and states’ RTW status is not surprising. Not only do RTW states have lower unionization rates, leaving workers with less leverage to bid up wages, but they often maintain other anti-worker policies. Combined, these anti-worker policies tilt bargaining power toward employers, weakening workers’ ability to garner higher wages. States without RTW laws are more likely to have a wide array of stronger labor standards in place that raise pay, protect workers’ ability to switch jobs, and support those who lose one. These same states—where unionization rates have not been suppressed by RTW laws—have a better safety net and stronger labor standards in part because stronger unions give workers more political power and the ability to better lobby for pro-worker policies writ large (Shierholz et al. 2026).

Figure A displays the incidence of these labor standards for states with and without anti-union RTW laws.{{2}} This is far from a complete list of relevant policies that may correlate with RTW status and wage levels, but they are illustrative.

Figure A

Prevailing wage

Prevailing wage laws apply to certain contracts for public works and require that contractors and subcontractors pay the employees performing such work a prevailing wage rate. The prevailing wage rate is defined as the average wage paid to similarly employed workers in a specific occupation in the area of intended employment. Federal law requires this of federal contracts while state and local laws may apply to state and local contracts.

Prevailing wage laws are more common in non-RTW states. Figure A shows that nearly all (96.0%) of non-RTW states have prevailing wage laws, while only 23.1% of RTW states have these protections for workers on state contracts.{{3}} Without prevailing wage requirements, contractors may reduce their workers’ wages to win bids on government contracts—putting contractors who pay their workers fair wages at a disadvantage.

Restrictions on noncompetes

Noncompete agreements are employment provisions that ban workers at a company from joining a competing business or starting their own for a set period after leaving a job. Noncompetes undermine economic dynamism and innovation; they depress business formation and labor mobility, hurt productivity and growth, raise prices, shrink workers’ wages, and restrict workers’ freedom (Shierholz 2024).

One of the most effective ways workers achieve higher wages is by securing another offer, and then either renegotiating their current salary or leaving for the new higher-paying job. It is not surprising then that workers in states that enforce noncompetes are paid less than similar workers in states that do not enforce noncompetes (Starr 2019). States without anti-worker RTW laws are nearly four times as likely to restrict noncompetes as RTW states (44.0% vs. 11.5%).{{4}} In this analysis, we include only full bans on noncompetes and states with income-based restrictions. States with only small carve-outs for certain occupations provide a much weaker standard and are, by definition, less broad, so those aren’t included.

Minimum wage

Raising the minimum wage is particularly important for lower-wage workers. While the federal minimum wage has sat at $7.25 an hour since 2009, losing over 30% of its purchasing power, workers in 31 states enjoy a higher wage floor (Zipperer 2026). Non-RTW states are three times as likely to have minimum wages higher than the federal minimum compared with RTW states (92.0% vs. 30.8%). This is not a small difference. As of 2026, the average value of the minimum wage in non-RTW states is 60% higher—$5.36 more per hour—than in RTW states.{{5}}

Pay transparency

Pay transparency laws prevent employers from lowballing wage offers by requiring them to include wage information in job postings. While they vary, all laws include some requirement that employers provide salary information in job postings or if directly requested by applicants. Lack of knowledge about wages and benefits keeps potential workers in the dark and limits their ability to learn about and apply for better paid opportunities. A study on Colorado’s newly passed wage transparency law found that wages increased 4.2% faster than those in neighboring states without such laws (Shedge 2025). Further, wage transparency has the potential to reduce gender- and race-based discrimination by arming jobseekers with more information and limiting employers’ ability to pay different amounts to similarly qualified candidates. Only two RTW states have a pay transparency law, compared with 64% of non-RTW states.{{6}}

Unemployment insurance

Access to unemployment insurance (UI) also shapes workers’ ability to seek better wages. When a worker is laid off and their household income falls, unemployment insurance supports them and their family until they find another job. UI is a joint federal-state program that relies on state UI systems to effectively deliver benefits to unemployed workers. States differ in their rules and approaches under the federal UI framework, leading to wide variation in the share of unemployed workers receiving UI benefits (the UI recipiency rate).

