Today, 40 organizations led by the Economic Policy Institute, We Build Progress, the AFL-CIO Tech Institute, and Workshop delivered the letter below urging Congress to center workers in federal AI legislation. 

Dear Member of Congress: 

Employers’ increasing use of AI systems has the potential to affect the lives and livelihoods of workers across the country. Without appropriate guardrails, employers’ integration of these technologies may jeopardize workers’ rights, put workers at risk of discrimination, violate privacy rights, and dramatically impact the economic stability of working families.

These risks posed by technological change are not new. For years, employers have used algorithmic or automated systems and similar technologies in ways that harm workers. Now, the pervasive and growing integration of AI into the workplace is amplifying these risks. These impacts on workers are further exacerbated by persistent power imbalances in the labor market that favor employers. 

It is urgent that Congress take action. It has been nearly two years since the Bipartisan Senate AI Working Group released its roadmap for AI policy, but the Senate has yet to consider comprehensive legislation. AI adoption is moving forward at breakneck speed, and America’s workers cannot afford to wait. 

We applaud members of Congress who have introduced worker-focused legislation addressing issues like civil rights, surveillance in the workplace, and improvements to labor market data. Efforts at broader federal reform must also center the impacts of AI on workers. Under these circumstances, we urge the newly formed House Democratic Commission on AI and the Innovation Economy to center the recommendations of members with expertise in workers’ need for strong labor protections and AI’s impact on the economy. 

The urgency of this moment is further compounded by the Trump administration’s decision to prioritize corporate capture over the public good. In December, after Congress again declined to preempt critical state efforts to regulate AI, President Trump issued an Executive Order that purports to block states from protecting their own residents—a move that blatantly infringes on states’ rights while offering no federal alternative. The administration has doubled down with a national AI legislative framework that would severely curtail states’ ability to regulate AI. Rather than respecting states’ authority to protect their own residents, the administration is doing the bidding of tech oligarchs. 

The AI industry, venture capitalists, and lobbyists spent hundreds of millions of dollars last year pressuring Congress to pass legislation that would prevent state lawmaking. These attempts have failed multiple times because a significant number of members across both parties recognize the dangers posed by AI, while industry actors continue to push for deregulation. 

This is not what the public wants. Recent polling shows a bipartisan consensus in support of AI safety measures: 88% of Democrats and 79% of Republicans favor maintaining existing rules for AI security. Many people want more guardrails on AI: Majorities of both parties are in favor of new regulations to protect society, including 63% of Democrats and 59% of Republicans. 

Federal action is necessary, but it must also leave states room to innovate. Not all states are taking action, so Congress must provide a baseline of protection for people across the country, with a core focus on workers’ rights and livelihoods. 

But federal legislation should be a floor, not a ceiling. Locking the U.S. into a static, insufficient federal framework would guarantee that protections will swiftly become obsolete. It’s important that policymakers do not build a framework that is so narrow or rigid that it fails to keep up with constantly changing AI risks and shifting economic conditions, leaving workers vulnerable to new risks from new tools and practices.

A strong federal framework can create a reinforced system of guardrails to help working people navigate the growing use of AI. Congress has a responsibility to act now—the well-being of our workers and communities depends on it.

Sincerely,

AFL-CIO 

AFL-CIO Tech Institute 

AFT 

American Federation of State, County and Municipal Employees

Americans for Responsible Innovation 

California Initiative for Technology and Democracy

California School Employees Association 

Care in Action

Center for Democracy & Technology 

Center for Oil & Gas Organizing

The Century Foundation 

Communications Workers of America (CWA) 

Consumer Federation of America 

Data & Society 

Economic Policy Institute 

Encode AI 

Future of Life Institute

Interfaith Center on Corporate Responsibility

Jobs With Justice 

The Leadership Conference on Civil and Human Rights 

Legal Aid Justice Center 

Louisiana Progress 

National Action Network 

National Association of Voice Actors 

National Black Worker Center 

National Domestic Workers Alliance 

National Employment Law Project 

National Employment Lawyers Association 

National Institute for Workers’ Rights 

National Partnership for Women & Families 

National Women’s Law Center 

Open MIC (Open Media and Information Companies Initiative) 

Oxfam America

Public Citizen 

Service Employees International Union (SEIU) 

TechTonic Justice 

United Church of Christ Media Justice Ministry

United Food and Commercial Workers International Union 

We Build Progress 

Working Partnerships USA 

Workshop 

Writers Guild of America West

The Council was preparing to discuss a wage adjustment in June 2025 when the chair resigned. It is expected to take up the issue when the governor names a new chair, which has yet to happen. Given that almost two years have passed since the initial setting of the $20 wage standard—a year and a half that has seen continued inflation—the Council should prioritize this cost-of-living adjustment in 2026 to prevent rising prices from erasing the gains made by fast-food workers. One impediment to this adjustment is opposition from fast-food restaurant operators, who argue that raising workers’ pay to $20 damaged their businesses and that they cannot absorb any further increases.

This debate in California between fast-food workers and employers highlights the importance of regular and automatic adjustments to wage standards (like minimum wages) that ensure inflation-adjusted living standards for low-wage workers do not erode over time.

Indexation is often an afterthought in debates over wage standards. But it can turn out to be the most important part of any policy that sets a wage standard. This report examines salient issues related to indexing wage standards and offers recommendations for policymakers. Its key arguments are:

• Wage standards are necessary and efficient because of unbalanced power in labor markets.
• Wage standards that are fixed in nominal terms and have no automatic adjustment (like the federal minimum wage) get weaker every single year that passes without a legislated increase. The cumulative erosion of inflation-adjusted wage standards often exceeds the initial legislated increase.
  • For example, in inflation-adjusted terms, the federal minimum wage today is lower than it was in 2007, the last time a new standard was passed into law.
• Mandating higher wages for any group of workers will set off a chain of adjustments elsewhere in labor and product markets. What these adjustments eventually mean for relative incomes, prices, and employment is an empirical question.
  • Thankfully, minimum wage increases are some of the most well-studied events in economics, and the weight of empirical evidence is that they do not measurably increase overall inflation or lead to significant job loss, but they do raise the inflation-adjusted pay of targeted workers.
• Adjusting wage standards only for increases in inflation is actually a conservative policy in the sense of minimizing potential burdens on low-wage employers. More ambitious targets for adjustment—like wages or even productivity—could be preferable depending on the specific case.
  • In the case of the California Fast Food Council, providing a price-based adjustment to account for inflation since the initial adoption of the $20 minimum wage in April 2024 is an appropriate and modest step.
  • A 3.5% increase in the wage standard—the maximum adjustment the Council can recommend—is also conservative because it will only partially offset the actual 4.2% cost of living increase since April 2024 and because it does not account for ongoing productivity improvements in the sector.
• Over the past decade—and continuing since April 2024—the inflation rate faced by lower-income households has been higher than the overall inflation rate, largely because housing is a higher share of lower-income households’ budget. This means indexing based on the average inflation rate would fail to fully restore the affordability lost to fast-food workers since the enactment of the $20 wage standard, making such an adjustment even more modest (and even more necessary).

Wage standards are necessary because of unbalanced labor market power

Modern labor markets—particularly those that low-wage workers participate in—are characterized by significant employer power. Low-wage employers rarely if ever negotiate pay with workers, instead posting take-it-or-leave-it wage offers. Further, when a given employer lets its own wages lag 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.

The seminal source for modeling labor markets as situations where employers have substantial wage-setting power is Manning (2003), who describes this situation as one of “monopsony” power in labor markets.The literal definition of monopsony is a market with a single buyer. At points in history (think 19th century “company towns” in rural and isolated areas) this kind of literal monopsony may have existed. But Manning and those who have built on this work point to several features and frictions in real-world labor markets that make it hard for workers to effectively search for better jobs. These job search barriers effectively grant employers excess market power over workers even when there are numerous employers. Some of these frictions include things like lack of information about wages and other policies of alternative employers, transportation restrictions that require workers to look for jobs only in places near their home or public transit nodes, child care considerations that require a job’s location be compatible with picking up kids at a regular time, along with many other factors.

Employers use these barriers to employees finding better outside options to “mark 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, making pay levels inefficient.

At the level of the total economy, the excess power of employers in labor markets and their ability to markdown wages can be seen in the gap between economy-wide productivity (the amount of income generated in an average hour of work in the economy) and the hourly pay (including benefits) of typical workers.

Figure A

Wage standards—like minimum wages—can correct for this excess employer power. This leaves low-wage workers with higher pay and living standards and moves the economy to a more efficient allocation of workers across jobs. It can in theory even lead to an increase in employment. This degree of employer power in labor markets and the inefficiency of labor market outcomes without wage standards help explain the general empirical finding that minimum wage increases in the United States have not caused significant employment declines, a finding that is counter to what one would expect if labor markets were competitive.1

California’s fast-food minimum wage has had minimal employment effects

Current evidence suggests that California’s fast-food minimum wage is no different in that it has raised wages without causing large, negative employment reductions. There are four studies on the specific wage and employment effects of the California fast-food minimum wage. Three studies show both sizable earnings effects and limited-to-no employment changes. One analysis, in contrast to the other three studies, shows moderately negative employment effects, but also found the policy raised the total earnings of fast-food workers.

Schneider, Harknett, and Bruey (2024) surveyed fast-food workers in large chains and showed that relative to other states, the California policy raised wages and had no effect on the usual number of hours of fast-food workers in the quarter after the minimum wage change. With data from Equifax, Hamdi and Sovich (2025) compared fast-food establishments within large firms across different states and found that California fast-food establishments raised wages by about 12% and increased employment by a statistically insignificant 2%. Sosinskiy and Reich (2025) used data from the Quarterly Census of Employment and Wages (QCEW) to study employment and earnings trends in fast-food restaurants in California relative to those in other states and to full-service restaurants in California, which are not directly bound by the fast-food minimum wage. The authors’ preferred specification estimated a wage increase of about 7% and an employment decline of just under 1% that was statistically indistinguishable from zero. Finally, Clemens, Edwards, and Meer (2025) used QCEW data and estimated a similar wage increase of about 8%, but also a statistically significant employment decline of over 3%.2

In interpreting employment changes from a minimum wage increase, it’s best to compare the size of estimated wage effects with the estimated employment effects. The ratio of these two estimates—the own-wage elasticity of employment3—helps to gauge whether any employment changes were small or large relative to how much the policy actually raised wages. When the ratio is more positive than -1, total fast-food worker earnings rose even after accounting for potential employment losses. Standardizing the estimates by dividing employment and wage effects also allows us to make consistent comparisons across these studies and with studies of other minimum wage increases.

For the three studies where it is possible to calculate them, the own-wage elasticities are 0.19 (Hamdi and Sovich 2025), −0.12 (Sosinskiy and Reich 2025), and −0.40 (Clemens, Edwards, and Meer 2025). The first two are consistent with small or no employment impacts, but the last one moves into “medium negative” territory. All three studies’ estimates imply that the policy increased the aggregate earnings of fast-food workers, but the last study implies that employment losses caused fast-food workers to receive only about 60% of the potential earnings increase spurred by the minimum wage hike.

Even though Sosinskiy and Reich (2025) and Clemens, Edwards, and Meer (2025) use similar data, one important difference is that the Sosinskiy and Reich (2025) study controls for population changes. Net immigration rapidly fell after the implementation of the policy, disproportionately affecting California’s population levels. For example, according to the latest Census estimates, California’s resident population did not grow in 2025, whereas the rest of the country’s population grew by about 0.5%. Not accounting for these different population trends between California and elsewhere could cause an analysis to overstate any employment declines stemming from the policy, particularly if fast-food employment levels are sensitive to falling labor supply or a shrinking customer base. In their appendix, Sosinskiy and Reich (2025) find that ignoring population changes causes their estimates to be more negative.

In addition, when selecting a comparison group for fast-food workers, Clemens, Edwards, and Meer (2025) use fast-food workers in other states and high-wage industries in California, but they do not directly compare the California fast-food sector with the California full-service sector, which is not covered by the policy. Comparing the two sectors would be especially useful for capturing underlying economic trends if slowing population growth is driving declines in both fast-food and full-service employment levels. Indeed, Clemens, Edwards, and Meer (2025) show that the policy did not raise wages in the California full-service sector, but full-service employment in California declined by close to 2%. Failing to account for this decline in full-service employment also causes the Clemens, Edwards, and Meer (2025) estimates to be more negative.

Regardless of the source of these differences, the average own-wage elasticity across the three studies is −0.11, suggesting that the fast-food policy was successful in raising wages without causing sizable job losses. This point estimate is very similar to the median elasticity of all published minimum wage studies on restaurants (see Dube and Zipperer 2025). However, even if the policy were associated with larger employment reductions, measured job losses may still overstate the consequences for low-wage workers. First, lower headcount employment in the fast-food sector does not automatically translate into reduced employment or lower wages for low-wage workers if they move to other low-wage jobs, like retail, where they must be paid at least the California $16.90 minimum wage. Second, a measured decline in headcounts in a high turnover sector like fast-food is more likely to manifest as more weeks in between jobs rather than being shut out of work completely; in that case, some fast-food workers would indeed be working less but earning more money over the course of the year due to higher hourly wage rates.4

Why wage standards need to be automatically adjusted

If wage standards stay fixed in nominal terms, they are reduced in real (inflation-adjusted) terms every year inflation is nonzero. When there is a burst of rapid inflation, these real wage cuts get large very quickly. In fact, steady inflation can combine with policy inaction to leave wage standards lower in real terms than they were the last time a legislated increase happened.

Take the example of the federal minimum wage. Its current value of $7.25 came into effect in 2009. Today’s inflation-adjusted value of the federal minimum wage is almost 40% lower than its historic peak. It reached this peak in 1968, in an economy where productivity (the income generated in an average hour of work in the economy) was just 46% as high as it is in 2025. 5

Figure B

Adjusted for inflation, the 2025 value of the federal minimum wage is in fact lower than it was in 2007 when the U.S. Congress and president last signed a legislated increase into law. Put simply, without effective and automatic indexation, higher wage standards can be eroded almost entirely over time.

Today’s debate over the cost-of-living adjustment to the California Fast Food Council’s minimum wage often frames such adjustments as imposing new burdens on low-wage employers. But inflation since April 2024 means that the real minimum wage paid to California’s fast-food workers has been steadily cut since then. From April 2024 to January 2026, as measured by the consumer price index for all urban wage earners (CPI-W), this cut amounts to 4.2%. Without indexation, any burden on employers from this wage standard has fallen considerably since its adoption, providing a windfall to low-wage employers at the expense of their frontline employees. A failure to regularly index for inflation is essentially a backdoor method for unraveling the wage standard that policymakers passed into law.

Price indexing wage standards is a necessary and conservative policy

Raising wage standards each year by an amount equal to inflation holds low-wage workers’ living standards steady at the level that prevailed when the wage standards were set. For example, the $20 minimum wage for fast-food workers in large chains in California came into effect in April 2024. If these wages are indexed regularly to account for inflation since then, this will keep California fast-food workers’ living standards frozen at April 2024 levels going forward.

This is a clear improvement compared with outright erosion of living standards. But it remains the case that price indexing wage standards is a conservative policy in the sense that it minimizes any potential burdens on low-wage employers. It is a conservative policy for two reasons: (1) indexed wage changes are very small relative to the initial phase-in of wage standards, and (2) indexing for prices allows productivity growth in the wider economy to steadily reduce any potential burden or need for adjustment imposed by wage standards.

Price indexations are very small increases to wage standards

The increases to wage standards that result from price indexation are significantly smaller than the increases that result when the standards are initially phased in. For example, say that the last federal minimum wage increase in 2009 also indexed for subsequent price changes. The initial phase-in of the higher federal minimum wage saw it rise from $5.15 to $7.25 between 2007 and 2009. This constituted an average annual change of 19% for these two years. The average annual inflation rate (measured by the consumer price index for all items) between 2007 and 2024 was just 2.5%.

If the initial introduction of higher wage standards does not cause problematic outcomes, then it is very hard to see how the much smaller changes spurred by indexation for price changes would cause any.

The research on minimum wages provides very little reason to worry that changes in the United States in recent decades have caused any such problematic outcomes. The most commonly expressed worries about minimum wage increases are employment losses and upward price pressure.

We noted earlier that studies looking specifically at the California wage standard continue a common pattern in research on the employment effects of phased-in minimum wages: Employment declines caused by these minimum wage changes tend to be extremely modest or even zero on average. If one applied the modest measured employment losses stemming from the large initial increase in fast-food wages to the much smaller indexed adjustments, these already small employment losses become totally trivial.

The same logic holds regarding potential upward price pressures stemming from indexation: Compared with the initial setting of wage standards, indexed changes are very small and therefore unlikely to push up prices.

It is a fact that one person’s income is another person’s cost, so as low-wage workers’ pay rises, this raises costs for their employers. These employers could pass on these costs (in part or in full) to their customers by raising prices. But even if the entirety of the wage increases driven by price indexing wage standards was passed on in the form of price increases, overall price pressures would be extremely modest and low-wage workers would still unambiguously come out ahead.

Say that low-wage workers’ pay constitutes a third of labor costs in the fast-food sector, and that labor costs in turn constitute a third of total costs of fast food.6 If low-wage workers’ pay rises by 3.5% due to price indexing, this would increase prices the employer charges customers by less than 0.4% even if the full amount was passed on as price increases. Because fast food accounts for less than 3% of the overall inflation consumption basket, even a 0.4% increase in fast-food prices would raise overall prices by only 0.01%.

Price indexing still sees reductions in low-wage workers’ relative pay and allows productivity growth to steadily erode any potential burden on low-wage employers

We noted earlier that price indexing a minimum wage essentially holds low-wage workers’ pay frozen thereafter at the level that prevailed when the wage was introduced. Again, this is better than allowing inflation to erode the real value of pay, but average incomes throughout the overall economy are not frozen over time in real terms. Instead, they rise faster than prices over any reasonable period. Inequality often keeps this growth in average living standards from reaching many (or even most) workers and families in the economy, but the potential for living standards to rise is generated every year of positive economic growth.

This means that even when wage standards are indexed to prices, low-wage workers’ relative standing in the economy still falls over time. Further, because low-wage workers’ earnings are a cost to their employers, this means that even with price indexing, any potential burden of wage standards on low-wage employers slightly declines any year that productivity rises. In this sense, price indexing of wage standards—providing regular cost-of-living adjustments based on price growth—is a conservative policy that allows the costs and benefits of wage standards to slowly erode over time relative to developments in the larger economy.

A quick example can help make this point. Say that pay for low-wage workers at a particular employer amounts to 20% of the final price of the firm’s output. Say that productivity (how much output is generated with each hour of work) rises by 2% per year. If low-wage workers’ pay rises only with inflation (and not with productivity) and all other firm costs rise with inflation and productivity, this implies that over 10 years the share of low-wage workers’ pay in total costs would fall to just 16.4% of total costs. Employers could use this decline in real costs to either lower their prices to consumers or raise their profit margins. Either way, so long as there is any growth in productivity, the burden of low-wage workers’ pay to employers falls even when this pay is indexed to inflation.

Price indexing is not the only option for adjusting wage standards. One could, for example, index growth in minimum wages to growth in wages at other parts of the wage distribution—growth in the median wage for example.7 An even more ambitious indexing choice would be to match wage changes to changes in average wages or even economy-wide productivity.8

The obvious benefit of using these alternative wage indexations would be faster wage growth and higher living standards for low-wage workers. The potential downside is that they do not allow any potential burden from higher wages for employers to relent over time—meaning that if the initial setting of wage standards is high enough to cause problematic outcomes (job losses or rapid price increases), then this would not smooth over time with wage indexation. Price indexation, conversely, would actually allow any higher than optimal initial wage standard to become less binding over time. In this sense, it is a conservative choice that is highly responsive to the pressures faced by low-wage employers.

In the case of the California Fast Food Council, the $20 minimum wage enacted in 2024 was an admirably ambitious standard. There is little persuasive evidence that it is too high in that it has caused any problematic outcomes on either the employment loss or price increase fronts. Yet it was high enough to provide a significant wage boost for affected workers. For these types of ambitious standards, indexing to prices seems necessary to protect workers’ gains yet conservative in that it puts declining pressure on low-wage employers over time. Further, since 2019, the limited-service restaurant sector has seen significant productivity growth—roughly 2% per year—which should allow any price indexation to be easily absorbed with no wrenching adjustments for employers.9

Different groups face different inflation rates: The case for discretion in indexing

The benefit of indexing wage standards for inflation is the protection it provides for the living standards of low-wage workers. The costs are the various adjustments or burdens forced onto employers. Because the group of low-wage workers and employers are heterogenous, and because inflation is measured by the aggregation of price changes across the entire economy, there remains room for judgement and discretion in balancing these costs and benefits.

The California Fast Food Council has some discretion, as they can either index wages up to 3.5% for inflation or they can decline to index these wages and let them be eroded.

We noted before that indexing only for prices (as opposed to indexing for wages or productivity growth) results in a steady reduction in any economic burden wage standards might place on employers. So long as these employers see any growth in productivity (the efficiency with which each hour of labor generates output), then having some portion of their wage costs fixed in real terms will see these costs become a progressively smaller share of total output over time. In this sense, simply choosing to index by prices means the cost of wage standards to employers is set to shrink consistently over time.

In terms of the benefits to low-wage workers, recent years have seen a large jump in the overall price level. Any given episode of inflation is likely to have uneven effects across groups in the economy. For example, the inflation of the 1970s was actually accompanied by an increase in real wages, even for low-wage workers.10 The inflationary spike in 2021 and 2022, conversely, was largely driven by large increases in profit margins, which meant that real wages for most workers fell in those years.

More systematically, inflation faced by various groups in the economy can diverge if they have different consumption baskets that skew average price growth in a predictable way. For example, housing makes up a larger share of consumption spending for lower-income households than higher-income households, and in recent decades the price of housing as measured by the consumer price index has slightly outpaced overall price growth. This implies that inflation faced by lower-income households has likely been systematically higher than that faced by higher-income households. This makes the overall CPI that informs discussions of wage indexation inadequate for fully protecting lower-wage workers from inflation in recent years.

Concretely, the CPI-W, which is the price index the Council can target, has risen by 4.2% since April 2024. This means that a 3.5% cost of living adjustment—the largest that can be granted by the Fast Food Council—would not quite neutralize the affordability losses experienced by workers since the $20 minimum wage was enacted. Research from the Federal Reserve Bank of New York (2025) indicates that households in the bottom 40% of the income distribution saw inflation between April 2024 and August 2025 (the most recent data point available) that averaged 0.2% higher than overall inflation. This means actual inflation faced by many fast-food workers in California exceeded 4% since the introduction of the $20 wage standard.

The bias in actually experienced inflation stemming from housing runs even deeper. The housing component of the CPI essentially assumes everybody is paying market rent for their housing. There are good reasons for this decision, but it means that discretion and judgement must enter into using the CPI for different purposes. Well over half of the U.S. population owns their homes, and these people have significantly higher incomes on average than renters. Homeowners either have no monthly housing payment or pay a mortgage that is fixed over time and therefore experiences no inflation. By assuming these homeowning households experience the average amount of rental inflation each month the CPI overstates actually experienced inflation for homeowners.

This means when weighing the interests of low-wage workers against other economic actors—including consumers facing potential price increases stemming from wage standards—the real gap in living standards growth is likely larger than what would be implied by assuming all households face the same CPI inflation. Given this, there is a strong case for policymakers to use their discretion to put a countervailing thumb on the scale by boosting low-wage workers’ pay.

Notes

1. For a review of the estimates of employment loss caused by minimum wage increases, see Dube and Zipperer 2025.

2. There is an additional study by Pandit (2026) that estimates the fast-food minimum wage caused an 8% decline in staffing intensity based on long-duration visits from mobile phone location data. However, the study finds almost all of the estimated effect occurred before the actual policy went into effect, with little-to-no change in the proxy for employment activity after the effective date of the minimum wage increase on April 1, 2024. It is hard to believe that in a very high turnover industry like fast food—where employers can adjust employment levels rapidly by reducing hiring—that businesses would reduce staffing levels several months before being compelled to pay higher wages, but then not change employment levels at all after actually being required to increase wages. The study also provides no evidence on wage changes, cannot distinguish between headcounts and hours reductions, and excludes new businesses that may have started during the policy period.

3. For an explanation of the importance of the own-wage elasticity in interpreting studies of the minimum wage’s effect on employment, see Dube and Zipperer 2024.

4. See Cooper, Mishel, and Zipperer 2018 for the importance of accounting for turnover rates when assessing the likely implications of any measured employment decline.

5. See Economic Policy Institute 2025a for data on productivity levels over time.

6. Both of these assumptions are likely close to true or overstate the actual price pressure that would be experienced from price indexing wage standards in fast food. For the leisure and hospitality sector—the larger sector in which fast-food (or limited-service) restaurants are embedded—aggregate weekly payrolls are roughly $10 billion. To estimate low-wage workers’ aggregate pay, we took the number of leisure and hospitality sector workers making less than $17 per hour in 2024 (5.7 million) and multiplied this by $17 and by 35 hours per week. All of these (the high $17 threshold for defining “low-wage”, the assumption that all making under $17 were making exactly $17, and the 35 hours per week) likely increase the estimate of low-wage workers’ wage bill in the sector. Making these generous assumptions yields a weekly wage bill of roughly $3.4 billion, or just over a third of the total wage bill in the sector. For total labor costs as a share of total output in the sector, we used the Composition of Gross Output by Industry table from the GDP by Industry accounts of the Bureau of Economic Analysis.

7. Growth in wages has been used to index some labor standards. Under the overtime rule enacted by the Obama administration the salary threshold for being granted automatic rights to overtime protections was set at the 40th percentile of annual earnings in the lowest-wage region of the country.

8. Social Security uses an “average wage index” to deflate workers’ past earnings to calculate their initial Social Security benefit amount. This implicitly credits recipients for overall economic growth (overwhelmingly determined by productivity) over the course of their working life.

9. This figure calculated from data provided by the Detailed Industry Productivity database from the Bureau of Labor Statistics.

10. See Economic Policy Institute 2025b, specifically the wages for workers at the 10th percentile.

References

Bivens, Josh. 2022. “Corporate Profits Have Contributed Disproportionately to Inflation: How Should Policymakers Respond?” Working Economics Blog (Economic Policy Institute), April 21, 2022.

Clemens, Jeffrey, Olivia Edwards, and Jonathan Meer. 2025. “Did California’s Fast Food Minimum Wage Reduce Employment?” NBER Working Paper no. 34033, July 2025.

Cooper, David, Larry Mishel, and Ben Zipperer. 2018. Bold Increases in the Minimum Wage Should Be Evaluated for the Benefits of Raising Low-Wage Workers’ Total Earnings. Economic Policy Institute, April 18, 2018.

Dube, Arindrajit, and Ben Zipperer. 2024. “Own-Wage Elasticity: Quantifying the Impact of Minimum Wages on Employment.” NBER Working Paper no. 32925, September 2024.

Dube, Arindrajit, and Ben Zipperer. 2025. Minimum wage own-wage elasticity repository, Version 2025.9.1., https://economic.github.io/owe.

Economic Policy Institute. 2026. “The Productivity-Pay Gap” (web page). Last updated January 16, 2026.

Economic Policy Institute. 2025a. State of Working America Data Library, “Productivity and pay levels – Productivity and pay, real dollars per hour (2024$).”

Economic Policy Institute. 2025b. State of Working America Data Library, “Minimum wage – Real minimum wage (2024$).”

Economic Policy Institute. 2025c. State of Working America Data Library, “Hourly wage percentiles – Real hourly wage (2024$).”

Federal Reserve Bank of New York. 2025. “Economic Heterogeneity Indicators.” Accessed January 2026.

Hamdi, Naser, and David Sovich. 2025. “The Wage and Employment Effects of California’s Fast-Food Minimum Wage.” SSRN, March 28, 2025.

KFF. 2025. “Population Distribution by Citizenship Status.” Accessed January 23, 2026.

MacDonald, Daniel, and Eric Nilsson. 2016. “The Effect of Increasing the Minimum Wage on Prices: Analyzing the Incidence of Policy Design and Context.” Upjohn Institute Working Paper no. 16-260, June 2016.

Manning, Alan. 2003. Monopsony in Motion: Imperfect Competition in Labor Markets. Princeton, N.J.: Princeton Univ. Press.

Pandit, Hitanshu. 2026. “Simply Can’t Wait: Evaluating the Effect of California’s Fast-Food Minimum Wage Increase.” SSRN, February 23, 2026. 

Schmitt, John. 2013. Why Does the Minimum Wage Have No Discernible Effect on Employment? Center for Economic Policy and Research.

Schneider, Daniel, Kristen Harknett, and Kevin Bruey. 2024. Early Effects of California’s $20 Fast Food Minimum Wage: Large Wage Increases with No Effects on Hours, Scheduling, or Benefits. The Shift Project, October 2024.

Sosinskiy, Denis, and Michael Reich. 2025. “A $20 Minimum Wage: Effects on Wages, Employment and Prices.” Institute for Research on Labor and Employment Working Paper, September 2025.

United States Census Bureau (Census). 2026. Annual Estimates of the Resident Population for the United States, Regions, States, District of Columbia and Puerto Rico: April 1, 2020 to July 1, 2025. NST-EST2025-POP. Accessed January 27, 2026.

Zipperer, Ben. 2024. “Most Minimum Wage Studies Have Found Little or No Job Loss.” Working Economics Blog (Economic Policy Institute), September 9, 2024.

Immigration has been a source of strength for the U.S. economy and has great potential to boost it even more, but the current U.S. immigration policy regime squanders too many of its potential benefits by depriving immigrants of their full rights as workers and granting employers too much power to manipulate the system. It is crystal clear that immigration expands U.S. gross domestic product and is good for growth. And immigration overall has led to better, not worse, wages and work opportunities for U.S.-born workers. Yet, it is also clear that when workers are denied full and equal labor and employment rights, as some immigrants are when their immigration status is used against them—it makes immigrant workers’ lives more precarious and can harm the people with whom they work side-by-side in the same industries.

Even in the face of our unjust policy regime, immigration today provides numerous benefits to the U.S. economy. The nation could benefit even more, and the benefit could be shared more widely with a smarter set of immigration policies that benefit all workers.

The benefits are too often overlooked, and the challenges it poses for policymakers and U.S.-born workers are often grossly exaggerated. In this paper, we assess the evidence on immigration’s effect on a number of economic outcomes given the policy status quo. We then go on to highlight how immigration status impacts wages and working conditions, and finally, offer recommendations on how to craft a better immigration policy regime—one that grants immigrants their full rights as workers in U.S. labor markets, generating broad benefits for U.S. and foreign-born workers alike. Our key findings are: 

• Immigration is enabling economic growth despite the sharp deceleration in the growth of the U.S.-born workforce. Immigrants have risen as a share of the U.S. population in recent decades. Because immigrants are younger and more likely to be working age than the U.S.-born population, they have risen even faster as a share of the labor force. These trends are driven as much by the sharp deceleration of growth of the U.S.-born workforce as they are by a sharp increase in the absolute scale of immigration flows. Even when available data are adjusted to better reflect the notable increase in immigration in recent years, the growth of the foreign-born population since 2022 is not out of line with other historical periods (and is lower than what prevailed in the late 1990s). What is unprecedented in recent years is the rapid deceleration of growth in the U.S.-born population—a population that will soon start seeing outright contraction. Without immigration, the prime-age workforce (between the ages of 25 and 54) would have seen essentially no growth at all in the past quarter century, dramatically constricting the ability to grow our economy and staff key industries. 
• Immigration does not reduce the number of jobs available for U.S.-born workers and has provided a source of deflationary pressure in recent years. Macroeconomic policymakers like the Federal Reserve and Congress have both the necessary tools and the obligation to ensure that demand in the economy is strong enough to absorb willing workers in a reasonable amount of time. Historical periods when job opportunities have become scarce were not driven by increased immigration flows but instead by the failures of macroeconomic policymakers to respond in a timely fashion to weakness in demand for labor. Immigration flows generally do not make policymakers’ task of matching macroeconomic supply and demand any harder because immigrants add to both economywide supply (by increasing labor supply) and demand (by buying goods and services). Overall, immigrants likely tilt this balance more toward supply relative to U.S.-born workers, but only by a bit. This net supply increase is sometimes quite useful—by boosting supply a bit more than demand over the 2021–2023 period, immigration provided a source of deflationary pressure that helped bring inflation down in those years, while also assisting in preventing a recession. 
• Immigration’s effects on U.S. wages overall range from neutral to slightly positive. Immigrant workers tend to complement rather than substitute for U.S.-born workers of similar educational levels. The educational mix of immigrants is much more similar to that of U.S.-born workers than is often recognized, hence limiting the scope for competition from immigrants to lower wages for any U.S.-born workers of any particular educational background. Additionally, while the overall wage effects from immigration are neutral to slightly positive for U.S.-born workers, it is true that our flawed immigration policy regime often sees employers use the precariousness of some immigrant workers to suppress wage growth in particular sectors. Immigration reforms that remove this precariousness would aid the wage growth of both immigrants and the U.S.-born workers they work alongside in these labor markets. 

• Immigration is clearly positive for the balance of taxes and spending at the federal level. Immigrants pay substantial amounts of tax (nearly $100 billion in the most recent years, and almost $60 billion in federal taxes) and yet are often excluded from drawing on key benefits. At the state and local level, the balance is much closer. The types of goods and services provided by these governments (public education and infrastructure, for example) are generally more available to immigrant families regardless of their legal status. Taken together, the net of federal and state/local taxes and spending stemming from immigration flows is positive, with immigrants paying more in taxes relative to what they draw in spending than U.S.-born households.

• Increased housing costs are caused not by immigration, but by the failure of U.S. housing markets to meet any new demand with more housing units rather than higher prices. Immigration, like any source of population growth, can put upward pressure on housing prices on the demand side. Yet the rapid rise in housing costs over the past decade has come at a time when overall population growth has slowed significantly—indicating strongly that the root of the problem is a dysfunctional housing supply side that translates any housing demand increase into sharp price increases rather than newly constructed housing. This can be seen most clearly in the very large run-up in home prices in 2020 and 2021, just as rates of immigration were plummeting. Further, higher rental costs stemming from the intersection of population growth and supply-side rigidity actually boost housing wealth for homeowners, who are disproportionately U.S.-born. Finally, immigrants are a key source of labor that boosts the supply side of housing markets. Anti-immigrant reforms (like mass deportations) done in the name of preserving housing affordability for U.S.-born workers could well see large increases in housing costs due to resulting shortages of workers employed in residential construction.

• People who immigrate into the United States increase the economy’s stock of human capital and ideas, two crucial ingredients for long-run economic growth. Immigrants can provide a key source of innovation and economic vitality. If immigration into the United States allows talented individuals a better chance of realizing their full potential and contributing to economic activity at their maximum level than they would have had in their country of origin, this immigration flow can be highly beneficial for both the U.S. and global economies. Research on past waves of immigration shows that they had positive causal effects on long-run economic growth in the United States. 

• Policy reforms that grant immigrants full rights in the labor market spur benefits for everyone in the U.S. economy. The evidence is clear that workers who lack formal immigration status or have precarious or temporary immigration statuses experience degraded pay and conditions. Thus, the most obvious and pressing policy priority is to provide a quick and broad path to legal status and a green card (and eventually citizenship) for the currently unauthorized immigrant population. Regularization would provide a near instant wage boost for these workers, which would in turn bid up wages in the labor markets they share with U.S.-born workers. The same logic, backed by research, suggests that future immigration flows should rely much more on green cards than temporary work visas. Green cards give immigrant workers the ability to shop around for the best fit for their skills and abilities, which breaks employers’ power over these workers and helps lift standards for all workers. Allowing temporary migrant workers to control their own visas rather than being tied to specific employers would also be a win-win for both foreign-born and U.S.-born workers.
• Underresourced labor standards enforcement is enabling low-road employers to abuse immigrant workers with impunity and flout basic worker protections. Rather than spending tens of billions of dollars per year for immigration enforcement, what’s needed are more resources and staffing for labor standards enforcement agencies and a more strategic focus on labor standards enforcement that does not take immigration status into account. Lawbreaking employers who abuse workers of any status should face much higher chances of being caught and much higher penalties than they currently do.

Immigration to the U.S. provides many economic benefits, and those benefits would increase substantially if immigration policy was improved to guarantee equal and enforceable labor and workplace rights. With proper alignment with other policy spheres, such as investment in physical and human infrastructure, win-win opportunities abound for the United States with respect to future immigration flows or even expansions. 

Current and historical immigration and where immigrants are in the U.S. economy

Immigrants in the economy and recent growth in the immigrant population

Migrants and immigrants who resided in the United States in 2022 accounted for 13.8% of the population, as measured by the American Community Survey (ACS). The ACS is perhaps the most commonly used source for the size of the foreign-born population. However, we can get slightly more timely data from a different data source, the Current Population Survey (CPS). The CPS is also the most cited source of data on labor market trends in the United States. Because of this, we also show CPS-based trends in the annual foreign-born shares of the U.S. population in Figure A. In 2023, this share was 14.9%.1

Figure A

Because so many debates around immigration center around its effect on U.S. labor markets, in Figure B we also present the number and share of immigrants in the overall labor force, defined as those ages 16 and older who are working or seeking jobs. As of last year, the U.S. workforce had 31.0 million immigrants, or 18.6% of the total U.S. labor force. Since 2000, immigrants have been more likely to be in the labor force than U.S-born workers, largely reflecting the fact that immigrants are younger and more likely to be of working age than the U.S. born population.2

Figure B

Immigrant flows have fluctuated greatly in recent years. During the beginning of the Trump administration and also the beginning of the coronavirus pandemic, immigration slowed due to border closures and other restrictions. As those restrictions were lifted, a strong U.S. labor market and high employer demand for workers coincided with a reestablishment of the administrative elements of the immigration system and higher immigration flows. Immigration flows over the last two years have been relatively high according to some estimates, prompting calls for more immigration enforcement at the U.S. southern border and restrictions on asylum, and discussions about the number of new arrivals that the U.S. economy can absorb.

However, Figure C shows that while recent flows of migrants and immigrants into the United States have been high, the number of immigrants residing in the United States at any given point is growing at a rate similar to other periods of high immigration like the late 1990s. The decennial censuses estimated that the foreign-born (immigrant) population grew by 4.6% annually between 1990 and 2000. Between 1994 and 2000, the immigrant population in the CPS grew by 5.6% annually. In contrast, recent data from the CPS show that the 2022–2023 immigrant population growth rate was 3.7%.

Figure C

Recently there has been some debate as to whether the CPS may underestimate recent growth in immigration levels.3 After making an adjustment to account for this undercount, the CBO (2024) estimated significantly higher immigration levels using CPS data, but even with this adjustment, the modified CBO/CPS flows between 2022 and 2023 are still lower than estimates from the late 1990s. The modified CBO/CPS immigrant population growth between 2022 and 2023 was 4.8%, compared with an annual rate of 5.6% between 1994 and 2000 using the unmodified CPS.4

The rates in Figure C also show how U.S.-born population growth has been declining, from 0.9% annually in the 1990s to 0.2% in 2022–2023. Figure D examines this in more detail, focusing on the prime-age labor force—i.e., those who are ages 25 through 54 and employed or seeking work. Over the last three decades, the total size of the U.S.-born prime-age labor force has essentially stayed the same. In fact, were it not for immigration, the total prime-age U.S. labor force would have stagnated: Over 95% of the cumulative growth of the labor force in the past three decades is due to immigration.

Figure D

Occupations that rely heavily on immigrants

In 2023, immigrants were about 18.6% of U.S. employment, and many immigrants are disproportionately concentrated within a select number of occupations. Table 1 shows the top-ten major occupation groups with the highest shares of immigrants. These occupations have immigrant shares of employment ranging from about 21% to 40%, and together they comprise just over half (54%) of all employed immigrants in 2023.

Table 1

Immigrants are often associated with agricultural occupations—and they do comprise two-thirds of the workforce in crops (Fung et al. 2023), but because of the relatively small total size of agricultural occupations overall, most immigrants work in other types of jobs. For example, in 2023, 2.2 million immigrants were employed in “Building and grounds cleaning and maintenance” occupations, like janitors, custodial workers, and landscapers, accounting for 40% of employment in the occupation. For the purpose of comparison, the 2.2 million immigrants in the occupation is the roughly the same number of all hired farmworkers, including immigrants and U.S.-born citizens (USDA 2024). Immigrants were also about one out of every four (24%) workers in “Healthcare support occupations,” which include jobs like home care aides and nursing, dental, and medical assistants.

While immigrants work in some high-paying occupations, many of the occupations in which immigrants are disproportionately concentrated typically pay low wages. Table 1 shows that a majority of the top-ten occupation groups with higher shares of immigrants have median hourly wage rates that are substantially lower than the national median wage. For example, building and grounds cleaning and maintenance workers, who had a national median wage of $17 an hour in 2023, were paid 32% less than the 2023 national median wage of $25 per hour.