Better UI not only cushions workers and their families but also provides workers with additional leverage in the labor market. When UI access and generosity fall, job seekers are forced to settle for lower wages and employers leverage this by offering less (Dahl and Knepper 2026). On the flip side, research shows a positive relationship between more UI generosity and wages (Rinz and Wasser 2026). The UI recipiency rate is significantly higher in non-RTW states compared with RTW states (32.8% vs. 18.7%).{{7}} This is not surprising because non-RTW states have stronger unions, which fight to increase UI eligibility, ease of access, benefit levels, and benefit duration—to the benefit of all workers, union and nonunion (Hertel-Fernandez and Gould-Werth 2020). As with other pro-worker policies, better UI access can contribute to higher wages in non-RTW states.

Other policy levers

Not only do workers in non-RTW states enjoy higher wages, but they also benefit from increased economic security thanks to higher health insurance rates, higher public education spending, paid family and medical leave access, and no preemptions against higher labor standards. Every non-RTW state expanded Medicaid compared with fewer than two-thirds of RTW states (61.5%).{{8}} Unsurprisingly, the uninsured rate is higher in RTW states than in non-RTW states (8.9% vs. 6.2%).{{9}}

Investments in public education deliver a more productive workforce and more informed and engaged society. Increased funding for schools not only improves educational attainment, but also increases wages and family incomes (Jackson, Johnson, and Persico 2016). Non-RTW states spend 45% more on public education than RTW states ($24,084 vs. $16,620).{{10}} Paid family and medical leave provides essential benefits that help workers maintain their livelihoods while taking care of themselves and their families. These laws are also shown to increase labor force participation, job retention, and earnings (Glynn 2020). Only one RTW state has comprehensive paid family and medical leave provisions for private-sector workers, compared with 56% of non-RTW states.{{11}}

Finally, some cities, counties, and other local governments enact policies that raise standards for working people, but some state legislatures force those standards back down with preemption—the use of state law to void local ordinances. Preemptions apply to policies such as minimum wage, prevailing wages, paid leave, and fair scheduling (EPI 2025). Preemption laws interfere with local governments’ ability to set job quality standards. They suppress wages, exacerbate racial inequities, and reduce worker power (Sherer, Cohn, and Ahdoot 2025). These preemptions are more common in RTW states.

Workers in RTW states are paid less

Our analysis of the relationship between wages and RTW status puts states into three categories: long-term RTW states, always non-RTW states, and “switcher” states that have changed their RTW status since 2011.{{12}} Appendix Table 1 provides descriptive statistics for these three groups of states, and Appendix Figure A provides a map of states by RTW status. Our estimates focus on comparisons between long-term RTW states and always non-RTW states.

A simple comparison of hourly wages, the primary variable of interest, reveals that average wages are 22.9% higher in non-RTW states ($40.28 in non-RTW states vs. $32.78 in RTW states). Median wages are 19.2% higher in non-RTW states ($28.79 vs. $24.16). Because there are differences between worker-, job-, and state-level characteristics in RTW and non-RTW states, and since some of these characteristics will directly impact workers’ wages, it is important to control for these factors in a multivariate regression model. This allows us to more accurately identify the relationship between RTW status and wages.

We estimate log wage equations using Bureau of Labor Statistics Current Population Survey Outgoing Rotation Group (CPS-ORG) data for 2023–2025. We pool three years of data to minimize any spurious year-specific economic relationships, thereby helping us achieve more precise estimates. The total sample consists of 234,155 workers, ages 16+, who earn wages and salaries. About 42% of the sample lives in states with RTW laws (see Appendix Table 1 for the sample’s full demographic breakdown).{{13}}

In Table 1, we construct a regression model, starting with an unadjusted framework and building up to a model that controls for the full range of explanatory variables. The dependent variable is the natural log of hourly wages.

The variable of interest is an indicator variable when the worker lives in a long-term RTW state. We also include an indicator in the regression for a switcher state, though that is not our focus of interest (see Appendix Table 2 for a complete set of regression results, including the coefficient for switcher states). There are currently 26 states with RTW laws on the books and 25 without them including Washington, D.C. (see Appendix Figure A). There are five switcher states that have adopted RTW in recent years: Indiana (2012), Michigan (2013), Wisconsin (2015), West Virginia (2016), and Kentucky (2017). In 2023, Michigan became the first state to repeal its RTW statute. We separate the five switcher states from the analysis to better isolate the relationship between wages and the long-run effects of RTW.

The naïve model, without any controls, suggests that workers in RTW states are paid 17.8% less than workers in non-RTW states (Model I in Table 1).{{14}} But many factors contribute to wage differences: demographic characteristics such as age, race/ethnicity, gender, and level of education are relevant as well as job-related factors such as work hours, occupation, and industry. Since these vary across states, they reduce the measured relationship between RTW status and wages. Model II shows that when the analysis compares more similar workers with each other, the RTW penalty drops to 14.0%.{{15}}

There are other factors that are different in RTW states, aside from these individual and job-related characteristics. Most importantly, the cost of living. The cost of living in non-RTW states is higher than in RTW states, as shown in Appendix Table 1. In prior research, Gould and Kimball (2015) showed that choice of price indicator makes no material difference on the measured relationship between RTW and wages. Here we include the Bureau of Economic Analysis’ Regional Price Parity index: The regional price parity index is higher in non-RTW states.{{16}} Not surprisingly, this correlates with wage levels and therefore is important to include in the model. As with earlier studies, we also include state unemployment rates to control for differences in economic conditions.

After including the full set of demographic, individual-, and state-level labor market and cost of living controls (Model III), our analysis finds that workers in RTW states are paid 6.7% less than similar workers in non-RTW states. For the average worker, this is a difference of $2.70 per hour, or more than $5,600 for full-time workers at 2,080 hours over the year. However, because the average is skewed upward due to wage inequality, an assessment at the middle of the wage distribution provides a more conservative but perhaps more robust estimate of the RTW wage penalty. Acknowledging that this is a likely understatement of the full relationship, the RTW wage penalty for the median worker at full-time hours is over $4,000 for the year.

Table 1

This does not mean that RTW laws are a causal explanation for this 6.7% wage penalty. As demonstrated earlier in this report, RTW status correlates strongly with a number of other anti-worker policies. Other research has attempted to calculate the causal effects of RTW laws on wages. Using the switcher states in the most recent period, Fortin, Lemieux, and Lloyd (2022) found smaller wage penalties, which is to be expected given the shorter time frame for measurement post-adoption. Our wage penalty for switcher states—shown in the final column of Appendix Table 2—is much smaller (closer to zero) than the coefficient for long-term RTW states, though not statistically significant at conventional levels. Dasgupta and Merchant (2023) also examine states recently adopting RTW laws and find a statistically significant decline in annual wages by almost $1,900. Again, these states are more recent RTW adopters, so the full effects we find may take time to materialize.

We find that this relationship between long-term RTW status and wages remains economically and statistically significant under alternative specifications of our econometric model. For instance, removing major industries and occupations changes the coefficient to -0.071. Restricting the model to RTW status as of 2026—removing the separate indicator for states that switched since 2011—yields a coefficient estimate of -0.060. Doing the same but dropping Michigan (since its status changed in the measurement period) results in a coefficient of -0.064. In our final model, we do not include an indicator for unionization, unlike Gould and Kimball (2015). If we reintroduce that variable, the coefficient is -0.063.

In each of our regressions discussed above, we include a three-year average as it improves data reliability and allows us to run smaller groups of the data (see next section). When we run our analysis using just 2025, our results do not change. In fact, when we run our three-year final model for the periods 2021–2023, 2022–2024, and 2023–2025, our coefficients remain in the -0.070 to -0.068 range.

Wage penalties are steeper for women, Black, Hispanic, and nonunion workers in RTW states

The RTW penalty is not uniform across demographic groups and other labor market characteristics. Table 2 below shows the results of a series of final-model regressions (comparable to Model III in Table 1) for a set of demographic groups.