Health care support occupations are also paid a wage that is much lower than the median national wage (28% lower). These occupations, however, are expected to be among the fastest growing over the next decade according to the Bureau of Labor Statistics (BLS 2023). If BLS employment projections turn out to be correct, and if we assume that immigrant workers will continue to fill the same share of jobs in the occupation that they do now (24%), then that implies over 250,000 more immigrants will be needed for health care support jobs alone between 2023 and 2032 (BLS 2024b). Low wages and large workforce requirements suggest that we should ensure that immigrants who fill these jobs arrive with full and equal employment rights and a path to citizenship. The second-fastest growing occupation according to BLS will be in computer and mathematical occupations, an occupation that is expected to grow by 15.2% (BLS 2023) and in which 27% of current workers are immigrants.

Economic effects and impacts of immigration

This section provides a review of arguments and evidence on how immigration can affect economic outcomes. We focus on a range of issues in which particular concerns about the effects of immigration have been commonly raised in public debates: jobs, wages, fiscal effects, housing, and long-run economic growth. We also highlight in each section how policy changes can loosen any particular trade-off that might bind from balancing the benefits of higher immigration and the economic welfare of U.S.-born residents.

This focus on the economic welfare of U.S.-born residents is not because it is the only population we care about. It is instead because this is a politically salient issue, and because disentangling the effects on U.S.-born residents helps inform sound policymaking to promote broadly shared prosperity.

For example, calculations are occasionally presented regarding the “economic impact” of immigration that simply tally up all spending done in the U.S. by immigrants. This is a large-sounding number (on the order of hundreds of billions of dollars at least). But increasing overall gross domestic product (GDP) or overall consumer spending by increasing the number of people (and workers) in an economy does not mean much in terms of impacting the welfare of any individual (whether immigrant or native-born) in that country. What matters far more for human welfare is income or spending per capita, and how it grows over time.

An equally useless exercise would be to remove a particular set of immigrants from calculations of per capita GDP—note that this removal increases per capita GDP—and suggest that this tells us something about the economic effect of immigration. Removing lower-income segments will inflate the average income of any group simply through composition effects. This removal does nothing to boost any real person’s income or living standards; it is simply an arithmetic quirk.

In what follows, we aim to provide a serious discussion of the substantive effects of immigration on various economic outcomes.

Employment effects of immigration

A common trope in immigration debates is that immigrants take jobs away from native-born workers. The weakness in this argument is fairly obvious: It presumes there are a fixed number of jobs that does not respond at all to rising immigration flows. But that’s not how employment generation works. Take the most obvious example of why this reasoning is wrong: There are 70 times more people in the United States as there are in Norway (a similarly rich country). There are also almost 70 times as many jobs in the United States as there are in Norway. Is it just a lucky coincidence that the United States was able to generate 70 times as many jobs as Norway and give decent employment outcomes for our much larger population?

Obviously not. When an economy is being managed well, it should be providing a job for essentially all people in the economy who want to work regardless of how large or small that number of people is. Immigration does not change this, and it does not even put much pressure at all on the mechanisms that are supposed to ensure near full employment of every person who wants a job.

In the jargon of macroeconomists, unemployment rises when growth in aggregate demand lags behind growth in potential output. Aggregate demand is simply all desired spending in the economy; consumption spending by households, investment spending by businesses, and public spending by governments. Potential output is a measure of how much the economy could produce if nearly all of the economy’s willing workers were employed. When aggregate demand falls beneath the economy’s potential output, then unemployment rises.

This logic might at first glance seem to buttress fears about increased immigration in generating unemployment—a larger labor force boosts the economy’s potential output, and if this boost pushes it above the economy’s aggregate demand, then unemployment results. But the fault in this story is essentially never that potential output is rising too fast; it is instead that aggregate demand is falling too rapidly (or rising too slowly).

The key insight here is that aggregate demand moves far more quickly than potential output, and it swings up and down much more in compressed periods of time. In short, recessions and recoveries are clearly driven by these quick and large movements in aggregate demand, not in the steady rise of potential output.

So long as policymakers recognize any change in potential output in a reasonably short time frame—like an increased labor supply stemming from immigration—then they can adjust aggregate demand to ensure enough jobs are created to keep unemployment rates from rising. Aggregate demand can be boosted through either monetary or fiscal policy interventions, with the Federal Reserve using monetary policy tools like interest rate cuts to boost demand and Congress and the president setting taxes and spending at levels that boost demand when that is needed (in practice, fiscal policy is the more powerful tool if used effectively). Support for the idea that policy can quickly restore falls in aggregate demand is provided by the U.S. economic recovery from the COVID-19 recession. In late 2020, there was a burst of job creation as the first wave of the pandemic eased and the economy reopened. Nevertheless, in December of that year, the unemployment rate remained high at 6.7%, and job growth had turned negative. But thanks to fiscal recovery packages passed in December 2020 and March 2021, by the end of 2021 the unemployment rate had already fallen below 4.0%—essentially matching the immediate pre-pandemic level.5

Further, policymakers in charge of aggregate demand really do not have to try all that hard to keep aggregate demand matched to potential output as immigration flows rise. Immigrants do boost potential output—but they also boost aggregate demand. That is, immigrants are not only workers, but they are also consumers.

Relative to U.S.-born residents, immigrants do tend to boost potential output a bit more than they boost aggregate demand, mostly by virtue of being younger and hence more likely to work than U.S.-born residents.6 Because immigration boosts potential output a bit more than aggregate demand, the net effect is slightly deflationary. Overall, this deflationary slant of immigration is neither good nor bad. Sometimes deflationary influences in the economy need to be counterbalanced by policymakers, but sometimes deflationary influences just help reach targets policymakers are already aiming for.

The obvious example of the latter is the beneficial deflationary effect of a rebound in immigration following the immigration collapse of 2020. The pandemic led to a very sharp fall in net immigration in that year (in part because of slowdowns of visa processing at U.S. consulates around the world and mass layoffs in sectors with a heavy immigrant worker presence). As the pandemic also led to a sharp rise in inflation as the economy recovered, policymakers began looking for ways to reduce these inflationary pressures in late 2021 and 2022. The rebound in immigration flows in 2022 and 2023 clearly went in the correct direction to help tamp down inflation, though this effect can be exaggerated. The overall dynamics of inflation over the whole 2021–2023 period were generally not driven by the labor market. Wage growth consistently muffled instead of amplified price inflation, and it was the extraordinary growth in corporate profit margins that was a key part of the initial inflation surge in 2021.7 Though inflation was not caused by labor markets, the rebound of immigration flows in 2022 and 2023 provided some measure of deflationary relief in those years.

It is also worth noting that immigration is not the only challenge facing policymakers in charge of aggregate demand. Similar challenges exist for all sorts of economic influences—technological change, exchange rate movements driven by events abroad, sharp changes in household wealth driven by financial market excesses, and many others. There is nothing uniquely challenging about sustaining job creation in the face of immigration flows if policymakers have the will to do it. If any significant number of U.S.-born workers are finding it hard to find work, it is macroeconomic policymakers, not immigrants, who are to blame.

As important as it is for policymakers to keep unemployment low for all workers, this consideration should be mostly irrelevant when considering optimal immigration policy. U.S. immigration policy should decide the level and composition of immigrant flows that the U.S. should seek to absorb—taking into consideration the need for new workers or to meet humanitarian obligations, etc.—and in terms of the effect of that immigration policy on jobs and employment for U.S.-born workers, macroeconomic policymakers should just seek to keep aggregate demand at levels that keep unemployment low and jobs plentiful for the number of people entering the workforce.

It is worth reemphasizing how much the experience of the U.S. labor market in the years after the pandemic highlights the point that there is no trade-off between immigration and job opportunities for U.S.-born workers. Even as there has been an increased inflow of immigrants and immigrant workers in recent years following the dip driven by the Trump administration and pandemic shutdowns, unemployment rates for U.S.-born workers are historically low, and the share of prime-age U.S.-born workers with a job is historically high. While we remain concerned that in the medium- to long-term, there are potential significant challenges to having millions of new workers entering the workforce who only have temporary, precarious statuses, it is nevertheless an odd historical moment indeed to raise the specter that rapid growth of immigrant workers is hurting the job prospects of U.S.-born workers. Very simple policy changes—namely, providing these workers with full workplace rights and the ability to participate in society that comes with a green card—could blunt those challenges and maximize the economic benefits these new workers can bring.

Finally, we should note that while immigration poses no real challenge to creating enough jobs for U.S.-born workers, one particularly damaging form of immigration policy could profoundly affect these jobs: mass deportation. One example comes from a recent study by East et al. (2023), which looked at past episodes of significant deportations and found that for each 1 million immigrants seized and deported from the United States, 88,000 jobs were lost for U.S.-born workers. These deportations reduced local demand for output, which led to job loss, and they led to employers having to shut down all production in the face of key bottlenecks incurred by losing their immigrant workforce, including production that supported employment of U.S.-born workers.

Immigration and wages

One of the major concerns about immigration is that it could reduce the wages of U.S.-born workers. Yet across the economy broadly, a large body of empirical evidence suggests this concern is misplaced. A review by Peri (2014) of more than 270 estimates from 27 published studies found that the average effect of immigration on native-born wages is essentially zero. Two-thirds of studies were clustered around zero, finding small positive or small negative effects. The comprehensive review of immigration, wages, and employment by the National Academies of Science, Engineering, and Medicine (NASEM) found that “when measured over a period of more than 10 years, the impact of immigration on the wages of natives overall is very small” (NASEM 2017). Below we explain some of the mechanisms behind why increased immigration does not tend to harm the wages paid to U.S.-born workers, at least at the broader nationwide level.

The conventional warning about how immigration can reduce the wages of U.S.-born workers is a supply-side story in which large flows of immigrants push down the market wages of otherwise comparable native-born workers. In this story, the labor market employs some mix of workers of different education levels: workers with college degrees and workers without college degrees. Under this theory, the relative supply of and demand for these different types of workers determine their relative wages. Immigration can then change the wage structure by changing the mix of low-education and high-education workers. In particular, if immigration increases the relative supply of workers without a high school degree, there could be downward pressure on market wages for all low-education workers, including U.S.-born workers.

To a rough approximation, however, immigration in the United States in recent decades has not dramatically changed the relative mix of the workforce that has or doesn’t have college degrees. Figure E shows that the share of foreign-born workers with a college degree is similar to that of U.S.-born workers: About 4 in 10 people in the U.S. labor force has college degrees, whether or not they were born in the United States or in another country. This empirical fact greatly reduces the scope of any substitution between types of workers and therefore also limits any resulting wage effect of immigration on U.S.-born workers. Historically, increasing the number of immigrants has had little effect on the educational mix of U.S. workers.

Figure E

Wages are, of course, not solely determined by a simple model of college and noncollege labor, and the education distribution of immigrants is not completely identical to U.S.-born workers. There are relatively more foreign-born workers with no high school degree, but also relatively more foreign-born workers with advanced degrees. But adding more real-world complications to a simple model of wage determination also further limits the scope of immigration’s effect on the overall wages of U.S.-born workers.

For example, immigration to a particular area also increases demand for goods and services. As immigrants purchase food and housing, they raise the overall demand for labor, offsetting the potential wage decreases induced by increased labor supply. In addition, investment will eventually respond to the increased demand and also to the increased population growth that raises the relative productivity of physical capital. Once capital can adjust to the change in population, there is little to no long-run effect of immigration on average wages (Card 2012).

The naïve story about college and noncollege workers above implicitly assumes that at least within education categories, foreign-born workers are essentially identical to U.S.-born workers. However, empirical evidence like Peri and Sparber’s (2009) makes clear that even within education categories, U.S.- and foreign-born workers often have different skills and perform different tasks at work.

It is also useful to point out here that the idea or claim that most immigrants work in low-wage jobs is inaccurate. As depicted in Figure F, Kallick and Capote (2023) found that among persons who work full-time and year-round, the share of immigrants with higher-wage jobs that pay more than twice the median earnings level (17%) is the same as the share of U.S.-born workers. They also found that 65% of immigrant workers who work full-time and year-round earn at least two-thirds of median earnings.

Figure F

Immigrants may be overrepresented in some jobs and underrepresented in others, but the difference between the U.S.- and foreign-born shares is rarely as dramatic as is often assumed. Immigrants are strongly represented in some high-wage jobs and play a significant role in many middle-wage jobs. For example, 24.6% of dental, nursing, and health aides are immigrants, as are 37.7% of computer software developers—well above immigrants’ 18.6% share of the labor force.8 While immigrants are overrepresented in low-wage occupations, immigrants are a significant part of the top, middle, and bottom of the economic ladder in the workforce.

A final reason the wage impact of immigration is smaller than one might expect is that recent immigrants are very geographically mobile. Basso and Peri (2020) show that newly arrived immigrants with less than ten years in the United States are more likely to move across states and different labor markets than U.S.-born workers. When employment demand in a local labor market is declining (or increasing), recently arrived immigrants are much more likely than native-born workers to move away from (or into) that locality. Because of this exit option, employers may have somewhat less leverage over immigrants than they would otherwise. Recently arrived immigrants therefore reduce the monopsony and wage-setting power employers have over immigrants and also U.S.-born workers who are viewed as substitutes.

At the same time, research has shown that some incumbent foreign-born workers are more likely than U.S.-born workers to be substituted by employers for newly arrived immigrants. As a result, the wage effects of increased immigration on previous immigrants may be more negative. Ottaviano and Peri (2012) found that while immigration had small positive effects on the wages of U.S.-born workers, it had negative effects on the wages of foreign-born workers. Policies should thus be implemented to ensure that wages are not degraded for immigrants who have already resided in the United States for many years.

Immigration, taxes, and public spending

The preponderance of the research and evidence shows that immigration does not create a fiscal burden. Differences between the U.S.-born and immigrant populations in the taxes they pay and the public services and benefits they receive improve the fiscal position of the federal government while, to a lesser extent, putting financial pressure on state and local governments. As detailed in the 2017 NASEM report and related research, the net effect is positive over the lifetimes of immigrants, their children, and grandchildren (NASEM 2017; Orrenius 2017).

The fact that immigration is a boon to the federal government’s fiscal balance but a strain on state and local governments’ balances can be addressed by increasing the federal share of jointly funded programs. Since the federal government already transfers nearly a trillion dollars annually to state and local governments through numerous grant programs, these transfers could simply be made more generous. A perhaps bigger challenge is that the short-term fiscal impact of immigration can be negative even if the long-term impact is positive.

Relative to U.S.-born residents at equivalent income levels, immigrants put less strain on other taxpayers because they have high employment rates and do not qualify for many public services and benefits. Further, immigrants have little or no impact on the cost of national defense, foreign aid, interest on the national debt, and similar expenses known as public goods. However, if an equal share of these costs is assigned to immigrants, the NASEM report found that they and their children and grandchildren will pay less in taxes than they will receive in services and transfer payments over their lifetimes, as is also true of U.S.-born residents (keeping in mind that the United States consistently runs a budget deficit in the consolidated government sector).

This does not mean that immigration is costly to U.S.-born residents, however. To the contrary, immigrants will pay more in taxes than the additional public spending they generate. That is, immigrants’ net contribution is enough to reduce the taxes paid by others—just not enough to cover a proportionate share of public goods like national defense and interest on the debt (NASEM 2017; Orrenius 2017).

Most immigrants to the United States arrive as adults, and some 30–40% later return to their countries of origin, often following a drop in earnings (NASEM 2017; Migration Policy Institute 2022; Akee and Jones 2024). As a result of these patterns of migration and return migration, much of the cost of immigrants’ upbringing and education, and some of the cost of supporting them in disability or old age, are borne by their families and the governments in their countries of origin. More of their time spent in the United States, meanwhile, is during prime working-age years when people tend to pay more in taxes than they receive in public services and benefits.

Immigrants have somewhat larger families than U.S.-born residents, though not large enough to stem a decline in the under-20 U.S. population in recent years (SSA 2023). In 2017, immigrant women were estimated to have 2.18 children over their lifetimes, versus 1.76 for U.S.-born woman (Peri 2020). Given the low U.S. birth rate, immigration itself and larger immigrant families were the only factors preventing the U.S. population from shrinking.

As discussed elsewhere in this report, immigrants are more likely than U.S.-born residents to lack high school degrees, but also more likely to have advanced degrees and to work in STEM fields (“STEM” refers to science, technology, engineering, and mathematics). Research shows that immigrants tend to be highly innovative and entrepreneurial, and their presence boosts productivity growth. For example, Hunt (2011) found that immigrants were roughly twice as likely as U.S.-born residents to be granted patents, including patents that are licensed or commercialized, and Peri (2012) found that productivity is higher in areas that attract immigrants based on historical settlement patterns and distance from the Mexican border (factors that are used as statistical instruments to differentiate between higher productivity attributable to immigrants versus immigrants being drawn to higher-productivity areas).

While lifecycle patterns tend to benefit countries attracting immigrants, economic migration of less-educated immigrants from poorer to richer countries can increase spending on means-tested benefits. This may be less true in the United States than in many other destination countries, however, because the United States has a less generous safety net than most wealthy countries and because many immigrants to this country are ineligible for, or dissuaded from, accessing benefits (Lacarte, Gelatt, and Podplesky 2024). Moreover, people who take on the costs and risks of migration, even if they hail from countries with lower educational attainment, tend to have a high capacity for gainful employment, often supporting less healthy or skilled family members back home (Feliciano 2020).

Immigration of working-age adults slows the rise in the age dependency ratio—the size of the population 65 and older relative to that of the prime working-age population. Immigrants’ relatively high birth rate, conversely, slows the decline in the youth dependency ratio—the under-20 population relative to the working-age population. The net effect on the working-age share of the U.S. population is positive, though the impact on public finances also depends on the relative cost of different programs.

Generally, faster population growth will tend to increase the tax rates needed to fund education and other programs that primarily benefit young people while reducing the tax rates needed to fund Social Security, Medicare, and other programs that primarily benefit older people. In addition to the relative cost of these different programs, timing matters. As a result of the decline in birth rates that followed the post-World War II baby boom, the United States, like many countries, has an aging population, such that funding Social Security and Medicare is a bigger challenge than funding education. This demographic imbalance is even more acute in other countries, notably Japan, which admits too few immigrants to allow a shrinking workforce to support its aging population.

Age, family size, and earnings potential are not the only factors affecting the net impact of immigration on public finances. For example, recent research indicates that unauthorized immigrants pay just under $60 billion per year in federal taxes and over $37 billion in state and local taxes (Davis, Guzman, and Sifre 2024). Yet many benefits, notably Medicare, are not accessible if an immigrant returns to their home country even if they are naturalized U.S. citizens or were legal residents who had paid into the program and would have been eligible for benefits if they had remained in the United States. Other benefits, such as Social Security, are portable but are not available to unauthorized workers. Though some means-tested benefits are currently available to unauthorized immigrants—a few states extend health and other benefits to low-income families regardless of legal status—take-up by immigrants is low since many fear it could jeopardize their ability to remain in the country or apply for citizenship (KFF 2023).

Millions of unauthorized immigrant workers contribute to Social Security using numbers that are not their own. This reduces Social Security’s projected shortfall but does so at the expense of a disadvantaged population. Immigration reform that allows immigrants to fully participate in society would help future immigrants receive the benefits to which they contribute. It could also potentially help some immigrants access benefits that can be linked to past contributions through Individual Taxpayer Identification Numbers (ITINs).

As discussed elsewhere in this report, immigration reform that includes a broad regularization for those who lack an immigration status would increase earnings, reported earnings, and the taxes paid on these earnings, including contributions to social insurance programs such as Social Security and Medicare. However, the net impact on these contributory programs as well as some programs funded by general revenues, such as Medicaid, could be negative in the short run if immigrants who paid into these programs gain access to benefits. Factoring in both increased revenues and increased costs, however, the Congressional Budget Office estimated in 2013 that legislation that would have significantly increased the number of immigrants living and working legally in the United States would have decreased federal budget deficits by nearly $200 billion over a 10-year period (Elmendorf 2013).

In the long run, society generally gains when workers have full legal rights and protections and are able to achieve their full potential. However, there is always the potential that a large influx of new migrants or immigrants earning low wages, if they were eligible for most government benefits, could tilt the balance toward a negative fiscal impact in the short run. This should not be the only consideration, however, in decisions about whether to welcome or extend government benefits to vulnerable immigrants, just as humanitarian policies should not be expected to pay for themselves. It is worth noting, in any case, that even in the case of refugees and asylees who tend to have low educational attainment and are eligible for the same government services and benefits as U.S. citizens, a recent report found that their fiscal impact is positive in the long run (Ghertner, Macartney, and Dost 2024).

Studies that find negative fiscal impacts from immigration of lower-wage workers typically ignore dynamic effects on prices, productivity, and profits that improve U.S. residents’ living standards. Correcting for one of these omissions, a study accounted for increased tax revenue from higher profits attributable to immigration and found that the net fiscal impact of immigrants without college degrees changed from negative to positive when this revenue was taken into account (Clemens 2022).

Immigrants’ positive fiscal impact reflects their strong motivation and ability to improve their lives. Uprooting one’s life to travel to a new country to live and work is costly and risky, so people with a higher-than-average earning potential are more likely to emigrate, even if their formal education is limited (Feliciano 2020). As a result, after an initial adjustment period, immigrants tend to have high employment rates compared with the general population. They also tend to be more geographically mobile in response to economic conditions within the United States, boosting productivity growth by better matching workers and jobs (Basso and Peri 2020).

The success of immigrants extends to their families and can have spillover effects on the general population. The children of immigrants, most born in the United States, are more likely to have college degrees and are less likely to be poor than other Americans (Pew 2013). The presence of foreign-born and second-generation students has a positive effect on the academic achievement of other students, especially those from disadvantaged backgrounds (Figlio et al. 2024).

Immigration and housing affordability

One of the most pressing issues regarding immigration and housing in the current moment is providing adequate and humane shelter for the significant number of migrants who have entered the United States in recent years—many of whom are applicants for asylum who are awaiting a hearing. These migrants often have steep barriers to finding employment or other means of earning income, and they have had to rely on already inadequate systems to help the unhoused population in the United States.

In the shorter term, there are a range of stopgap and ameliorative measures that could be taken to provide humane housing options for recent arrivals. Solf, Guerrero, and Sherzad (2024) outline the scope of the problem and offer some short-term solutions, and a report by Eikenberry and Obser (2023) for the Women’s Refugee Commission provides an in-depth discussion about the key challenges and offers best practices for managing and housing large numbers of new arrivals at the federal, state, and local levels. 

In the longer run, the larger issue of homelessness in the United States would be greatly alleviated if measures that broadly reduced the cost of rent and homeownership were implemented. Given existing shelter capacity for the unhoused population, policies that increased housing affordability in the long run would also create more slack that could help absorb temporary surges in demand for short-term shelter. 

Further, the issue of housing affordability is not just an issue of people becoming unhoused; it has become a prime concern for working families up and down the income scale in the United States. Since the mid-1990s, housing cost inflation has outpaced overall price inflation significantly, particularly in a subset of major cities that provide promising employment opportunities.

The root cause of the housing affordability crisis is simply that housing supply has not kept up with rising demand, especially in areas experiencing strong economic and population growth. Essentially, new housing supply is not responsive enough to increased demand, so demand surges mostly are resolved with higher prices rather than greater housing output.9

This unresponsiveness of housing supply to demand surges means that any demand change can cause distress in local housing markets. For example, during the pandemic, millions of people realized they would be able to work from home (at least partially) for an extended period of time. This led to a strong change in preferences regarding what they wanted from a home. More space to accommodate home offices was desired, while the benefits of a short commute eroded. These changing preferences led to a surge of population away from city centers and into surrounding suburbs, and, to a lesser extent to increased intercity movements. Given the unresponsiveness of housing supply generally, this sharp change in demand patterns led to a historically large increase in housing costs—even in years when immigration from abroad dropped sharply.10

The constraints on housing supply are varied, but a major factor is restrictive land-use policy, generally set by local governments. The policy agenda to alleviate these constraints is multifaceted and beyond the scope of this paper but suffice to say that such a policy agenda would provide huge relief to U.S.-born residents if it was enacted  regardless of the level of immigration flows.11 As we noted before, housing demand shocks can (and do) happen constantly, even during times of low immigration flows (like during 2020 and 2021), and they lead to sharp affordability problems.

It is certainly true that given our dysfunctional housing supply side in the United States, any source of housing demand will put upward pressure on housing costs. Both long-term immigrants and new migrants—like U.S.-born residents—need housing and hence, higher levels of immigration will boost housing demand. Under our current housing supply-constrained regime, this does put some upward pressure on housing costs. Saiz (2007) and Mussa, Nwaogu, and Pozo (2017) find that an increase in immigration equal to 1% of the existing population leads to an increase of roughly 1% in housing costs. However, as we showed earlier, increased immigration flows in recent decades in the United States have largely not even made up for declining growth in the U.S.-born population. If our housing supply regime in the United States cannot even provide affordable housing for a population whose overall growth is steadily decelerating, it seems there is clearly a more important issue to address.

Another key issue is the difference between the cost of housing services and the price of homes. Generally, economists looking to measure inflation or the burden of housing costs use measures of housing rents. These rents are seen as the consumption value of housing services. Home prices bundle the cost of housing services as a consumption good with the cost of purchasing an investment good (a home is a valuable asset). In a country like the United States in which most families live in a home that they own, this raises some measurement challenges. But aside from this, it also introduces some distributional conflict regarding housing costs. For a family that must rent a home, rising rents are a pure cost burden that lowers consumption possibilities on nonhousing items. For a family that owns a home, rising rents are an implicit income that drives up the value of their assets and hence their net worth. Additionally, because most homeowners either own their home outright or have fixed-rate mortgages that pin down their monthly payment, the only significant effect of rising rents they face is a rising value of their home. In short, anything—including higher levels of immigration—that raises rental costs for housing also raises the value of homeowners’ wealth. Given that homeownership rates are substantially higher among U.S.-born residents than foreign-born households, rising rents stemming from higher immigration flows are a direct transfer from immigrant households to native households.

There is one final important intersection between immigration and housing: Immigrants are exceedingly overrepresented in the construction workforce, and particularly so in residential construction (see Bivens 2014). They are a key source of skilled labor for this sector. Policy changes that sharply slowed (or reversed) immigration flows could quickly reduce the labor supply in construction, restrict housing output, and raise the cost of building new homes. Howard, Wang, and Zhang (2024) examined the staggered rollout of a national immigration enforcement increase to see if it affected costs and output in residential construction. They found that counties that saw greater immigration enforcement effort saw reductions in the residential construction workforce, fewer homes built, and higher prices, all consistent with immigration enforcement suppressing labor supply in this sector. 

Of course, one could argue the same thing about any policy change that raised wages in construction—and part of our policy recommendations includes measures (like providing a path to citizenship) that would raise wages for immigrants in construction. However, the rise in wages stemming from boosting immigrants’ bargaining power vis-à-vis employers (by regularizing their legal status) would largely come from a pure redistribution from business income. Employers’ power in labor markets—particularly relative to unauthorized immigrant workers—means they can impose a “markdown” on wages and earn supernormal profits.12 Curtailing this power will overwhelmingly show up in lower supernormal profits and not in higher construction prices. Further, evidence has shown that the higher wages that accompany normalizing immigrants’ legal status are often “financed” largely through higher productivity. With more secure, permanent, and lawful immigration status, immigrants are incentivized to invest in productivity-enhancing training and education and are freer to enter occupations with higher productivity that require more stringent scrutiny of their immigration status. 

The direct effect of reduced labor supply, on the other hand, is to immediately lower construction output, and this would almost surely raise home prices.

All in all, it is mechanically true that any population growth puts stress on the nation’s dysfunctional system of housing supply. Recent decades have seen very little change in overall population growth, so the inability of the nation’s housing supply sector to accommodate this (slow and unchanged) growth without creating affordability problems is glaring. It is this dysfunctional supply side of the housing market, not anything—including immigration—on the demand side that should be the focus of policymakers. Further, the overall effect of immigration on U.S. housing should take into account the effects of immigration flows on the housing wealth of U.S.-born residents and on the construction industry’s ability to produce new homes at reasonable costs. Tallying these up leads to very little reason to think that increased immigration flows are a first-order challenge in U.S. housing policy. 

Immigration and long-run growth

The size and growth rate of aggregate GDP will obviously be directly affected by immigration in coming decades. But when economists talk about growth, they mostly are referring to growth in per capita GDP or even growth in productivity—the amount of output generated in an average hour of work in the economy. Productivity growth is the basis of rising living standards—the level of productivity (and not the level of aggregate GDP) is what determines if a country is rich or not. Bangladesh, for example, has an aggregate GDP roughly 4 times larger than Norway, yet it is Norway’s tenfold advantage in GDP per capita that makes it rich relative to Bangladesh.

Accordingly, when economists assess the effect of any influence—including immigration—on economic growth, they are mostly talking about the effect on productivity growth. It is true that if immigrants are younger than U.S.-born residents, that an influx of immigrants can boost per capita GDP simply by virtue of having higher employment rates. But for most of the following section, we will abstract from that when discussing the literature on immigration and growth (the greater employment rates of immigrants, however, are of first-order importance in their effect on the nation’s fiscal balances, as we discuss in a later section).

Theories of economic growth have enough variety and complexity that it is nearly impossible to make firm predictions about the effect of immigration on this long-run growth, but we will note just a few things in this section.

First, in all growth models that posit a negative relationship between growth in per capita income and population growth (including immigration), the influence of faster population growth that slows per capita income growth is its effect in suppressing growth in the capital-to-labor ratio. But the evidence over business cycles shows little relationship between slower growth of the capital-to-labor ratio and immigration flows in the U.S. data. The scatterplot in Figure G shows the change in the overall capital-to-labor ratio in the U.S. economy over various decades, along with the change in the immigrant share of the population. If the worries about immigration and growth are to be sustained, there should be a clear negative relationship: A higher immigrant share should be associated with a smaller increase in the capital-to-labor ratio. But there is no obvious relationship (if anything, it appears to be slightly positive). This is not dispositive evidence, obviously, but it does highlight once again that rising shares of immigrant workers in the U.S. economy are mostly matched by declining growth rates of the native-born population and labor force.

Figure G

Further, richer models of economic growth consider other determinants of growth besides just the rate of physical capital accumulation. Many models highlight that human capital (the skills and training of the workforce) not only matters for growth but also generates externalities (positive economic returns that are not fully captured by the workers creating them). Given this, immigration that increases human capital in the destination country might actually boost per capita income growth.

A related (but not identical) idea is that today’s economic growth depends on the existing stock of ideas (including ideas from decades or even centuries ago). If the total stock of ideas is related strongly to the simple size of the population, then rising population can be an aid to growth rather than a drag. Because ideas are often fluid across national borders, the link between a larger population and faster growth might hold more tightly at the global rather than the national level, but if any parts of idea formation are nationally “sticky,” then it might hold at the national level as well (and patents and copyrights and other forms of intellectual property protection probably do make ideas’ economic benefits “sticky” in the country in which they are formed, for good or for ill).

A recent paper from Bernstein et al. (2022) highlights the disproportionate success of immigrants in patenting new inventions from the period 1990–2016. Immigrants constituted 16% of inventors applying for patents in the United States, but accounted for 23% of the total patents, and 25% of the most cited patents (a proxy for the importance of the innovation).

Sequeira, Nunn, and Qian (2017) provide examples of how past immigration flows into the United States brought ideas that surely contributed strongly to growth:

One example of such an innovation is the suspension bridge, which was pioneered by John A. Roebling, a German-born and trained civil engineer, who built numerous suspension bridges, including the Niagara Fall Suspension Bridge and the Brooklyn Bridge (Faust, 1916, p. 10). Other notable engineers include… John F. O’Rourke, an Irish engineer, who built seven of the tunnels under the East and Hudson Rivers, and six of the tunnels of the New York subway systems (Wittke, 1939, pp. 389–390). Another example is Alexander Graham Bell …[who] developed an acoustic telegraph…. Other notable inventors include David Thomas (Welsh), who invented the hot blast furnace; John Ericsson (Swedish), who invented the ironclad ship and the screw propeller; Conrad Hubert (Russian), who invented the flashlight; and Ottmar Mergenthaler (German), who invented the linotype machine (Kennedy, 1964, pp. 33–34). Immigrants also made important contributions to the educational system of the United States. For example, the concept of kindergarten, which has been shown to have had important economic effects, was brought to the United States by German immigrant Friederich Fröbel (Paz, 2015, Ager, Cinnirella and Jensen, 2016). Recent research by Paz (2015) finds that the presence of kindergartens during the kindergarten movement (1890–1910) resulted in an average of 0.6 additional years of total schooling by adulthood and six percent higher income…The State University system… was modeled after the Prussian… system… (Faust, 1916, pp. 10–11). 

These crosscutting theoretical effects of immigration on growth are likely why the empirical estimates of immigration on growth yield mostly neutral results. For example, Borjas (2019) uses cross-state estimates of income growth and immigration flows and finds largely neutral results for the causal effect of immigration flows on per capita income growth. Kane and Rutledge (2018) also use cross-state data to assess the link between immigration flows and growth. In the raw data, there is an unambiguous positive correlation between immigration flows and state growth, but some of this is due to immigrants settling in states with strong economies. Using techniques to isolate the causal effect of immigration flows on per capita GDP growth, they find modestly negative growth effects, but no effect on levels, which they interpret as indicating that the negative growth effects are transitory. Using historical data, Sequeira, Nunn, and Qian (2017) find that past waves of immigration in the United States had significantly positive short- and long-run causal effects on economic growth.

Finally, it is worth noting again that the possible win-win economic effects of immigration are often squandered by bad immigration policies and a system that doesn’t work well enough for workers. It is very likely that the returns to innovation and human capital that would accrue to many immigrants within the United States are artificially suppressed because the immigrant workers would not be able to claim the same benefits from innovation that U.S.-born workers might. If a U.S.-born worker has an idea for a great invention, they can start their own company and retain full control over the resulting patent. Many highly skilled and educated temporary migrants must remain attached to a specific employer as a condition of residence and employment in the United States, and this means that they may see their employer able to claim many of the benefits of their innovation instead of getting it themselves.

The artificially created labor market monopsony facing many temporary migrant workers in the United States almost surely stifles many potential innovations. Take a couple of recent examples: One of the key inventors of the mRNA vaccines was Katalin Karikó, a Hungarian citizen working in the United States on an academic visa. When Karikó tried to change jobs and move to another lab with better working conditions, her academic adviser threatened to have her deported. The resulting legal entanglements led to the offer from the other lab being withdrawn (Shrikant 2023). Another example is Sandeep Maganti whose situation is detailed in a recent Bloomberg investigative news report on the H-1B visa program. Maganti is a software engineer who, while on a student visa, created a business that grew “into a million-dollar company that employed six people” (Fan et al. 2024). U.S. “visa rules wouldn’t allow Maganti to sponsor himself” for an H-1B visa so that he could remain and run his company. Without a viable path to a temporary H-1B visa—a visa that would require Maganti to be sponsored by an employer and become an employee—he thus “had to sell his stake and stop working at his own company” (Fan et al. 2024).

Immigration pathways and statuses in the United States

While the previous sections of this paper have focused on the migrant and immigrant population and what existing research says about their impact on the economy, wages, and housing, it has so far only briefly mentioned how people arrive to the United States from abroad, and how those pathways and the resulting immigration statuses—or lack thereof—impact the economic outcomes of immigrants themselves and labor standards for all workers writ large. In order to provide some background, this section summarizes the broad immigration statuses for migrants and immigrants in the United States, and the different rights they have depending on their particular pathway and status. The following section will focus on why the differences in immigration status matter to the economy and workplaces across the country.

Pathways into the United States and immigration statuses

There are multiple pathways that persons abroad take to arrive in the United States. The following is a brief summary:

Green cards: One of the available lawful pathways includes adjusting to lawful permanent resident (LPR) status—also commonly referred to as obtaining a permanent immigrant visa or “green card”—which can be through the family-based (FB) or employment-based (EB) preference categories, through a humanitarian pathway such as a refugee resettlement program or by qualifying as an asylee, through the Diversity Visa (DV) lottery, or through one of the lesser known narrower categories, which account for a small share of total green cards. 

The total number of green cards granted averaged just over one million per year between 2003 and 2022 (1,021,966). One thing to keep in mind with respect to this total number is that not all green cards granted go to newly arriving immigrants from abroad; in most years, half to two-thirds of green card recipients were already present in the United States in some other status and adjusted to lawful permanent resident status.13 Persons with green cards are eligible to apply for citizenship after five years, or three years if they have been married to a U.S. citizen. In terms of employment, green card holders may work for nearly any employer, except for positions that require citizenship, and may change jobs or employers without requiring authorization from the U.S. government.

Temporary, nonimmigrant visas: The other major pathway into the United States is as a temporary visitor, student, trainee, diplomat, exchange visitor, or employee acquiring a “nonimmigrant” visa or status.14 Nonimmigrant visas and statuses are temporary, meaning that the foreign-born person to whom the visa is issued must depart the United States after the visa expires unless they adjust to LPR status by acquiring either one of the green cards described or another valid nonimmigrant status. Many nonimmigrant visa classifications authorize the visa or status holder—sometimes referred to as the visa beneficiary—to be employed in the United States. Employed nonimmigrants are also often referred to as temporary foreign workers, temporary migrant workers, or guestworkers, but no one definitive term has been agreed upon (for the purposes of this report, we refer to them as temporary migrant workers). Most temporary migrant workers are only permitted to work for one employer, the employer that sponsored their visa. In 2023, nearly 1.8 million new temporary work visas were issued to temporary migrant workers and their accompanying family members.15 The total number of temporary migrant workers in the United States who are employed in a given year is estimated to be just over 2 million (because some temporary visa statuses can last for more than one year) (Costa 2021).

Humanitarian pathways (refugees, asylees, and asylum seekers): As noted above regarding green cards, some persons obtain green cards through the U.S. refugee resettlement program or after being granted asylum (a person granted asylum is referred to as an asylee). The refugee pathway comes from the United Nations Refugee Convention of 1951, which defines a refugee as someone fleeing his or her country because of persecution or “owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside of the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country.” In the United States, the Immigration and Nationality Act (INA), as amended by the Refugee Act of 1980, implements the Refugee Convention and its 1967 Protocol into U.S. law, and authorizes and governs the admission and resettlement of refugees into the United States.16 The number of refugees admitted to the United States in 2023 was just over 60,000 (Migration Policy Institute 2024). Refugees are permitted to work for any employer, just like green cards holders (except for positions that require U.S. citizenship).

Asylum is another form of humanitarian legal protection for persons who have been forcibly displaced and fear harm and persecution, similar to the protections available to those who qualify as refugees, although under U.S. law, there is a separate process for persons who apply for asylum (i.e., asylum seekers or asylum applicants) and those applying for refugee status. Like refugees, asylum seekers must meet the definition of a refugee; however, a major difference between applying for refugee status or asylum is that to apply for asylum, the applicant must already be in the United States or apply for asylum at an official U.S. port of entry, rather than applying from abroad as refugees do.17 Successful asylum applicants who become asylees are eligible and on the path to a green card. However, processing of asylum claims through the relevant government agency or in immigration court often takes years before they are adjudicated.

Asylum may also be claimed as a defense to deportation for persons who are detained by immigration authorities. During that time, asylum applicants (aka asylum seekers) do not have a formal immigration status but are authorized to remain in the United States while their claim is adjudicated.

Asylum seekers are authorized to work in the United States, but they are subject to applicable rules and wait times. They must first be issued an Employment Authorization Document (EAD) from U.S. Citizenship and Immigration Services (USCIS), but may only file for an EAD 150 days after they have applied for asylum, and then become eligible to receive an EAD once their asylum application has been pending for a total of 180 days. There were an estimated 1.5 million asylum seekers employed in the United States in 2023.18

Irregular migration: The other pathway involves migrants who are present in the United States but who do not have an authorized immigration status; such individuals are sometimes referred to as unauthorized, undocumented, or irregular migrants (being in an irregular status), and sometimes (derogatorily) called illegal migrants or illegal immigrants. Unauthorized migrants either entered into the United States without inspection by the appropriate authorities—often referred to as entering without inspection (EWI) in government documents and data—and may have done so in a clandestine manner. Other unauthorized migrants may have originally entered the United States lawfully with a nonimmigrant visa or through the Visa Waiver Program and after an inspection by government authorities, but then lost their temporary immigration status. The loss of status may have occurred because of a violation that led to the revocation of their visa or status or the simple expiration of a nonimmigrant visa that was temporarily valid for a set time, usually a period of authorized travel, employment, or study.

There are a few existing estimates on the number of persons in the United States who lack an immigration status, the most recent of which estimate the population at between 11 and 11.3 million as of 2022.19 The vast majority of the unauthorized migrant population is settled, with 79% being long-standing residents who arrived in the United States in 2010 or earlier (Baker and Warren 2024). Unauthorized migrants are not authorized to be employed in the United States, but an estimated 8.3 million were employed in the U.S. labor force in 2022, accounting for just under 5% of the total U.S. labor force (Passel and Krogstad 2024).