We find that women’s wages in RTW states are penalized at a higher rate (7.3%) than men’s (6.3%). The wage penalty also persists across all racial and ethnic groups, but it is almost twice as large for Black (9.7%) and Hispanic (10.0%) workers as for similar white workers (5.4%) and more than twice as large as for Asian American and Pacific Islander (AAPI) workers (4.6%). Young workers between the ages of 16 and 24 face a greater penalty (8.4%) than older workers. And workers without a college degree experience a much higher penalty (7.8%) than workers with a college degree (4.7%).

Table 2

It is particularly important to note that RTW laws have a statistically significant negative effect on the wages of nonunion workers, as shown in the last row of Table 2. Our analysis indicates that nonunion workers in RTW states have wages that are 6.3% lower, on average, than their counterparts in non-RTW states. Union members likely experience a smaller penalty (3.6%) because their wages tend to be higher than those in nonunionized workplaces, regardless of the state’s RTW status (McNicholas et al. 2025). However, the wage impact of unions extends beyond those directly covered by a union contract. Unions lift wage standards across entire industries, meaning that nonunion workers in states with higher union density are more likely to receive better pay.

As the gap in wages and labor standards between RTW and non-RTW states has grown, so has the wage penalty

Using our final model (Model III in Table 1) with pooled 2023–2025 data, we find a 6.7% RTW wage penalty—a larger penalty than we found in earlier examinations. For example, we found a 3.2% wage penalty using data for 2009 (Gould and Shierholz 2011) and pooled 2010–2012 data (Gould and Kimball 2015). Some of the increase in the current estimate of the RTW wage penalty reflects small changes (likely improvements) in our methods. But part of the higher estimate persists even when we apply our current methods to earlier data: Applying these methods to the 2010–2012 data yields an RTW wage penalty of 4.7%.

One method change concerns “switchers”—states that changed from non-RTW to RTW status (or back) since 2019. In the 2010–2012 period, there were no switchers to affect the data.{{17}} In our analysis, we isolate switchers from the long-term RTW group, but the RTW penalty remains negative and significant no matter how one deals with the issue of switcher states. For example, recategorizing the five switchers to match their 2012 status (all non-RTW) lowers the measured RTW wage penalty in the 2010–2012 period to 4.1%, but it remains significant.

Other changes relative to our earlier research include controlling for public-sector status; removing union as an independent variable from the regression; using the full working age range of 16 and up (as opposed to 18–64); and correcting an issue with nonrandom missing values to the metropolitan variable. We’ve further clustered standard errors at the state level as suggested by Douglas (2024), which does not affect the size of the coefficients, only the size of their standard errors and resulting statistical significance.{{18}}

Setting those minor model differences aside, a consistent comparison over time sees the RTW wage penalty rising from 4.7% in 2010–2012 to 6.7% in 2023–2025. One possible reason why this RTW wage penalty grew could be the increasingly divergent policy trajectories of RTW and non-RTW states over this period.

For example, the gap between minimum wages in RTW states and non-RTW states has grown since 2010, largely because policymakers in most RTW states have refused to increase their minimum wage despite the fact that the federal minimum wage has been stagnant in nominal terms since 2009. In that year, the average minimum wage in RTW states was $7.25 while the average minimum wage in non-RTW states was $7.54. That gap has grown enormously, from just $0.29 per hour in 2009 to $5.36 per hour by 2026 ($14.24 in non-RTW states vs. $8.89 in RTW states).

There were other potentially relevant policy changes between 2012 and 2025. The first state-level pay transparency law was passed in Maryland in 2020. Seventeen states followed suit in subsequent years, 15 of which were non-RTW states. On the flip side, six states have repealed their prevailing wage laws since 2012—all of which were RTW. Only one state, Michigan, reinstated its prevailing wage law. It did so in 2023, the same year it repealed RTW. While this is far from an exhaustive list, it’s not surprising that the measured RTW wage penalty has grown over time.

Repealing anti-union “right-to-work” laws and strengthening labor law is key to raising wages

The wage penalty for workers in RTW states has grown since the last time we conducted this research a decade ago. Long-standing and more recently enacted anti-union RTW laws in 26 states, in tandem with other anti-worker policies, have further eroded workers’ collective bargaining power, suppressed wages, and lowered labor standards. These outcomes are linked to clear policy choices, and lawmakers at every level of government have the tools they need to restore workers’ union rights, strengthen labor standards, and lift wages. Across the U.S., 56 million workers say they want a union in their workplace, yet only 10% of all workers have one (Shierholz et al. 2026). Current federal and state labor laws—including anti-union RTW laws—place too many obstacles in workers’ path to unionizing and give employers too much power to interfere with workers’ free choice.

First and foremost, states with RTW laws should follow Michigan’s recent lead and repeal them. Repealing state RTW laws will enable workers to unionize and raise wages across the country. Indeed, recent EPI research finds that equalizing collective bargaining rights across all states by repealing RTW laws that affect private employees and ensuring collective bargaining rights for public employees would increase national union density by almost 50% (from 9.9 to 14.4%) (Shierholz et al. 2026). States should also consider passing constitutional amendments that affirm collective bargaining rights and explicitly bar RTW-style restrictions, such as the 2022 Workers’ Rights Amendment in Illinois (Sherer 2026).

At the federal level, lawmakers should pass the Protecting the Right to Organize (PRO) Act. The PRO Act is designed to address major weaknesses of the National Labor Relations Act (NLRA), which has been severely eroded by amendments and court decisions. Among many other reforms, the PRO Act would eliminate the option for states to maintain anti-union RTW laws—restoring full bargaining rights to workers in all states as intended when the NLRA was originally passed in 1935 (McNicholas, Poydock, and Rhinehart 2021). Since its first introduction in 2019, the PRO Act has passed the House of Representatives twice with bipartisan support (Shierholz et al. 2026).

In addition to proposed PRO Act reforms that would enable newly unionized workers to pursue arbitration to achieve a first contract (if an employer fails to negotiate in good faith), federal labor law should set a minimum standard that such contracts include a cost-of-living adjustment (COLA). To further expand the benefits of collective bargaining, Congress should amend the NLRA to require default collective bargaining at any firm where CEO-to-worker pay ratios exceed 100 times what a typical worker makes in their industry (as this ratio would be one proxy for whether conditions at an employer are particularly unequal) (Shierholz et al. 2026). 

Legislators at both the state and federal level should also look to the list of anti-worker laws that are typically found in RTW states for more opportunities to rebalance unequal power in the labor market and improve wages and working conditions. These policies include:

• Raising the minimum wage;
• Banning noncompete agreements;
• Strengthening federal unemployment insurance guidelines and state UI systems;
• Removing state restrictions that preempt local governments from raising wages and strengthening labor standards.

These are among just a few of the policy levers states can use to raise wages and standards (EPI 2026b).

Our analysis suggests that repealing anti-union RTW laws and enabling workers to increase unionization levels are key to raising wages, both because RTW laws directly suppress unionization rates (and unions raise wages), and because weakening unions limits workers’ collective ability to shape other public policies.

Acknowledgements

The authors are grateful to Josh Bivens, Jennifer Sherer, Hilary Wething, and Ben Zipperer for their helpful comments and expertise.

Appendix

Appendix Figure A

Appendix Table 1

Appendix Table 2

Notes

{{1.}} See Appendix Figure A for RTW status. As of 2026, these states are Alabama, Arizona, Arkansas, Florida, Georgia, Iowa, Idaho, Indiana, Kansas, Kentucky, Louisiana, Mississippi, North Carolina, North Dakota, Nebraska, Nevada, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, Wisconsin, West Virginia, and Wyoming.

{{2.}} States are categorized as “RTW” or “non-RTW” based on their status as of 2026.

{{3.}} Authors’ analysis of prevailing wage laws from DOL (2023) and state legislative websites.

{{4.}} Authors’ analysis of noncompete restriction policies from EIG (2026).

{{5.}} Authors’ analysis of minimum wage data from EPI (2026c).