Precarious immigration statuses: There are a few types of temporary protections or quasi-statuses, which are not technically immigration statuses and which do not directly lead to any form of permanent immigration status. For example, while unauthorized immigrants lack a formal immigration status, some may nevertheless be in an authorized period of stay in the United States in which they are temporarily “lawfully present,” which can occur if they have qualified for some sort of temporary immigration relief like deferred action, humanitarian parole, parole-in-place, or Temporary Protected Status (TPS).

Because of their temporary nature, we refer to them as precarious statuses and other analysts like the Migration Policy Institute have referred to them as “twilight,” “liminal,” or “limbo” statuses.20 Persons may either enter the United States with one of these precarious protections already in place—for example if they qualified for the parole program for Cuba, Haiti, Venezuela, and Nicaragua—or they may have entered without inspection and later qualified for TPS, deferred action, or parole-in-place (a form of parole granted to persons who are already in the United States). Persons with these precarious statuses are usually eligible to obtain an Employment Authorization Document (EAD), also known as a work permit, which allows them to work lawfully. There were an estimated 2.8 million persons in precarious statuses as of early 2024.21

The U.S. foreign-born population by immigration status

The pathways described above, and the resulting statuses, are generally categorized into four major categories of immigration status for immigrants (or lack thereof): Either immigrants have obtained lawful permanent residence (having a green card), or they have naturalized, meaning they have become U.S. citizens, or they have a temporary lawful nonimmigrant status, or they lack a lawful immigration status. For the purpose of population estimates—in part because of data limitations—the migrants who have qualified for some sort of precarious temporary immigration relief like deferred action, parole, or TPS, or who are pursuing an asylum claim, are counted as unauthorized migrants or immigrants, even though they technically have a form of authorized stay.22

In terms of the shares of persons in these statuses, according to an analysis done by the Pew Research Center (represented in Figure H), as of 2022, roughly half of all immigrants (49%) were naturalized U.S. citizens, one-quarter (24%) were lawful permanent residents, 4% were temporary lawful residents, and 23% were unauthorized immigrants (lacking an immigration status or having a precarious form of temporary protection) (Passel and Krogstad 2024).

Figure H

The impact of immigration status on wages and workplace rights

All persons in the United States have—at least on paper—basic labor and employment rights under U.S. law, which in theory, should protect them from lawbreaking employers. However, the extent to which those rights are able to be exercised, and the extent to which they are enforceable in practice, depends very much on immigration status because of the power that employers have over workers vis-à-vis that immigration status and because of how employers can exploit that power. This section briefly discusses the connection between immigration status and workplace rights and the existing research on the impact that immigration status can have on wages and labor standards.

Naturalized citizens

All persons who become naturalized citizens enjoy all of the same labor and employment rights as native-born U.S. citizens, including being able to work for any employer and in any position—(except they lack the ability to become elected to the office of president of the United States). Naturalized citizens are not subject to removal from the United States, which is a fear that many migrants and immigrants in other statuses have (unless they are found to have committed fraud during the naturalization process). Naturalized citizens also have access to the same social safety net benefits as native-born citizens and are the only group of immigrants that can vote in federal elections—meaning they are allowed to fully participate in political life.

These rights, benefits, and protections—in particular, not fearing deportation—result in better economic outcomes, including higher wages, than persons in other immigration statuses or compared with persons who lack an immigration status. Below are just a few examples of research showing the impact of naturalization on economic outcomes.

Peri and Zaiour (2021) modeled a number of scenarios for a legalization program that would provide status to unauthorized immigrants, to all or a subset of them, depending on certain characteristics. In their first scenario, which would consist of a legalization and path to a green card and citizenship for all unauthorized immigrants, they found that in the short run—meaning five years, in which unauthorized immigrants would have a lawful immigration status but would not yet become naturalized U.S. citizens—workers who were previously unauthorized would see a wage gain of 10%, amounting to $4,300 per year. Peri and Zaiour then assume that all those who gained an immigration status would become naturalized citizens after five years—and in the long run, meaning five to ten years—the newly naturalized immigrants would see an increase in their annual wages of 32.4%, amounting to an increase of $14,000 per year. They also found a positive wage impact beyond just the workers who naturalized, estimating that all other workers would see an annual increase in wages of 1.1%, or $700. Their estimate on the broader impact of a pathway to citizenship was that it would increase gross domestic product by up to $1.7 trillion over the next decade and create hundreds of thousands of new jobs.

Lynch and Oakford (2013) estimated the wage gains for both legalization as a first step and the additional wage gains that would follow after naturalization. They estimated first, the same increase in income that unauthorized immigrants saw from legalization after the Immigration Reform and Control Act (IRCA), as measured by the U.S. Department of Labor, of 15.1% after five years (see more discussion in the next section on green cards). They then calculate the effect of moving from legalization to citizenship, estimating an additional 10% increase in income after acquiring citizenship. In total, they estimate that “the full effect of granting legal status and citizenship to unauthorized immigrants is an income gain of 25.1%. Of this boost in income, about three-fifths comes from legalization and about two-fifths is attributable to transitioning from legal status to citizenship.”

Pastor and Scroggins (2012) also found that citizenship would boost individual earnings of workers by 8% to 11%, “leading to a potential $21–45 billion increase in cumulative earnings over ten years that will have ripple effects on the national economy.”

Shierholz (2010) observed a difference between foreign-born adults who had become naturalized citizens and those who had not become naturalized, in terms of both economic outcomes and poverty rates. Naturalized citizens in 2007 had a median annual family income that was very similar to that of native-born U.S. citizens, even earning slightly more than the native-born. But noncitizen immigrants had an annual median income that was 33.2% below that of naturalized citizen immigrants. In terms of poverty rates, they were much higher for noncitizen immigrant adults: In 2007, the poverty rate for noncitizen immigrants was 20.0%, more than twice the 9.2% rate for naturalized citizen immigrants and the 9.8% rate for native-born U.S. citizens. The poverty rate difference was starkest for Latinos, with the poverty rate for naturalized citizen Latinos being 4.3 percentage points lower than the poverty rate of Latino noncitizens.

Sumption and Flamm (2012) made very similar findings, showing that “naturalized citizens earn more than their noncitizen counterparts, are less likely to be unemployed, and are better represented in highly skilled jobs.” They further found that:

Most of the gap between citizens’ and noncitizens’ outcomes is explained by the fact that naturalized immigrants have higher levels of education, better language skills, and more work experience in the United States than noncitizens. Even after accounting for these differences, however, there is some evidence that the naturalized may earn a wage premium of at least 5 percent.

Green cards

All persons who obtain green cards—whether through the family-based, employment-based, humanitarian, or Diversity Visa pathways—enjoy nearly all of the same labor and employment rights as U.S. citizens, including being able to work for any employer and in any position except for those that explicitly require citizenship. For the most part, green card holders are not subject to removal from the United States, although unlike naturalized citizens, they can be removed if they commit certain crimes that make them deportable or if they are deemed to have abandoned their permanent resident status.

Under current federal law, green card holders must wait five years before using most public support and social insurance programs. Green card holders are not eligible to vote in federal elections, although some jurisdictions allow them to vote in local elections for positions like school boards and city council seats and to run for some local elected offices.

Being able to work lawfully without fearing deportation results in better economic outcomes for green card holders, including higher wages, relative to persons in other lawful immigration statuses or compared with persons who lack an immigration status. Below are just a few examples of research showing the impact of naturalization on economic outcomes. (Estimates of the economic benefits obtained by green card holders depend on their status before they obtained their green cards. The two types discussed here are persons who were previously on a temporary work visa and persons who previously lacked an immigration status.)

One example of research assessing the economic and workplace impact of having a green card compared with a temporary work visa comes from Mukhopadhyay and Oxborrow (2012), who looked at college-educated migrant workers on temporary visas. They found that “H-1B workers are paid less than native workers” and that the wage gain associated with obtaining an employment-based green card was $11,860 per year. The authors noted that their “result shows that the current process of acquiring a green card gives too much power to employers and hinders job mobility among highly skilled immigrants.”

Apgar (2015) reviewed Mexican Migration Project survey data and compared employment outcomes for Mexican males who were on temporary visas, unauthorized immigrants, and green card holders. Controlling for other factors, Apgar found that the hourly wages of temporary migrant workers were about 11% less than those of immigrants with green cards and that unauthorized immigrant workers earned about 13% less than immigrants with green cards.

There is extensive research on the impact of unauthorized immigrants who regularized their status by becoming eligible for green cards going as far back as the 1990s, including research on the effects of the 1986 legalization program implemented after enactment of IRCA, which was signed into law by President Ronald Reagan. This research is particularly relevant for measuring the impact of obtaining a green card because nearly all migrants legalized under IRCA were eligible to receive a green card after 18 months. The discussion here only offers a small sampling of the existing literature.

Smith, Kramer, and Singer (1996), in a report conducted for the U.S. Department of Labor (DOL), found that after four or five years, the real hourly wages of persons who had their immigration status regularized increased on average by 15.1% by 1992 (13.2% for men and 20.5% for women). Kossoudji and Cobb-Clark (2000) found that, according to the same data set, 38.8% of Mexican men had moved on to higher-paying occupations by 1992.23

Hinojosa-Ojeda (2010) summarizes other related research and notes that “[t]he findings of these researchers vary according to their economic models, but the results show uniformly positive results for IRCA beneficiaries,” including measured wage increases that occurred even after controlling for factors such as education and English proficiency, and broader changes in the economy that might have impacted wages more generally.24

Temporary work visas and temporary migrant workers

Although they are legally authorized to work, temporary migrant workers are among the most exploited laborers in the U.S. workforce because employer control of their visa status leaves many virtually powerless to defend and uphold their rights. The vast majority of temporary migrant workers do not have a path to permanent residence through their nonimmigrant visa, and under current federal law, are ineligible for most federally funded public benefits. 

The first major temporary migrant worker programs in the United States were the Bracero programs, which were negotiated as bilateral agreements between the United States and Mexico in 1917 and 1942 (Martin 2020). Since then, numerous cases of abuse and exploitation of migrant workers employed with temporary visas have come to light through the media, reports from advocates and labor unions, and government audits.25 Many of the abuses occur because of the structure of the visa programs, which have few rules and inadequate protections and oversight by federal or state labor standards authorities.

Temporary migrant workers have good reason to fear retaliation and deportation if they speak up about wage theft, workplace abuses, or other working conditions like substandard health and safety procedures on the job—not because they don’t have a valid immigration status, but because their visas are almost always tied to one employer that owns and controls their visa status. That visa status is what determines the worker’s right to remain in the country; if they lose their job, they lose their visa and become deportable (Bauer and Stewart 2013). That leaves them afraid to speak up and complain to their employer or government authorities if their wages are stolen or other workplace violations take place. This arrangement results in a form of indentured servitude.26 Further, employers can punish temporary migrant workers for speaking out by not rehiring them the following year or by telling recruiters in countries of origin that they shouldn’t be hired for other job opportunities in the United States (effectively blacklisting them).27

Although on paper, temporary migrant workers have labor and employment rights and access to remedies that are on par with those of U.S. workers—including immigrants who are green card holders and naturalized citizens—the specter of retaliation makes it understandably difficult for temporary migrant workers to hold lawbreaking employers accountable through complaints to their employers or to government agencies about illegal conditions like unpaid wages and substandard working conditions. Private lawsuits against employers who break the law are also an unrealistic avenue for enforcing rights for two reasons: First, most temporary migrant workers are not eligible for federally funded legal services under U.S. law, and second, those who have been fired are unlikely to have a valid immigration status permitting them to stay in the United States long enough to pursue their claims in court. Because of the conditions created by tying workers to a single employer through their visa status, temporary work visa programs have been dubbed by some as “close to slavery” or “the new American slavery,” and government auditors have noted that increased protections are needed for temporary migrant workers.28

It is also well established that many, if not most, temporary migrant workers pay hefty fees to obtain their temporary jobs in the United States (CDM 2013 and 2020). Those who are in debt after paying recruitment fees are anxious to earn enough to pay back what they owe and hopefully make a profit, exacerbating their situation and making them even more unlikely to speak up at work when things go wrong on the job.

When it comes to wages, there is abundant evidence that the laws and regulations governing major temporary work visa programs—such as H-2B and H-1B—permit employers to legally pay their temporary migrant workers much less than the local average wage for the jobs they fill.29 For example, in the H-1B program—which has a prevailing wage rule that is intended to protect local wage standards—60% of all H-1B jobs certified by the DOL in 2019 were certified at a wage that was below the local average wage for the specific occupation (Costa and Hira 2020). However, most work visa programs have no minimum or prevailing wage rules at all—and while employers are still required by law to pay temporary migrant workers at least the state or federal minimum wage, that’s often far less than the true market rate, or the local average wage, for the occupation in which they are employed.30 Evidence also exists of rampant and systematic wage theft committed against temporary migrant workers.31

Considering how the wage rules or lack thereof in these programs operate, and the situation workers are left in, perhaps it is no surprise that according to one study, there is evidence that temporary migrant workers in low-wage jobs earn approximately the same wages, on average, that unauthorized immigrant workers do for similar jobs, despite being in a technically lawful status (Apgar 2015). In other words, temporary migrant workers may have little financial incentive to work legally through visa programs since there is a small or zero wage premium to be gained for it—and, in fact, authorized temporary migrant workers can end up worse off economically than unauthorized workers because of the debts they incur through fees paid to recruiters, and the fact that many have no family or social networks to rely on. This can incentivize workers to migrate without authorization, rather than using available legal channels.

In essence, these visa programs give employers dangerous levels of monopsony power over workers—similar to company towns in which an employer has enormous leverage over workers because it is the only employer in town.32 A growing body of research, some of it focused on temporary work visa programs, has shown that even modest amounts of employer monopsony power are corrosive to workers’ ability to bargain for better wages.33 

Unauthorized immigrants

Unauthorized immigrants, who make up nearly 5% of the U.S. labor force, contribute to the economy in vital industries and pay billions in taxes and contributions to the social safety net (Davis, Guzman, and Sifre 2024; Fernández Campbell 2018). But these nearly eight million workers are not fully protected by U.S. labor and employment laws because they lack an immigration status. Unauthorized immigrants are also vulnerable to exploitation by employers because they are largely ineligible for federally funded public support and social insurance programs because of their immigration status (as discussed earlier in this report) (Lacarte, Gelatt, and Podplesky 2024). There are a few narrow exceptions to this rule, which allow immigrants to be eligible for certain benefits regardless of immigration status; those programs include emergency Medicaid (if otherwise ineligible for their state’s Medicaid program), programs that provide immunizations and/or treatment of communicable disease symptoms, and school breakfast and lunch programs.34 Under federal law, all children have equal access to public education at the elementary and secondary level regardless of their immigration status or the status of their parents. 

On paper, unauthorized immigrants have labor and employment rights and access to remedies that are on par with those of U.S. workers35—including immigrants who are green card holders and naturalized citizens—but the specter of retaliation makes it understandably difficult for unauthorized immigrants to hold lawbreaking employers accountable. These workers are often afraid to complain about unpaid wages and substandard working conditions because employers can retaliate against them by taking actions that can lead to their deportation. Thus, complaints to their employers or to government agencies about illegal conditions like unpaid wages and substandard working conditions seem like risky options—and perhaps with a low chance of success given how underresourced and understaffed labor standards enforcement agencies are (Costa 2022).

This imbalanced relationship gives employers extraordinary power to exploit and underpay these workers, also making it more difficult for other workers to improve their wages and working conditions. The exploitation described here is not theoretical. A landmark study and survey of 4,300 workers in three major cities by Bernhardt et al. (2009) found that 37.1% of unauthorized immigrant workers were victims of minimum wage violations, compared with 15.6% of U.S.-born citizens. In other words, unauthorized immigrants are more than twice as likely to be the victims of wage theft compared with U.S.-born citizens. Immigrants who had a lawful immigration status (labeled as “authorized immigrants” in the study) were also more likely to be the victims of wage theft, but the gap with U.S.-born citizens was significantly narrowed, compared with unauthorized immigrants: 21.3% of authorized immigrants were the victims of wage theft versus 15.6% of U.S.-born citizens. Bernhardt et al. (2009) also looked at overtime pay violations and found that an astounding 84.9% of unauthorized immigrants were not paid the overtime wages they worked for and were legally entitled to, compared with 67.2% of authorized immigrants and 68.2% of U.S.-born citizens. And finally, beyond just the higher rates of wage theft, research related to IRCA by Kossoudji and Cobb-Clark (2002) also examined the value of the wage penalty for being an unauthorized immigrant worker, estimating that it ranged from 14% to 24%.

Migrants with precarious temporary immigration protections and work permits

When it comes to migrants with precarious temporary immigration protections like TPS, parole, and deferred action, or who are going through the asylum application process, most of the migrants who qualify for those temporary protections are also eligible to receive an Employment Authorization Document, which is often referred to simply as a work permit or EAD. All foreign-born persons who are not authorized to work by virtue of being a naturalized citizen, green card holder, or temporary migrant worker must first obtain an EAD from U.S. Citizenship and Immigration Services (USCIS) within the U.S. Department of Homeland Security (DHS) before they can be authorized to work.36 To do so, they must belong to an eligible group, including, for example, certain foreign students, or the spouses of temporary migrant workers in certain visa programs, such as H-1B or J-1, or they may obtain an EAD by qualifying for one of the forms of administrative immigration relief, such as TPS, deferred action, Deferred Enforced Departure, humanitarian parole, or parole-in-place. EADs are also available for other categories of individuals such as asylum applicants (after a waiting period of six months), certain applicants for adjustment of status, asylees, and refugees who have not yet obtained green cards, as well as persons granted withholding of removal.37

For workers who lack a permanent or more durable immigration status, obtaining a temporary EAD can mean having enforceable workplace rights that the individual would otherwise not have. While all workers have some labor and workplace rights under U.S. law—regardless of immigration status, as discussed in other parts of this section in this report—enforcing them in practice becomes virtually impossible because of the threat of deportation, which prevents workers who lack an immigration status or an EAD from calling out lawbreaking employers and demanding that they comply with the law, or from reporting workplace violations to labor enforcement agencies. But having protection from deportation through temporary administrative immigration protections accompanied by an EAD means that, in practice, workers can report workplace violations to government officials without fear of retaliation that can lead to deportation. It also means that a migrant worker with an EAD can be employed by just about any employer and change jobs or employers, unlike temporary migrant workers who must be employed by the sponsor of their visa.

There are some examples of research showing the important economic contributions that hundreds of thousands of migrants with precarious statuses and EADs are able to make thanks to their temporary protections. For example, when the TPS population was approximately 354,000 in 2021, AIC (2023) estimated that “TPS holders contributed more than $2.2 billion in taxes, including almost $1 billion to state and local governments,” as well as “held $8 billion in spending power.” Another estimate by Moriarty (2024) found that TPS-eligible individuals “annually contribute some $31 billion in wages to the national GDP.” The total number of TPS holders in 2024 is now roughly 864,000; thus these totals are likely to be much higher now.

Research has also quantified some of the contributions made by persons who have qualified for Deferred Action for Childhood Arrivals (DACA). DACA was created by the Department of Homeland Security during the Obama administration in 2012, and recipients are eligible for protections from deportation and EADs that are valid for two years and renewable. More than 835,000 persons have benefitted from DACA, and more than 500,000 are currently still protected by DACA (President’s Alliance 2024). Svajlenka and Truong (2021) found that DACA recipient households “pay $6.2 billion in federal taxes and $3.3 billion in state and local taxes each year,” and “after taxes, these households hold $25.3 billion in spending power,” and that DACA recipient families “own 68,000 homes, making $760 million in mortgage payments and $2.5 billion in rental payments annually.”

When it comes to measuring the workplace impact and economic benefits of being issued an EAD for the workers themselves, there are limited examples, but three are worth citing here. One is an annual survey of DACA recipients that was conducted in 2024 for the ninth time. The most recent survey, conducted by Wong et al. (2024) and published by the Center for American Progress, showed that DACA has been an essential tool to improve the economic and educational outcomes of recipients. In terms of the impact that deferred action and an EAD have had on the employment of DACA recipients: 59.1% of respondents moved to a job with better pay; 47.3% moved to a job with better working conditions; 47.5% moved to a job that “better fits [their] education and training”; 49.6% moved to a job that “better fits [their] long-term career goals”; 57.3% moved to a job with health insurance or other benefits; and 19.6% of respondents obtained professional licenses.

Wong et al. also measured the impact of DACA and EADs on wages, finding that “[d]ata from the past nine years show that DACA has had a significant and positive effect on wages: Recipients’ average hourly wage more than doubled from $11.92 to $31.52 per hour—an increase of 164.4 percent—after receiving DACA.” These significant wage increases are no doubt a result of the labor and workplace rights and stability that DACA recipients gain from having an EAD.

Orrenius and Zavodny (2014) examined the wage and employment impact of another form of temporary immigration protection, that of TPS—which allows those who are eligible to also be granted an EAD. They looked specifically at migrants from El Salvador, finding that having TPS increased employment rates, and that less-educated Salvadoran men who were employed earned 13% more if they had TPS. They note that “As a whole, the results suggest that less-educated Salvadoran men who receive TPS are able to move into better jobs and become more selective about the jobs they hold, increasing their earnings but also their job search and unemployment incidence.”

One other analysis that assesses the wage impact of being issued an EAD comes from Kallick (2023), which looks specifically at asylum seekers in New York and nationwide. Relying on previous methodologies for measuring the impact of a lawful immigration status being granted to unauthorized immigrants, Kallick estimates that asylum seekers who are granted EADs increase their wages by 10%.

While the relative benefits of precarious and temporary immigration protections and EADs to migrant workers and the broader economy are clear, it is important to note here that the protections and EADs are only temporary and will end if renewals are not approved. This may occur either because the migrant no longer qualifies or because the program that authorized them has ended or not been renewed (for example if DHS decides not to renew a TPS designation or if DACA or another program is ended by DHS or found to be unlawful according to a federal court ruling). Since migrants with precarious immigration protections and EADs do not have a direct path to a green card, they may never obtain one unless Congress passes reforms to provide them with such a path—keeping them in a precarious status indefinitely until there is a policy change or their protections are not renewed.

In fact, if we include asylum seekers with EADs, there are approximately 4.3 million migrants who have protections through a precarious immigration status as of early 2024.38 But more than half are in danger of losing their protections and EADs if the current U.S. president or a future president decides to end any or all of the administrative immigration relief programs or if a court rules them invalid. Thus, millions who have been issued temporary protections and an EAD could potentially lose the workplace rights and protections that a work permit provides, leaving them vulnerable to exploitation by employers. In the case of asylum seekers, many can and will lose their ability to remain lawfully in the United States along with their work permits, depending on how their case is adjudicated in immigration court or by USCIS.

Immigration enforcement and the impact of inadequate resources for labor standards enforcement

The inability of both unauthorized immigrants and temporary migrant workers to hold employers accountable through regular channels, as discussed in this report, is exacerbated by the current immigration enforcement regime, which prioritizes removing unauthorized immigrants, detaining migrants, and detecting persons who attempt to enter the United States without authorization. This section briefly discusses the U.S. government’s lopsided enforcement priorities, the resulting challenges faced by labor standards enforcement agencies, and the impact on all workers.

Labor standards enforcement agencies are underfunded and short-staffed

In order to carry out its immigration enforcement priorities, as Figure I shows, U.S. immigration enforcement agencies received $30.2 billion from Congress in fiscal year 2023. All U.S. labor standards enforcement agencies that protect workers, on the other hand, only received $2.2 billion (also Figure I). That gap between the amounts appropriated for immigration enforcement as compared with labor standards enforcement means that immigration enforcement agencies are now funded at a rate that is nearly 14 times higher than the budgets of all federal labor standards enforcement agencies combined. This is up from 12 times as much in 2021—and when it comes to staffing, immigration enforcement agencies had eight times as many staff as labor standards enforcement agencies in 2021 (Costa 2022). The ultimate result of these disparities is to increase the fear that unauthorized immigrants already have when considering whether to report workplace violations, thereby making it even less likely that labor standards enforcement agencies will know about employer lawbreaking and be able to adequately respond.

Figure I

Hinojosa-Ojeda (2010) provides a useful summary of a 2008 report from the Atlanta Federal Reserve (Brown, Hotchkiss, and Quispe-Agnoli 2008) that examined the negative impact that the current immigration enforcement regime has on low-wage labor markets:

The enhanced [immigration] enforcement regime moves unauthorized workers further underground, lowering their pay, and ironically, creating a greater demand for unauthorized workers. A 2008 report from the Atlanta Federal Reserve analyzes how this vicious cycle is activated and expands as firms find themselves forced to compete for the supply of cheaper, unauthorized labor. When a firm cuts costs by hiring unauthorized workers for lower wages, its competitors become more likely to hire unauthorized workers for lower wage, as well, in order to benefit from the same cost savings.

This impact of escalating immigration enforcement, coupled with a lack of funds for labor standards enforcement, leaves workers vulnerable and largely unprotected from employer lawbreaking. There are some clear examples in particular industries in which enforcement is inadequate, but violations are common, for example, in agriculture, as we discuss in the next subsection—in which more than half of the workforce is comprised of unauthorized immigrants and temporary migrant workers (Costa and Martin 2023).

First, some background is appropriate regarding how one of the main federal worker protection agencies, the Wage and Hour Division (WHD), part of DOL, has been impacted by inadequate staffing and funding. WHD is responsible for enforcing provisions of several federal laws related to minimum wage, overtime pay, child labor, federal contract workers, work visa programs, migrant and seasonal agricultural workers, family and medical leave, and more. Yet, despite this broad portfolio and the 167 million workers who are covered by these protections (WHD 2023),39 funding for WHD has not kept pace with the growth of the U.S. labor force.

Figure J shows that, in inflation-adjusted 2023 dollars, WHD’s budget in 2006 was $250 million, and in 2023, $260 million—an increase of just $10 million over nearly two decades. As Figure J also shows, this trend has been consistent with appropriations for two other key worker protection agencies, the Occupational Safety and Health Administration (OSHA) and the National Labor Relations Board (NLRB). At both OSHA and the NLRB, inflation-adjusted appropriations were significantly lower in 2023 compared with 2006.

Figure J

In addition to funding levels that have barely kept up with inflation at WHD and the 167 million workers WHD has a mandate to protect, the number of WHD investigators that the agency employs, who are primarily responsible for ensuring that federal wage and hour laws are obeyed by employers across all 50 states and U.S. territories, is near an all-time low. Figure K shows that there were only 733 WHD investigators at the end of 2023 to enforce all federal wage and hour laws, 79 fewer than in 1973, the first year for which data are available, and 499 fewer than the peak year of 1978 when there were 1,232 WHD investigators. Meanwhile, the number of workers that WHD has a mandate to protect has increased sharply. The average number of workers in 2023 was 167.1 million, which amounts to 227,989 workers for every wage and hour investigator. Compare this with 1973 when there were 72,588 workers for every wage and hour investigator.40 Investigators are now responsible for more than triple the number of workers as in 1973.

Figure K

The value of back wages recovered by WHD, the number of employees who received back wages as a result of WHD actions, and the total number of hours the WHD spent on investigations “all dropped in fiscal year 2022 compared to the year prior” according to WHD data reported on by Bloomberg Law in December 2022 (Rainey 2022). Despite WHD’s stated intention to hire 100 new investigators during the Biden administration, a heavy workload and inadequate funding from Congress appear to be hindering WHD from hiring enough staff for the tasks at hand. As the same Bloomberg Law report noted, WHD has “struggled to recruit new investigative staff.”

Inadequate labor standards enforcement in agriculture exemplifies the impact of underfunded and short-staffed worker protection agencies

What has been the result of underfunded and short-staffed worker protection agencies like WHD and OSHA in an industry like agriculture in which half of workers lack an immigration status and roughly one-tenth have a precarious status through temporary H-2A visas? There has been a clear downward trend in the number of closed investigations of agricultural employers by WHD over the past two decades, from more than 2,000 a year in the early 2000s, to 1,000 or fewer a year during the last three fiscal years. In fiscal year 2023, WHD closed only 831 investigations of agricultural employers—a record low during the 2000 to 2023 period—amounting to an average of 68 a month in 2023 (WHD 2024). Eight hundred thirty-one investigations in 2023 is just over a third of the 2,431 agricultural investigations closed in 2000, the peak year for WHD agricultural investigations (Costa and Martin 2023). 

The low number of investigations means that most farms are never investigated by WHD; in fact fewer than 1% of agricultural employers are investigated per year (Costa and Martin 2023). Since farm operators know there is a very low likelihood that they will ever be investigated, some may feel emboldened to have a business model that relies on wage theft and other forms of lawbreaking. In addition, when it comes to health and safety violations on farms, reporting from ProPublica suggests that farmworkers are unwilling to come forward to report employer violations because of a perception that OSHA doesn’t have adequate resources to investigate (Sanchez and Jameel 2023).

These issues extend beyond agriculture, but the agricultural industry is a useful microcosm that exemplifies the broader issues. Another problem that was recently identified that extends beyond agriculture is that even when WHD detects and can confirm employer violations, as another report from Bloomberg Law revealed, WHD “cannot litigate every case due to resource issues,” citing violations in the H-2B visa program where employers violated the law but were not punished (Rainey 2023).

Yet another enforcement gap has to do with governance of temporary work visa programs. WHD data show that violations of the H-2A visa program—which allows farm employers to hire temporary migrant farmworkers—now account for the vast majority of back wages owed and civil money penalties assessed on employers (73%) (Costa and Martin 2023). Thus, nearly three-fourths of all penalties in agriculture now result from violations of work visa program rules. WHD also recently reported that in the previous five fiscal years, “in 88 percent of WHD’s H-2A investigations, WHD found employers in violation of the law” (ETA and WHD 2024). That means that when WHD investigates H-2A violations, they nearly always detect them. These data strongly suggest that violations of the H-2A visa program in agriculture are not limited to a few employers who are “bad apples,” but instead that widespread and systematic violations of H-2A rules are ongoing.

In addition, in all U.S. temporary work visa programs, employers that have been found to violate the law—whether it be wage and hour, labor, health and safety, discrimination, or civil rights laws—are nevertheless allowed to continue to hire through visa programs. As numerous investigative reports have shown, even some of the worst violators are allowed to keep hiring, even after they have been sanctioned for lawbreaking and extreme abuses of their workers.41 Changing and enforcing a rule to prohibit them from hiring would also likely require a significant increase in funding to DOL’s Office of Foreign Labor Certification (OFLC), which certifies the applications of employers seeking to hire temporary migrant workers. But funding and staffing levels there have not increased proportionally with the rise in the number of applications for work visas and prevailing wage determinations over the past decade. (OFLC’s workload has increased by nearly 50%, while funding has declined by 4% since 2012 after adjusting for inflation (Costa and Hira 2024)). WHD would also need increased funding to help enforce the rule to bar employers that violate the rules.

Taken together, all of these realities further embolden lawbreaking employers to victimize their employees who lack an immigration status or only have a temporary or precarious status because employers know that unauthorized migrants and temporary migrant workers are unlikely to complain and report violations. And even when workers are brave enough to do so, employers know that labor standards enforcement agencies don’t have the resources or capacity to respond adequately. The ultimate result is that employers can often steal wages and violate labor and employment laws with impunity, which degrade labor standards for all workers, including the U.S.-born workers who work alongside foreign-born workers.

How to reform immigration policies to maximize benefits for workers and minimize challenges

So far in this report we have discussed existing analyses of how immigration impacts the economy and how immigration status impacts wages and worker rights. But we wish to reiterate that most, if not all, of the negative impacts or potential negative impacts that immigration could have on the economy, wages, or labor standards, are the result of how the immigration system is structured and its legal framework. In particular, the rights and protections that migrants and immigrants have depending on their immigration status, or lack thereof, and the employment relationships that result between migrants and immigrants and their employers, are the result of policy choices. A severe power imbalance between workers and employers that is tilted almost entirely in favor of employers is what leads to worker exploitation. Temporary and precarious immigration statuses, or the lack of status, can embolden employers to break laws while making it difficult to hold them accountable.

This section briefly discusses some of the most important policy reforms that should be made to update the immigration system in order to level the playing field between workers and employers, and thereby lift standards and raise wages economywide. Successfully doing so would maximize the economic benefits of immigration and bring credibility to the immigration system, allowing all workers to see that the system is not being misused by employers to degrade wages and labor standards. That credibility is necessary in order to build public support for higher levels of immigration and to assuage the concerns of those who have valid critiques of the current pro-employer U.S. immigration policy framework.

In sum, an immigration system that leads to shared prosperity for all workers means lifting standards and ensuring that all workers have equal and enforceable labor and workplace rights regardless of immigration status—as well as having a flexible and data-driven immigration system that can adjust based on the needs of the economy.

A broad and quick path to a green card for the unauthorized immigrant population and those with precarious statuses

Perhaps unsurprisingly, the first and most important reform needed is a quick and broad legalization for the current unauthorized immigrant population—roughly 5% of the total U.S. workforce—that leads to lawful permanent residence (green cards), making beneficiaries eventually eligible for citizenship. Having 5% of the U.S. workforce vulnerable to exploitation and workplace abuse only benefits lawbreaking employers—and it drags down labor standards for all workers, including  the U.S.-born, naturalized citizens, and green card holders. This reality has existed for far too long and should be an urgent priority for Congress to address.

To maximize economic gains that come with the path to a green card and citizenship—which include raising wages significantly, especially for workers in low-wage industries—the path must be as inclusive as possible. The pathway should not be long and arduous like those that have been proposed in major immigration reform bills of the past, such as S.744 from 2013, which required a 10-year path to a green card and another three years for citizenship eligibility. As discussed earlier, the 1986 IRCA legalization made beneficiaries eligible for green cards after 18 months; the same timeline should be the goal. However, one valid critique of IRCA was that the law’s legalization program was not nearly broad enough; it failed to legalize millions, which left behind the core of today’s unauthorized population (Chishti and Kamasaki 2014).

Having a long and difficult pathway with onerous requirements (which may be designed to reduce the number of eligible beneficiaries for green cards) will have the effect of leaving potential beneficiaries who are awaiting green cards vulnerable to abuse by employers. For example, a requirement that workers be employed for a set number of days per year in a particular industry, such as agriculture, could have the effect of leaving them in a quasi-indentured state. S.744 required potential applicants for green cards to pay processing fees and back federal taxes, as well as meet minimum income and employment tests—requirements that would have led to a significant share of the total unauthorized immigrant population never obtaining a green card—dooming the nation to repeat history.

A long and difficult pathway to permanent status also delays potential wage gains for immigrants and other workers, including U.S.-born workers and earlier immigrants, and the additional tax revenue associated with increased earnings. It makes no sense to restrict the wage gains and other societal benefits that come with a green card, which have been measured time and time again through rigorous research. And it would delay the ability of immigrants to fully integrate and participate in civic life, and make them less likely to make the kind of longer-term investments in their future that would also benefit the broader economy, whether it be purchasing a home, starting a business, or investing in job training and education.

The executive branch should use existing authority to expand temporary rights and protections and issue work permits to protect workers and improve labor standards

The number of people in precarious statuses through temporary protections like parole, deferred action, or Temporary Protected Status (TPS), has increased. There were almost 2.8 million migrants in those statuses as of early 2024, nearly all of whom are eligible for work permits, which are formally known as Employment Authorization Documents (EADs).42 There has also been a large increase over the past decade in the number of asylum seekers who are lawfully employed with EADs, over 1.5 million as of late 2023, who are also in a precarious quasi-status while they pursue their claims.43

In fact, the recent increase in precarious statuses has resulted in close to one-third of the entire unauthorized immigrant population having some form of temporary protection and the workplace rights that come with an EAD.44 As discussed in this report, the workers in precarious statuses who possess EADs are in a vastly better situation compared with being unauthorized with no EAD—because work permits allow them to have workplace rights that can be enforced in practice—even if most are unable to access an eventual path to a green card and citizenship. The current status quo in which one-third of an exploitable population has enforceable workplace rights—however precarious and temporary—is vastly preferable over the prior reality where nearly all had rights that only existed on paper and were virtually impossible to enforce in practice.

Whether through the use of TPS and humanitarian parole—which are based in statute—or deferred action and other forms of prosecutorial discretion, the executive branch has demonstrated across multiple presidential administrations that it has the requisite tools and legal authority necessary to expand temporary rights, protections, and work permits for the millions of migrants who lack an immigration status. (Many of whom have resided in the United States for decades and are deeply integrated into communities across the United States.)

The executive branch should thus take immediate action to improve standards for all workers through new and expanded TPS designations, additional use of humanitarian parole and parole-in-place, and grants of deferred action—at least until Congress acts and passes the most essential reforms necessary. Doing so would instantly improve wages and working conditions for both foreign-born and U.S. born workers in countless industries.

Green cards instead of temporary work visas

There are now at least 2 million temporary migrant workers in the United States who are employed through temporary work visas. The numbers of visas issued and workers in the programs have grown exponentially since the Immigration Act of 1990, while the number of employment-based green cards has remained relatively flat due to the annual numerical limit for employment-based green cards (Costa 2020). Temporary work visas are often the only viable employment-based option for migrant workers seeking jobs in the United States, but as discussed earlier in this report, temporary work visas leave those workers indentured and often (legally) underpaid compared with similarly situated U.S. workers—creating a two-tiered system of rights and standards in the workplace. Moreover, most temporary work visas do not offer a viable or direct path to permanence.

To ensure that all workers have equal rights and to restore the balance of power between employers and workers, the U.S. immigration system should be tilted away from temporary work visa programs and toward providing more green cards for workers. Instead of arriving in a quasi-indentured status and having limited rights, the vast majority of migrant workers should be able to arrive with a green card in hand, giving them the freedom to change jobs if they can earn more or be more productive working for another employer or starting their own business. This will also allow them to benefit from the economic gains associated with green cards and citizenship and to make long-term investments in their future that benefit the U.S. economy.

Reforming work visa programs

Given the critiques reiterated in this report and explained with extensive evidence through various EPI publications, we strongly believe that the U.S. immigration system should shift away from temporary work visa programs because of their inherent flaws that allow employers to exert a coercive amount of power over workers, which allows them to exploit and underpay workers and to violate labor and employment laws with impunity. The bargaining power of all U.S. workers is undercut when more than 2 million temporary migrant workers—1.2% of the U.S. labor force—are underpaid by employers and cannot safely complain to labor standards enforcement agencies or sue employers that exploit them because their visa status is owned and controlled by their employer.

The issue is not that all employers of temporary migrant workers break laws or have bad intentions, but that the current structures and legal frameworks of temporary work visa programs facilitate worker exploitation and are coupled with inadequate mechanisms in place for oversight and accountability. Reforming the system would ensure an even playing field for employers and raise wages and improve standards for both temporary migrant workers and U.S. workers in adjacent occupations and industries.

To achieve this, we propose several key reforms that are necessary to protect workers and modernize the U.S. immigration system:45

1. Congress should require employers to recruit and offer jobs to qualified U.S. workers before being allowed to recruit workers abroad.
2. Congress should regulate foreign labor recruiters to protect migrant workers and ensure transparency in recruitment chains, and hold end-user employers jointly liable for the actions of their recruiters.
3. Employers should be prohibited from hiring through temporary work visa programs if they have violated labor and employment laws, as well as civil rights, anti-discrimination and anti-trafficking laws.
4. Temporary migrant workers should be paid fairly according to U.S. wage standards based on the specific occupation and region.
5. Temporary migrant workers should be allowed to change employers and never be indentured to their employers through their visa status.
6. Temporary migrant workers should be allowed to self-petition for permanent residence after a short period in temporary status; ideally no longer than 18 months.
7. Much more funding should be appropriated to the U.S. Department of Labor to enforce an updated work visa system and strengthen the department’s mandates to conduct adequate oversight and debar employers, as well as to conduct random audits of employers.
8. Transparency in temporary work visa programs should be improved to protect workers and aid anti-trafficking efforts, in particular through the systematic release of more and better government data on temporary work visa programs.
9. An independent commission on employment-based migration and immigration should be established to make the system more flexible and data-driven and to depoliticize the adjustment of numerical limits in temporary work visa programs and for employment-based green cards.

Making U.S. labor migration more transparent, data-driven, and flexible with an independent commission on immigration

The U.S. immigration system needs to be much more flexible and data-driven in order to be responsive and able to adjust to the needs of the economy. Adjusting annual visa caps for both employment-based green cards and temporary work visa programs requires congressional action, which can be contentious because it is influenced by lobbying and opaque political considerations rather than facts, and too slow to keep up with changing economic conditions. Most permanent and temporary employment-based visa quotas have not changed since 1990, despite a vastly different U.S. economy and workforce three and a half decades later.

EPI has long proposed an independent commission on immigration and the labor market that would report regularly to Congress and the president, proposing new quotas on an annual or semiannual basis—based on economic needs and conditions, and issue public reports citing the evidence for its recommendations, which would be based on methodologies that are credible and transparent. The commission would consider the many trade-offs inherent in immigration policymaking in its recommendations, and Congress would ultimately decide which policies to adopt or reject. Basing quotas on evidence, data, and changing economic realities would depoliticize the process of setting numbers and provide an evidence base for decisions that can be inspected by all.46 Other expert organizations and groups have also called for a similar commission model.47

Expanding and strengthening humanitarian migration pathways

The world is experiencing the largest global displacement crisis in modern history, and the forced migration seen on the Western Hemisphere is a major component of the broader global trend (UNHCR 2024). Given the history of interventions in its own hemisphere, the United States government has a special responsibility to at least accept more refugees and asylum seekers from the region,48 but should also accept more from other regions. Existing research already shows that humanitarian migrants see wage gains and integrate well into communities across the United States and make important economic contributions, leaving little to fear from an economic perspective.

Existing humanitarian pathways should be expanded to meet the current need of persons fleeing persecution, conflict, and environmental change by (1) increasing the annual refugee ceiling quotas and investing in a more robust and durable network for resettlement assistance and support and (2) making the U.S. asylum system more welcoming, including by broadening the definition of asylum, either by statute or executive action where possible, including the creation of pathways for migrants who are displaced by the climate crisis and armed conflict. In addition, humanitarian pathways should be improved through strengthening worker protections for newly arriving refugees, asylees, and asylum seekers through know-your-rights trainings. And employers that regularly recruit and hire refugees, asylees, and asylum seekers, should be required to commit to providing fair and decent working conditions, including a commitment to labor neutrality.

To further facilitate workforce integration, public assistance should be made available to new arrivals; work permits should be expedited to boost economic gains and reduce the workload of social services providers, including shelters; and government agencies should partner formally with unions, workers’ centers, and assistance providers on job training and matching workers with employment opportunities.

Adequately funding labor standards enforcement and stricter penalties for lawbreaking employers

As EPI has reported numerous times over the years and as discussed in this report, the U.S. government now appropriates nearly 14 times more for immigration enforcement than on all labor standards enforcement combined ($30.2 billion vs $2.2 billion). Rather than spending tens of billions of dollars per year for immigration enforcement, what’s needed are more resources and staffing for labor standards enforcement agencies, and a more strategic focus on labor standards enforcement that does not take immigration status into account. A shift away from immigration enforcement will result in removing barriers to organizing in workplaces across the country—which will boost economic outcomes through the adoption of collective bargaining agreements—that also serve to close racialized and gendered gaps in wages and working conditions.

At present, labor standards agencies are underfunded, short-staffed, and ill-equipped to hold lawbreaking employers accountable. While immigration benefits the U.S. economy overall as discussed in this report, when worker protection agencies cannot adequately do their jobs, employers can exploit the millions of migrant workers who lack an immigration status or who only have a temporary or precarious status in ways that degrade standards for all workers.

U.S. labor standards enforcement agencies must therefore be staffed and funded adequately to protect the rights of all 167 million workers in the United States, including migrant and immigrant workers, so that lawbreaking employers who abuse workers of any status will face much higher chances of being caught and much higher penalties than they currently do. Congress must make major investments to achieve this, by at least tripling the funds appropriated to worker protection agencies like the Wage and Hour Division and the Occupational Safety and Health Administration in the U.S. Department of Labor, and the National Labor Relations Board.

Family-based immigration should remain an important pathway

Family-based immigration accounts for roughly two-thirds of all green cards and should remain a robust and important immigration pathway. While EPI’s immigration analyses and proposals mainly focus on the employment-based aspects of immigration, we fully support family reunification remaining the dominant pathway in the U.S. immigration system. Family reunification is in the national interest because families are our most basic learning and support systems and therefore greatly facilitate the assimilation of immigrants into American life. Many, if not most, family-based immigrants also enter the labor market and make important economic contributions to the United States, even if these contributions have not been analyzed and measured to the same extent as those of other migrants and immigrants.

Conclusion

As the body of evidence in this report shows, immigration to the United States has contributed greatly to growing the economy, and foreign-born workers have been complementary to U.S. workers and expanded opportunities for them. From an economic and labor force perspective, continuing or increasing immigration levels is not something to be feared if the right polices are in place and governance is improved. The challenges and potential pitfalls that must be addressed are mainly the result of workers lacking equal rights in the workplace due to their immigration status.

The share of those workers in the U.S. labor force, who either lack an immigration status or only have a temporary and precarious status that can expire or be rescinded, is large and growing. A workforce with so many workers in this situation results in a massive power imbalance that benefits employers and allows them to keep workers in fear of calling out workplace violations like wage theft and various other forms of lawbreaking. The precariousness of migrant workers makes it nearly impossible for them to exercise the labor and employment rights they ostensibly have on paper, which leads to degraded workplace conditions for all workers, compared with an ideal scenario in which all workers have full and equal rights. The reforms and policy interventions we have outlined are needed to fix this.

Nevertheless, despite an unjust immigration policy regime and the abuses enabled by the status quo, immigration has still benefitted the economy greatly, including most foreign-born and U.S.-born workers. But if the recommendations outlined here were implemented, leading to a guarantee of full and equal workplace rights for all workers and an improved and expanded labor standards enforcement regime—then future immigration flows and even expansions will result in a fairer and more broadly shared prosperity for all workers and increased dynamism for the U.S. economy.

Notes

1. In this report, the statistics we present from the American Community Survey and Current Population Survey follow the convention that immigrants are synonymous with what the U.S. Census Bureau calls the “foreign-born population”, which are defined as those people in the U.S. who were not citizens at birth. The small number of U.S. citizens born abroad to U.S. parents are not included in the foreign-born population and in this report, are treated as “U.S. born”.

2. The labor market participation rate in 2023 was 66.6% for immigrants and 61.8% for U.S.-born workers (BLS 2024a).

3. CBO (2024) in particular adjusts for the possibility that growing survey nonresponse in the CPS is causing an undercount of the size of the foreign-born population. Correcting this possible undercount also helps to make sense of a recent gap between the household-based CPS and the establishment-based Current Employment Statistics survey, as explained by Edelberg and Watson 2024 and Tedeschi 2024. At the same time, Butcher et al. 2023 argue that the CPS may be overstating the size of the foreign-born population.

4. Appendix B of CBO 2024 reports that their adjustments to the CPS increase the estimated foreign-born population share to 15.6% in 2022 and 16.2% in 2023. In Figure 2 of the CBO report, the total Social Security area population estimates are about 335.5 million in 2022 and about 338.4 million in 2023. Together, these estimates imply that the foreign-born population grew by about 4.8%.

5. See Bivens 2022 for an overview of the U.S. economic situation before the American Rescue Plan passed and the law’s subsequent effect on labor markets.

6. Immigrant workers at a given income level also likely have smaller propensities to consume out of current income than U.S.-born residents because they are much more likely to send remittances abroad to family members in their origin countries. This is another reason why a larger share of immigrants in the population is likely deflationary.

7. For an overview of the evidence on the weak causal link between labor market developments and subsequent inflation, see Banerjee and Bivens 2023.

8. Authors’ analysis of the 2023 Current Population Survey basic monthly microdata. “Dental, nursing, and health aides” refers to the occupations Home Health Aides, Personal Care Aides, Nursing Assistants, Orderlies and Psychiatric Aides, Occupational Therapy Assistants and Aides, Physical Therapist Assistants and Aides, Massage Therapists, Dental Assistants, and Medical Assistants.

9. See Saiz 2010 for evidence on the lack of responsiveness of supply to shifts in demand for housing.

10. See Kmetz, Mondragon, and Wieland 2022 for the effect of the pandemic shock on housing demand.

11. See Schuetz 2022 for an overview of this agenda.

12. For a discussion of monopsony power and wage markdowns, see Manning 2020.

13. Authors’ analysis of Baugh 2023 and Department of Homeland Security Statistics, Yearbook of Immigration Statistics (various years).

14. The fact that these visas are called “nonimmigrant” in U.S. law has to do with the fact that foreign nationals who apply for them are not allowed to immigrate permanently to the United States. U.S. law presumes that all foreign nationals wish to reside permanently in the United States, and in order to be issued a nonimmigrant visa, the applicant for the visa has to prove to the U.S. government that they do not intend to reside permanently in the United States.

15. Authors’ analysis of State Department nonimmigrant visa statistics and U.S. Citizenship and Immigration Services petition data.

16. The Refugee Act of 1980, S.643-Refugee Act of 1979 (title upon introduction), 94 Stat. 102, Public Law 96-212 (March 17, 1980), 96th Congress (1979–1980), Congress.gov.

17. There are also separate provisions in the Immigration and Nationality Act for the granting of asylum on a case-by-case basis to persons who are physically present in the United States or who arrive in the United States and who meet the definition of a refugee. For more background, see Bruno 2019.

18. Authors’ analysis of USCIS Form I-765 data for fiscal years 2022 and 2023.

19. See Ruiz Soto, Gelatt, and Hook 2024, Passel and Krogstad 2024, Baker and Warren 2024, and Warren 2024.

20. See for example, Chishti and Bush-Joseph 2023, Chishti, Bush-Joseph, and Putzel-Kavanaugh 2024, Ruiz Soto, Gelatt, and Hook 2024.

21. See Table A1 in Batalova, Gelatt, and Fix 2024, updated by authors with latest TPS population estimate in Wilson 2024. The 2.8 million total cited here excludes asylum seekers whom we also consider to be in a precarious status. As noted above, there were 1.5 million asylum seekers who held valid EADs in 2023, but the total number of asylum seekers is greater but unknown.

22. See for example, discussion of data in Passel and Krogstad 2024, Baker and Warren 2024, and Ruiz Soto, Gelatt, and Hook 2024.

23. See discussion in Hinojosa-Ojeda 2010.

24. Hinojosa-Ojeda 2010 summarizing and citing Rivera-Batiz 1999; Amuedo-Dorantes, Bansak, and Raphael 2007; and Kossoudji and Cobb-Clark 2002. Kallick 2013 also provides a useful review and commentary on the studies that measure the impact of granting status to unauthorized immigrants (see Appendix A).

25. See for example, Galarza 1956, Meissner 2004, and Costa 2021.

26. See for example, Lapinig 2017.

27. See for example, Bauer and Stewart 2013.

28. See for example, Bauer and Stewart 2013, Garrison, Bensinger, and Singer-Vine 2015, and GAO 2017.

29. See for example, Costa and Hira 2020, Costa 2016, Hira 2015, and Costa 2017.

30. For example, see discussion in Costa 2021.

31. See for example, Hira and Costa 2021 and Costa and Martin 2023.

32. Bivens and Shierholz broadly define “monopsony power” as “the leverage enjoyed by employers to set their workers’ pay.” See Bivens and Shierholz 2018.

33. See for example, Gibbons et al. 2019 and Naidu, Posner, and Weyl 2018 for estimates of monopsony power’s effects in wage suppression in the United States.

34. For more background, see Broder and Lessard 2023.

35. With the exception of back pay under the National Labor Relations Act due to the Hoffman Plastics decision of the U.S. Supreme Court, see for example WHD 2008.

36. For more discussion and background, see Kolker and Morton 2023.

37. For the full list of EAD eligibility categories, see USCIS 2024.

38. This total includes 2.8 million migrants in precarious statuses like DACA, TPS, and parole, and the 1.5 million asylum seekers with an approved EAD (see discussion in previous section).

39. The document cited from WHD does not list a date but was last viewed by the authors in mid-2023, meaning it was likely citing the number of workers in the civilian labor force in 2022; it is referenced as evidence of the number of workers who are protected by WHD. We cite an updated number for the number of workers in the civilian labor force in 2023, as the number for which WHD is responsible for protecting. The number of covered workers is derived from the annual averages reported for the total civilian labor force, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Series Id: LNU01000000, Not Seasonally Adjusted, Series title: (Unadj) Civilian Labor Force Level, ages 16 and over [data tables], U.S. Department of Labor.

40. To derive this estimate, the number of workers in 1973 and 2023 was divided by the number of WHD investigators in those years. The number of covered workers is derived from the annual averages reported for the total civilian labor force, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Series Id: LNU01000000, Not Seasonally Adjusted, Series title: (Unadj) Civilian Labor Force Level, ages 16 and over [data tables], U.S. Department of Labor.

41. See for example, Costa, Martin, and Rutledge 2020, Bensinger, Garrison, and Singer-Vine 2016, and Cotsirilos 2023.

42. Authors’ analysis of Table A-1 in Batalova, Gelatt, and Fix 2024 plus additional updated numbers for the TPS population reported in Wilson 2024.

43. Authors analysis of USCIS n.d. for employment authorization category “C085, applicant/pending asylum.”

44. Authors’ estimate based on the total population of persons in precarious statuses as a share of the total unauthorized immigrant population, which also relies on estimates by Passel and Krogstad 2024 and Batalova, Gelatt, and Fix 2024.

45. For more discussion, see Costa 2021.

46. See further discussion in, for example, Marshall and Eisenbrey 2010, Marshall 2009 and 2011, and Ruhs and Martin 2013.

47. See for example Papademetriou et al. 2009 and Gelatt and Chishti 2024.

48. See for example, Faux 2017, Borger 2018, and Shesgreen 2018.

References

Akee, Randall, and Maggie R. Jones. 2024. “Return Migration Decisions and Declining Earnings.” Journal of Human Resources 59, no. 3, May 2024.

American Immigration Council (AIC). 2023. The Contributions of Temporary Protected Status Holders to the U.S. Economy (fact sheet). September 19, 2023.

Amuedo-Dorantes, Catalina, Cynthia Bansak, and Steven Raphael. 2007. “Gender Differences in the Labor Market: Impact of IRCA’s Amnesty Provisions.” American Economic Review 97, no. 2 (2007): 412–416, May 2007.

Apgar, Lauren A. 2015. Authorized Status, Limited Returns: The Labor Market Outcomes of Temporary Mexican Workers. Economic Policy Institute, May 2015.

Baker, Bryan, and Robert Warren. 2024. Estimates of the Unauthorized Immigrant Population Residing in the United States: January 2018–January 2022. Office of Homeland Security Statistics, U.S. Department of Homeland Security, April 2024.

Banerjee, Asha, and Josh Bivens. 2023. Lessons from the Inflation of 2021–202(?). Economic Policy Institute, April 2023.

Basso, Gaetano, and Giovanni Peri. 2020. “Internal Mobility: The Greater Responsiveness of Foreign-Born to Economic Conditions.” Journal of Economic Perspectives 34, no. 3: 77–98, Summer 2020.

Batalova, Jeanne, Julia Gelatt, and Michael Fix. 2024. How Immigrants and Their U.S.-Born Children Fit into the Future U.S. Labor Market. Migration Policy Institute, April 2024.

Bauer, Mary, and Meredith Stewart. 2013. Close to Slavery: Guestworker Programs in the United States. Southern Poverty Law Center, February 19, 2013.

Baugh, Ryan. 2023. U.S. Lawful Permanent Residents: 2022. Annual Flow Report. Office of Homeland Security Statistics, U.S. Department of Homeland Security, November 2023.

Bensinger, Ken, Jessica Garrison, and Jeremy Singer-Vine. 2016. “Employers Abuse Foreign Workers. U.S. Says, By All Means, Hire More.” BuzzFeed News, May 12, 2016.

Bernhardt, Annette, Ruth Milkman, Nik Theodore, Douglas Heckathorn, Mirabai Auer, James DeFilippis, Ana Luz González, Victor Narro, Jason Perelshteyn, Diana Polson, and Michael Spiller. 2009. Broken Laws, Unprotected Workers: Violations of Employment and Labor Laws in America’s Cities. Center for Urban Economic Development, National Employment Law Project, and UCLA Institute for Research on Labor and Employment, September 21, 2009.

Bernstein, Shai, Rebecca Diamond, Abhisit Jiranaphawiboon, Timothy McQuade, and Beatriz Pousada. 2022. “The Contribution of High-Skilled Immigrants to Innovation in the United States.” Working Paper no. 30797, National Bureau of Economic Research, December 2022.

Bivens, Josh. 2014. The Short- and Long-Term Impact of Infrastructure Investments on Employment and Economic Activity in the U.S. Economy. Economic Policy Institute, July 2014.

Bivens, Josh. 2022. “Will Friday’s Wage Growth Numbers Stop the Fed from Snatching Defeat out of the Jaws of Victory?” Working Economics Blog (Economic Policy Institute), July 7, 2022.

Bivens, Josh, and Heidi Shierholz. 2018. What Labor Market Changes Have Generated Inequality and Wage Suppression?: Employer Power Is Significant but Largely Constant, Whereas Workers’ Power Has Been Eroded by Policy Actions. Economic Policy Institute, December 2018.

Borger, Julian. 2018. “Fleeing a Hell the US Helped Create: Why Central Americans Journey North.” The Guardian, December 19, 2018.

Borjas, George J. 2019. “Immigration and Economic Growth.” Working Paper no. 25836, National Bureau of Economic Research, May 2019.

Broder, Tanya, and Gabrielle Lessard. 2023. “Overview of Immigrant Eligibility for Federal Programs.” National Immigration Law Center, updated October 2023.

Brown, J. David, Julie L. Hotchkiss, and Myriam Quispe-Agnoli. 2008. Undocumented Worker Employment and Firm Survivability. Working Paper no. 2008-28. Federal Reserve Bank of Atlanta, December 2008.

Bruno, Andorra. 2019. Immigration: U.S. Asylum Policy. Congressional Research Service, R45539, February 19, 2019.

Bureau of Labor Statistics (BLS). 2023. “Employment Projections and Occupational Outlook Handbook News Release” (news release). U.S. Department of Labor, September 6, 2023.

Bureau of Labor Statistics (BLS). 2024a. “Labor Force Characteristics of Foreign-Born Workers Summary” (news release), U.S. Department of Labor, May 21, 2024.

Bureau of Labor Statistics (BLS). 2024b. “Table 1.1 Employment by Major Occupational Group, 2023 and Projected 2033.” U.S. Department of Labor [data table], n.d. Last accessed August 1, 2024.

Butcher, Kristin, Lucas Cain, Camilo Garcia-Jimeno, and Ryan Perry. 2023. “Immigration and the Labor Market in the Post-Pandemic Recovery.” Chicago Fed Letter, no. 486, October 2023.

Card, David. 2012. “Comment: The Elusive Search for Negative Wage Impacts of Immigration.” Journal of the European Economic Association 10, no. 1: 211–215, February 1, 2012.

Centro de los Derechos del Migrante (CDM). 2013. Recruitment Revealed: Fundamental Flaws in the H-2 Temporary Worker Program and Recommendations for Change. January 17, 2013.

Centro de los Derechos del Migrante (CDM). 2020. Ripe for Reform: Abuses of Agricultural Workers in the H-2A Visa Program. April 2020.

Chishti, Muzaffar, and Kathleen Bush-Joseph. 2023. “In the Twilight Zone: Record Number of U.S. Immigrants Are in Limbo Statuses.” Policy Beat, Migration Information Source. Migration Policy Institute, August 2, 2023.

Chishti, Muzaffar, Kathleen Bush-Joseph, and Colleen Putzel-Kavanaugh. 2024. “Biden at the Three-Year Mark: The Most Active Immigration Presidency Yet Is Mired in Border Crisis Narrative.” Policy Beat, Migration Information Source. Migration Policy Institute, January 19, 2024.

Chishti, Muzaffar, and Charles Kamasaki. 2014. IRCA in Retrospect: Guideposts for Today’s Immigration Reform. Migration Policy Institute, January 2014.

Clemens, Michael A. 2022. “The Fiscal Effect of Immigration: Reducing Bias in Influential Estimates,” IZA Discussion Paper no. 15592, September 2022.

Congressional Budget Office (CBO). 2024. The Demographic Outlook: 2024 to 2054. January 18, 2024.

Costa, Daniel. 2016. The H-2B Temporary Foreign Worker Program: For Labor Shortages or Cheap, Temporary Labor? Economic Policy Institute, January 2016.

Costa, Daniel. 2017. “H-2B Crabpickers Are So Important to the Maryland Seafood Industry That They Get Paid $3 Less Per Hour Than the State or Local Average Wage.” Working Economics Blog (Economic Policy Institute), May 26, 2017.

Costa, Daniel. 2020. “Temporary Migrant Workers or Immigrants? The Question for U.S. Labor Migration.” RSF: The Russell Sage Foundation Journal of the Social Sciences 6, no. 3: 18–44, November 2020.

Costa, Daniel. 2021. Temporary Work Visa Programs and the Need for Reform: A Briefing on Program Frameworks, Policy Issues and Fixes, and the Impact of COVID-19. Economic Policy Institute, February 2021.

Costa, Daniel. 2022. Threatening Migrants and Shortchanging Workers: Immigration Is the Government’s Top Federal Law Enforcement Priority, While Labor Standards Enforcement Agencies Are Starved for Funding and Too Understaffed to Adequately Protect Workers. Economic Policy Institute, December 2022.

Costa, Daniel, and Ron Hira. 2020. H-1B Visas and Prevailing Wage Levels: A Majority of H-1B Employers—Including Major U.S. Tech Firms—Use the Program to Pay Migrant Workers Well Below Market Wages. Economic Policy Institute, May 2020.

Costa, Daniel, and Ron Hira. 2024. “EPI Comment on DOL’s RFI Regarding Schedule A Modernization.” Economic Policy Institute, May 13, 2024.

Costa, Daniel, and Philip Martin. 2023. Record-Low Number of Federal Wage and Hour Investigations of Farms in 2022: Congress Must Increase Funding for Labor Standards Enforcement to Protect Farmworkers. Economic Policy Institute, August 2023.

Costa, Daniel, Philip Martin, and Zachariah Rutledge. 2020. Federal Labor Standards Enforcement in Agriculture: Data Reveal the Biggest Violators and Raise New Questions About How to Improve and Target Efforts to Protect Farmworkers. Economic Policy Institute, December 2020.

Cotsirilos, Teresa. 2023. “The Dark Side of America’s Sheep Industry: Sheepherders Face Wage Theft, Isolation, Hunger and Alleged Abuse.” High Country News, October 2, 2023.

Davis, Carl, Marco Guzman, and Emma Sifre. 2024. Tax Payments by Undocumented Immigrants. Institute on Taxation and Economic Policy, July 30, 2024.

East, Chloe N., Annie L. Hines, Philip Luck, Hani Mansour, and Andrea Velásquez. 2023. The Labor Market Effects of Immigration Enforcement. Journal of Labor Economics 41, no. 4, October 2023.

Edelberg, Wendy, and Tara Watson. 2024. New Immigration Estimates Help Make Sense of the Pace of Employment. The Hamilton Project. The Brookings Institution, March 2024.

Eikenberry, Diane, and Katharina Obser. 2023. Opportunities for Welcome: Lessons Learned for Supporting People Seeking Asylum in Chicago, Denver, New York City, and Portland, Maine. Women’s Refugee Commission, November 2023.

Elmendorf, Doug. 2013. “CBO Releases Two Analyses of the Senate’s Immigration Legislation.” Congressional Budget Office Blog, June 18, 2013.

Employment and Training Administration and Wage and Hour Division (ETA and WHD). Improving Protections for Workers in Temporary Agricultural Employment in the United States. Final Rule. Published in the Federal Register, 89 Fed. Reg. 33898, April 29, 2024.

Fan, Eric, Zachary Mider, Denise Lu, and Marie Patino. 2024. “How Thousands of Middlemen Are Gaming the H-1B Program: New Data Reveal How Companies that Farm Out IT Workers Exploit Flaws in the Visa Lottery While Other US Businesses and Talented Immigrants Lose Out.” Bloomberg, July 31, 2024.

Faux, Jeff. 2017. “How US Foreign Policy Helped Create the Immigration Crisis.” The Nation, October 18, 2017.

Feliciano, Cynthia. 2020. “Immigrant Selectivity Effects on Health, Labor Market, and Educational Outcomes.” Annual Review of Sociology 46: 315–334, July 2020. 

Fernández Campbell, Alexia. 2018. “Trump Says Undocumented Immigrants Are an Economic Burden. They Pay Billions in Taxes,” Vox, October 25, 2018.

Figlio, David, Paola Giuliano, Riccardo Marchingiglio, Umut Ozek, and Paola Sapienza. 2024. “Diversity in Schools: Immigrants and the Educational Performance of U.S.-Born Students.” Review of Economic Studies 91, no. 2: 972–1006, March 2024.

Fung, Wenson, Kimberly Prado, Amanda Gold, Andrew Padovani, Daniel Carroll, and Emily Finchum-Mason. 2023. Findings from the National Agricultural Workers Survey (NAWS) 2021–2022: A Demographic and Employment Profile of United States Crop Workers. Research Report no. 17. JBS International for the Employment and Training Administration, U.S. Department of Labor. September 2023.

Galarza, Ernesto. 1956. Strangers in Our Fields. 2nd ed. Washington, D.C.: Joint United States-Mexico Trade Union Committee.

Garrison, Jessica, Ken Bensinger, and Jeremy Singer-Vine. 2015. “The New American Slavery: Invited to the U.S., Foreign Workers Find a Nightmare.” BuzzFeed News, July 24, 2015.

Gelatt, Julia, and Muzaffar Chishti. 2024. A New Way Forward for Employment-Based Immigration: The Bridge Visa. Migration Policy Institute, February 2024.

Ghertner, Robin, Suzanne Macartney, and Meredith Dost. 2024. The Fiscal Impact of Refugees and Asylees at the Federal, State, and Local Levels from 2005 to 2019. Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services, February 2024.

Gibbons, Eric M., Allie Greenman, Peter Norlander, and Todd Sørensen. 2019. “Monopsony Power and Guest Worker Programs.” IZA Institute of Labor Economics, Discussion Paper no. 12096, January 2019.

Government Accountability Office (GAO). H-2A and H-2B Visa Programs: Increased Protections Needed for Foreign Workers. GAO-15-154, reissued May 30, 2017.

Hinojosa-Ojeda, Raúl. 2010. Raising the Floor for American Workers: The Economic Benefits of Comprehensive Immigration Reform. Center for American Progress, January 7, 2010.

Hira, Ron. 2015. “New Data Show How Firms Like Infosys and Tata Abuse the H-1B Program.” Working Economics Blog (Economic Policy Institute), February 19, 2015.

Hira, Ron, and Daniel Costa. 2021. New Evidence of Widespread Wage Theft in the H-1B Visa Program: Corporate Document Reveals How Tech Firms Ignore the Law and Systematically Rob Migrant Workers. Economic Policy Institute, December 2021.

Howard, Troup, Mengqi Wang, and Dayin Wang. 2024. “How Do Labor Shortages Affect Residential Construction and Housing Affordability?” SSRN, February 16, 2024.

Hunt, Jennifer. 2011. “Which Immigrants Are Most Innovative and Entrepreneurial? Distinctions by Entry Visa.” Journal of Labor Economics 29, no. 3 (2011): 417–457, July 2011.

Kaiser Family Foundation (KFF). 2023. “Key Facts on Health Coverage of Immigrants.” September 17, 2023 (updated June 26, 2024).

Kallick, David Dyssegaard. 2013. Three Ways Immigration Reform Would Make the Economy More Productive. Fiscal Policy Institute, June 4, 2013.

Kallick, David Dyssegaard. 2023. “’Let Us Work’: The Wage Gain When Asylum Seekers Gain Work Authorization.” Immigration Research Initiative, September 7, 2023.

Kallick, David Dyssegaard, and Anthony Capote. 2023. Immigrants in the U.S. Economy: Overcoming Hurdles, Yet Still Facing Barriers. Immigration Research Initiative, May 1, 2023.

Kane, Tim, and Zach Rutledge. 2018. “Immigration and Economic Performance Across 50 U.S. States from 1980–2015.” Hoover Institution Economics Working Paper no. 18112, July 2018.  

Kmetz, Augustus, John Mondragon, and Johannes Wieland. 2022. “Remote Work and Housing Demand.” Federal Reserve Bank of San Francisco Economic Letter no. 2022-26, September 26, 2022.

Kolker, Abigail F., and William R. Morton. 2023. Noncitizen Eligibility for Employment Authorization and Work-Authorized Social Security Numbers (SSNs). Congressional Research Service, R47483, March 22, 2023.

Kossoudji, Sherrie A., and Deborah A. Cobb-Clark. 2000. “IRCA’s Impact on the Occupational Concentration and Mobility of Newly-Legalized Mexican Men.” Journal of Population Economics 13, no. 1 (2000): 81–98, 2000.

Kossoudji, Sherrie A., and Deborah A. Cobb-Clark. 2002. “Coming Out of the Shadows: Learning About Legal Status and Wages from the Legalized Population.” Journal of Labor Economics 20, no. 3 (2002): 598–628, July 2002.

Lacarte, Valerie, Julia Gelatt, and Ashley Podplesky. 2024. Immigrants’ Eligibility for U.S. Public Benefits: A Primer. Migration Policy Institute, January 2024.

Lapinig, Christopher. 2017. “How U.S. Immigration Law Enables Modern Slavery.” The Atlantic, June 7, 2017.

Lynch, Robert, and Patrick Oakford. 2013. The Economic Effects of Granting Legal Status and Citizenship to Undocumented Immigrants. Center for American Progress, March 20, 2013.

Manning, Alan. 2020. “Monopsony in Labor Markets: A Review.” ILR Review 74, no. 1. June 2020.

Marshall, Ray. 2009. Immigration for Shared Prosperity: A Framework for Comprehensive Reform. Economic Policy Institute, 2009.

Marshall, Ray. 2011. Value Added Immigration: Lessons for the United States from Canada, Australia and the United Kingdom. Economic Policy Institute, 2011.

Marshall, Ray, and Ross Eisenbrey. 2010. “Commission Needed to Solve Immigration.” The Hill, June 10, 2010.

Martin, Philip. 2020. “Mexican Braceros and US Farm Workers.” Wilson Center, July 10, 2020.

Meissner, Doris. 2004. “U.S. Temporary Worker Programs: Lessons Learned.” Migration Information Source. Migration Policy Institute, March 1, 2004.

Migration Policy Institute. 2022. Age-Sex Pyramids of U.S. Immigrant and Native-Born Populations, 1970-Present. Data tool. 2022.

Migration Policy Institute. 2024. U.S. Annual Refugee Resettlement Ceilings and Number of Refugees Admitted, 1980–Present. Data tool, n.d. [accessed July 1, 2024].

Moriarty, Andrew. 2024. “Temporary Protected Status (TPS): 5 Things to Know.” Policy Brief. FWD.US, February 29, 2024.

Mukhopadhyay, Sankar, and David Oxborrow. 2012. The Value of an Employment-Based Green Card,” Demography 49: 219–237, February 2012.

Mussa, Abeba, Uwaoma Nwaogu, and Susan Pozo. 2017. “Immigration and Housing: A Spatial Econometric Analysis.” Journal of Housing Economics 35: 13–25, March 2017.

Naidu, Suresh, Eric A. Posner, and E. Glen Weyl. 2018. Antitrust Remedies for Labor Market Power. University of Chicago Coase-Sandor Institute for Law and Economics Working Paper, September 21, 2018.

National Academies of Sciences, Engineering, and Medicine (NASEM). 2017. The Economic and Fiscal Consequences of Immigration. Washington, D.C.: The National Academies Press, 2017.

Orrenius, Pia. 2017. “New Findings on the Fiscal Impact of Immigration in the United States.” Federal Reserve Bank of Dallas Working Paper no. 1704, April 2017.

Orrenius, Pia, and Madeline Zavodny. 2014. “The Impact of Temporary Protected Status on Immigrants’ Labor Market Outcomes.” Federal Reserve Bank of Dallas Working Paper no. 1415, December 2014.

Ottaviano, Gianmarco .I.P., and Giovanni Peri. 2012. “Rethinking the Effect of Immigration on Wages.” Journal of the European Economic Association 10, no. 1: 152–197, February 1, 2012.

Papademetriou, Demetrios, Doris Meissner, Marc R. Rosenblum, and Madeleine Sumption. 2009. Aligning Temporary Immigration Visas with U.S. Labor Market Needs: The Case for a New System of Provisional Visas. Migration Policy Institute, July 2009.

Passel, Jeffrey S., and Jens Manuel Krogstad. 2024. “What We Know About Unauthorized Immigrants Living in the U.S.” Pew Research Center, July 22, 2024. 

Pastor, Manuel, and Justin Scoggins. 2012. Citizen Gain: The Economic Benefits of Naturalization for Immigrants and the Economy. Center for the Study of Immigrant Integration, University of Southern California, December 2012.

Peri, Giovanni. 2012. “The Effect of Immigration on Productivity: Evidence from U.S. States.” Review of Economics and Statistics 94, no. 1 (2012): 348–358, February 2012.

Peri, Giovanni. 2014. “Do Immigrant Workers Depress the Wages of Native Workers?” IZA World of Labor 42: 1–10, May 2014.

Peri, Giovanni. 2020. “Immigrant Swan Song.” Finance & Development Magazine, International Monetary Fund, March 2020.

Peri, Giovanni, and Chad Sparber. 2009. “Task Specialization, Immigration, and Wages.” American Economic Journal: Applied Economics, 1, no. 3: 135–169, July 2009.

Peri, Giovanni, and Reem Zaiour. 2021. Citizenship for Undocumented Immigrants Would Boost U.S. Economic Growth. Center for American Progress, June 14, 2021.

Pew Research Center (Pew). 2013. “Second-Generation Americans: A Portrait of the Adult Children of Immigrants.” February 7, 2013.

President’s Alliance on Higher Education and Immigration (President’s Alliance). 2024. Breakdown of Dreamer Populations—Both with and Without DACA. Updated May 23, 2024.

Rainey, Rebecca. 2022. “Wage Division Enforcement Declines Again in Wake of Hiring Woes.” Bloomberg Law, December 28, 2022.

Rainey, Rebecca. 2023. “Inadequate Labor Department Resources Stymie Enforcement Efforts.” Bloomberg Law. November 7, 2023.

Rivera-Batiz, Franciso L. 1999. “Undocumented Workers in the Labor Market: An Analysis of the Earnings of Legal and Illegal Mexican Immigrants in the United States.” Journal of Population Economics 12, no. 1 (1999): 91–116, February 1999.

Ruhs, Martin, and Philip Martin. 2013. “On Migration, the US Should Copy the UK.” Financial Times, February 18, 2013.

Ruiz Soto, Ariel G., Julia Gelatt, and Jennifer Van Hook. 2024. “Diverse Flows Drive Increase in U.S. Unauthorized Immigrant Population.” Commentaries. Migration Policy Institute, July 2024.

Saiz, Albert. 2007. “Immigration and Housing Rents in American Cities.” Journal of Urban Economics, 61, no. 2: 345–371, March 2007.

Saiz, Albert. 2010. “The Geographic Determinants of Housing Supply.” Quarterly Journal of Economics 125, no. 3: 1253–1296, August 2010.

Sanchez, Melissa, and Maryam Jameel. 2023. “OSHA Investigates Small Dairy Farms So Rarely That Many Worker Advocates Don’t Bother to Report Deaths and Injuries.” ProPublica, November 13, 2023.

Schuetz, Jenny. 2022. Fixer-Upper: How to Repair America’s Broken Housing Systems. Brookings Institution Press, February 22, 2022.

Sequeira, Sandra, Nathan Nunn, and Nancy Qian. 2017. “Migrants and the Making of America: The Short- and Long-Run Effects of Immigration During the Age of Mass Migration.” National Bureau of Economic Research Working Paper no. 23289, March 2017.

Shesgreen, Deirdre. 2018. “How US Foreign Policy in Central America May Have Fueled the Migrant Crisis.” USA TODAY, December 21, 2018.

Shierholz, Heidi. 2010. The Effects of Citizenship on Family Income and Poverty. Economic Policy Institute. February 2010.

Shrikant, Aditi. 2023. “Nobel Prize Winner Katalin Karikó Was ‘Demoted 4 Times’ at Her Old Job. How She Persisted: ‘You Have to Focus on What’s Next.’ ” CNBC.com, October 6, 2023. 

Smith, Shirley J., Roger G. Kramer, and Audrey Singer. 1996. Characteristics and Labor Market Behavior of the Legalized Population Five Years Following Legalization. U.S. Department of Labor, 1996.

Social Security Administration (SSA). 2023. 2023 OASDI Social Security Trustees Report. (Table V. A3.—Social Security Area Population on July 1 and Dependency Ratios, Calendar Years 1941–2100.) n.d., Accessed July 1, 2024.

Solf, Benedicta, Lindsey Guerrero, and Selena Sherzad. 2024. “Global Affordable Housing Shortages Can Harm Migrant Reception and Integration.” Migration Information Source. Migration Policy Institute, March 20, 2024.

Sumption, Madeleine, and Sarah Flamm. 2012. The Economic Value of Citizenship for Immigrants in the United States. Migration Policy Institute, September 2012.

Svajlenka, Nicole, and Trinh Q. Truong. 2021. “The Demographic and Economic Impacts of DACA Recipients: Fall 2021 Edition.” Center for American Progress, November 24, 2021.

Tedeschi, Ernie. 2024. “Immigration and the U.S. Economy Since the Pandemic: An Accounting Exercise: Immigrants Have Been an Important Part of America’s Recent Extraordinary Economic Growth.” Briefing Book (Substack), April 1, 2024.

United Nations High Commissioner for Refugees (UNHCR). Global Trends: Forced Displacement in 2023. Statistics, Data Science, and Survey Section, UNHCR Global Data Service, June 13, 2024.

U.S. Citizenship and Immigration Services (USCIS). 2024. “Who May File Form I-765?” in Instructions for Application for Employment Authorization, Form I-765, OMB No. 1615-0040, Expires February 28, 2027. U.S. Department of Homeland Security, August 28, 2024.

U.S. Citizenship and Immigration Services (USCIS). n.d. “Form I-765 Application for Employment Authorization, All Receipts, Approvals, Denials Grouped by Eligibility Category and Filing Type.” Fiscal years 2022 and 2023. U.S. Department of Homeland Security, n.d.

U.S. Department of Agriculture (USDA). 2024. 2022 Census of Agriculture. National Agricultural Statistics Service [full report and online data tables], February 2024.

Wage and Hour Division (WHD). 2008. Fact Sheet #48: Application of U.S. Labor Laws to Immigrant Workers: Effect of Hoffman Plastics Decision on Laws Enforced by the Wage and Hour Division (fact sheet). U.S. Department of Labor, revised July 2008.

Wage and Hour Division (WHD). 2023. About the Wage and Hour Division (fact sheet). U.S. Department of Labor, n.d. Accessed July 1, 2024.

Wage and Hour Division (WHD). 2024. “Agriculture Data Table.” U.S. Department of Labor, n.d. Accessed July 1, 2024.

Warren, Robert. 2024. After a Decade of Decline, the US Undocumented Population Increased by 650,000 in 2022. Center for Migration Studies, January 28, 2024.

Wilson, Jill H. 2024. Temporary Protected Status and Deferred Enforced Departure. Congressional Research Service, RS20844, updated May 28, 2024.

Wong, Tom K., Ignacia Rodriguez Kmec, Diana Pliego, Karen Fierro Ruiz, Silva Mathema, Trinh Q. Truong, and Rosa Barrientos-Ferrer. 2024. 2023 Survey of DACA Recipients Highlights Economic Advancement, Continued Uncertainty amid Legal Limbo. Center for American Progress, March 25, 2024.

• A spike in profit margins contributed significantly to inflation in the early part of the pandemic recovery, and likely contributed to even more persistent inflationary pressure by helping spur a countervailing rise in nominal wage growth. For example, rising profits explained well over 40% of the rise in the price level between the end of 2019 and mid-2022, compared with profits normally accounting for about 11-12% of prices.
• The profit spike was overwhelmingly due to pandemic distortions (shifting demand rapidly across sectors) and supply chain snarls (exacerbated by the Russian invasion of Ukraine) that granted many producers temporary monopoly power in key sectors.
  • Contrary to many influential economic writers and commentators, it is simply wrong to label the correlation between high profit margins and high inflation as simple evidence of an overheated economy. The overwhelming post-World War II evidence is that profit shares fall, not rise, as economies heat up.
• Corporate power absolutely conditioned how the post-pandemic inflation happened.
  • Corporate concentration likely did not increase during the post-pandemic recovery, and concentration over the previous decade was unlikely to by itself explain much of the post-pandemic inflation.
  • But the economic and policy context of the post-pandemic recovery saw corporate power dramatically change how it was deployed to maintain and expand profits—instead of suppressing wages, they raised prices. If this episode increases public support for measures that constrain excess corporate power, that would be good even if it has little relevance for inflation in the future.

Corporate profits largely explain the initial rise in inflation

Figure A below shows the share of price increases in the nonfinancial corporate sector that could be accounted for by rising profits, as well as the long-run share of profits in corporate output (a useful benchmark). In each case, the calculation starts from the last quarter of 2019 and moves to the quarter identified in the bar of the chart. This contribution has dropped steadily since 2021 but remains quite high relative to historic norms. Even as of the second quarter in 2024, corporate profits could explain roughly a third of the growth in the price level since the end of 2019, still much higher than the long-run average of just 11.5%.