{{6.}} Authors’ analysis of pay transparency laws from GovDocs (2026).

{{7.}} Authors’ analysis of UI recipiency rate data from DOL-ETA.

{{8.}} Authors’ analysis of Medicaid expansion policies from KFF (2026).

{{9.}} Authors’ analysis of uninsurance rates by state, 2023 through 2025, from Carter (2025).

{{10.}} Authors’ analysis of per-pupil education spending data from NCES-NPEFS (2024).

{{11.}} Authors’ analysis of paid family and medical leave policies from Wielk (2026).

{{12.}} Prior to 2012, the last state to change its status was Oklahoma in 2001.

{{13.}} This analysis builds on earlier work by Gould and Kimball (2015); see the report for a full description of the methodology and list of controls.

{{14.}} Interpreting the results from these semilogarithmic functions requires utilizing the exponential function on the coefficient. Specific to the binary variable coefficient (β1) for RTW, the percent change in workers’ wages resulting from a state being RTW can be calculated by the formula: 100*[exp(β1)-1]. Typically, the result of this equation will be very close to the coefficient itself but will differ more as the coefficient becomes larger. We show the coefficient in the first row of Table 1 and the exponentiated result as a percent in the third row.

{{15.}} Full regression results for covariates in all three models can be found in Appendix Table 2.

{{16.}} Because the 2025 BEA RPP data were not yet available at the time of publication, we assign an average of 2022–2024 data to all years in our sample (2023–2025).

{{17.}} Gould and Kimball (2015) classify Indiana as non-RTW as its RTW law had just taken effect.

{{18.}} We do not control for state-level fixed effects, as Douglas (2024) suggested, because state-level fixed effects perfectly correlate with RTW status and would deem our results uninterpretable.

References

Bureau of Economic Analysis, Regional Price Parities by State and Metro Area: All items (BEA-RPP). Various years [Release date: February 19, 2026]. Public data series accessed via the Regional Data interactive tool. Accessed August 1, 2026. 

Carter, Caitlin. 2025. “Health Insurance Coverage by State: 2023 and 2024.” U.S. Census Bureau, September 11, 2025. 

Childers, Chandra. 2024. The Evolution of the Southern Economic Development Strategy: Rooted in Racism and Economic Exploitation: Part One. Economic Policy Institute, May 2024. 

Dahl, B. Gordon, and Matthew Knepper. 2026. “Unemployment Insurance, Starting Salaries, and Jobs: Evidence from Multi-state Firms.” The Review of Economic Studies, rdag056. https://doi.org/10.1093/restud/rdag056.

Dasgupta, Kabir, and Zofsha Merchant. 2023. “Understanding Workers’ Financial Wellbeing in States with Right-to-Work Laws.” FEDS Notes (The Federal Reserve), September 8, 2023.

Department of Labor (DOL). 2023. “Dollar Threshold Amount for Contract Coverage Under State Prevailing Wage Laws.” Last modified January 1, 2023. 

Department of Labor, Employment and Training Administration (DOL-ETA). Various years. Unemployment Insurance Chartbook. Public data series accessed via the interactive tool. Accessed August 6, 2026. 

Douglas, Christopher C. 2024. Right-to-Work States Do Not Have Lower Wages. Mackinac Center, December 2024. 

Economic Innovation Group (EIG). 2026. “State Noncompete Law Tracker.” Last modified March 24, 2026. 

Economic Policy Institute (EPI). 2025. “Workers’ Rights Preemption in the U.S.” Last modified February 2025. 

Economic Policy Institute (EPI). 2026a. Current Population Survey Extracts, Version 2026.7.8, https://microdata.epi.org.

Economic Policy Institute (EPI). 2026b. Holding the Line: State solutions to the U.S. Worker Rights Crisis.

Economic Policy Institute (EPI). 2026c. “Minimum Wage Tracker.” Last Modified July 1, 2026. 

Fortin, Nicole, Thomas Lemieux, and Neil Lloyd. 2022. “Right-to-Work Laws, Unionization, and Wage Setting.” National Bureau of Economic Research Working Paper no. 30098, June 2022. http://www.nber.org/papers/w30098. 