Figure A

Figure B below highlights how front-loaded the profit spike was in the recovery. It shows a rough measure of profit margins—profits in the nonfinancial corporate sector (NFC) divided by the sum of labor and non-labor costs. This “markup” of profits over other costs rose to historically high levels early in the pandemic recovery. It stopped rising relatively quickly, but still has not retreated all that much from these historic highs. This means that since its peak in mid-2021, profit margins cannot explain much of the subsequent price inflation, but they failed to be a source of disinflationary pressure. If profit margins fall to pre-pandemic norms and this disinflationary pressure does eventually begin, it will be good news for the future path of inflation.

Figure B

Corporate profits spiked because of historically large but temporary supply chain disruptions—not an overheated economy

The coincidence of high profit margins and inflation early in the recovery led to many theories about what linked them. The most convincing theories centered on the role of pandemic-era supply disruptions granting some sellers temporary monopoly power that allowed them to raise prices without the normal fear that this would lead to customers flocking to competing firms. This profit shock to prices led to countervailing increased nominal wage demands from workers as they tried to insulate themselves from real income losses.

Further, once the higher profit margins were set, many firms seemingly used the episode to tacitly collude with competitors and keep margins high even as supply disruptions abated and there should have been more price competition between firms. Because collusion and oligopolistic behavior are hard to sustain in normal times, it seemed natural to expect these high profit margins would move back to pre-pandemic norms before too long. But this hasn’t happened yet, which has been a real surprise from this episode.

Many influential economic writers and some economists cast contempt at analyses highlighting the coincidence of high profit margins and fast inflation early in the recovery. While it was possible to overstate the causal chain running from profits to inflation, these writers often made an even worse mistake—proclaiming confidently that high profits are always and everywhere a natural occurrence when the economy heats up (i.e., when measures of aggregate demand catch up to and surpass measures of the economy’s potential output). This just isn’t true. In every single business cycle since World War II, in fact, the opposite pattern has held—as the recovery from recessions sees the economy “heat up” (sees lower rates of unemployment) the profit share falls, not rises. 

Figure C below shows the one-year change in profit shares and a measure of the economy’s “heat”. Specifically, we calculate the difference between the unemployment rate and estimates of the long-run natural rate of unemployment. This difference is sometimes referred to as the “unemployment gap” and it measures how close actual unemployment is to an estimate of the lowest unemployment rate the economy can sustain without seeing inflation accelerate.

The economy is running hot when this unemployment gap is low (or negative), while the economy is cooling when the gap is high. Using data since 1948, the figure shows a clear and extremely strong positive relationship, with the profit share rising as the economy cools and falling as it heats up. This is exactly the opposite relationship that would hold if the confident proclamations that high profits are simply a sign of a hot economy were true.

Figure C

Besides rebutting claims that high profits are a clear sign that inflation is coming from an overheated economy, this relationship in Figure C should have informed discussions of monetary policy more than it did. It was often claimed in 2022 and early 2023 that the labor market was unambiguously overheated in unprecedented ways. Some measures of the labor market were historically strong—like the record high number of vacancies during this time. But other measures—like real wage growth and the profit share of income—were actually showing values much more consistent with slack labor markets. This should have at least shaken confidence that the labor market needed immediate cooling to normalize inflation, and raised questions about whether very standard macroeconomic reasoning should be applied seamlessly to such a strange period of economic history.

In the end, the macroeconomic lessons to be learned from recent years’ inflation mostly boil down to avoiding mammoth supply disruptions and sectoral shocks. The traditional macroeconomic diagnoses of inflation—monetary and fiscal policies that are too stimulative—explain very little about post-pandemic inflation.

The inflation of 2021-2023 had huge distributional consequences—which means it was indeed about corporate power

Excess corporate power is the backdrop of nearly all economic trends in recent decades. However, for most of the post-1979 period until the pandemic, this corporate power was mostly leveraged to suppress wage growth for workers rather than raising prices charged to customers.

The circumstances of the post-pandemic recovery changed the easiest path to profitability for companies—unit labor costs rose quite fast in historic terms, but profit growth ran faster, and the combination of fast-rising unit labor costs and thickening profit margins led to rapid price inflation. The end result was largely the same as in past recoveries from steep recessions—there was a rapid increase in the share of total income claimed by profits rather than going to workers’ pay.

It should be noted again that not every inflationary episode is associated with this kind of sharp distributional change. As inflation gained momentum steadily through the late 1960s and then spiked sharply in 1973 and 1979, there was no large redistribution toward profits. In the early 1980s, high unemployment led to this redistribution toward profits, but inflation did not. In short, there is something different this time around in the profits/prices interplay, and it deserves more analysis than it has generally gotten.

If this later round of redistribution toward profits helps build support for measures like more robust anti-trust enforcement that tamp down excess corporate power, then some good might result. While it is true that corporate concentration did not directly lead to the inflation of recent years and that corporate power more broadly did not necessarily increase over this period, it is also true that lack of inflation is not a sign that corporate power has been tamed. Again, the 2010s saw extremely subdued inflation even as income was redistributed away from typical workers through wage suppression.

Conclusion

Profits don’t explain everything about recent years’ inflation. But ignoring trends in profits over this time makes inflation analyses much weaker. Further, a future normalization of profit margins could be another key source of disinflationary pressure in coming years.

Being relatively sanguine about the effect of technology and AI on labor markets does not imply that we think labor markets have been working well for U.S. workers. On the contrary, unemployment has been too high and wage growth too slow for decades. But the roots of labor market dysfunction—both past and future—have very little to do with technological changes. Instead, this dysfunction is driven by the concerted policy push to exacerbate the extreme imbalance of power between typical workers and the corporate managers and capital-owners who hire them.

It is important to get the facts and analysis right on the questions of why labor markets have not delivered enough jobs or acceptable wage growth, and what the real threats are to decent jobs in the future. Faddish debates about AI distract attention away from the more fundamental problem of imbalanced power in labor markets, pulling policy in less useful directions.

More specifically, we argue:

• Interpretations of past episodes of rising wage inequality—whether they were driven by changes in technology or changes in policy, institutions, and norms—differ enormously based on one’s assessment of employers’ ability to exercise power in labor markets. If this power is great, then policy, institutions, and norms have great scope to influence wage inequality. If instead employer power is limited, technological change becomes the major force driving inequality.
• Technology manifests most directly in measured economic statistics as an increase in productivity—the amount of output generated in an average hour of work in the economy. Productivity growth has not historically been associated with higher unemployment or higher inequality, meaning that worries that technological change could be driving a jobless future have yet to materialize.
• Economic research claiming that the very rapid rise in wage inequality in the 1980s through the mid-2000s was caused by the rapid introduction of new technologies (mostly the spread of personal computers and other information and communications technologies) has not stood the test of time; few economists today would highlight the impact of technology alone as a driver of this inequality.
• While it is possible that technology can reduce the demand for specific jobs, these job losses can be more than counterbalanced by expanding employment in other sectors, as long as we maintain aggregate demand.
• In labor market models that allow for employer power, technological change in and of itself is largely neutral in its effect on the distribution of economic growth. But when employers exercise unbalanced power in wage-setting, they are often able to use new forms of technology to claim more of a firm’s output at the expense of typical workers. However, it is the unbalanced power that is the root of this problem—not technological change per se, which could easily boost workers’ wages if deployed in more balanced labor markets.
• Given this history of technology and labor markets, there is very little AI-specific labor market policy that will do much to help workers. Instead, policymakers should focus on broader policy levers to boost workers’ leverage in wage bargaining that will aid workers in claiming the potential gains spurred by AI in the future and reclaiming lost ground from past periods of economic growth. AI-specific provisions in workplace negotiations and collective bargaining agreements, of course, make lots of sense. How AI—or any technological tool—can be deployed to raise productivity and foster broad-based wage growth instead of increasing employer control will be a crucial question for many workplaces. But the best policy support for this process that can be given by national policymakers is strengthening worker voice and power, not trying to micromanage how AI is used in specific workplaces.

Background on past waves of concern regarding technology and labor markets

Concerns that technological changes can cause labor market distress for workers has a long history. The term “Luddite,” for example, has its origins in a movement of British textile workers in the early 19th century who opposed the introduction of new machinery they feared would displace their jobs.

In more recent decades, there have been waves of popular concern regarding technological advances as either a direct threat to workers’ well-being or an enabler of other threats (like globalization). In the early 1980s, for example, the rise of personal computers raised fears of “technological unemployment.”1 In the early 2000s, IT-enabled growth of “white-collar offshoring” was cast as a major threat to U.S. workers.2 In the 2010s, the introduction (real or imagined) of robots and autonomous vehicles was argued to imminently threaten huge swathes of the workforce.3 And, of course, in the last year or two, advances in AI have spurred a multifaceted debate about its impact—including its potential labor market effects.

Much of this concern over technological changes in recent decades has coincided with undeniably bad outcomes for most workers in the U.S. labor market. The post-1979 period has seen unemployment at excessively high levels for extended periods and wage growth for typical workers slow dramatically relative to what prevailed in the first three decades following World War II. Wage growth has slowed even relative to the much slower pace of economywide productivity growth that has characterized the post-1979 period. Slow wage growth for most workers has led to sharply higher levels of wage inequality, along with a shift of income away from labor compensation and toward business incomes (particularly corporate profits).4

However, despite the concern about the effect of technological change on labor markets—and even despite the objectively poor performance of labor markets for most U.S. workers in recent decades—the effect of technological change has been generally positive when looked at from the perspective of the U.S. working class writ large. The anemic wage growth since 1979 for the typical worker would have been far smaller, and perhaps even negative, had there been no technological advances and no corresponding increase in labor productivity since that time.

But the full potential boost to living standards that technology could have provided has been more than swamped by the declining leverage and bargaining power of typical workers over this same period. The shift in labor market power away from typical workers and toward employers is the result of intentional changes in public policies, institutions, and norms. Key examples include failures to protect workers’ right to organize unions from growing employer hostility, raise the federal minimum wage for long periods of time, and the maintain extended periods of very low unemployment. It is these intentional policy decisions, not technological progress itself, that have redistributed so much income away from typical workers and toward corporate profits and those at the very top of the pay scale (CEOS and other corporate managers, for example).5

Before walking through the economics and data supporting this statement, it is important to note one powerful piece of anecdotal evidence regarding the technological dog that didn’t bark. We highlighted waves of concern about technological advancements in the 1980s, early 2000s, 2010s, and today. We could not find any serious wave of concern from the mid- to late 1990s. This should be strange. The 1990s saw the technological advance with by far the greatest effect on the economy in several decades—the introduction of the Internet and the rise of e-commerce—often at the expense of brick-and-mortar retailers. Unlike the other waves of popular concern surrounding technological changes, the rise of the Internet in U.S. economic life really did show up in key statistics (Figure A clearly shows a sharp uptick in productivity growth in the 1990s business cycle, for example). Yet the late 1990s (even in real time) was generally seen as a period of broad-based prosperity and healthy labor markets.6 The explanation for this perception is simple: for the first time in decades, unemployment was driven low enough to generate opportunities for many who had been shut out of job markets and spur genuinely healthy wage growth for most workers. In short, technology—even the significant acceleration of technological advance in the late 1990s—was never really a headwind to decent labor market performance. Instead, the headwinds were all poor policy choices and changing some important ones (like allowing an extended period of very low unemployment) improved labor market performance radically, even in the midst of the most rapid technological change in decades.

Getting the story right on technology, inequality, and labor market dysfunction is crucially important for making the right policy decisions. Efforts to blame inequality and unemployment on bloodless, apolitical forces like “technology” constitute a convenient alibi for those social forces supporting the concrete policy changes that actually drove these outcomes. This technology alibi has been extraordinarily effective in distracting attention away from the major causes of rising inequality and anemic wage growth. As the debate over AI’s potential effect on labor markets begins, this history of technology-as-alibi needs to be kept front and center in the minds of analysts and policymakers alike.

A concrete example illustrates how myopic focus on new technological trends can divert attention away from the root causes of labor market dysfunction. In the mid-2010s, long-term unemployment (unemployment spells exceeding six months) was particularly high and had been for years. Around this time, many employers were using automated data systems to sort through job applications. As the automated systems ranked job applicants, they were frequently programmed to instantly reject applicants who had not worked in the previous six months. This obviously exacerbated the problem of re-employment facing the long-term unemployed, and proposals were floated to bar employers from undertaking this kind of application sorting.7

But this proposed solution was severely flawed relative to the optimal response. The reason why long-term unemployment was high in the 2010s was because overall unemployment was high. Aggregate demand (spending by households, businesses, and governments) was too low to absorb enough willing workers to meaningfully push down unemployment (either short- or long-term). Policy efforts to boost aggregate demand could have quickly lowered overall unemployment, and long-term unemployment would have quickly followed suit. We know this is true because as unemployment fell steadily (if slowly) into the late 2010s, long-term unemployment fell even more rapidly.8

Essentially, a severely damaged macroeconomy was inundating employers with far more applications for each job than they felt capable of processing efficiently, so these employers used a technological advance (automated hiring software) as a shortcut for sorting applications based on long-term unemployment. Barring employers from using this coping strategy for dealing with the excess of applications over job vacancies would not have solved any society-wide problem. Employers still would have faced too many applicants per job and would likely have just moved onto some other application sorting shortcut. A common one was ratcheting up educational credentials required for the job despite the underlying work not really demanding these credentials.9

Crucially, while barring employers from using unemployment duration as a criterion in their automated application sorting processes might have resulted in some long-term unemployed worker getting a job, this job would have come directly at the expense of another worker who was also unemployed. Again, the main labor market problem in the mid-2010s was too few jobs per jobseeker. Changing how these too few jobs were allocated would have done little to improve aggregate human welfare over this period. But generating more jobs through expansionary macroeconomic policy would have solved this underlying problem and improved aggregate human welfare enormously. Focusing on the technological fad (automated hiring systems) and missing the deeper economic problem (a shortfall of aggregate demand) led to a much less constructive policy debate.

We worry that concerns about AI’s potential effects on labor markets will prompt a rush to construct targeted AI-specific policies—as has happened over and over again in U.S. policy debates on technological change. These policies mostly will not materialize at all because policymakers will soon be distracted by the next fad. Even if some policies do get constructed, they would be mostly ineffective in making labor markets better for workers and will divert valuable attention away from other policies that would actually improve labor market functioning.

Is it possible we’re wrong and AI will be the technological change that finally drives bad labor market outcomes for the vast majority? It’s possible. But there’s no evidence of it doing that yet and the nature and history of how technology affects labor markets argues that it is policies bolstering typical workers’ bargaining position in labor markets—not the newest development in AI—that should preoccupy policymakers who aim to deliver better labor market outcomes for workers in the years to come.

Key definitions, issues, and questions about technological change and labor markets

The remainder of this report will focus on key concepts, definitions, issues, and questions about technological change and labor markets.

In section 2, we provide a brief overview of two competing models of the labor market. The choice of which model best describes the functioning of real-world labor markets is crucial in assessing how technological changes can affect labor market outcomes, and which influences (technology or institutional change) have driven historical trends in wage growth and inequality.

In section 3, we explain how economists tend to measure technological progress—as movements in productivity growth.

In section 4 we assess broad empirical correlations between faster productivity growth (or an increased pace of technological progress) and overall labor market outcomes.

In section 5 we evaluate the claims of some economists that particular forms of technological change have altered the relative demand for large classes of workers in competitive labor markets, and hence have driven much of the rise in inequality of pay seen in recent decades. We find these claims lacking in key evidence.

In section 6, we note some arguments surrounding the effect of technological change on labor markets that have not received enough attention from mainstream economists: the role of technology as a tool for employers to boost their leverage in pay-setting versus typical workers. However, we note that the root cause of this problem is unbalanced labor market power, not technology qua technology, which could in theory be just as easily used to boost workers’ power as degrade it.

Finally, we sum up what this analysis implies for policy and what should preoccupy policymakers looking for real solutions to boost workers’ pay and improve their labor market outcomes.

Competing models of the labor market

In recent decades, a key debate in labor economics has been determining which changes in the economy are responsible for the large rise in wage inequality since 1979. Since the late 1970s, only workers at the top of the wage distribution (those earning more than 90% of all other workers) have seen growth in wages that approaches growth in economywide productivity. Wage growth for workers below the 90th percentile has substantially lagged productivity growth.10

Two competing explanations for this rise in wage inequality are: first, technological change that has decreased the relative demand for less credentialed labor (sometimes called skill-biased technological change, or SBTC) and second, institutional changes (like the decline of unions, the erosion of the federal minimum wage, and a change in macroeconomic policy priorities) that undercut typical workers’ leverage and bargaining power in the labor market. 11

It is often underrecognized that the outcome of the debate over the sources of wage inequality hinges almost entirely on what one assumes is the correct underlying model of the labor market: one where labor markets are competitive and power is roughly balanced between workers and employers, or one where employers have structurally greater power than typical workers.

Those who emphasize technological change as the root of wage inequality are invariably working with a model that assumes labor markets are competitive. In these models, workers and employers are equally powerless, and wages and employment are set by the intersection of demand and supply curves in competitive markets for labor, with very little scope for the economy to diverge from these competitively determined levels without adverse consequences. Crucially, this means that only one employment level is consistent with a given wage level and vice-versa—wages and employment are jointly determined by the same underlying forces, and this means that any influence that affects one of these necessarily affects the other. Given this model of the labor market, it is natural to react to large changes in wages or employment for any group of workers by postulating that something must have shifted either relative labor demand or supply.

“Relative” labor demand or supply means demand or supply of one type of labor relative to other types of labor. So, for example, if employers decided that college-educated workers were growing more productive and valuable over time (say because they had more facility with new forms of technology), relative labor demand would increase for workers with college credentials, while relative labor demand would decrease for those without these credentials. The result would be both wage and employment levels rising for college workers and falling for noncollege workers.

Much of the economic research making strong claims that the rise in wage inequality over recent decades is driven by technological change relies on competitive models of the labor market. It is important to realize the strong role that this assumption of a competitive labor market plays in this research. Real-world trends in relative labor supply are easy to observe in data on the size of the workforces with and without college degrees. The relative wage can also be seen in the data—it’s the ratio of average wages for workers with a four-year college degree to the average wages of other workers. However, these two observable datapoints are often combined with the assumption of competitive labor markets to infer trends in relative demand for different types of labor. Often these inferences of trends in labor demand are incredibly influential in public debate. For example, the claim that the introduction of personal computers drove inequality in the 1980s and 1990s is often a direct statement about the inference that technology shifted the relative demand for workers without a college degree. Yet direct evidence of economic influences that reliably shift demand or the timing of when they might have happened is extremely thin.12

Those who emphasize the importance of institutional change for wage inequality are nearly always working with a labor market model that includes substantial employer-side market power. The source of this power may vary. It can include traditional monopsony power stemming from too few employers, dynamic monopsony power stemming from informational and logistic frictions associated with job changing and search, employer choice in how effort is elicited from workers (through costly monitoring or higher wages), or some other source.

Frictions and unbalanced power in labor markets mean that a range of influences besides workers’ own productivity affect wage levels and their evolution over time. Manning (2003) has argued that frictions in real-world labor markets make changing jobs costly to workers, and hence effectively grant employers substantial “monopsony power” over their employees. Some of these frictions that make job changes more costly include things like researching and applying for alternative jobs, changing commuting schedules, rejiggering child care arrangements, switching health insurance plans, and breaking social ties with work colleagues.13 No single one of these frictions imposes costs that are high enough to prevent any job switching from happening, but the accumulated drag of some or all of them can substantially blunt the potential of labor market competition to boost workers’ wages. Further, even quite small reductions in competition spurred by these frictions can lower wages significantly.

Of course, a literal labor market monopsony would be one in which only one single employer existed, which would obviously keep competitive pressures from working to help workers bargain for higher wages. The Manning (2003) model does not require just one employer or even some arbitrarily small number of employers; it only requires that some employers are able to exploit the real-world fact that the costs of switching jobs for workers is nonzero. If this cost of job switching is a part of the baseline model of labor markets, it grants employers considerable power.

Besides this baseline reality of costly job changing, other forms of employer power stem from realities of the production process within firms in capitalist economies.

For example, Bowles (1985) points out that employers must hire workers to produce output, but must also elicit effort from these workers. Employers’ main leverage to elicit effort is the threat to fire workers found to be shirking. Firms have two main instruments to maximize leverage from this threat: they can monitor workers intensely—so that any shirking is highly likely to be detected—or they can pay high wages to intensify the pain of losing a job. Either higher monitoring or higher wages can incentivize workers to expend more effort and shirk less. Both strategies are costly to employers: to implement the monitoring strategy, they must hire managers who do not contribute directly to production, but instead just oversee workers’ effort, whereas to implement the high-wage strategy, they must increase the pay of workers directly involved in production. In some cases, if the “outside” wage available to workers is one generated in a labor market characterized by substantial monopsony power, a high-wage strategy adopted by a firm to elicit effort can counterbalance the depressing wage effects of this monopsony power.

Regardless of the source of market power, recent cutting-edge research has demonstrated how far from the competitive ideal most labor markets truly are. The key effect of this employer-side power is to make the range of possible wage-employment level combinations set in the labor market much wider than is possible in competitive models. A given employment level can be consistent with a wide range of wage levels. This “range of indeterminacy” can explain, for example, why large increases in mandated minimum wages are often found in empirical studies to have no significant effect on employment levels.14 This noneffect of minimum wages on employment, conversely, would be hard to explain with competitive models where a single combination of wage and employment level is determined jointly by the intersection of demand and supply curves. Hence, the potential role for institutional change to significantly drive inequality—even absent any change to underlying demand and supply for labor—is much larger in models of labor markets with employer-side power.

For decades, the assumption that labor markets are best represented by competitive models was widely adopted across the economics profession, and this naturally channeled much research about rising inequality into searches for “demand-shifters” like technology. More recently, models with employer-side power have become much more prominent in labor market debates, and the possible scope for institutional change to drive trends in wages and inequality has been more widely recognized.15 Now, debates over the drivers of wage inequality in recent decades require a much higher empirical burden of proof than they did in the past, when the assumption of competitive labor markets lead almost inevitably to the conclusion that technology played a key role.

To put our cards on the table, we believe the evidence strongly supports a view of labor markets where employer power is significant, and that direct evidence of technological change having first-order effects in changing relative demand for labor is extremely thin (we highlight some of this evidence and its thinness in a later section). But putting this debate front and center when discussing the potential effects of technological change on wages and employment is a useful practice going forward regardless.

How economists typically measure technological progress: productivity growth

Economists generally measure technological progress as an increase in economywide productivity. There are two main ways that productivity and productivity growth are measured. First, labor productivity is the amount of income generated in an average hour of work in the U.S. economy. This income includes wages, but also business income (including corporate profits), rents accruing to landlords, and other forms of income. Second, total factor productivity (TFP, sometimes also called multifactor productivity) growth measures how much output has grown after accounting for the growth of all measurable inputs, such as labor and contributions from capital services (like factories and machines). Economists often focus more on TFP as a measure of pure technological change. However, in the rest of this paper, we will focus more on labor productivity and argue that it maps more directly onto popular conceptions of how technology might influence economic outcomes.

Labor productivity—or the income generated in the average hour of work in the U.S. economy—rises consistently over time. These increases are why the current generation is on average so much richer than their ancestors—the average hour of work in the economy of 2023 generated far more income than the average hour worked in (say) 1960. Labor productivity has grown steadily—if inconsistently—for the last century or more in the United States.

This paper focuses more on trends in labor productivity, because we think most people understand the broader determinants of growth in labor productivity as being reflective of technological change.16 For example, most people would see the introduction of digital scanners and computers as a key way that technological progress has changed how people perform work in recent decades. In contrast, many people would find it odd or too limiting to hold constant the effect of computers when assessing the influence of technological change on the labor market.

Figure A highlights trends in labor productivity growth and the contribution of its drivers over U.S. business cycles since World War II. The most striking finding from this analysis is that productivity growth over the most recent business cycles has been historically slow, not fast. This alone provides key context for current debates about how the economy can absorb technological progress: any technology-induced job destruction allowing a given hour of work to produce more income—and hence substitute more sharply for labor—has substantially slowed in recent decades. Yet breathless reporting on today’s technological advances often ignores this, or even outright claims the opposite.

Figure A

By far the biggest slowdown in the contributors to labor productivity growth has been in the category of TFP growth—or the “pure” form of technological change. The first implication of these trends is obvious: if rapid technological progress is feared to cause labor market problems, were these labor market problems more pronounced in past business cycles, when this technological progress ran faster? The next sections address this question.

Can accelerating technological progress cause mass joblessness?

If we define technological progress as the ability to produce more output in a given hour of work, this often raises an obvious concern: Won’t less labor be needed over time, causing mass joblessness?

The answer is a clear “no.” While it is true that the level of unemployment at any given point in time is in part a function of the economy’s productivity, there is another variable—aggregate demand—that policymakers have significantly more control over and which can be adjusted to keep unemployment low, regardless of productivity trends.

Unemployment rises when the economy’s potential output exceeds aggregate demand. Potential output is a measure of how much an economy could produce if nearly all the economy’s willing workers were fully employed.17 A key determinant of potential output is productivity—any given employed workforce can produce more if productivity is higher. Aggregate demand is the amount of spending by households, businesses, and governments. When aggregate demand falls beneath the economy’s potential output, then unemployment rises. Say that there is a hotel with staff and rooms for 50 parties. If only 45 parties offer to rent these rooms, then five rooms and the workers to staff them will be unneeded. If this deficiency of demand is widespread across most sectors of the entire economy, then unemployment will rise as unneeded workers are laid off and not reemployed in other sectors.

Figure B shows estimates of the economy’s potential output, as well as actual measures of gross domestic product (GDP)—the value of the nation’s output and income in a given period. When actual GDP falls beneath potential, this means that aggregate demand is running more slowly than growth in the economy’s supply side, resulting in rising unemployment. Recessions are indicated by grey shading in the figure and they are defined by actual GDP falling beneath potential output.

Figure B

This logic might at first glance seem to buttress fears about technological progress generating unemployment: technological progress boosts the economy’s potential output, and if this boost pushes it above the economy’s aggregate demand, then unemployment can result. But the data show clearly that sharp changes in potential output (which is how technology-driven productivity jumps would show up in this data) is not behind the mismatches in aggregate demand and potential output that lead to recessions.

The determinants of potential output move slowly. The size of the labor force and the nation’s capital stock (and its state of technological sophistication) do not whipsaw around year to year. Instead, they tend to grow at a slow and predictable rate over time. Aggregate demand is far more volatile and can whipsaw quickly from year to year. For example, when the bubble in home prices began deflating in late 2006 and 2007, households immediately began spending less money and saving more to make up for the lost value of wealth, leading quickly to the severe 2008–2009 recession.18 Similarly, in early 2020, the labor force available to firms in the face-to-face services sector did not disappear and cause an employment collapse. It was customers who disappeared as fears of COVID-19 spread, and it was this demand shock that led to mass layoffs in the early part of that year.19

But just as aggregate demand can fall quickly, policy efforts can boost it quickly to ensure recessions are short-lived and recoveries are rapid. Aggregate demand can be boosted through either monetary or fiscal policy interventions to boost demand, with the Federal Reserve using monetary policy tools (like interest rate cuts), and Congress and the president setting taxes and spending at the levels needed (in practice, fiscal policy turns out to be the more powerful tool). Support for the statement that policy can quickly restore falls in aggregate demand is provided by the U.S. economic recovery from the COVID-19 recession. In December 2020, after the low-hanging jobs created by simply reopening the economy after the first wave of the pandemic had been restored, the unemployment rate was 6.7% and job growth had turned negative. Absent further policy efforts, there was a real possibility of stagnation at this high rate of unemployment. But due to further fiscal recovery packages passed in December 2020 and March 2021, by the end of 2021, the unemployment rate had already fallen below 4% again—essentially matching the immediate pre-pandemic level.20

Table 1 highlights this point about which variable—aggregate demand or potential output—moves more quickly (and in the right direction) to cause periods of joblessness. It shows growth rates for both actual and potential GDP in the year before recessions have hit the U.S. economy, and then over the subsequent recession. It then calculates the “swing” in these growth rates—how much they changed as the economy entered recession. Crucially, any sharp divergence of real GDP from potential output is caused by changes in aggregate demand.

In all cases, real GDP growth has swung sharply from positive to negative in the first year of recessions, by an average of 4.6% in the five business cycles before the COVID-19 pandemic (the COVID-19 recession was so extreme that we will set it aside for now). Estimates of potential output slowed as well, but only by an average of 0.4% over these same business cycles. Further, slowing potential output growth can reduce unemployment if it represents a slowdown in productivity growth, so this slowdown in estimated potential output puts downward—not upward—pressure on joblessness. In short, the wrenching change that causes recessions and rising unemployment is not an acceleration of technological progress making labor unnecessary—again, potential output decelerated in each of these periods. Instead, the pronounced change is the rapid deceleration and outright fall of real GDP, which, given trends in potential GDP, must by definition have been caused by a fall in aggregate demand.

Table 1

Figures C and D provide some slightly more systematic looks at the relationship between productivity growth and joblessness. In both figures, average values over an entire business cycle peak—from one peak to the next—are assessed. The dates on the dots in the figure mark the beginning of the business cycle. Figure C shows the average rate of productivity growth and the average rate of unemployment across business cycles since World War II. Contrary to worries about tradeoffs between fast productivity growth and low unemployment, fast productivity growth is associated with lower average rates of unemployment across business cycles.

Figure C

Figure D shows the relationship between average productivity growth and the change in unemployment rates between business cycle peaks. That is, it looks to answer the question: On average, fast productivity growth may be associated with lower unemployment, but does fast productivity growth over a business cycle keep unemployment from falling as fast as it could have? Again, there is no systematic relationship between the average pace of productivity growth and the decline of unemployment over an entire business cycle.

Figure D

Over the last completed business cycle (from 2007–2019), productivity growth averaged roughly 1.5%. The most highly optimistic projections for how much AI can boost the pace of productivity growth are about 1% per year (most other projections are quite a bit lower). This would move productivity growth from 1.5 to 2.5%—a level that the U.S. economy saw for decades following World War II, and which was accompanied by lower average unemployment than has persisted in recent decades.21 In short, there is nothing in either the historical relationship between productivity growth and unemployment or in projections of AI’s impact on productivity growth that indicate that this technological change will prevent policymakers from sustaining low rates of unemployment—should they choose to do so.

Does technological change ever displace jobs?

None of this is to say that specific jobs are not threatened by technological progress. Rapid technological change concentrated in any specific sector can reduce employment in those sectors. The analysis above simply says that the aggregate number of jobs and the overall rate of unemployment is unlikely to be threatened by an acceleration of technological progress, as long as policymakers respond appropriately by boosting aggregate demand.

As productivity rises following an acceleration of technological progress, job losses within sectors experiencing the productivity increase will be counterbalanced (or more than counterbalanced) by expanding employment in other sectors as long as aggregate demand is maintained. Autor and Salomons (2018) empirically estimate how employment responds to a sectoral productivity shock. They find that the own-effect of a productivity shock within a sector is indeed modestly negative, with the reduction in hours of work needed to produce output in the sector not fully offset by the rise in demand for the sector’s output, made cheaper by productivity growth. However, the cross-effect of productivity growth within a sector—the effect of its own productivity growth on employment in other sectors—is strongly positive, and outweighs the negative own-effect in terms of aggregate employment trends.

Take the example of a 1% increase in productivity in a specific economic sector like manufacturing. Autor and Salomons (2018) find that the average first-order effect of a sectoral shock (the own-effect) is to decrease employment in that sector by 0.1%. This is the intuitive effect most people think about when they worry that introducing more automation into production might displace human labor in that sector.22

But this 1% rise in sectoral productivity means that more income is being generated in each hour of work in that sector, and this extra income boosts employment when it is spent in other sectors. This positive “final demand effect” on jobs alone almost completely counterbalances the direct effect, adding almost 0.1% to employment. Additionally, the combination of productivity growth and competition in product markets lowers the prices of goods from the sector that has seen the positive productivity shock. In the example of manufacturing, this would provide a boost to employment in sectors that use manufactured goods as intermediate inputs (for example, a falling price of computers makes it less expensive to produce accounting services). These “upstream effects” boost employment by almost twice as much as the direct effects reduce it. Overall, the economywide net impact of these effects is an increase in overall employment stemming from productivity growth within a given sector.

How to reduce damage from sectoral job displacements in labor markets—whatever its cause

It is certainly true that some individual workers may suffer from sector-specific job displacements, even if aggregate unemployment or employment is unaffected. The labor market is not frictionless, and it may take some painful time before comparable employment in a new sector is obtained. Some workers (particularly older workers) may never find a specific job as good as the one they lost. Yet much of this individual suffering could be ameliorated with broad policies that provide better protective social insurance, more widespread collective bargaining, and sustained high-pressure labor markets—policies that are highly desirable regardless of the pace of technological change.

One reason specific jobs are occasionally highly valued in the U.S. labor market is because they come bundled with nonwage compensation—like health and retirement benefits, which are not universally available. But if these benefits were universally available through more protective social insurance systems, the damage done by the loss of any particular job would be greatly lessened. Another key social insurance system—unemployment insurance (UI)—is too stingy in the U.S., causing large income losses while workers search for alternative employment. Boosting the protectiveness of UI would be a key win for those looking to reduce the pain caused by the loss of particular jobs.

Another reason some specific jobs can be highly valued in the U.S. economy is because they are unionized. This should not be as rare as it currently is, but recent decades have seen a combination of employer hostility and policy indifference lead to a near shutdown of organizing unions in newly created jobs at any large scale. This means that sectors today that remain unionized do so largely because of a historical legacy that saw their unions formed decades ago; the chances of workers leaving this sector finding a unionized job elsewhere are slim indeed. In short, there are only rare pockets of unionized jobs in the U.S. economy and new ones are not being created fast enough. Hence, anything (including technological progress) which leads to the destruction of today’s unionized jobs are likely to leave many of their former holders worse off.

Additionally, the U.S. economy has spent much of the past four decades with excessively high unemployment, which radically increases the cost of losing a job. When unemployment is low and vacancies are high, a worker who has lost their job can quickly find alternative employment, putting employers under constant pressure to keep job quality high enough to retain and attract new employees.

If U.S. policymakers created a more protective social insurance system, restored the effective right to organize unions, and maintained high-pressure labor markets with low unemployment, then a large part of the damage done by technologically induced job displacements would disappear.

Finally, despite all the possible challenges faced by workers who are displaced from specific sectors by technological change (or anything else), it is possible to both overstate how widespread these challenges are, and underestimate the value of new jobs and the higher productivity created by technological change.

How widespread is sector-specific “churn” caused by technology and has it increased?

Were technology responsible for a reallocation of jobs toward certain industrial sectors or occupations, we should expect to see an increased amount of employment flows with workers increasingly separating from jobs, and certain sectors losing and gaining shares of employment in the labor market. The U.S. labor market has always and everywhere been characterized by tremendous rates of job “churn”—workers separating from employers either voluntarily or involuntarily. For example, in the last year before the COVID-19 pandemic, 3.7% of all workers left their jobs each month. Similarly, 3.9% of all workers were newly hired each month (for a net change in employment each month of roughly 0.2%). Over a year, this is a huge amount of churn, with more than a third of the entire workforce changing jobs (or changing their employment status) each year. Yet this churn has been a feature of the U.S. labor market for decades, and most data indicates that it has actually slowed, not increased, in the 2000s—despite the proliferation of the Internet and large advances in computer hardware and software.

Figure E, reproduced from Bivens and Mishel (2017), clearly emphasizes this point. It shows the sum of the (absolute) change in occupational employment shares over various decades. To construct this metric, Bivens and Mishel examined the shares of total employment for 250 occupations at the beginning and end years of each decade and computed the changes in these shares. The metric shown in the figure is half of the sum of the absolute value of changes in occupational employment shares (taking only half of the sum avoids double counting gains and losses). This metric measures the share of total employment exchanged between occupations—or the measure of job churn between occupations—for each decade.

The decadal rates of occupational employment shifts, starting in the 1940s, are shown in Figure E. The rate of change was fairly uniform over the 1940–1980 period, and far more rapid than for any period since 1980. The period since 2000 has seen the lowest rate of change—half the rate of change of the 1940–1980 period.23

Figure E

Were technology causing massive displacements or reallocation, the data would have exhibited the opposite pattern. One important reason for the lack of widespread displacements is that technological increases can complement the tasks of workers, rather than permanently substitute away from particular occupations or industries. As a result, even though large shares of the labor market may be exposed to new technologies, much smaller shares of jobs would be destroyed entirely by automation. Indeed, some observers have in fact argued that AI provides “an opportunity to complement worker skill and expertise” (Acemoglu, Autor, and Johnson 2023).

Does technology reduce demand for workers without the right credentials or skills?

We argued in the previous section that technological change and increased productivity has not led to aggregate job loss or increased unemployment. Moreover, even in recent years, these forces have not even led to more rapid occupational churn in the labor market. However, many economists have argued that technological change was a major cause of growing wage inequality in the U.S. labor market in the post-1979 period, and that this technology-induced rise in inequality was the result of technological changes that boosted relative demand for workers with higher skills (almost always proxied by a four-year college degree). This technology narrative has been extraordinarily influential among policymakers, even as cutting-edge research increasingly casts doubt on it.

This shift in relative demand toward college workers, sometimes called skill-biased technological change (SBTC), has been a major focus of economic research in understanding the growth of U.S. wage inequality. The SBTC-based explanation of inequality relies on a model of competitive labor markets, where wages and employment of workers of different skill levels have their relative wages and employment levels set by the intersection of supply and demand. The SBTC theory claims that technological change has caused an increase in relative employer demand for college workers (presumably because these allegedly more skilled workers have greater facility with using new forms of technology), and this rise in turn led to higher relative wages (or a higher college wage premium) over the last several decades.

This stylized story simply does not fit the data. First, basic estimates of the relative demand for college labor suggest that the bulk of the growth in the college wage premium in the 1980s and 1990s is not due to an acceleration in employer demand for college labor, but a slowdown in the supply of college labor (therefore raising the price or wage of college labor). As Autor, Goldin, and Katz (2020) explain, “rapid and disruptive technological change from computerization, robots and artificial intelligence is not to be found” during these periods of massive innovation in computing technology. These authors (and others) often present this set of facts as demonstrating that inequality is the result of a “race between technology and education,” with technology presumed to raise relative demand for college graduates, while education conditions the supply. However, recent decades have clearly seen much more marked changes in the education/supply side of this race—and that leads to a narrative about the driver of inequality that departs significantly from stories that center technological change as the driving force.

Second, compared with earlier time periods, there has been little change in wage inequality between college and noncollege workers since 2000. Figure F shows the annual college wage premium over 1979–2023, controlling for demographic differences in the college versus noncollege population within each year. There was a sharp increase in the college wage premium in the 1980s and 1990s, but a much smaller rise since 2000, during the widespread adoption of computing at the workplace. In fact, there has been essentially zero change in college/noncollege wage inequality since 2010, so if anything, these wage patterns suggest a decline in the relative demand for college labor over the last one to two decades.

Figure F

In a preview of recent concerns over AI, the early recovery from the COVID-19 recession saw many expressing worries that employers would respond to the organizational changes they made in the era of social distancing to replace workers with technology. Casselman (2021), for example, wrote that:

An increase in automation, especially in service industries, may prove to be an economic legacy of the pandemic. Businesses from factories to fast-food outlets to hotels turned to technology last year to keep operations running amid social distancing requirements and contagion fears… But some economists say the latest wave of automation could eliminate jobs and erode bargaining power, particularly for the lowest-paid workers, in a lasting way.

As support, Casselman (2021) pointed to a 2021 working paper from the International Monetary Fund that argued: “Our results suggest that the concerns about the rise of the robots amid the COVID-19 pandemic seem justified” (Sedik 2021).

And yet, almost three years on, the post-pandemic labor market has actually been a huge source of strength for low-wage and low-credential workers. Autor, Dube, and McGrew (2023) show that after accounting for changes in the demographic composition of the workforce, the college/high school wage premium fell during the last two years. Instead of technological change widening the gap between those with more or fewer credentials, a tighter labor market during the 2021–2023 period compressed wages. Young, noncollege workers saw significant wage increases because the tighter labor market provided more opportunities to switch to higher paying jobs.

Employment rates for workers without a college degree are still worse than they were decades ago, but in aggregate, they are largely not determined by technological changes. An easy way to see this is comparing the United States to other advanced economy countries who have faced similar technological shocks but who have very different macroeconomic policy and social support systems. Figure G shows that in the United States, the share of the population with a high school but no college degree that is employed has dropped dramatically since 2000. In contrast, low-credentialed employment in other G7 countries has not experienced such falls. In some cases, like in Austria, Germany, and Great Britain, employment rates have grown for workers with just a high school education.

Figure G

Technology is a tool—the balance of labor market power determines who it helps

While most economic analysis of technology’s potential role in generating greater inequality in recent decades has focused on its effect in shifting demand and supply curves in competitive labor markets, this is often not how most informed consumers of news about the U.S. economy think about the effect of technology.

Instead, many media reports about technology’s role in the workplace—and how it might exacerbate inequality—focus on how it might be used as a tool for employer monitoring and speed up of workflow. For example, a well-known example of technology-enhanced monitoring is the “black box” installed in long-haul cargo trucks. Originally, these boxes were installed to validate that long-haul truckers were meeting mandated rest requirements for safety reasons. Now, however, one of the main appeals of the box for employers is to ensure that they only pay truckers for the time they spend actually moving cargo forward. Skott and Guy (2013) note that the producer of one of these black boxes boasts on their website that for trucking company managers, this technology “is like being able to sit next to every one of your drivers every second they drive.”

Another example is robots in an Amazon warehouse essentially setting the pace for human workers when processing packages. As Anway (2022) writes: “The clock was always ticking…As soon as she’d filled a rack, she’d press a button and one robot would zip it away while another robot would bring a new one to fill.” This high pace of work has been implicated in high rates of injury in these warehouses.

These examples have nothing to do with technology’s effect in shifting demand or supply curves for labor in competitive markets, and yet show technology as enabling exploitation and degrading job quality. A much smaller body of economic research highlights labor market models that shed light more directly on these situations.

Earlier, we described models where employers had market power and could choose between monitoring or high wages as strategies to elicit effort from workers. One could imagine technological change that reduces the cost of monitoring. This could induce firms to lean more heavily on the intensive monitoring strategy and less on the high-wage strategy. This in turn would lead to more workers having their wages set directly by the labor market outside the firm—an outside labor market which itself might be riven with employer-side power—and hence lower wages. This heavier reliance on monitoring would lower measured productivity, as more work hours in the firm would be spent monitoring other workers rather than producing output for sale. In this case, technological change would not be boosting productivity (it would in fact be lowering it) and would instead only be leading to a zero-sum (or even negative-sum) redistribution away from workers and toward employers and managers.

However, technological advance is not the only—and likely not the primary—determinant of whether firms choose a high-wage or a high-monitoring strategy. A bigger determinant is the relative bargaining power of workers. If the workers at a given firm manage to organize a union, for example, the choice is essentially decided: wages will be higher, and workers will value the unionized job more than what they can get in the outside labor market (and hence will expend more effort).

The example of the trucking black box monitoring technology makes this clear. Originally the box was thought necessary to keep truckers from breaking safety rules regarding how much rest they got between spells of driving. In more recent incarnations, it is marketed as a device to ensure trucking companies do not have to pay for any time spent that does not move cargo forward. However, what this example makes clear is that it is the underlying power relationship, not the new technology, that determines wages and job quality. For example, if workers were paid sufficiently for the entire time commitment of hauling cargo (and not just for time actively driving), the worry that they would skimp on rest requirements to earn more money would be blunted. Further, while the black box is often referenced as a tool for employer control, the underlying technology could in theory solve a pressing problem for truckers: proving that large swathes of time they’re not actively hauling cargo is in fact necessary “company time,” as they are forced to wait to pick up loads at ports.

In the supply chain breakdown of 2021 and 2022, a key bottleneck to moving goods from producers to consumers was a backlog at ports. This backlog led to truckers often having to wait long hours (or even days) idling in a queue waiting to have their cargo loaded. And often, this wait time went uncompensated. One employer objection to paying for this wait time could be verification—the company only “knows” when the trucker is really working for them when the load is transferred to the truck. But that’s obviously not true—the same technology that verifies whether long-haul truckers are spending enough time actively driving to meet their contractual demands could also verify that truckers are indeed in an active queue waiting for cargo to be loaded.

The real reason why truckers have not been compensated for these wait times in recent decades is not the technical impossibility of verifying wait times, but instead simply the power of employers. Trucking was once a highly unionized, high-wage job. The push to deregulate and deunionize beginning in the 1970s substantially eroded the relative wages in this sector. One imagines that if the black box had been invented in the 1950s and employers tried to force its adoption by a more heavily unionized trucking workforce, it would have been successfully rejected by the then considerably more powerful Teamsters union.

In short, it is true that technology exists that might aid employers engage in zero-sum redistributions away from typical workers. But the exact same technology used by employers to wring more effort and profit out of workers could often in theory be used by workers to wring higher wages and productivity out of employers. The same robots that are implicated in a work pace that is injurious to workers at Amazon could be a genuine boon to worker safety if robots handled all heavy loads and did so at a pace that did not put undue stress on human workers. This pace is not dictated by technology, it is set by employers, too often in the context of highly unbalanced power. In short, the problem is almost never in the technology itself, and nearly always in the relative power relationships.

Recent flashpoints about hiring discrimination and IRS audits highlight that unbalanced power is the root problem

Discrimination in hiring processes and in federal tax enforcement are two key examples highlighting that it is power—of bosses and policymakers—that determines whether or not technology (including variants of AI) are used to ameliorate or exacerbate existing inequalities in U.S. society.

It is known that automated processes for employer hiring can use embedded discriminatory criteria when sorting applicants.24 This is obviously a real problem. Yet discriminatory criteria do not appear by magic in automated data processing systems; their logic is explicitly or implicitly programmed somewhere along the way. The best response to this issue has very little to do with the automated process itself: it is making firms legally responsible for the outcomes of their own hiring software’s decisions and providing regulators enough access and information to perform audits and measure the magnitude of bias in the hiring process. There are even reasons to believe that discriminatory criteria embedded in automated hiring systems will be easier to detect and solve than old-fashioned employment discrimination that largely happened inside the heads of hiring managers.25 Again, the underlying problem here is not technology, it’s the broader social context this technology operates in, which could in fact benefit from the use of technology in combatting some of its problems.

A final example highlights a potential danger of focusing on technology as the problem rather than the more foundational decisions embedded in technology. A recent paper looked at IRS audit rates for Black and non-Black taxpayers. They found Black taxpayers audited at substantially higher rates. They found this disparity (and other key features of IRS audits) could possibly be explained if the IRS was picking taxpayers for audits based on an algorithm that sought to maximize the share of underreported income that was accounted for by refundable tax credits (like the Earned Income Tax Credit).26

This is an odd target to maximize if you thought the point of audits should be to simply generate as much appropriate revenue as possible. Potentially, however, it is an understandable thing to maximize if you are an agency that has been swayed by unrelenting Republican attacks on refundable tax credits in the name of minimizing “fraud” perpetrated by low-income taxpayers. The authors also find that the audits fall heavily on returns with zero business income. This is again odd if you want to maximize unclaimed revenue, as business income is rife with underpayment. But this choice might make sense for an agency that has been starved of resources—business income returns require a lot more resources to audit than individual returns.

Both plausible maximization goals (focusing on refundable tax credits and not focusing on returns with business income) cut sharply toward increasing the share of Black households that would be selected for audits. They suggest (implicitly) a much better maximization goal to guide the audit selection algorithm: maximize underreported income, period. This goal would not only raise more revenue (the larger point of audits), but would also erase the race-based disparity in current audits.

Again, the issue is not algorithms or automation per se, it is the human choices behind them. More broadly, is there any question that advances in information processing (AI or otherwise) could be a hugely helpful tool for using IRS audits to maximize revenue if that is what the agency wanted? To put it simply, banning or severely constraining the use of AI or any other information-processing tool in the conduct of tax enforcement would be a huge win for wealthy tax cheats looking to escape taxation. It may be fanciful to worry about such bans, but given that Republicans in Congress have routinely sought to hamstring tax enforcement for decades, if fears and generalized bad feeling about AI becomes widespread across society, it may provide an opening for such destructive proposals.

Conclusion and policy implications—looking through the latest technological fad to see the real threats to workers’ well-being

It is no doubt a useful exercise to make sure public policies are tailored to specific forms of significant new technology that arise. So, a recent spate of proposed legislative and regulatory activity around AI has many sensible elements. But it is also extremely easy to focus too much on the latest form of technology and get distracted away from more structural reasons for why U.S. workers struggle to secure a decent living in the labor market.

Addressing these structural issues—the too-thin social insurance systems of the U.S., the impediments to organizing unions, and the failure to sustain low rates of unemployment—would not only boost workers’ wage growth across the board. It would also address most of the stress on smaller groups of workers experiencing job displacement due to technological change.

There is real harm to public analysis and policymaking that focuses so much attention on each new mini wave of technological advance as a cause for workers’ problems. The most obvious harm is that it’s a clear misdiagnosis as source of wage suppression. If one could somehow completely ban progress on AI, this would do nothing to improve workers’ lot in the future. If we could go back in time and ban research on robots or autonomous vehicles, wages today for workers would be no higher. Yet AI and robots and autonomous vehicles have sucked up more attention than the structural issues we referenced above from many who should sincerely be concerned with how U.S. workers are faring. The attention of policymakers, researchers, and advocates is a scarce resource, and every minute they are convinced they need to be constructing plans around the newest technological fad is a minute they are not working on issues of deeper concern to workers.

When this is recognized and how technology is used by workplaces becomes a focus of empowered workers, smarter workplace policy can result. Key examples of clear-eyed stances toward AI can be seen in recent negotiations between the AFL-CIO and Microsoft, and the negotiated role of AI in contract agreements between the Writers Guild of America (WGA) and the Alliance of Motion Picture and Television Producers (AMPTP). The AFL-CIO and Microsoft have recently come to an agreement that Microsoft would remain neutral in future organizing campaigns and agreed to future discussions about how AI can be used to improve workplace productivity and workers’ pay and working conditions (rather than be used as a cudgel to reduce workers’ leverage and bargaining power). The WGA contract with the AMPTP states that writers can use AI as a tool in their own work, but that AI cannot be used to undermine writers’ claims to credit for what they produce. There will clearly be unforeseen issues that will arise going forward, but these are encouraging first steps that clearly show that, in a balanced labor market (like when a union is present), issues regarding AI (and any other technological change) have a strong chance of being settled in ways that benefit workers.

Notes

1. See Marcus (1983) for a contemporaneous account of fears concerning “technological unemployment.”

2. See Engardio, Bernstein, and Kripalani (2003) for an example of fears being raised over the prospect of “white-collar offshoring.”

3. See Gilbert (2013) for a piece detailing the alleged threats robots pose to human employment.

4. See Bivens and Mishel (2015) for a review of the historical interplay between wage growth for typical workers and productivity, and for a decomposition of where the wedge between these workers’ pay and productivity growth went.

5. See Bivens and Mishel (2021) for a review of the research supporting the case that it is this policy-induced degradation of typical workers’ labor market power that drove the sharp slowdown in wage growth for these workers and the resulting increase in wage inequality.

6. See Krueger and Solow (2002) for a deep examination of the “Roaring Nineties.”

7. See Ghayad (2013) for documentation of this employer sorting by duration of unemployment.

8. See Bivens and Shierholz (2014) for real-time evidence on how deficient demand, not worker skills or employer behavior, was the real cause of elevated long-term unemployment.

9. See Modestino, Shoag, and Balance (2020) for evidence of this type of employer “upskilling” during periods of too-slack labor markets.

10. See Gould (2019) for an overview of U.S. wage trends since 1979.

11. For a time in the 1990s, globalization—not institutional change—was generally seen as the main competitor to SBTC as the dominant driver of wage inequality in the U.S. Since then, however, economists have increasingly settled on (what we consider to be) the correct view that globalization has had significant effects on wage inequality but remains insufficient to explain most of the rise in inequality. See Bivens (2017a) for a review of some of this debate on globalization.

12. Card and DiNardo (2002) and Schmitt, Shierholz, and Mishel (2013) provide extremely detailed examinations of the direct evidence supporting the SBTC view and find it lacking.

13. See Bassier, Dube, and Naidu (2022) on how individual firms do have discretion over wage levels and cannot simply hire as many workers as desired at exogenously set “market wages.”

14. Lester (1952) first coined the term “range of indeterminacy” to describe the situation where a single wage might be consistent with many different employment levels. Schmitt (2013) has reviewed the research on the many margins of adjustment available to accommodate increases in mandated minimum wages.

15. For example, Ashenfelter, Card, Farber, and Ransom (2022) edited a symposium in the Journal of Human Resources on “Monopsony in the Labor Market,” and Mishel (2022) edited a symposium in the Journal of Law and Political Economy on “Not So Free to Contract: The Law, Philosophy, and Economics of Unequal Workplace Power.”

16. There is also the important issue that labor productivity is much more straightforward to measure and interpret than is total factor productivity. Because it measures increases in output growth after accounting for all observable inputs, in many measures, total factor productivity is simply a quantity representing what we cannot truly explain—it has been labelled a “measure of our ignorance.”

17. We say “nearly all” instead of “all” because potential output is not actually the maximum feasible output an economy could produce (say under conditions of wartime and price controls). Instead, it’s how much an economy can produce without spurring accelerating inflation. As aggregate demand gets extremely high relative to potential output, unemployment can be driven so low that workers’ wage demands exceed productivity growth, spurring inflation. This level of unemployment that maps onto maximum output that can be produced without accelerating inflation is sometimes called the “natural rate” of unemployment.

18. See Baker (2010) for the best macroeconomic narrative of how the housing bubble’s burst reduced aggregate demand and caused the Great Recession.

19. This dichotomy between determinants of potential output growth and aggregate demand growth is not quite as strict as this section indicates. Long periods of time when aggregate demand is depressed, for example, can actually reduce productivity growth and labor force growth as businesses invest less in labor-saving technologies and potential workers stay on the sidelines if wage growth is sluggish (see Bivens (2017b) for evidence on some of these links). Yet in well-functioning economies with responsible policymakers, policy decisions can effectively make the determinants of potential output and aggregate demand mostly separate.

20. See Bivens (2022) for an overview of the U.S. economic situation before the American Rescue Plan passed and the law’s subsequent effect on labor markets.

21. A very useful discussion and possible scenarios for AI’s effect on productivity growth over the next decade is provided by Briggs and Kodnani (2023). They estimate a 1.5 percentage point potential annual productivity growth rate boost due to the adoption of AI in the U.S., which would be likely significantly dampened by AI’s substitution away from other technologies and the possibility of a slower adoption period (say 20 years, rather than 10 years).

22. The numbers referenced here for direct, final demand, and upstream effects are very rough estimates taken from Figure 1B in Autor and Salomons (2018).

23. There is some suggestive evidence that some measures of churn—like job-to-job moves—have been increasing since roughly 2015, and that churn jumped enormously in response to the COVID-19 shock. However, as Figure E shows, even a pronounced uptick in churn relative to the recent past will likely not approach past historical peaks.

24. See, for example, the discussion of AI and employment discrimination in Kim and Bodie (2021).

25. See Mullainathan (2019) for this argument that it may be easier to correct discrimination occurring by algorithm relative to discrimination occurring by personal decision-making.

26. See Hadi et al. (2023). Note that they obviously could not assess the true IRS algorithm as this was kept confidential. Instead, they constructed their own algorithms and assessed them for how closely their predicted outcomes matched actual audit patterns.

References

Acemoglu, Daron, David Autor, and Simon Johnson. 2023. Can We Have Pro-Worker AI? Choosing a Path of Machines in Service of Minds. MIT Shaping the Future of Work Initiative, September 2023.

Anway, Nicholas. 2022. “Amazon’s Approach to Robotics Is Seriously Injuring Warehouse Workers.” OnLabor (blog post), May 5, 2022.

Ashenfelter, Orley, David Card, Henry Farber, and Michael Ransom, eds. 2022. “Special Issue: Monopsony in the Labor Market.” Journal of Human Resources 57, supplement 2022: S1–S10.

Atkinson, Robert D., and Wu, J. John. 2017. False Alarmism: Technological Disruption and the U.S. Labor Market, 1850–2015. Information Technology & Innovation Foundation ITIF, May 2017.

Autor, David, Arindrajit Dube, and Annie McGrew. 2023. “The Unexpected Compression: Competition at Work in the Low Wage Labor Market“. National Bureau of Economic Research Working Paper # 31010, November 2023.

Autor, David, Claudia Goldin, and Lawrence Katz. 2020. “Extending the Race Between Education and Technology.” American Economic Review, AEA Papers and Proceedings 110 (May): 347–51.

Autor, David, and Anna Salomons. 2018. “Is Automation Labor-Displacing? Productivity Growth, Employment, and the Labor Share.” Brooking Papers on Economic Activity: BPEA Conference Drafts, March 2018.

Baker, Dean. 2010. False Profits: Recovering from the Bubble Economy. Sausalito, CA: PoliPoint Press.

Bassier, Ihsaan, Arindrajit Dube, and Suresh Naidu. 2022. “Monopsony in Movers: The Elasticity of Labor Supply to the Firm Wage Policies.” Journal of Human Resources 57 (S): s50–s86.

Bivens, Josh. 2017a. Adding Insult to Injury: How Bad Policy Decisions Have Amplified Globalization’s Costs for American Workers. Economic Policy Institute, July 2017.

Bivens, Josh. 2017b. A “High-Pressure Economy” Can Help Boost Productivity and Provide Even More “Room to Run” for the Recovery. Economic Policy Institute, March 2017.

Bivens, Josh. 2022. “Will Friday’s Wage Growth Numbers Stop the Fed from Snatching Defeat out of the Jaws of Victory?” Working Economics Blog (Economic Policy Institute), July 2022.

Bivens, Josh, and Lawrence Mishel. 2015. Understanding the Historic Divergence Between Productivity and a Typical Worker’s Pay: Why It Matters and Why It’s Real. Economic Policy Institute, September 2015.

Bivens, Josh, and Lawrence Mishel. 2017. The Zombie Robot Argument Lurches On: There Is No Evidence that Automation Leads to Inequality or Joblessness. Economic Policy Institute, May 2017.

Bivens, Josh, and Lawrence Mishel. 2021. Identifying the Policy Levers Generating Wage Suppression and Wage Inequality. Economic Policy Institute, May 2021.

Bivens, Josh, and Heidi Shierholz. 2014. Lagging Demand, Not Unemployability, Is Why Long-Term Unemployment Remains So High. Economic Policy Institute, August 2014.

Bowles, Samuel. 1985. “The Production Process in a Competitive Economy: Walrasian, Neo-Hobbesian, and Marxian Models.” American Economic Review 75, no.1 (March): 16–36.

Briggs, Joseph, and Devesh Kodnani. 2023. “Upgrading Our Long-Run Growth Forecasts to Reflect the Impact of Generative AI.” Global Economics Analyst (Goldman Sachs), October 29, 2023.

Bureau of Economic Analysis (BEA). 2023. “Real Gross Domestic Product, Chained Dollars” [Table 1.1.6. ], National Income and Product Accounts (NIPA). Accessed December 2023.

Bureau of Labor Statistics, Current Population Survey (BLS-CPS). 2023. “Labor Force Statistics from the Current Population Survey” (web page). Accessed December 2023.

Bureau of Labor Statistics, Office of Productivity and Technology (BLS-OPT). 2023. “Productivity,” (web page). Accessed December 2023.

Card, David, and John DiNardo. 2002. “Skill Biased Technological Change and Rising Wage Inequality: Some Problems and Puzzles.” Journal of Labor Economics 20, no. 4: 733–783.

Casselman, Ben. 2021. “Pandemic Wave of Automation May Be Bad News for Workers.” New York Times, July 3, 2021.

Congressional Budget Office (CBO). 2023. Potential GDP and Underlying Inputs, August 2019.

Engardio, Peter, Aaron Bernstein, and Manjeet Kripalani. 2003. “The New Global Job Shift: Is Your Job Next?,” Businessweek Economics (Bloomberg), February 3, 2003.

Fernald, John. 2023. Utilization-Adjusted Quarterly-TFP Series for the U.S. Business Sector, Federal Reserve Bank of San Francisco. Accessed December 2023. 

Ghayad, Rand. 2013. “The Jobless Trap.” Working Paper. 

Gilbert, James. 2013. “Robot Job Growth Is an Accelerating Problem, Researchers Tell 60 Minutes.” Huffington Post, January 14, 2013.

Gould, Elise. 2019. “Decades of Rising Economic Inequality in the U.S.” Testimony before the U.S. House of Representatives Ways and Means Committee, March 27, 2019.

Hadi, Elzayn, Evelyn Smith, Thomas Hertz, Arun Ramesh, Robin Fisher, Daniel Ho, and Jacob Goldin. 2023. “Measuring and Mitigating Racial Disparities in Tax Audits.” Stanford Institute of Economic Policy Research Working Paper, January 2023.

Kim, Pauline, and Matthew T. Bodie. 2021. “Artificial Intelligence and the Challenges of Workplace Discrimination and Privacy.” Journal of Labor and Employment Law 35, no. 2: 289–315.

Krueger, Alan, and Robert Solow, eds. 2002. The Roaring Nineties: Can Full Employment Be Sustained? New York, N.Y.: Russell Sage Foundation.

Lester, Richard. 1952. “A Range Theory of Wage Differentials.” ILR Review 5, no. 4 (March): 483–500.

Manning, Alan. 2003. Monopsony in Motion: Imperfect Competition in Labor Markets. Princeton, NJ: Princeton University Press.

Marcus, Steven. 1983. “Technology: As Computers Eliminate Jobs.” New York Times, June 2, 1983.

Mishel, Lawrence, ed. 2022. “Special Issue: Not So Free to Contract: The Law, Philosophy, and Economics of Unequal Workplace Power.” Journal of Law and Political Economy 3, no. 1.

Modestino, Alicia Sasser, Daniel Shoag, and Joshua Balance. 2020. “Upskilling: Do Employers Demand Greater Skill When Workers Are Plentiful?” Review of Economics and Statistics 102, no. 4: 793–805.

Mullainathan, Sendhil. 2019. “Biased Algorithms Are Easier to Fix Than Biased People: Racial Discrimination by Algorithms or by People Is Harmful – But That’s Where the Similarities End.” New York Times, December 6, 2019.

Organization for Economic Cooperation and Development (OECD). 2024. “Employment by Education Level” (indicator), https://doi.org/10.1787/26f676c7-en (accessed on February 4, 2024).

Schmitt, John. 2013. Why Does the Minimum Wage Have No Discernible Effect on Employment? Center for Economic Policy Research, February 2013.

Schmitt, John, Heidi Shierholz, and Lawrence Mishel. 2013. “Don’t Blame the Robots: Assessing the Job Polarization Explanation of Growing Wage Inequality.” Economic Policy Institute-Center for Economic Policy Research Working Paper, November 2013.

Sedik, Tahsin Saadi. 2021. “Pandemics and Automation: Will the Lost Jobs Come Back?” International Monetary Fund Working Paper No. 2021/011, January 2021.

Skott, Peter and Fred Guy. 2013. “Power, Luck, and Ideology: Technological and Institutional Parameters of the Agency Problems for CEOs.” Review of Radical Political Economics 45, no. 3.

Stansbury, Anna M., and Lawrence H. Summers. 2020. “Declining Worker Power and American Economic Performance.” Brooking Papers on Economic Activity: BPEA Conference Drafts, 2020.

Key takeaways:

• Productivity and pay once climbed together. But in recent decades, productivity and pay have diverged: Net productivity grew 59.7% from 1979-2019 while a typical worker’s compensation grew by 15.8%, according to EPI data released ahead of Labor Day.
• If median hourly compensation had grown at the same rate as productivity over the 1979-2019 period, the median worker would be making $9.00 more per hour.
• This divergence has been primarily driven by intentional policy choices creating rising inequality: both the top 10% and especially the top 1% and top 0.1% gained a much larger share of all compensation and labor’s share of income eroded.
• Public policies which restore worker power and balance in the labor market can provide robust, widely shared wage growth.

The growth of inequalities is the central driver of the widening gap between the hourly compensation of a typical (median) worker and productivity—the income generated per hour of work—in recent decades. Specifically, this growing divergence has been driven by the growth of two distinct dimensions of inequality: the surge of compensation received by the top 10%—particularly the top 1.0% and top 0.1%—and the erosion of labor’s share of income and the corresponding growth of capital’s share. This post documents these trends by presenting an updated account of the U.S. productivity-pay divergence originally analyzed in both Mishel and Gee 2012 and Bivens and Mishel 2015. 

The key metric, as explained below, is the lag between the growth of net productivity (taking into account depreciation and evaluated using consumer prices) and hourly compensation (wages and benefits) of a typical or median worker. Between 1979 and 2019, net productivity grew 59.7% while a typical (median) worker’s compensation grew by 15.8%, a 43.9 percentage point divergence driven by inequality. The effects have been felt broadly: During this period, 90% of U.S. workers experienced wage growth (26%) far slower than the economywide average, while workers in the top 1% (mostly highly credentialed professionals and corporate managers) saw 160% wage growth (Mishel and Kandra 2020) and owners of capital reaped large rewards made possible only by this anemic wage growth for the bottom 90%.

This divergence between net productivity and a typical worker’s compensation means that neither slow productivity growth nor inevitable economic forces can explain the poor wage growth of America’s workers. In fact, wage problems are a “failure by design” (Bivens 2010), engineered by those with the most wealth and power. The dynamics are primarily located in the labor market and the strengthening of employers’ power relative to their rank-and-file workforce (which increasingly includes those workers with a four-year college degree). As Mishel and Bivens (2021) recently documented, wage suppression was generated by policy choices that resulted in excessive unemployment, eroded unionization, corporate globalization, lower labor standards (e.g., lower minimum wage), new imposed contract terms (e.g., noncompetes), and corporate structures changes that pushed down wages and profits in supply chains to the benefit of large firms.

The gap between productivity and median hourly compensation growth

The last four decades have seen a systematic divergence between the growth of economywide productivity and the growth of hourly compensation (wages and benefits) for typical workers. We proxy the wages of “typical” workers as either wages for nonsupervisory workers (roughly 80% of the private-sector workforce) or wages for the worker earning the median wage. As a starting point, we compare this to the growth of net productivity for the entire economy (rather than only nonfarm businesses or any subset of the economy).

Figure A below shows the growth of net productivity (gross productivity minus depreciation) adjusted for inflation according to consumer prices (rather than inflation in all types of output including investment goods and government), which is labeled ‘effective productivity’ as it is available for household consumption. Figure A compares this growth with the typical worker’s hourly compensation since 1948, using the hourly compensation of production/nonsupervisory workers because that is the only series available for the entire period since 1948 (it corresponds closely to median wage growth). Further detail on the underlying data and measurement issues can be found on the newly expanded EPI page providing information on the productivity-pay gap.

Figure A

The starting point in the analysis below is net productivity measured at output prices since that is how statistical agencies measure productivity and it allows us to illustrate the difference between net productivity measured at consumer rather than output prices. Net productivity measured at output prices and a typical worker’s compensation grew roughly in tandem over the 1948–1973 period, but strongly diverged thereafter and split entirely after 1979. Specifically, this measure of productivity grew 112.5% from 1948 to 1979 with a corresponding 90.2% growth in a worker’s compensation. In contrast, productivity grew 85.1% (1.55% annually) further between 1979 and 2019, but a typical worker’s compensation grew by only 13.2% (0.31% annually). Again, this calculation is for net productivity measured at output prices rather than consumer prices.

This divergence was first pointed out in the early 1990s (Mishel and Bernstein 1994) to demonstrate that stagnant wages for the typical worker over the previous decade or so could not be explained solely or even mainly by the slowdown of productivity growth. This section updates our 2015 analyses of the wedges between typical workers’ pay and productivity and the decomposition of the main factors generating it, drawing on previous work (Mishel and Gee (2012) and the decomposition framework developed by the Centre for the Study of Living Standards: Sharpe, Arsenault, and Harrison 2008a; Sharpe, Arsenault, and Harrison 2008b; and Harrison 2009. This decomposition enables a breakdown of the productivity-pay divergence into the three wedges between compensation and productivity: (1) growing inequality of compensation; (2) the erosion of labor’s share of income; and (3) the divergence of the growth of output prices (used to measure productivity) and consumer prices (used to measure worker compensation trends).

These three wedges are illustrated in Figure B. The first is the area between net productivity deflated by the implicit price deflator (output prices in net domestic product) and net productivity deflated by consumer prices (CPI-U-RS), which is labeled “net effective productivity.” The second wedge is the gap between net effective productivity and average compensation (also deflated by consumer prices), reflecting changes in labor’s share of income. The third wedge is the area between average compensation growth and median hourly compensation growth, which reflects growing inequality of compensation.

Figure B

The top panel (A) in Table 1 below provides the basic trends required to compute the divergence and the wedges in each of the subperiods starting in 1973 and for the entire 1979-2019 period. The 1979-2019 period is chosen as the period for analysis (though there are data through 2020) since this is the period of rising inequality and 1979 and 2019 are both cyclical peaks (years of low unemployment) and therefore suitable for comparison. We compute the wedges between net productivity (productivity net of capital depreciation, which is a better metric of the income available from productivity growth than is gross productivity) in row 2 and median hourly compensation (wages and benefits) presented in row 6 but provide data on median wages and gross productivity for completeness. We shift to the compensation of the median worker rather than for nonsupervisory workers because data for the median worker are available starting in 1973 (the results are comparable, regardless of measure).

Table 1

Over the 1979-2019 period, net productivity (row 7) grew 1.36% annually while median hourly compensation (row 2) grew just 0.38% annually, a sizeable divergence (row 8) of 0.99% each year. Panel B provides the annual impact of each of the three wedges (rounded to two decimals, with all calculations made using entire values) and the bottom panel shows the share of the net productivity-median hourly compensation growth gap that can be accounted for by each wedge.

The most important finding is that the gap between the growth of net productivity and median hourly compensation was primarily driven by factors associated with growing inequality—the decline of labor’s share (especially since 2000) and the growing inequality of compensation, as shown in rows 9 and 10. These inequality factors together can explain 81.0% of the growth of the productivity–median hourly compensation gap (row 14) over the entire 1979-2019 period, with the remaining portion due to difference in the growth of producer (used to measure productivity) and consumer (used to measure compensation) prices (row 15). In the most recent period, from 2000 to 2019, the inequality factors together can explain 92.0% of the productivity-pay divergence (44.5% due to rising compensation inequality and 47.5% from the erosion of labor’s share) and the difference in deflators was much less important (only 8.0%).

There is some controversy about whether the consumer–output price divergences should be included in analyses of the productivity-pay divergence. In this analysis we consciously sidestep this issue below by examining both net productivity and median hourly compensation adjusted for inflation by consumer prices (CPI-U-RS index) so their gap only captures the two inequality wedges. We label net productivity deflated by consumer prices as “net effective productivity” to indicate that its growth is readily available for consumption. That said, we are agnostic whether the output–consumer price divergences are simply a technical matter or reflect non-technical matters of political economy. There has not been serious research into what has caused these price divergences in the various subperiods presented in Table 1. In that regard, it is curious to us that rapid improvements in information equipment investment productivity, resulting in rapidly falling prices, have not seemed to flow through to consumers.

The net effective productivity–median hourly compensation divergence, exclusive of the price deflator differences, reflects the rising inequalities. Over 1979-2019, as noted earlier, net productivity grew 59.7% while median hourly compensation grew 15.8%, a 43.9 percentage point divergence. Net productivity, in other words, grew more than 3.5 times as fast as median hourly compensation. The compensation of the median worker would have risen 0.80 percentage points faster each year (summing rows 9 and 10) had median compensation not lagged net effective productivity. As shown in Figure C below, this resulted in a $9.00 loss in compensation for the median worker.

Figure C illustrates the increased net effective productivity–median hourly divergence in inflation-adjusted ($2019) dollars over the 1979-2019 period where both net effective productivity and median hourly compensation are inflation-adjusted with consumer prices (eliminating the consumer versus output price wedge). Median hourly compensation rose from $20.46 in 1979 to $23.72 in 2019, up 15.8%. However, if median hourly compensation had grown at the 59.7% pace of net effective productivity, then median hourly compensation would have reached $32.71, $9.00 higher than what was actually attained. In other words, rising inequality of compensation and the erosion of labor’s income share over the 1979-2019 period cost the median worker $9.00 an hour.

Figure C

The decline in labor’s share of income

One of the trends that alerted analysts to the erosion of worker bargaining power and the corresponding strengthening of employer bargaining power in recent years is the erosion of labor’s share of income in the 2000s (as shown in Table 1). These metrics reflect the shift in income shares for the whole economy, including the government and nonprofit sectors which have no profits. The distributional conflict between workers (labor) and employers (capital) is best examined in the corporate sector, where all income is divided between compensation going to workers and income accruing to owners of capital. Focusing on the corporate sector hence avoids issues of having to decide whether some other form of income—”proprietor’s income,” or income of noncorporate businesses—is labor or capital (see Bivens 2019 for measurement details) or the influence of trends in sectors which have no profits.

EPI’s Nominal Wage Tracker provides data on the “workers’ share of corporate income.” The data clearly show a decline in labor’s share in the 2000s. For example, labor’s share fell from 82.4% in 2000 to 77.9% in 2007, the last year before the Great Recession. By 2016, when unemployment had reached levels comparable to what had prevailed in 2006 and 2007, it remained roughly 2.5 percentage points below its 2007 level, equivalent to an 8.4% across-the-board cut in compensation for every employee. Equivalently, it would require an across-the-board compensation boost of 9.1% to restore labor’s share to its 2000 level.

This computation may exaggerate the impact of labor’s falling share, since 2000 was a near historic high because of the lowest unemployment rate in decades (4.0%). However, the unemployment rate in 2018 and 2019 also averaged below 4.0%, and labor’s share ended 2019 at 77.3%, well below 2000’s level. This shift toward greater capital income and returns is even more impressive given that real interest rates have fallen sharply in recent years, a development that should (all else equal) be accompanied by a lower return to capital (Farhi and Gourio 2018).

Conclusion

In the 40 years since 1979, the net productivity growth vastly exceeded the growth of compensation for the median or typical worker. This was not the case in the 1948-1973 period. This productivity-pay divergence has been primarily driven by rising inequality, both the top 10% and especially the top 1% gained a much larger share of all compensation and labor’s share of income eroded since 2000. In other words, the factors driving inequalities in the labor market are responsible for workers’ inability to make gains commensurate with productivity growth.

Whether workers make wage gains commensurate with future productivity growth will depend on whether we prevent this ongoing, and eminently preventable, growth in wage inequality. These inequalities, in turn, are the consequence of conscious policy decisions made on behalf of the rich and corporations, as documented in Mishel and Bivens (2021), such as failing to achieve full employment, weakening unions, reducing labor standards, pursuing corporate globalization, employers imposing contract terms (such as noncompete agreements), and large firms using domestic outsourcing to redistribute wages and profits up from the supply chain. Public policies which restore worker power and balance in the labor market can provide robust, widely shared wage growth. What is broken is politics and policy, not the economy itself.

This work has benefitted from a long-term collaboration with EPI director of research Josh Bivens, data computations by EPI research assistant Jori Kandra, and tables and figures from EPI research assistant Melat Kassa.

Introduction: Prosecutors are increasingly pursuing employer crimes against workers

Increasingly, district attorneys (DAs), state attorneys general (AGs), and other criminal prosecutors1 are bringing charges against employers for wage theft,2 misclassification and payroll fraud,3 workplace safety hazards, sexual assault, and human trafficking, among other crimes against workers. This development represents a shift, because historically, crimes against workers have not generally been prosecuted. More often, the criminal justice system has intervened to protect employers; for example, a worker stealing from an employer would likely face charges, while an employer committing wage theft likely would not. Yet state and local prosecutors have unique tools and an important role to play in protecting workers. Many are taking on this function as an enforcement priority, and more should get involved in this area.

The involvement of prosecutors is timely and has the potential for significant impact. As explained in this report, violations of workplace laws are widespread; state and federal labor agencies face serious limitations from a lack of resources, limited authority, and more; and private lawyers are often blocked from bringing cases because workers have been forced to sign arbitration provisions waiving their right to sue in court.

To familiarize prosecutors and worker advocates with this important work, this report provides:

• background on the increased involvement of criminal prosecutors in workers’ rights enforcement, the context in which such activity occurs, and a discussion of the rationale for such prosecutions
• descriptions and examples of the types of cases that have been brought
• discussions of several considerations related to such cases, including applicable statutes, sources of case referrals, criminal justice concerns, and funding sources
• appendices that include sample pleadings from recent cases, compilations of case reports, more detailed information about two state funding mechanisms, and tips for prosecutors and worker advocates on getting started in this work

Background: The growing involvement of prosecutors in addressing employer misconduct emerges in the context of widespread violations of workers’ rights and fits squarely within a prosecutor’s function

Increased involvement of prosecutors in workers’ rights violations has taken form in several ways. Offices have brought various types of cases. Some offices have created dedicated units or subunits to do this work, while others have handled individual cases as they have arisen. Increased prosecutor activity has emerged within a landscape in which violations of workers’ rights are widespread and avenues for redress are inadequate. In this context, there are numerous reasons for prosecutors to actively pursue employer crimes against workers.

Criminal prosecutors across the country are addressing a wide range of employer crimes against workers

A set of federal and state laws extend to most employees in the United States a bundle of protections covering wages paid and hours worked (wage and hour laws), safety hazards in the workplace (safety and health laws), economic security in the event of injury or unemployment (workers’ compensation and unemployment insurance laws), discrimination and harassment (equal opportunity laws), and other workplace conditions. In recent years, a powerful new enforcer has entered the picture in numerous jurisdictions: District attorneys and other prosecutors have brought cases involving employer-committed crimes against workers4—crimes including wage theft; labor trafficking; creating conditions causing predictable, preventable workplace fatalities and serious injuries; payroll fraud, including failure to pay unemployment insurance (UI) taxes and/or to procure workers’ compensation insurance, and/or misclassification of workers; prevailing wage violations;5 retaliation and witness intimidation; and workplace sexual assault (CPR n.d.; HPM Digital Team 2018; Graves 2020; Kashinsky 2019; Wash. AG 2018; Reyes 2021; Vosseller 2019; Mass. AG 2019; Byars 2017).

These cases have been brought in a range of jurisdictions, including in California, Colorado, Maine, Massachusetts, Michigan, Minnesota, Montana, New Jersey, New York, Ohio, Pennsylvania, Rhode Island, Texas, Virginia, and Washington (Intarasuwan 2018; Pace 2019a, 2019b; Byars 2017; Kashinsky 2019; Byrne 2019; Mass AG 2019; AP 2019; Graves 2020; Christian 2019; N.J. AG 2019; Byfield 2019; Warsmith 2019; Pa. AG 2019; R.I. AG 2019; R.I. AG 2020; HPM Digital Team 2018; Haynes 2019; Wash. AG 2018).

Several district attorneys have created dedicated units or subunits specifically for this function, including San Diego DA Summer Stephan in 2021; Queens (N.Y.) DA Melinda Katz and San Francisco DA Chesa Boudin in 2020; Philadelphia DA Larry Krasner in 2019; and more. (Christian 2020; S.D. DA 2021; S.F. DA 2020; Reyes 2019).

In Nassau County (N.Y.), the Labor Unit is located within the Rackets and Enterprise Crime Bureau (Nassau DA n.d.) and in Brooklyn (N.Y.) the Labor Fraud Unit is located within the Frauds Bureau. (Brooklyn DA n.d.). In the New York State AG’s office, the Labor Bureau has a dedicated criminal section (N.Y. AG n.d. ). Staff in these units include lawyers with labor law and/or criminal prosecution experience; some also have access to investigators and forensic auditors.

By creating a dedicated unit, DAs and AGs enable assigned lawyers to develop expertise in the subject matter and handling of these cases, which require a different approach than many other criminal cases. These cases often entail building relationships with worker organizations, conducting extensive interviews with workers who may be reluctant witnesses for a variety of reasons (including fear of retaliation or potential immigration consequences for themselves or family members), and thoroughly reviewing and auditing payroll and other employer records. Creation of a dedicated unit also allows for lawyers to build relationships with other government agencies (for example, state and local labor departments) and stakeholders (such as unions and other worker organizations) that are potential sources of cases. Finally, lawyers in a dedicated unit are able to develop legal expertise in the overlap between labor and criminal law, as well as knowledge regarding common violations and problematic industries. A dedicated unit also institutionalizes the work within an office, thus promoting the likely longevity of a DA’s office involvement in such prosecutions (Gerstein 2020). Professor Cesar Rosado Marzán examined the topic and specifically recommended specialized prosecutors for this work, given “the vulnerable nature of the workers who seek their aid” and potential immigration and other consequences that could result from nonspecialized prosecutors handling such cases (Rosado Marzán 2020).

Even without a dedicated unit, prosecutions of employer crimes can be added to existing divisions or bureaus, such as those handling economic or financial crimes. In Boulder, Colorado, for example, wage theft cases are handled within the Community Protection Division, which handles the office’s economic crimes cases. Indeed, offices without a dedicated unit have played a leadership role on this topic within their states. Within the last few years, Boulder County DA Michael Dougherty, along with the Colorado District Attorneys’ Council, co-hosted a training on prosecution of wage theft and human trafficking;6 he also played an instrumental role in promoting stronger anti-wage theft legislation in his state (Boyd 2019).