Glynn, Sarah Jane. 2020. “The Rising Cost of Inaction on Work-Family Policies.” Center for American Progress, January 21, 2020.

Gould, Elise, and Will Kimball. 2015. “Right-to-Work” States Still Have Lower Wages. Economic Policy Institute, April 2015.

Gould, Elise, and Heidi Shierholz. 2011. The Compensation Penalty of “Right-to-Work” Laws. Economic Policy Institute, February 2011. 

GovDocs. 2026. “Pay Transparency Laws by State and Province.” Last modified July 2026.

Hertel-Fernandez, Alexander, and Alix Gould-Werth. 2020. Labor Organizations and Unemployment Insurance: A Virtuous Circle Supporting U.S. Workers’ Voices and Reducing Disparities in Benefits. Washington Center for Equitable Growth, October 2020.

Jackson, C. Kirabo, Rucker C. Johnson, and Claudia Persico. 2016. “The Effects of School Spending on Educational and Economic Outcomes: Evidence from School Finance Reforms.” The Quarterly Journal of Economics 131, no. 1: 157–218. https://doi.org/10.1093/qje/qjv036.

KFF. 2026. “Status of State Medicaid Expansion Decisions.” Last modified May 21, 2026. 

McNicholas, Celine, Margaret Poydock, and Lynn Rhinehart. 2021. “How the PRO Act restores workers’ right to unionize” (fact sheet). Economic Policy Institute, February 4, 2021. 

McNicholas, Celine, Margaret Poydock, Heidi Shierholz, and Hilary Wething. 2025. Unions Aren’t Just Good for Workers—They Also Benefit Communities and Democracy. Economic Policy Institute, August 2025.

National Center for Education Statistics, National Public Education Financial Survey (NCES-NPEFS). 2024. Public data series accessed via the Common Core of Data database. Accessed July 1, 2026. 

Rinz, Kevin, and David Wasser. 2026. “Unemployment Insurance Generosity and Wage Determination.” Federal Reserve Bank of Cleveland Working Paper no. 26-11, May 2026. https://doi.org/10.26509/frbc-wp-202611.

Shedge, Nirav. 2025. Pay Transparency Laws and Wage Dynamics: Evidence from Colorado. Social Science Research Network, November 2025. 

Sherer, Jennifer. 2026. Rights to Unionize and Collectively Bargain: State Solutions to the U.S. Workers Rights Crisis. Economic Policy Institute, February 2026.

Sherer, Jennifer, Emma Cohn, and Ruby Ahdoot. 2025. “Updated EPI Tracker Shows More States Obstructing Progress on Workers’ Rights: Harmful Preemption Laws are Increasing Inequality and Repressing Democracy.” Working Economics Blog (Economic Policy Institute), March 6, 2025.

Sherer, Jennifer, Nina Mast, Elise Gould, and Emma Cohn. 2026. Everything You Need to Know About Anti-Union “Right-to-Work” Laws. Economic Policy Institute, forthcoming 2026.

Shierholz, Heidi. 2024. “Banning Noncompete Agreements.” Testimony before the U.S. Senate Banking, Housing, and Urban Affairs Subcommittee on Economic Policy, July 30, 2024. 

Shierholz, Heidi, Celine McNicholas, Josh Bivens, Jennifer Sherer, Ben Zipperer, and Margaret Poydock. 2026. The Case for Tripling Union Membership: How Rebuilding Union Power Would Strengthen Workers, the Economy, and Our Democracy. Economic Policy Institute, July 2026. 

Starr, Evan. 2019. “Consider This: Training, Wages and the Enforceability of Covenants Not to Compete.” ILR Review 72, no. 4 (August): 783–817. https://doi.org/10.1177/0019793919826060.

Wielk, Emily. 2026. “State Paid Family Leave Laws Across the U.S.” Bipartisan Policy Center. Last updated April 23, 2026. 

Zipperer, Ben. 2026. Setting High Standards for a Federal Minimum Wage. Economic Policy Institute, May 2026.