Similarly, Orange County (New YOrk) DA David Hoovler announced a new focus on labor crimes by his office, and held a training for fellow prosecutors on the topic (Yakin 2019; Mid Hudson News 2020). The Alameda (California) District Attorney’s Office started a labor trafficking task force in 2015 (H.E.A.T. Watch n.d.). The Westchester (New York) district attorney’s office has a multilingual hotline for the public to report a number of crimes typically affecting immigrants, including wage theft (Westchester DA 2021). The Manhattan District Attorney’s Office in 2017 used criminal forfeiture funds obtained through settlements with international banks to make grants to several organizations serving underserved communities, including over $1.5 million to the New York Committee for Occupational Safety and Health to provide outreach and training regarding wage theft, health and safety, and more (NY Cty. DA 2017).

In several jurisdictions, state attorneys general have used their criminal prosecution authority to pursue employers for wage theft and other crimes against workers (Gerstein 2020). In Rhode Island, for example, where the attorney general is the sole criminal prosecutor for the state, AG Peter Neronha proposed stronger anti-wage-theft and misclassification legislation during the 2021 legislative session; his office brought three wage theft criminal prosecutions in 2020, apparently the first ever brought within the state (R.I. AG 2021).

The increased involvement occurs in the context of high rates of violations of workplace laws and inadequate enforcement resources

Although violations of workplace law are widespread, resources to address such violations are grossly insufficient. Familiarity with this landscape—extensive, largely civil workplace violations with inadequate deterrence—helps one understand the scope of the enforcement chasm. The statistics below demonstrate the overall context in which employer crimes occur. Simply put, routine and widespread violations, inadequate enforcement resources, low union density, and other factors allow too many exploitative employers to operate with impunity. The statistics below do not suggest that all cases should be handled criminally, of course; criminal prosecutions should be reserved for the most serious violations and, of course, brought in situations in which intent and acts can be proven beyond a reasonable doubt.

• Wage theft. A 2017 study on minimum wage violations in the 10 most populous state found that each year, 2.4 million workers, or 17% of the low-wage workforce in these states, reported being paid less than the applicable minimum wage, losing an average of $3,300 per year (nearly a quarter of their earned wages) (Cooper and Kroeger 2017). State-specific studies of wage theft in Colorado, Iowa, New Mexico, New York, and Ohio have found similarly high rates of violations while examining a broader range of workplace infractions (Gordon et al. 2012; Schrank and Garrick 2013; Shields 2019; Sen 2016; Stiffler 2019). According to a 2020 Washington Center for Equitable Growth study, Black, Latinx, noncitizen, and women workers experience higher rates of wage theft (Fine et al. 2020).

• Workplace safety and health. The Occupational Safety and Health Act (OSH Act) was enacted in 1970 to ensure that workplaces are free of hazards that kill or injure workers. Even before the COVID-19 pandemic, workplace fatalities, many of them preventable, were common. In 2019, 5,333 workers were killed on the job, and hundreds of thousands experienced nonfatal injuries and illness (BLS-IFF 2019a and 2019b). A 2017 study by the National Employment Law Project reviewing U.S. Occupational Safety and Health Administration (OSHA) severe injury data from 29 states reported that 27 workers per day suffer amputation or hospitalization (Berkowitz 2017). As with the other workplace harms discussed in this report, there are racial disparities: the overall fatality rate of Black and Latino workers is higher than that of white workers (AFL-CIO 2020). Occupational safety and health risks and violations are even more stark in light of widespread workplace outbreaks during COVID-19.

• Misclassification and payroll fraud. A 2019 study of Washington state found that the proportion of employers that misclassify their workers as independent contractors (not including those who paid “off the books”) averaged 16% from 2013 to 2017 (Xu and Erlich 2019). An earlier analysis of state-level research found that between 10% and 20% of employers have misclassified at least one worker as an independent contractor, noting that employers who misclassify their workers avoid paying payroll taxes and workers’ compensation insurance, and often fail to comply with minimum wage and overtime pay requirements in the Fair Labor Standards Act (FLSA) (Carré
  2015). Misclassification is costly for workers, who lose significant money each year because of it (Shierholz 2020). Misclassification also costs the public. The Washington state study conservatively estimated that from 2013 to 2017, the state annually lost over $30 million in unemployment insurance taxes and more than $53 million in unpaid workers’ compensation premiums; losses are even greater when federal taxes are considered (Xu and Erlich 2019).Misclassification is a particularly acute problem in certain industries. A recent study found that in an average month of 2017, between 12.4% and 20.5% of the construction industry workforce nationwide was either misclassified as independent contractors or working “off the books,” and a report issued by the District of Columbia Attorney General’s Office found that construction contractors save between 17% and 40% by misclassifying workers (Ormiston, Belman, and Erlich 2020; D.C. AG 2019). Another study found extensive wage theft and misclassification in the construction industry in several midwestern states (Goodell and Manzo 2021).

• Workplace harassment, including sexual harassment. Data suggest that workplace harassment is extensive, despite federal, state, and often local equal employment opportunity (EEO) laws prohibiting employment discrimination (including harassment) on the basis of race, color, religion, sex, national origin, disability, age, and more. A 2016 report by the U.S. Equal Employment Opportunity Commission (EEOC)—the agency that enforces these laws at the federal level—noted that one-third of the approximately 90,000 charges the agency received in the prior year included allegations of workplace harassment. The report also suggested that harassment statistics of worker complaints likely seriously understate the extent of the problem, because “the least common response to harassment is to take some formal action—either to report the harassment internally or file a formal legal complaint” (Feldblum and Lipnic 2016). A number of surveys and reports, usually based in specific industries, have found extensive incidence of sexual harassment (ROC United and Forward Together 2014; Covert 2020; NASEM 2020; Chatterjee 2018). However, the U.S. Government Accountability Office in 2020 noted the scarcity of data on this issue and recommended further surveys (U.S. GAO 2020).

• Labor trafficking. Labor trafficking occurs when a person is compelled or coerced to provide labor or services, and often afflicts people who are vulnerable because of life circumstances and economic hardship (U.S. DOJ n.d.). The National Human Trafficking Hotline identified nearly 5,000 labor trafficking cases in 2019 based on its complaint line alone (NHTH 2019). Despite its frequency and severity, labor trafficking often goes undetected and is rarely prosecuted (Smith 2021).

• Employer retaliation against workers for exercising their rights. Although retaliation is illegal, employers commonly retaliate against workers for exercising their workplace rights. Illegal retaliation has been identified in a wide range of circumstances, including when workers report or file lawsuits challenging labor violations, and when workers join together to organize a union or engage in collective action (a right guaranteed under the National Labor Relations Act or NLRA). One study revealed that employers were charged with illegally firing workers or other retaliatory conduct (discipline, threats) in one-fifth to nearly one-third of union elections (McNicholas et al. 2019). A seminal 2009 study of working conditions in three major cities found that of workers who had complained to their employers about violations or tried to form a union in the prior year, 43% experienced retaliation (Bernhardt, Milkman, and Theodore 2009). In workplaces or communities with undocumented workers, a common form of retaliation for asserting workplace rights involves threats or acts related to immigration status. In many cases, laws are insufficient to adequately address or deter forms of employer retaliation (Huizar 2019; Rhinehart and McNicholas 2021).

Several factors play a role in enabling these widespread violations:

• Federal and state enforcement resources are inadequate. The shortcomings in enforcement occur at all levels, starting at the top. Federal resources for the enforcement of worker protections have declined while the U.S. workforce has grown. The U.S. Department of Labor’s Wage and Hour Division (WHD) enforces federal wage and hour laws. In 1978, WHD had one investigator for approximately every 69,000 workers; by 2018, that figure was one investigator per 175,000 workers. In many states, the federal WHD may be the primary or only government agency enforcing wage and hour laws (Costa, Martin, and Rutledge 2020). Similarly, in 1978, OSHA—the federal agency charged with protecting and enforcing workers’ rights to a safe workplace—had one compliance officer for approximately every 60,000 workers; by 2018 that number had almost tripled to nearly 180,000 (Hamaji et al. 2019).7 Meanwhile, at the state level, in 2018, seven states had no investigators at all whose responsibilities included enforcement of minimum wage and overtime laws, while most states had fewer than 10 on staff (Levine 2018). In addition to the low frequency with which penalties are imposed on violators as a result of inadequate resources, the amounts employers must pay are frequently modest, often limited to paying back what they should have paid in the first place. In addition, back wages and penalties are also often difficult to collect, even by civil labor enforcement agencies (Cho, Koonse, and Mischel 2013).

• Low union density leaves workers unprotected. Unions have traditionally helped ensure compliance with workplace laws, by serving as an on-site monitor in unionized workplaces, and by creating pressure on nonunionized workplaces to improve conditions (in order to compete for employees). However, as a result of several factors, including unfavorable federal labor laws and common employer retaliation for organizing, union density (meaning union membership as a share of employment) has diminished greatly over the past several decades (Rhinehart, Windham, and Mishel 2020). The national union membership rate of private-sector workers was only 6.3% in 2020 (BLS-CPS 2021).

• Forced arbitration blocks an increasing number of workers from suing in court, and hides misconduct from public view. Historically, attorneys in the private bar and public interest organizations have played a significant role in addressing wage theft, discrimination, and other workplace violations. As experts have noted, often underfunded public enforcement agencies are unable to address all violations of workers’ rights in the workplace. That is why the ability of workers to take their employers to court—and join together in doing so—to fight wage theft, discrimination, harassment, and other violations has been crucial to enforcing workplace protections (Hamaji et al 2019). However, private attorneys are increasingly unable to address workplace protections. As of 2017, more than half (56.2%) of all private-sector nonunion employees were required by their employer, as a condition of employment, to sign a forced arbitration agreement (Colvin 2018). Under these agreements, workers waive their right to take their employer to court and consent instead to resolving disputes in private arbitration—a secretive process heavily tilted toward the employer. These agreements often include class- and collective-action waivers, under which employees give up their right to sue on a collective basis. In 2017, 41.1% of private-sector nonunion employees covered by mandatory arbitration procedures were also subject to class action waivers (Colvin 2018). The share of private-sector nonunion workers blocked from going to court by forced arbitration clauses with class- and collective-action waivers is projected to exceed 80% by 2024. (Hamaji et al. 2019). Workers win less often in forced arbitration than in court, and when they do win in arbitration, they win less money than in court (Stone and Colvin 2015). Importantly, when workers are required to give up their right to bring a class action, it becomes extremely difficult for them to find a private attorney to bring their case, since bringing a solo arbitration case eliminates the economies of scale that make wage theft and discrimination cases economically feasible for private lawyers. Most often, when there are forced arbitration and class waiver provisions, workers never bring cases at all: one scholar estimated that hundreds of thousands of claims are not even brought each year because workers are subject to forced arbitration and never file cases; she referred to the “black hole” of forced arbitration (Estlund 2018). The secrecy of arbitration proceedings also allows violations to persist unabated, by preventing wrongdoing from coming to light as would occur in a court case.

• The growth of the “fissured workplace” leads to increased violations and creates challenges for effective enforcement. The “fissured workplace” refers to companies that subcontract, use temporary agencies, use a franchise model, or otherwise use business models in which they avoid bearing the legal responsibilities of an employer (Weil 2014). According to Weil, growth of fissured workplaces over the past several decades contributes to workplace law violations. Lower-level contractors are often less capitalized and may exist within the underground economy. Also, mid-level firms, such as temp agencies, must make a profit themselves, leaving smaller margins and pressure to cut corners to make a profit by paying less money to workers at the bottom level. In addition, effective enforcement of minimum wage requirements, overtime pay obligations, and other workplace standards is often more difficult in a fissured workplace, because it can be difficult for enforcers to impose liability on higher-level “up-chain” entities that drive working conditions and have the ability to bring about lasting compliance.8

The rationale for prosecution of wage theft and other employer crimes against workers is sound

Several reasons have been offered to support the position that violations of workers’ rights are an appropriate subject for criminal prosecution.

Employer crimes against workers cause significant harm

Although wage theft and other crimes against workers are—like most white collar and property crimes—typically nonviolent, they nonetheless cause significant harm to workers, to honest employers, and to neighborhoods and communities. The amount of money denied to workers because of wage theft dwarfs the amount stolen through many other forms of theft: One study found that in 2012, the total value of property taken in robberies reported to police across the United States was $341 million, compared with $933 million in wages recovered for known victims of wage theft. (Meixell and Eisenbrey 2014). Note that the wage theft figure includes actually recovered back wages, which is likely a significant undercount, given underreporting of wage theft for various reasons.

• Persistent wage theft impacts workers, their families, and neighborhood businesses (since underpaid workers have less money to spend), and it can affect the economic stability of neighborhoods and communities. As the cases in Section 3 demonstrate, employers’ crimes in wage theft or workplace safety cases often impact numerous victims. Moreover, given the social determinants of health, some public health experts have noted that wage theft may exacerbate adverse health impacts of low wages and low-income status by generating income insecurity. Individuals not paid for hours worked, or paid less than what they earned, may not reliably be able to pay rent or heating, buy groceries, or access transit. This in turn may result in increased crowding or homelessness, hunger, decreased mobility, and decreased ability to pay for child care or medical care—all having an adverse impact on health. Wage theft may also increase the number of hours or jobs worked, which may in turn decrease time spent with family, leisure time for physical activity, and sleep or rest (Minkler et al. 2014).
• Workers who are sexually harassed or assaulted in the workplace experience serious effects, including harm to mental and physical health, reduced opportunities for on-the-job learning and advancement, forced job change, unemployment, and abandonment of careers. Sexual harassment and assault in the workplace also harm employers, causing absences, turnover, reduced productivity, and litigation (Shaw, Hegewisch, and Hess 2018).
• Other crimes, like human trafficking and workplace fatalities, have clear and devastating lifelong impacts on victims and their families.
• Wage theft, payroll fraud, and related crimes also harm lawful businesses that comply with the law, since they must operate at a disadvantage relative to competitors that save money through breaking the law. When a business offers lower prices because it underpays employees, or when a company wins a contract because it cheats on unemployment insurance and other taxes, these acts create unfair competition for law-abiding employers (D.C. AG 2019). Widespread unaddressed violations by employers and corporations also undermine respect for the rule of law.
• Employer crimes harm the general public. Employers who evade unemployment insurance taxes deprive the system of resources needed to provide this critical safety net; those who lie in relation to their workers’ compensation insurance burden public health care resources and increase insurance costs for all employers. In addition, given that very low wages in certain sectors require full-time workers to rely on government assistance to survive (Cooper 2016), persistent underpayment of workers’ wages likely exacerbates reliance on public benefit programs.

The crimes are intentional

As many of the case examples provided in Section 3 demonstrate, crimes against workers generally result from conscious decision-making. Creating false payroll records; underreporting workers on unemployment insurance or workers’ compensation documents; shaving employee work hours or stealing tips; retaliating against employees who speak up when their rights are violated; paying no wages for an entire month or longer; assaulting minors or undocumented women workers; and eschewing critical and obvious workplace safety requirements are not inadvertent mistakes. Some employers are repeat violators (N.Y. AG 2014a). In addition, employer crimes often occur in clusters: The same employer who fails to pay workers is also evading UI and other taxes and violating workers’ compensation laws (Piore and Schrank 2018, 37–39). The Colorado legislature, in a recent statute targeting labor trafficking and wage theft, noted, “Persons who commit the crime of human trafficking often commit other crimes such as wage theft, tax evasion, and workers’ compensation fraud, which drains local and state resources, as well as denies the state its right to revenue.”9 Further, certain predatory employers specifically take advantage of young people, immigrants, or other particularly vulnerable worker populations.

Criminal prosecution of employer crimes against workers is likely to deter similar violations by other employers

Inefficacies in and weaknesses of our regulatory systems have been cited as necessitating criminal prosecutions against companies for workplace and other abuses (Steinzor 2015, 15–39). As described in Section 2B, civil labor enforcement agency resources are inadequate, limiting their effectiveness. DA involvement could have a particularly decisive effect in jurisdictions where state labor department enforcement is minimal or nonexistent. Elsewhere, too, criminal prosecutions are likely to have a significant deterrent impact on employer misconduct.

Prosecution of an employer can result in specific deterrence, meaning prevention of violations by the prosecuted employer through, for example, a plea agreement requiring ongoing compliance as a condition of probation or incorporating independent monitoring. While it requires further study (see Section 5 below), criminal prosecution of employers also appears likely to have a meaningful general deterrent impact on exploitative business models that treat civil enforcement as unlikely to occur and civil penalties as a modest “cost of doing business.”

Noncompliance with wage and hour laws (for example) has been described as a “rational” profit-maximizing decision made by unethical employers in response to low enforcement rates and deficient penalties. Scholars who have analyzed employer costs and benefits of noncompliance find that “employers will not comply with the law if the expected penalties are small either because it is easy to escape detection or because assessed penalties are small” (Ashenfelter and Smith 1979). Currently, as described in Section 2B, the likelihood of detection is low. The cost of detection is generally modest: Civil enforcement frequently recovers only back wages owed to employees, essentially converting workers’ wages into an interest-free loan to the employer. In this context, criminal prosecution could change an employer’s equation, by increasing the likelihood of detection through adding a visible and powerful new enforcer to the picture, as well as increasing the cost of detection (in the form of individual financial, reputational, and other costs). Media coverage of criminal prosecutions should also increase the perceived likelihood and cost of detection. Such publicity serves an additional deterrent purpose: A recent study showed that press releases about OSHA enforcement of workplace safety violations deterred other workplace safety violations (Johnson 2020), a conclusion likely applicable to other aspects of workplace compliance as well.

Prosecutions related to workplace conduct already routinely occur, indicating that workplace matters are not currently treated as solely civil in nature

Prosecutors routinely bring charges against employees who embezzle or otherwise steal from their employers. In addition, many prosecutors routinely pursue claimants who have fraudulently received workers’ compensation or unemployment insurance benefits.10 Given that the criminal justice system already addresses workplace matters when employers are harmed, it is appropriate to use the same jurisdiction and power when workers are harmed. Fairness in administration of justice would seem to require examination of employer misconduct, perhaps particularly urgently so during a time when workers deemed essential often experience serious workplace dangers and violations of law. (Rosenthal 2021). Moreover, fraud or wage theft by one employer will often cause a greater magnitude of harm than fraud perpetrated by an individual worker; for example, one employer that evades UI taxes year after year will generally cheat the UI program of significantly more money than one individual fraudulent UI claimant. In addition, criminal prosecution of a single employer who is a serial violator is likely to result in justice for many workers, past and future.

More resources are needed to enforce workplace protections

It is insufficient to rely on civil enforcement agencies or workers themselves to enforce labor standards protections on their own. Public enforcement resources dedicated to civil enforcement agencies are too limited, and there are too many impediments to private enforcement, including forced arbitration and workers’ realistic fears of retaliation. Such challenges are more acute in cases involving particularly egregious employer conduct, where fraudulent behaviors may impede or prevent civil enforcement agencies from readily identifying violations. Dedicating criminal enforcement resources to enforcing workplace protections would fill a significant currently unmet need.

Prosecution of employers for crimes against workers is consistent with a reform-oriented approach to criminal justice

Along with a number of more traditional prosecutors, several prosecutors associated with the “progressive prosecutor” movement have taken on this work, and they have articulated how they believe it fits within a criminal justice reform framework.11 For example, Philadelphia District Attorney Larry Krasner and San Francisco District Attorney Chesa Boudin both appointed labor liaisons within their offices for this purpose, and Minnesota Attorney General Keith Ellison created a Wage Theft Unit. These officials have cited the power employers wield over workers, and the vulnerability of workers, as a basis for committing to these prosecutions (Reyes 2019; S.F. DA 2020; Minn. AG 2019). Attorney General Ellison and Boulder County District Attorney Michael Dougherty both noted the inequity of pursuing other kinds of theft but not wage theft (Minn. AG 2019, Boyd 2019). Moreover, prosecuting extreme employer abuses aligns with the work of advocates and worker organizations on the ground (CTUL 2019; Svoboda 2011; Colorado General Assembly 2019). Relationships between prosecutors and organizations that are accountable to workers can also help the criminal justice system become more accountable to marginalized communities more broadly. As Professor César Rosado Marzán has noted, “The value of criminalization lies in its unambiguous moral condemnation of wage theft, in its capacity to shame employers who abuse their power, and in the real threat of imprisonment. Criminalization will help activists to co-enforce wage and hours laws, likely deter wage theft, and better resolve a dire problem affecting some of the most powerless individuals in U.S. society” (Rosado Marzán 2020). Prosecution of wage theft also involves reorienting enforcement resources “toward aggressive enforcement of the predations of powerful economic entities” (Bhargava and Hertel-Fernandez 2020).

Case types and examples cover a range of worker abuses and span the country

Prosecutors have brought charges to address numerous different types of employer crimes against workers. The following are descriptions and examples of some common types of cases, pulled primarily from publicly available news media articles and agency communications.12 Often, infractions occur in clusters: the same employer commits wage theft and evades UI taxes, commits workers’ compensation insurance fraud, has unsafe working conditions, and retaliates against workers who complain about any of these problems. The discussion below is organized by violation type in the interest of clarity, but many cases involve multiple violations. In addition to the discussion below, Appendix A contains links to a number of court-filed documents such as indictments, plea agreements, and more. Appendix B links to resources including a database and spreadsheet containing information and news reports of additional cases.

Prosecutors have used various statutes and charges to pursue wage theft

Wage theft takes many forms but essentially involves not paying workers what they are owed, whether it is for hours worked, or at levels required under the statutes that govern required minimum wages and overtime pay.13 Employers may avoid paying a worker for all hours worked, or avoid paying overtime, through a variety of methods: altering time cards, maintaining two sets of payroll records (one real and one false), or requiring people to work before clocking in or after clocking out. Some employers pay workers in cash, to avoid creating a record of underpayments (as well as to avoid paying unemployment insurance taxes and workers’ compensation premiums). In certain instances, including in the restaurant, construction, and home health care industries, prosecuted employers have failed to pay any wages at all for a period of time, including up to one or two months. There have been cases of employers taking permissible deductions from workers’ pay for health insurance, union dues, or other benefits—but then simply keeping the money for themselves. While the majority of wage underpayment situations that are reported are handled by the civil legal system, a growing number of prosecutors have brought criminal charges in appropriate situations.

DAs and AGs bring criminal prosecutions under state laws. The relevant state statutes, and therefore the charges brought in these cases, vary. Some state statutes explicitly address workplace conduct and may even use the term “wage theft.” Other prosecutors have brought these cases using theft of services, theft by swindle, larceny, scheme to defraud, or similar statutes. Prosecution of wage theft, prevailing wage and payroll fraud cases typically requires not only witness testimony, but also audits of employer documents—evidence similar to that needed to pursue other types of economic crimes.

Sample wage theft cases

• Nonpayment of overtime. In 2021, the Santa Clara County Office of the District Attorney brought charges against a flooring company owner who was accused of owing workers nearly $1 million of overtime wages, and also accused of workers’ compensation premium fraud (Green 2021).
• Keeping payroll deductions intended for specific purposes. In 2021, the Worker Protection Unit of the Philadelphia Office of the District Attorney charged a plumbing company’s owner with multiple counts of theft for deducting union dues from workers’ pay and keeping the money instead of remitting the dues to the union as required (Phil. DA 2021). In a similar case, Michigan’s attorney general brought criminal charges against an employer who allegedly withheld more than $52,000 from workers’ paychecks for deferred retirement contributions but failed to deposit the funds into their accounts or pay the employer match (Mich. AG 2019).
• Failing to pay wages. In 2020, Colorado’s 5th Judicial District Attorney’s Office charged a contractor with tax evasion and theft from a person; the contractor ultimately pleaded guilty to one count of each. The investigation, which began in late 2019, was conducted by the DA’s office in collaboration with the Colorado Department of Revenue Criminal Tax Enforcement Division after past employees of the defendant complained about not having been paid (Colo. 5th DA 2020, Lotshaw 2021).
• Writing bad checks to undocumented workers. In 2019, Colorado’s 20th Judicial District Attorney’s Office (in Boulder County) brought charges against an employer accused of giving undocumented workers bad checks for various projects and threatening to report them to immigration when they asked for payment (Byars 2019a, 2019b). The employer was sentenced to four years of probation and ordered to pay restitution (Oravetz 2019).
• Failing to pay for work performed. In 2018, the Harris County (Texas) district attorney charged the owner of a high-value home with theft of service for failing to pay a painter for repair work completed after a hurricane (McPherson 2018).
• Wage theft of elder care employees. Numerous DAs in California have brought charges against elder care homes for wage theft, tax fraud, and other criminal violations (Gartrell 2016).
• Not paying wages to immigrant workers. In 2016, the San Diego district attorney, partnering with the California labor commissioner, secured a criminal jury trial conviction for felony grand wage theft by false pretenses. A San Diego restaurant owner was sentenced in 2016 to two years in jail for paying immigrant workers only in tips. The San Diego County Superior Court also ordered the employer to repay $20,000 in stolen wages and tips to six of the restaurant workers (CSLEA 2016).
• Failing to pay overtime and other wages. The New York attorney general’s office has brought wage theft-related cases in several industries. For example, in 2015, the office charged a Papa John’s franchisee who was accused of continuing to underpay workers even after being investigated for wage theft by the U.S. Department of Labor. The charges included creating fictitious worker names to conceal overtime hours worked and filing fraudulent quarterly state tax returns. The employer pleaded guilty to failure to pay wages and falsifying business records (NY AG 2015a, 2015b). In 2012, the New York AG’s office announced a guilty plea by the owner of a home health agency in a case that involved failing to pay workers $300,000 in wages (NY AG 2012b).

Crimes related to prevailing wage laws involve cheating employees, government agencies, and the public

Prevailing wage laws require contractors on government-funded public works projects (typically government construction contracts) to pay their workers at least the locally prevailing wages and fringe benefits paid on similar projects in the area (Mahalia 2008). In addition to the federal prevailing wage law (the Davis-Bacon Act of 1931), many states have enacted state-level prevailing wage laws. Prevailing wages are set based on a worker’s location and occupation, and they are often considerably higher than the applicable minimum wage in a given jurisdiction. To allow government contracting agencies to ensure compliance, prevailing wage laws generally require public contractors to routinely submit “certified payroll records” of hours worked and wages and benefits paid on a given contract. When employers violate these laws, they are not only underpaying workers, but also cheating the government agency and taxpayers, since the agency awarded the contract based on the premise that workers would be paid prevailing wages. As a method of violating prevailing wage laws, employers may, for example, create and submit false certified payroll records that fraudulently demonstrate compliance by showing artificially inflated pay amounts, artificially deflated hours, or even listing as employees individuals who were not on the job at all.

Some prevailing wage laws directly include criminal sanctions for violations. Violations such as submitting false payroll records may also result in other charges, such as filing a false instrument or maintaining false business records.

Sample prevailing wage cases

• Cheating on fringe benefit requirements. In 2021, the Pennsylvania attorney general’s office charged a contractor with extensive and complex violations of prevailing wage laws in what the AG described as “the largest prevailing wage criminal case on record.” The contractor was charged with appropriating retirement benefits owed under the law, and falsely inflating the amount of money paid for health benefits in order to pay workers less. These charges came on the heels of another separate criminal prevailing wage case by the Pennsylvania AG’s office that resulted in a guilty plea. (Pa. AG 2021, Rushton 2021).
• Failing to pay required prevailing wages. In 2020, the Queens (New York) district attorney announced the guilty plea to a prevailing wage labor law violation by a contractor for the New York City School Construction Authority and the New York City Department of Education, who “pocketed more than $1.5 million that should have gone to employees.” (Queens DA 2020).
• Falsifying records and underpaying workers. In 2019, New Jersey’s attorney general announced the guilty plea of a construction contractor for falsifying payroll records to conceal his underpayment, and in some cases nonpayment, of wages to workers, many of whom were immigrants (N.J. AG 2019).
• Underpayment, falsifying records, and demanding kickbacks of money back from workers. In 2013, New York’s attorney general charged a contractor for the Port Authority with prevailing wage violations. The contractor, who ultimately pleaded guilty to grand larceny and prevailing wage violations, created false business records demonstrating compliance with the law and issued checks to workers which would demonstrate compliance; he then made his workers cash the checks at his bank and kick back, or return, most of the cash to him. Under the plea agreement, the contractor was sentenced to five years of probation, was banned for five years from working on public projects in New York, and was ordered to pay $200,000 in restitution to workers (N.Y. AG 2013, 2014b).

Payroll fraud cases include worker misclassification, workers’ compensation fraud, and nonpayment of unemployment insurance taxes

Numerous employers have been criminally prosecuted for payroll fraud, including crimes resulting from misclassifying workers as independent contractors, or from paying workers in unreported cash “off the books.” In such cases, employers often falsely underreport the number of workers on unemployment insurance (UI) tax returns filed with the state or on workers’ compensation insurance applications. These actions can lead to charges of filing a false document or maintaining false business records, or specific workers’ compensation- or insurance-related charges. Additional charges may stem from an employer’s failure to pay UI taxes or to procure required workers’ compensation insurance, acts which themselves have criminal consequences in some states. In addition, as noted above, often employers committing these offenses engage in wage theft as well.

Sample payroll fraud cases

• Underreporting employees on workers’ compensation forms. In 2020, the Hennepin County Attorney (Minnesota) won convictions against the owners of a drywall company for insurance fraud and theft by swindle, based on allegations that the company underreported employees on workers’ compensation documents and wrongly treated workers as independent contractors. As part of a guilty plea, both owners of the Minnesota drywall company received five years of probation, 180 days of home monitoring, and 30 days of community service, along with a $30,000 fine and an order to pay $309,000 in restitution (Hennepin CA 2020a, 2020b).
• Failure to pay wages, UI taxes, and workers’ compensation insurance. In 2020, the Suffolk County (New York) DA’s office announced that eight people and nine businesses had been charged in a labor-related crackdown in Suffolk County. The alleged crimes collectively involved the theft of more than $250,000 in employees’ wages and benefits, nonpayment of more than $58,000 to the New York State Department of Labor for unemployment insurance fund contributions, and failure to pay more than $133,000 to the state insurance fund for workers’ compensation insurance premiums (Suffolk Cty. DA 2020b).
• Wage theft and workers’ compensation insurance crimes, related in part to underreporting payroll. In 2020, the Rhode Island attorney general brought charges against the former owner of a cleaning company (a contractor for the Community College of Rhode Island) for wage theft, failure to maintain workers’ compensation insurance coverage, and workers’ compensation insurance premium fraud. Among other things, the defendant allegedly falsely reported only $10,000 of payroll, instead of almost $400,000, lowering business expenses and gaining a competitive advantage in bidding (R.I. AG 2020).
• Workers’ compensation insurance fraud. In 2019, the Stanislaus County (California) DA’s office announced that a former temp agency owner had been convicted of workers’ compensation insurance premium fraud and ordered to pay close to $1 million in restitution. An insurance audit revealed that the agency owner had underreported payroll and the number of employees to obtain a lower premium (CSLEA 2019).
• Misclassifying workers to avoid paying overtime and UI taxes. In 2018, the New York state attorney general obtained guilty pleas (to grand larceny and falsifying business records) from three construction companies that had misclassified their workers as independent contractors to avoid paying overtime and unemployment insurance taxes (N.Y. AG 2018).

• Hiding the existence of workers to avoid workers’ compensation costs and payroll taxes. In 2017, the San Diego district attorney obtained a guilty plea from a married couple accused hiding the existence of at least 800 housekeeping and janitorial workers to avoid paying millions in workers’ compensation insurance rates and payroll taxes (Littlefield 2017).

Workplace safety and health cases target highly predictable, avoidable, and sometimes fatal workplace hazards

While most violations of workplace safety and health laws are addressed through civil enforcement by OSHA or an OSHA-approved state plan (U.S. DOL-OSHA n.d.a, n.d.b), prosecutors have brought various charges in several cases of highly predictable, easily avoidable workplace fatalities or serious injuries, such as unsecured roofers who fell to their deaths or workers killed when trenches that had not been shored up collapsed. State exercise of traditional police powers, such as criminal prosecutions of such cases, is generally not preempted by the OSH Act (Flanagan, Gerstein, and Smith 2020). Charges have included workplace or involuntary manslaughter, criminally negligent homicide, reckless endangerment, and assault. There is long-standing precedent for prosecution in this area; for example, in the 1980s, a program was established in Los Angeles involving collaboration between the district attorney and the state OSHA plan (McCluskey et al. 2016).

Sample workplace safety and health cases

• Roofing fatalities. In 2019, the Office of the Maine Attorney General charged a contractor with workplace manslaughter when a roofing worker without required protection against falls fell to his death (Flaherty 2019). In a similar case that year, the prosecutor from Summit County, Ohio, obtained an involuntary manslaughter guilty plea from a contractor in a roofing fatality case (Warsmith 2019).

• Trench collapses. Several employers have been prosecuted following workplace fatalities resulting from trench collapses, including in Boston; Brooklyn, New York; Fairfax County, Virginia; Granby, Colorado; Manhattan, New York; and Seattle. In all cases, the employers were charged with manslaughter, except in the Seattle case; there the charge was criminal negligence (NBC10 Boston 2019; Brooklyn DA 2019; Haynes 2019; Pace 2019a, 2019b; Chen 2016; Green 2016)
• Forklift fatality. In 2018, the San Francisco district attorney brought involuntary manslaughter charges against the employer of a worker crushed to death by a forklift; the victim was assigned to use the forklift, despite not being certified to do so, and a ramp at the worksite lacked a required curb to prevent the forklift from falling (Sernoffsky 2018).
• Crane-related injuries. In 2018, the Manhattan district attorney brought assault charges against a contractor on a construction site where two workers were gravely injured by the fall of a mini crane (NY Cty. DA 2018).
• Child labor. In the last decade, child labor charges were brought in two New York cases, both involving teenagers assigned to operate machinery prohibited at their age. A 14-year-old was killed on the job at a farm (Harris 2014; N.Y. AG 2019) and a 17-year-old’s arm was severed at a restaurant (N.Y. AG 2014c). Both employers pleaded guilty.

Labor trafficking cases involve extreme worker exploitation

Labor trafficking occurs when a person uses force, fraud, or coercion to obtain labor or services of another person (U.S. Department of State 2021). State statutes defining labor trafficking vary in their precise language, and some encompass a wider range of conduct than federal trafficking statutes (NCSL 2018). Although labor trafficking has been less commonly prosecuted than sex trafficking (Smith 2021), state or local criminal prosecutors have brought labor trafficking charges in a few notable cases. In some instances, a focus on trafficking has served as a pathway for prosecutors’ offices to get involved in broader worker exploitation issues.14

Sample labor trafficking cases

• Paying grossly subminimum wages and threatening workers. In 2020, the Suffolk County (New York) district attorney arrested a gas station owner for several charges, including labor trafficking, scheme to defraud, grand larceny, and retaliation. The employer was accused of paying grossly subminimum wages and no overtime for workweeks of 70 to 100 hours; he was also alleged to have threatened to file false police reports or call immigration authorities regarding any employees who complained about working conditions or cooperated in state labor department investigations. The case was referred to the DA’s office by the New York State Department of Labor (Suffolk Cty. DA 2020a).
• Coercing, underpaying, and threatening undocumented workers in unsafe conditions. In 2019, a contractor in the Twin Cities, Minnesota, region was sentenced to 270 days in jail and five years of probation for labor trafficking and insurance fraud, after pleading guilty on the eve of trial in a case brought by the Hennepin County Attorney. According to the criminal complaint, the contractor recruited workers for construction work, knowing that they were undocumented, and used that leverage to force them to work long hours at low pay and without adequate safety protection, allegedly also telling workers they would be fired and deported if they went to a doctor for injuries suffered on the job (Hennepin CA 2019, 2020c). The investigation was conducted by a state law enforcement agency after reports were made to the Hennepin County Attorney’s Office by a trade union and a local workers’ rights organization, Centro de Trabajadores Unidos en la Lucha (CTUL) (Feshir 2019).

• Underpaying, keeping passport, and threatening immigrant cleaning worker with deportation. In 2018, the Massachusetts attorney general’s office charged an employer with labor trafficking, among other offenses, in a case in which the defendant was accused of recruiting a worker from abroad and requiring her to perform cleaning work, for which she was paid subminimum wages; he allegedly retained the worker’s passport and threatened that if she tried to return to her home country of origin, she would be arrested by immigration authorities (Mass. AG 2018).
• Subjecting care workers to brutal conditions. In 2018, the California attorney general’s office brought human trafficking and other charges against four individuals who ran an adult residential and child care company. The complaint alleged that workers were forced to work around the clock, seven days a week, and sleep on floors and in garages, and that defendants also confiscated some workers’ passports and threatened to report workers to immigration authorities (Calif. AG 2018).
• Taking visas and passports from farmworkers and threatening harm. In 2018, a jury in Fresno County, California, found an individual guilty of human trafficking and extortion in relation to farmworkers. The defendant was accused of taking victims’ visas and passports, and of threatening to harm them and report them to immigration if they stopped working for him (Lopez 2018a, 2018b).

Sexual assault cases target extreme harms of harassment

The #MeToo movement, which gained momentum in 2017, has led to an increase in awareness and exposure of workplace sexual harassment. In certain instances, this conduct has risen to a level resulting in assault and even rape charges.15

Sample workplace sexual assault cases case summaries

• In a nationally high-profile case, the Manhattan DA office brought criminal charges against former Hollywood producer Harvey Weinstein, who in 2020 was found guilty of criminal sexual assault and rape, and sentenced to 23 years in prison (Ransom 2020a, 2020b).
• The owner of a Boulder, Colorado, ice cream company pleaded guilty to two counts of misdemeanor unlawful sexual conduct based on allegations that he sexually abused female employees, including two undocumented immigrants. In 2018, he was sentenced to six months in the Boulder County Jail followed by a year of work release, among other conditions of probation (Bear 2018).
• The owner of the country’s last Howard Johnson’s restaurant (located in Lake George, New York) was sentenced to six months in jail and six years’ probation  after being charged in 2017 with sexual abuse, unlawful imprisonment, and endangering the welfare of a child based on allegations that he sexually harassed about 15 female employees, including minors (AP 2017, 2018).

Retaliation and witness intimidation cases target actions that hamper investigations

Retaliation against workers who report violations can consist of termination, demotion, pay reduction, assignment to less desirable schedule or job assignment, threats of any kind (including calling immigration), or advising other employers not to hire a person. Retaliation is particularly harmful because of its potential to deter other workers from reporting violations or cooperating with an investigation. In some cases pursued by prosecutors, employers pressured workers to withdraw complaints or provide inaccurate testimony in wage-related proceedings. Other cases involved retaliation or conduct similar to witness tampering.

Sample retaliation and witness intimidation cases

• Retaliating against gas station workers who reported violations. In 2020, the Suffolk County (New York) District Attorney charged a gas station owner with, among other things, retaliation against workers who reported violations (Suffolk Cty. DA 2020a). (This case is also described in the labor trafficking case section above) (Suffolk Cty. DA 2020a).
• Intimidating laundry temp workers serving as witnesses in a wage theft case. In 2019, the Massachusetts attorney general’s office announced guilty pleas of owners of a temp agency for, among other things, witness intimidation and retaliation against workers placed at an industrial laundry facility. Workers placed by the temp agency in the warehouse were paid subminimum wages and no overtime for workweeks of 60 to 70 hours. The temp company owners were accused of threatening to terminate witnesses cooperating with the AG’s investigation, directing employees not to cooperate with investigators, and reducing the hours of workers who spoke with investigators during an on-site inspection. The investigation began when a local branch of the United Food and Commercial Workers International Union contacted the AG’s office (Mass. AG 2019).
• Dissuading witnesses of crimes related to a state construction subcontract. In 2014, the Orange County, California, district attorney’s office obtained a guilty plea from a construction subcontractor for taking workers’ wages on a public work project and dissuading witnesses from prosecuting a crime. Hired by the general contractor refurbishing a state hospital, the subcontractor required workers to turn over a portion of their paychecks to him. When workers contacted the DA’s office about their wages, leading to an investigation, the subcontractor invited workers to his house to receive their final paychecks, but instead attempted to dissuade them from acting as witnesses against him (Dobruck 2014).
• Witness tampering. In 2012, the New York attorney general’s office brought witness tampering charges against a garment factory owner accused of instructing a former employee to falsely testify that her work tenure was shorter than it was. The case was referred to the AG’s office by staff from the New York State Department of Labor and Industrial Board of Appeals (administrative hearing body) when they learned of the tampering before a hearing (N.Y. AG 2012a).
• Intimidating immigrant car wash workers. In 2010, the Los Angeles city attorney obtained a protective order against two car wash owners after they made immigration-related threats to workers amid an ongoing wage theft case. The protective order directed the employers not to “harass, intimidate or retaliate” against workers, and also not to “attempt to prevent or discourage any employee or named victim…from participating or cooperating” in the investigation, prosecution, or enforcement of the case (CA Superior Court 2010).

Past prosecutions of crimes against workers provide some guidance on common questions about bringing such cases

Although providing prosecutors with a road map on how to bring such cases is beyond the scope of this report,16 past prosecutions provide some initial guidance regarding several common questions:

• What statutes may be used to bring such cases?
• How can prosecutors learn about cases? What sources of potential referrals exist?
• How can prosecutors engage in this work in a manner that responds to racial equity, social justice, and similar concerns?
• What funding sources may be available to support this work?

A range of applicable statutes can provide prosecutors with authority to take on various workplace-related crimes

In some states, there may be labor or wage-specific criminal provisions. But as noted in the case studies section, many prosecutors have brought workplace-related crimes cases using a variety of generally applicable state statutes. Often, prosecutors may be able to use existing law to bring such cases, including statutes addressing the following conduct:

• Theft (including theft of services or theft by swindle)
• Larceny
• Scheme to defraud
• Check fraud or passing bad checks
• Filing false documents with government agencies
• Creating and maintaining false business records to conceal wage theft and other violations
• Witness tampering and retaliation
• Insurance fraud
• Unlawful activity related to unemployment insurance, workers’ compensation, and prevailing wage requirements
• Manslaughter and homicide
• Labor trafficking
• Criminal sexual assault
• Endangering the welfare of a minor
• Child labor

Examples of state statutes governing workplace crimes

Some states, such as New York, have relatively long-standing statutes specifically addressing workplace-related employer crimes. Other states, such as Colorado, Minnesota, and Texas, have passed specific statutes on wage theft in recent years. Colorado and Minnesota passed laws that would define wage theft beyond a certain monetary threshold as a felony. This designation makes such cases more appealing to prosecutors for various reasons, including that it affords them more options in plea bargaining situations. Also, in some states, such as California, prosecutors have a more robust set of tools to address felonies (as compared with misdemeanors), including search warrants and use of a grand jury.

• New York. Under New York Workers’ Compensation Law § 52(1)(a), failure to secure workers’ compensation for more than five employees is a Class E felony; under Labor Law 220(3)(d)(i), willful failure to pay prevailing wages totaling more than $25,000 is a Class E felony (with higher level felonies for larger underpayments).
• Colorado. Legislation passed in 2019 explicitly included within its statutory definition of theft an employer who “being able to pay wages or compensation and being under a duty to pay, willfully refuses to pay wages or compensation.” In Colorado, theft is a felony if the dollar amount involved is at least $2,000.
• Texas. In 2011, Texas enacted a wage theft law specifying that within the existing “theft of services” law, partial payment of wages is not sufficient to negate the intent to avoid payment by an actor (in this case, an employer). (Contemporaneous news articles noted that this was a common employer defense, see for example McPherson 2011.) Tex. Penal Code Section 31.04(d-3)(1, 2)(1994).
• Minnesota. Minnesota’s theft statute includes “wage theft” as a type of theft, and defines the term “wage theft” as occurring, among other things, “when an employer with intent to defraud: (i) fails to pay an employee all wages, salary, gratuities, earnings, or commissions at the employee’s rate or rates of pay or at the rate or rates required by law.” MN Statutes 2020 Section 609.52 subdivision (1)(13).

Proposed legislation

• Rhode Island. A bill proposed in Rhode Island in 2021 (Rhode Island Legislature 2021 Regular Session Senate Bill 195) would increase penalties for wage theft, making nonpayment of wages a felony if the value of the wages owed to an employee is at least $1,500, or if the violation was a knowing or repeat violation (R.I. AG 2021)
• California. A bill proposed in California in 2021 would increase criminal penalties for wage theft (Alvarez 2021).

Prosecutors may receive workers’ rights case referrals from a variety of sources

State and local prosecutors who have brought cases against employers for violating workers’ rights have received case referrals through a variety of sources. Accordingly, prosecutors wishing to receive referrals should build relationships with a number of organizations and offices. These relationships should be ongoing and systematic; referrals and successful collaborations are unlikely to result from one-off conversations or one-sided presentations.

Organizations and agencies representing or assisting workers are the most common case referral sources

• Worker advocacy groups. Worker centers, workplace safety and health advocates, and other worker advocacy groups regularly speak with and hear from workers seeking help for a range of workplace violations.
• Labor unions. In addition to representing their members, many unions are actively organizing in a range of workplaces and industries, and routinely speak with workers experiencing violations. Unions might bring forward cases involving employers that are violating workers’ rights, or winning government contracts and cheating on taxes owed and worker pay.
• State labor departments. As the state’s primary regulators and civil enforcers of workplace laws, many state labor departments receive a considerable volume of incoming complaints. While most do not have a systematic method for referring cases to prosecutors, labor departments of both California and New York have regular methods of ongoing referrals, which could readily be replicated elsewhere. In fact, in 2014, the California Labor Commissioner’s Office (the equivalent of the state labor department) created a “Wage Theft is a Crime” campaign, with materials including posters and radio spots (CA Lab. Com. 2021); the office also offered training to district attorney offices on how to develop and bring these cases (Ramirez 2018).17
• Labor advisory boards or councils. A noteworthy model is provided by the labor advisory boards or councils established by both the Queens and Suffolk County district attorneys in New York. The councils consist of unions, worker centers, worker advocacy groups, and others within their jurisdiction. The office holds quarterly meetings (in person prior to the COVID-19 pandemic), which allow for formal discussions as well as informal conversations, relationship-building, and case referrals. These formalized groups create a way for DA offices to engage systematically and regularly with the community. (Suffolk Cty. DA 2020b).

Government agencies and officials outside of labor agencies are additional potential sources for case referrals

• State agencies such as those overseeing workers’ compensation insurance are sources of referrals, as are any state inspectors general
• City or municipal labor standards offices, where they exist, or city departments of investigation may refer cases. Many cities, including Chicago, Denver, Minneapolis, New York, Philadelphia, San Francisco, Seattle, and more, have city-level labor offices devoted to protecting workers’ rights. Also, for example, the New York City Department of Investigation hosts inspectors general for the School Construction Authority, the New York City Housing Authority, and more; these inspector general offices have been an active source of cases for prosecutors in New York City.
• Elected officials may be sources of referrals, particularly those representing immigrant or low-income communities.
• The U.S. Department of Labor, especially the Wage and Hour Division and OSHA, may be a source of referrals, although generally they first refer cases to U.S. attorney offices.
• DAs, AGs, and labor enforcers in other states can refer cases in situations involving employers operating in multiple jurisdictions.

Nongovernmental organizations are another important source of potential case referrals. They include:

• legal services and other public interest law offices
• plaintiffs’ wage and hour lawyers, such as members of the National Employment Lawyers Association or its state or local affiliate
• organizations that serve victims of human trafficking
• immigrants’ rights organizations and lawyers/nonprofits representing immigrants
• media, including social media and foreign language media, whose coverage can provide leads on cases
• law school clinics
• companies that compete with employers who are violating the law

Traditional law enforcement sources, like police departments or sheriffs’ offices, may also be helpful, but may presently be more useful as supplements to referrals from and collaboration with worker-focused organizations and agencies described above

• Workers face barriers to reporting violations directly to law enforcement for a variety of reasons, including fear of retaliation, agency language access limitations, unfamiliarity with legal rights or avenues for complaints, and fear of potential immigration consequences, among others (Grittner and Johnson 2021). One study estimated that there are about 130 violations for every one complaint lodged overall, and that this ratio varies tremendously across industries (Weil and Pyles 2007). Thus, a lack of worker complaints does not indicate employer compliance.18
• Still, a prosecutor’s own intake phone number, hotlines, or other avenues for receiving calls and tips from the public can sometimes lead to cases, especially after an office has publicly communicated its involvement in worker issues. These public engagement resources may receive more intakes and calls from workers after media or other announcements about workers’ rights cases. However, based on the barriers to worker complaints, relying solely on already-existing passive intake systems is unlikely to lead to information about the most egregious violations, especially during an office’s initial stages of involvement on these issues.
• To date, few prosecutions appear to have been initiated by law enforcement, such as sheriffs’ offices or the police. Prosecutors may wish to consider offering trainings to such agencies on workers’ rights issues.
• In a recent development, the Los Angeles County Sheriff in February 2021 launched a wage theft task force “to protect undocumented and documented workers in Southern California.” The task force is a collaboration of the sheriff’s department with the state labor commissioner, the Los Angeles County Office of Immigrant Affairs, the Los Angeles County District Attorney’s Office, the Los Angeles County Federation of Labor, and several community groups (LACSD 2021a). The sheriff’s department will be receiving complaints and referring them for criminal prosecution, civil enforcement, or other handling, as well as playing a direct role in enforcement and collections (LACSD 2021b). The sheriff also authored a Washington Post op-ed about the task force (Villanueva 2021). In addition, the Travis County (Texas) Sheriff’s Office added “wage theft” to a form allowing for online reporting of certain crimes (Travis CSO 2021).

Criminal justice concerns should be considered when prosecuting workplace violations

Racial and economic inequities in the criminal justice system and vulnerabilities of immigrant workers raise important concerns about these criminal prosecutions, including the following:

• Charging only low-level supervisors may fail to punish those with real responsibility for and authority over workers’ conditions. Some prosecutors may lean toward pursuing low-level supervisors with limited authority, while taking no action regarding higher-level officials with greater decision-making power. Prosecutions should seek to avoid this focus only on the “low-hanging fruit,” and instead attempt to target those with greatest responsibility for causing violations and with genuine ability to stop or prevent the violations. Prosecuting actors higher up the hierarchy is likely to prove challenging at times, because it is necessary to demonstrate that the defendant had mens rea, or the requisite criminal intent, and also because the standard of proof in criminal cases—“beyond a reasonable doubt”—is higher than the standard in civil cases. Prosecuting higher-level officials is important, however, to place responsibility on those who truly can change workers’ conditions, and to more effectively deter violations. As University of Maryland School of Law Professor Rena Steinzor observes, “the law must authorize prosecutors to climb the managerial ladder to find those responsible for making such incidents inevitable” (Steinzor 2015, 92.) In addition, prosecutors should be aware of racial disparities that may exist in pursuing only lower-level actors.
• Certain convictions have collateral immigration consequences for defendants. Under current immigration laws, certain criminal convictions can have immigration consequences, including threat of deportation. Prosecutors may wish to seek to avoid such consequences in considering charges for these cases. In 2017, for example, Brooklyn District Attorney Eric Gonzalez hired immigration attorneys to help prosecutors in his office “tailor criminal charges and plea bargains to avoid placing immigrant defendants in jeopardy of deportation” (Ryan 2017).
• Prosecutors should consider certifying victims and witnesses or U visas, where appropriate. Immigration issues also arise in relation to victims and witnesses, who may fear coming forward because of perceived potential consequences. The U visa is an immigration benefit for victims of certain crimes who are currently assisting, have assisted, or are likely to be helpful in assisting law enforcement in the investigation or prosecution of a qualifying crime. The U visa provides eligible victims with temporary immigration status to remain in the United States while assisting law enforcement, and if certain conditions are met, the U visa holder can ultimately obtain lawful permanent resident status. Individuals seeking U visas must be certified by a qualifying law enforcement agency, a category that includes prosecutors, and the certification process is relatively uncomplicated. Qualifying crimes include, among other things, human trafficking, involuntary servitude, manslaughter, obstruction of justice, peonage (holding someone in debt in servitude), sexual assault, and witness tampering (U.S. DHS 2019; NILC 2010).
• Alternatives to incarceration may be appropriate. Our country is undergoing an extensive national conversation about systemic racial inequities involved in mass incarceration. While a discussion of the problem of mass incarceration is beyond the scope of this report, prosecutors may want to consider alternatives to incarceration in resolving these cases. They may also want to consider whether innovative approaches to sentencing might be more effective. Finally, some criminal prosecutions of workers’ rights violations have been brought against corporations, not individuals; while such prosecutions are often seen as having less of an impact, they obviously raise no concerns about incarceration.
• Case resolution should include measures to ensure future compliance. In cases involving employers that continue to operate, prosecutors should seek terms that include monitoring or other measures to ensure future compliance. Monitoring could be performed by third-party monitors paid by the employer, or in partnership with administrative enforcement agencies. Prosecutors should also seek asset forfeiture where appropriate. A powerful tool in prevailing wage cases is debarment, which prevents a company from bidding on public works contracts in a given jurisdiction for a set period, sometimes up to five years. In addition, prosecutors may want to consider whether a restorative justice approach may be warranted or appropriate in these cases.19
• Workers who are victims should be provided with the opportunity to submit victim impact statements. Victims are often permitted or encouraged to submit victim statements (in writing or orally) regarding the impact of the crime on their lives. This opportunity should be provided to workers, because it enables their voices to be included in the process. Having the opportunity to address the court orally or in writing can be meaningful for workers, educates employers about the human consequences of their actions, and helps fully inform courts about the impact of wage theft and other employer crimes against workers.
• Civil enforcement may be a good option for prosecutors in some states. In some states, district attorneys have the authority to bring not only criminal prosecutions, but also civil lawsuits. For example, the Los Angeles and San Francisco district attorneys recently filed a civil lawsuit against the platform cleaning company Handy (L.A. DA 2021). To the extent that district attorneys have civil authority, they can consider exercising it to enforce workers’ rights.

While most prosecutors have brought workers’ rights cases without dedicated funding, limited dedicated funding mechanisms exist in some jurisdictions

Most DAs and state AGs who have brought workers’ rights cases have done so without any specific or dedicated funding. Just as they regularly prosecute theft, larceny, fraud, manslaughter, and other cases without dedicated funding, they simply add labor-related cases to their caseload when they emerge. However, there are a few examples of dedicated funding mechanisms.

• Funding to combat human trafficking. The U.S. Department of Justice has awarded funding to combat human trafficking; a 2020 press release describes over $101 million in grants for, among other things, “enhancing the capacity of law enforcement and other stakeholders to identify victims and provide justice for those victims through the investigation and prosecution of their traffickers” (U.S. DOJ 2020).

• Examples of state funding programs that may be used to pursue employer crimes. California and New York both have unique programs that are not specifically devoted to prosecution of employer crimes, but that have routinely been used to prosecute them. The infractions that qualify for the funding include those related to payroll fraud (such as failure to carry workers’ compensation), workers’ compensation fraud, and failure to pay or accurately report unemployment insurance taxes. Prosecutors have often also brought wage theft charges as part of those cases, given that these offenses often occur in clusters by the same employers.

• • Funding created by state legislation in California. California’s Workers’ Compensation Insurance Fraud Program, established in 1991, came about as part of a legislative package making workers’ compensation fraud a felony, requiring insurers to report suspected fraud, and establishing a mechanism for funding enforcement and prosecution activities. The funding comes from an assessment on employers. The aggregate assessment in the 2017–2018 fiscal year was more than $62 million. The legislation also established a commission, with representatives from labor, employers, and insurers, to determine the level of assessments and award grants to prosecutors. (Calif. DOI n.d.a, n.d.b).

• • State funding without legislation in New York. New York’s Crimes Against Revenue Program (CARP) was established in 2004 as a program funded by the state’s Division of Criminal Justice Services. It provides grants to district attorneys’ offices across the state to fund investigations and prosecutions of tax crimes as well as Medicaid, public assistance, and workers’ compensation fraud. Under the program, local district attorneys’ offices partner with various state agencies in bringing prosecutions. CARP funds have been used to support prosecutions of cases involving violations of prevailing wage, unemployment insurance, and workers’ compensation laws. Successful cases under CARP allow the state to recoup the costs of the program through restitution, fines, and penalties. The program has been revenue-generating for the state (NYS DCJS 2015, 2020, 2021; DAASNY 2019).

Areas for further exploration include research on deterrence and questions about implementation

The incidence and impact of state and local criminal prosecutions of employers have not been extensively studied to date, leaving a number of questions for researchers, prosecutors, worker advocates, legal scholars, and others.

One set of key research questions relates to the impact of criminal prosecution. Does prosecution of one employer deter violations by others? If so, how can such deterrence be measured, and how does it compare with that of civil enforcement? Does deterrence stem from the greater likelihood of detection resulting from more enforcers addressing labor issues, from publicity and reputational harm, from the gravity of potential consequences, or from all of the above?

Questions related to implementation include: What is the capacity of district attorneys and state attorneys general to bring such cases and build practices in this area? What training needs do such offices have? How can current staff, including investigative staff or law enforcement, be trained to effectively handle cases that are often considerably different than many typical criminal prosecution cases?

There are also questions related to partnerships between prosecutors and other government and nongovernmental actors. What partnerships can and should be built with labor enforcement agencies, unions, or worker organizations? How should roles in partnerships with labor enforcement agencies be defined to ensure compliance with respective ethical obligations and agency priorities?

Conclusion and recommendations

Numerous DAs and state AGS have begun to prosecute wage theft, payroll fraud, and other crimes committed by employers. These state and local prosecutors are responding to egregious violations that harm workers and communities, make it difficult for honest employers to compete, and deprive public coffers of money needed for critical safety net programs. We recommend that state and local prosecutors, state legislatures, and worker advocates build on this valuable work by taking further action. More detailed tips for getting started are included in Appendix C. Some general recommendations are as follows.

Recommendations for state and local prosecutors

Become involved. If your office has not yet become engaged in protecting workers’ rights, begin to do so. Learn more about the issue, meet with relevant stakeholder groups, review your office’s authority and potentially applicable statutes, research pressing needs in your jurisdiction, and begin to map out a plan of action.

Increase involvement. Offices that have brought occasional prosecutions in this area should continue to develop and increase their involvement.

Establish dedicated units or build existing ones. DA and AG offices without dedicated workers’ rights units should consider creating such units, using existing staff and jurisdiction if necessary. District attorney offices may consider including such units within an economic crimes, white-collar crime, financial investigations, or community protection bureau, if such bureaus already exist within the office. Offices with dedicated units should continue and expand their work in this area.

Connect with other prosecutors involved in this area, to share best practices and learn from each other.

Recommendations for state legislatures: next steps

Review statutes to assess whether they adequately address wage theft, payroll fraud, retaliation, and other crimes against workers. If appropriate, states should strengthen laws protecting workers’ rights, including laws related to civil enforcement and criminal jurisdiction. They should also provide jurisdiction for labor enforcement, both civil and criminal, to state attorneys general.

Consider establishing funding mechanisms. California’s Workers’ Compensation Insurance Fraud Program and New York’s Crimes Against Revenues Program both provide funds for prosecutions that can include violations of workplace laws, and result in recoveries for the state. Prosecutions of workers’ rights cases can sometimes generate revenue, because employers who commit payroll fraud fail to pay unemployment and other taxes.

Recommendations for worker organizations and advocates

Engage with DA and state AG offices. Worker organizations and advocates—including unions, worker centers, advocacy groups, legal services providers, and others—should consider ways to engage with the local DA’s office as well as their state AG’s office, particularly where the DA or AG have expressed support for or concern about worker issues, low-income communities, or economic and/or racial justice.

Acknowledgments

The author thanks Mackenzie Bouverat, Daniel Perez, and Nikita Rumsey for research assistance, Lora Engdahl for editing assistance, and numerous worker advocates and DA and AG government lawyers for their insights. The author also thanks the Bernard and Anne Spitzer Charitable Trust and the Justice Catalyst for their support.

Endnotes

1. This report describes the role of criminal prosecutors at the state and local level, typically including district attorneys, county attorneys, and state attorneys general. They will collectively be referred to herein as “criminal prosecutors,” “prosecutors,” or “district attorneys.” This report does not include a discussion of federal prosecutions.

2. Wage theft is the practice of employers failing to pay workers the full wages to which they are legally entitled. It includes situations in which employers refuse to pay promised wages, pay less than legally mandated minimums, fail to pay for all hours worked, keep worker tips or deductions intended for worker benefits, or do not pay overtime. In some states, the term “wage theft” is defined in the law, but more commonly it is used as a colloquial and descriptive term to refer to a set of practices. See Rosado Marzán 2021 for a detailed description of wage theft.

3. When employers wrongly treat workers as independent contractors instead of as employees, this is known as misclassification. When employers pay workers in unreported cash “off the books,” this leads to payroll fraud. Both practices result in employer failure to pay unemployment insurance taxes or buy required workers’ compensation insurance; they are often also accompanied by various forms of wage theft. Misclassification and payroll fraud harm workers, deprive public coffers of revenue, and hurt honest employers who struggle to compete with lawbreakers.

4. See the “CPR’s Crimes Against Workers Database” (Center for Progressive Reform n.d.)

5. Prevailing wage laws exist in a number of states; they generally require contractors on government contracts to pay workers at least the locally prevailing wages and fringe benefits paid on similar projects in the area. This topic is discussed in greater detail in section Three B.

6. Training materials on file with author.

7. In about half of the states, workers’ OSH Act rights are enforced by state agencies that have state plans approved by OSHA (Rosenthal 2021, note 48).

8. The difficulty of holding companies accountable for complying with federal wage and hour requirements has been analyzed by EPI research on joint-employer standards (standards that guide when contractors and the firms that use them can be held jointly responsible for complying with the law). See, for example, Shierholz and Poydock 2021.

9. An Act Concerning Criminal Offenses for Failure to Pay Wages, and, In Connection Therewith, Implementing Recommendations from the Colorado Human Trafficking Council, CO H.B. 19-1267, sec. 1, § 3(a) (2019).

10. There are also many examples in U.S. history of arrests of workers who are striking or otherwise seeking better conditions, from striking garment workers in the early 1900s to striking janitors toward the close of the century. For example, when the owners of the Triangle Shirtwaist Factory in 1909 hired “thugs” to attack striking workers, the police ultimately arrested the strikers (Greenhouse 2019; Baker 1990).

11. One academic commentator, Professor Ben Levin, has raised concerns about prosecuting wage theft and other employer crimes; however, these objections are based on general critiques of the criminal justice system and opposition to incarceration, rather than anything specific about prosecution of employers for crimes against workers. A thorough discussion of such objections is beyond the scope of this report; for those who are particularly interested in more details regarding this critical appraisal, see Levin 2018a and 2018 b; Gerstein and Seligman 2018; and Migiel-Schwartz 2021.

12. This list of sample cases seeks to provide an overview of the types of cases pursued. Thus, some examples describe charges brought but not outcomes if, for instance, charges are announced when a case is started, but the agency does not issue a press release when the case is resolved.

13. See endnote 2 above for a discussion of the term “wage theft.”

14. In Colorado, a law strengthening penalties for wage theft stemmed in part from a report by the state’s human trafficking council; the bill’s legislative declaration notes that employers who commit human trafficking “often commit other crimes such as wage theft, tax evasion, and workers’ compensation fraud,” and also that “not all victims of wage theft are victims of human trafficking.” An Act Concerning Criminal Offenses for Failure to Pay Wages, and, In Connection Therewith, Implementing Recommendations from the Colorado Human Trafficking Council, H.B. 19-1267, sec. 1, § 3(a) (2019).

15. For information about workplace sexual assault in the agriculture and janitorial industries, see Frontline documentaries “Rape in the Fields” (Cediel and Bergman 2013) and “Rape on the Night Shift” (Altan, Cediel, and Bergman 2018).

16. For a manual with practical guidance on prosecuting occupational safety and health-related crimes against workers, see McCluskey et al. 2016.

17. Various legal ethics issues arise in relation to cases that may be either criminal, civil, or both. For example, ethical rules prohibit threatening criminal charges to gain advantage in a civil suit. Prosecutors and any civil agencies referring cases should carefully review and discuss these and other restrictions as part of the process of establishing any collaborations.

18. See “Strategic Complaint Response Matrix,” Figure 6.2, p. 84 (Weil 2010).

19. “Restorative justice is an approach that focuses on meeting the needs of those who have been harmed while inviting those who have caused harm into a process of active accountability” (CCI n.d.).

Appendix A: Sample legal documents from past prosecutions of employer crimes against workers

The following list contains samples of public record documents in a range of prosecutions of employers. For simplicity and ease of reference, the listing includes the name of the prosecuting office, the state (where not included in the office name), the type of document, and year the document was filed. Documents are organized according to the primary issue in a given case; however, as discussed throughout this report, the same case may often include a number of different charges and violations.

Child labor

New York State Attorney General, misdemeanor complaint, 2014

New York State Attorney General, supporting deposition, 2014

New York State Attorney General, appellate court decision, 2018

Fraudulent garment shop licensing

California Attorney General, felony complaint, 2018

California Attorney General, declaration in support of arrest warrants, 2019

Labor trafficking

California Attorney General, felony complaint, 2018

Minnesota, Hennepin County District Attorney, complaint, 2018

New York, Suffolk County District Attorney, Singh case. Four felony complaints were filed in this case: labor trafficking, labor trafficking, offering a false instrument for filing, and scheme to defraud, all 2020

Payroll fraud primarily

California Attorney General, amended felony complaint, 2010

California Attorney General, petition to preserve property and assets, 2011

California, Alameda County District Attorney, felony complaint, 2016

California, Contra Costa County District Attorney, felony complaint, 2020

California, Contra Costa County District Attorney, forfeiture and restitution order, 2020

California, Contra Costa County District Attorney, plea agreement, 2020

New York, Suffolk County District Attorney, felony complaint, 2020

New York, Suffolk County District Attorney, indictment, 2020

New York, Suffolk County District Attorney, information, 2020

[Wage theft and payroll fraud] New York State Attorney General, indictment, 2013

Prevailing wage violations

California, Yolo County District Attorney, preliminary hearing brief, 2014

California, Yolo County District Attorney, sentencing brief, 2019

Massachusetts Attorney General, joint proposed motion and order for agreed restitution amount, 2018

New York State Attorney General, felony complaint, 2017

New York State Attorney General, indictment, 2017

New York, Kings County District Attorney, debarment stipulation, 2017

New York, Kings County District Attorney, indictment, 2019

New York, Queens County District Attorney, indictment, 2019

New York, Suffolk County District Attorney, felony complaint, 2020

Pennsylvania Attorney General, complaint and affidavit of probable cause, 2021

Pennsylvania Attorney General, sentencing order, 2021

Pennsylvania Attorney General, complaint and affidavit of probable cause, 2019

Wage theft and payroll fraud

Wage theft

California, Contra Costa County District Attorney, TRO barring dissipation of assets, 2016

Minnesota, Hennepin County District Attorney, complaint, 2013

Washington State Attorney General, Sandoval case (multiple documents): defendant’s sentencing memorandum, judgment and sentence of corporate defendant, judgment and sentence of individual defendant, all 2018

Wage theft and payroll fraud

California, Alameda County District Attorney, complaint, 2016

California, Alameda County District Attorney, order barring dissipation of assets, 2017

California, Contra Costa County District Attorney, application for TRO barring dissipation of assets, 2015

California, Contra Costa County District Attorney, complaint, 2015

California, Contra Costa County District Attorney, felony complaint, 2014

California, Santa Monica City Attorney, complaint, discovery request, 2013

California, Santa Monica City Attorney, first amended complaint, 2013

California, Santa Monica City Attorney, terms and conditions of probation, 2013

Massachusetts Attorney General, memo to aid the court regarding sentencing, 2018

Massachusetts Attorney General, statement of the case, 2018

New York State Attorney General, complaint, 2015

New York State Attorney General, felony complaint, 2015

New York State Attorney General, indictment, 2018

Workplace safety and health

New York, Kings County District Attorney, corporate summons, 2019

New York, Kings County District Attorney, indictment, 2017

New York, Kings County District Attorney, indictment, 2018

New York, Kings County District Attorney, letter to court, 2020

Appendix B: Additional sources of information about cases

While no comprehensive list of all relevant cases has been compiled, the Center for Progressive Reform maintains a “Crimes Against Workers” database.

In addition, this informal spreadsheet contains information about numerous cases that have recently been the subjects of press releases or media coverage.

Appendix C: Getting started

While this report does not include a detailed roadmap for implementing a new criminal prosecution program in a jurisdiction, the following are some tips, mostly from front-line prosecutors engaged in this work, for district attorneys’ offices wanting to get started, as well as tips for worker advocates hoping to encourage their local DAs to begin prosecuting employer crimes against workers. Finally, there are some tips for both prosecutors and worker advocates about building their relationships.

Tips for prosecutors’ offices wanting to get started

Prepare to do the work

• Conduct initial research
  • Review your jurisdiction’s criminal, labor, and insurance fraud statutes and compile a list of all laws with criminal provisions that may apply in an employment setting (even if there is no precedent). Determine what elements need to be proven.
  • Think about what background information can inform the work: the shape of the local economy, which kinds of workers are most vulnerable, who handles these cases civilly, etc.
  • Connect with prosecutors from other jurisdictions to understand how they have brought cases (even under different laws).
  • Connect with labor-focused national organizations and think tanks (like the Economic Policy Institute, the National Employment Law Project, and others) that can provide background and orientation.
• Prepare the team
  • Identify lawyers who will staff the unit (if applicable) or handle the cases. Ideally, there would be at least two: one with labor background and one with a background in criminal prosecution, although this may be difficult. Another effective combination would be to identify an attorney from the state or local labor enforcement agency who can work in close partnership with the assistant DA handling cases.
  • Train lawyers on the team regarding finding and prosecuting cases.
  • Clarify who will be responsible for conducting your investigations—local police, DA investigators, attorneys, sworn or nonsworn investigative personnel from other agencies—and try to prepare some initial trainings or find other offices that can share training resources.

• Ask big picture questions
  • Consider setting some basic parameters for cases that you will consider for criminal prosecution (such as number of workers, amount of theft, evidence of discrimination). This will help set realistic expectations for worker advocates who refer cases and help maintain those relationships. If needed, you can make exceptions to the parameters for particularly egregious cases in which criminal charges are appropriate.
  • Consider the broad goals of the work. What would constitute a “success” for your office in this area: restitution to large numbers of workers? evidence of deterrence? How might you assess your office’s impact?
• Plan for implementation issues
  • Think about what it might take for your office to collaborate with vulnerable workers: language assistance? Certain kinds of investigators? Strategic collaborations? Be sure to ensure language accessibility by having interpreters on staff or on standby if needed for interviewing witnesses. Working closely with community-based and worker organizations can enhance trust with witnesses.
  • Set up a complaint form, phone line, and e-mail submission access.

Reach out to a wide range of stakeholders and partners

• Build relationships with organizations that engage with workers
  • Identify, reach out to, and build relationships with unions, worker centers, advocacy groups, and other community-based organizations that serve and advocate for workers, as well as private and nonprofit employment and labor lawyers. In addition to organizations focusing on workers, consider reaching out to social services, immigration services, or religious organizations; community or cultural centers; and consulates.
  • It is often helpful to meet in the offices of these partner organizations or attend events they hold, to learn about what they do and who they serve. Talk to their members and ask about their experiences.
  • Offer to do a presentation on wage theft and related crimes for their members, maybe highlighting cases in other jurisdictions that may have parallels in cases in your jurisdiction.
  • Keep in regular touch so that you are top of mind and staff at these organizations feel comfortable contacting you should a particularly egregious case come to their attention.
• Connect with other government agencies
  • Get to know state agency partners—and not just those in the labor department. Include insurance regulators, revenue departments, financial institution regulators, and others. Identify, reach out to, and build relationships with other law enforcement agencies that operate in the following workplace areas: wage and hour standards, employment tax issues, health and safety, and industry-specific areas (for example, public health departments may inspect nursing homes). Research which of the laws/regulations they enforce could constitute criminal violations.
  • Meet and greet and talk about what other jurisdictions have done, and explore how coordinated enforcement can make all your cases stronger. If you have investigative resources, offer them to support joint investigations. When a specific case is referred to you, reach out to them as experts for questions related to their agencies’ jurisdiction.
  • Connect with other law enforcers in your county/state from the offices of the attorney general, department of labor, human rights, etc.
• Reach out to the employer community
  • Reach out to business associations, the local Chamber of Commerce, and the management bar, and inform them of your office’s intention to start bringing cases in this area.
  • In some industries, law-abiding employers appreciate enforcement because they lose work and struggle to compete with businesses that gain a competitive advantage by violating laws.
  • This outreach also places the community that may face prosecution on notice. In fact, outreach itself can help drive legal compliance, as concerned employers may change practices.
• Consider a general outreach campaign
  • A general outreach campaign, along the lines of the California Labor Commissioner’s “Wage Theft is a Crime” media campaign, can be helpful for raising awareness.

Select and handle cases

• Choose initial cases carefully. While your office surely selects all cases carefully, it is wise to choose your first several cases in any new area extremely carefully, so that you begin by taking on cases with egregious facts and exceedingly strong evidence, as well as witnesses committed to the case, which can be facilitated through working with community-based and worker organizations.
• Set realistic expectations regarding outcomes. Set realistic expectations with workers who are victims and witnesses regarding case outcomes, including regarding restitution amounts and likelihood of incarceration.
• Plan your investigative steps
  • Visit and observe the place of employment if open to the public.
  • Conduct interviews with as many workers as possible. Learn the witnesses’ stories. Be sure to ask what they experienced on the job (pay rates, schedules, cash or check, type of work, etc.)
  • If applicable, work closely with the organization or advocates that referred the case to you, to help build a relationship of trust with the witnesses, facilitate a thorough investigation, and foster open communication throughout the course of the investigation.
  • If possible, work closely with an investigator and financial auditor. Try to enlist a forensic accountant, who can help follow the money and be able to explain how things went bad.
  • Serve grand jury subpoenas when warranted: recipients may include the employer, payroll services, banks, insurance companies that have issued liability or workers’ compensation policies, and unemployment insurance agencies.
  • Where appropriate, obtain a search warrant if needed when you have probable cause and reliable information as to the location of payroll and employee information.

Tips for worker advocates hoping to engage with their local DAs

Prepare to do the work

• Talk with workers or members about their needs and your organizational goals, and discuss the pros and cons of engaging with the criminal justice system for these cases.
• Reach out to advocates in jurisdictions where there have been successful criminal wage theft prosecutions. Talk to the community-based organization partners there to learn how they got started, what worked and didn’t work, what challenges they faced, what they would do differently. Ask for contact information for the prosecutors and administrative enforcement agencies they worked with.
• Reach out to the administrative enforcement agencies you already work with to ask whether they have ever considered referring cases for criminal prosecution. Offer to facilitate a conversation with their colleagues in other jurisdictions who have.
• Connect with labor-focused national organizations and think tanks (like the Economic Policy Institute, the National Employment Law Project, and others) that can provide background and orientation.

Meet with your local DA’s office

• Plan before the meeting: Consider how you might best educate prosecutors about what you see happening on the ground, including different kinds of sample cases you might share with them.
• Learn about the office’s structure, jurisdiction, and staffing. Are there economic crimes or consumer protection units, which might be a good fit? Does the prosecutor’s office also have civil jurisdiction in addition to criminal authority?
• Prepare to educate them about worker issues and workplace laws. These issues may be new to them, and they may not be familiar with the labor laws in your jurisdiction. Understand that what you are asking them to do may be different from the cases they have traditionally brought. Share information about cases brought in other jurisdictions, and share resources about the growing trend of criminal prosecution of employer crimes against workers.
• Ask prosecutors and investigators what information they need, and how and when they want it presented to them. Also try to learn where resources and the law are lacking. Ask how they work differently from civil attorneys or agencies.
• Share information about the working conditions you believe should be addressed by criminal prosecution. Share compelling, egregious stories, and also information about successful prosecutions in other jurisdictions. Explain why criminal prosecutions can be so powerful in terms of deterrence and compliance.
• Offer to connect them with prosecutors from other jurisdictions who have brought cases enforcing workplace rights.
• Consider inviting your local prosecutor to your space, so they can get to know your workers and vice versa. Or invite the prosecutor to one of your organization’s meetings or events, so they can hear firsthand about unlawful working conditions from the workers experiencing them. Offer them time at a meeting or event to speak about what their office does and how the office can help the community, beyond the specific issue of prosecution of work-related issues.

Make referrals

• Ask beforehand what kind of information the office would like to receive. Ask also when referrals should be made: Sometimes prompt referrals, such as when a construction project is still ongoing, can enable covert investigation.
• Remember sensitive aspects of criminal prosecution and government work. For example, emails may be subject to freedom of information laws, and witness statements must be provided to the defense in criminal cases. Ask if prosecutors would prefer an email or phone call to start a conversation about a referral.
• Understand constraints faced by the prosecutor such as statutes of limitation and the more stringent “beyond a reasonable doubt” burden of proof.
• Be mindful of relationships among different prosecutors, agencies, and regulators: they may prefer not to have the same issue referred to multiple government offices. If you are referring the same matter to more than one office, it’s helpful to alert them.
• One of the most useful things you can do in referring a case is to help an office connect with workers who can serve as witnesses, and to help keep track of worker witnesses as an investigation and case proceed. Other helpful steps include providing background information about an employer, and helping workers gather evidence such as pay stubs, paychecks, or photographs.

Tips for both prosecutors and worker advocates about building their relationships

Discuss priorities and concerns of advocates.

Discuss relevant aspects of the prosecutor’s office, including the following: investigative process, case selection criteria, enforcement priorities, and applicable statutes of limitation.

Share information about the prosecutor’s office and worker organization as a whole, beyond the potential for case referrals.

Learn about the full scope of each other’s functions and activities:

• What areas does the office or organization work in? What services does it provide, or for an organizing group, what activities does it engage in?
• What resource limitations or constraints exist? What is the staffing level and structure?
• What are current priorities? Recent innovations?
• Are there other issues on which there is potential for collaboration, such as fighting elder abuse, affinity group fraud, or fraudulent immigration service providers?

Discuss the process for case referrals

• What kinds of cases should be referred? Are there dollar or worker thresholds?
• When and how should cases be referred? What information should be included in a referral?
• To ensure that workers are willing to come forward and report violations, prosecutors’ offices should not ask workers about immigration status. This should be explicitly discussed so that everyone involved understands the office’s practices in this regard.
• Will there be a point person on both sides?
• What information can be shared with the worker organization referring the case? There is often a significant asymmetry in the information flow, as DA offices have significant limitations in what they can share with people outside of the DA’s office.
• What are the general steps in cases and what is the typical timeline?
• How will any media coverage be handled?
• What may be included in a resolution?
• Will there be an opportunity for workers to submit victim impact statements?
• What information must be kept confidential to avoid compromising the case?

Maintain regular contact

• Even though a DA office may not accept the first few case referrals, eventually there may be a referral that works.
• Remember that both offices share the goal of protecting workers.
• Stay in communication even if there’s no case yet, and explore opportunities for collaboration. For example, the worker organization could provide training about a specific industry with high rates of violation, or the prosecutor’s office could provide a know-your-rights presentation on an issue of interest to members.

References

AFL-CIO. 2020. Death on the Job: The Toll of Neglect, 2020. 29th edition, October, 2020.

Alvarez, Felicia. 2021. “New Bill Proposes Jail Time For Employers Who Participate in Wage Theft.” Sacramento Business Journal, February 23, 2021.

Ashenfelter, Orley, and Robert S. Smith. 1979. “Compliance with the Minimum Wage Law.” Journal of Political Economy 87, no. 2 (1979): 333–350. Accessed March 23, 2021. http://dx.doi.org/10.1086/260759

Associated Press (AP). 2017. “Owner of Last Howard Johnson’s Restaurant Charged with Sexual Abuse.” Boston.com, October 12, 2017.

Associated Press (AP). 2018. “Owner of Last Howard Johnson Restaurant Jailed for Sexual Harassment.” CBS News, November 1, 2018.

Associated Press (AP). 2019. “Michigan AG Nessel Issues 13 Felony Charges in First Payroll Fraud Case.” Fox47 News, August 28, 2019.

Altan, Daffodil, Andrés Cediel, and Lowell Bergman. 2018. “Rape on the Night Shift.” Frontline PBS and Univision Documentary. Aired January 16, 2018.

Baker, Bob. 1990. “Police Use Force to Block Strike March.” Los Angeles Times, June 16, 1990.

Bear, John. 2018. “Boulder Businessman Scott Roy Sentenced to Six Months in Jail for Unlawful Sexual Contact.” Daily Camera, January 12, 2018.

Berkowitz, Deborah. 2017. OSHA Severe Injury Data From 29 States: 27 Workers A Day Suffer Amputation or Hospitalization. National Employment Law Project, April 2017.

Bernhardt, Annette, Ruth Milkman, and Nik Theodore. 2009. Broken Laws, Unprotected Workers: Violations of Employment and Labor Laws In America’s Cities. National Employment Law Project, September 2009.

Bhargava, Deepak, and Alexander Hertel-Fernandez. 2020. “Looking Ahead: Enforcement For the Many, Not the Few.” Democracy, a Journal of Ideas, Fall 2020.

Boyd, Shaun. 2019. “Put the Exploiters In Jail’: Wage Theft Bill Cracks Down On Employers.” 4CBS Denver, April 2, 2019.

Brooklyn District Attorney’s Office. n.d. “Frauds Bureau,”( web page). Accessed April 19, 2021.

Brooklyn District Attorney (Brooklyn DA). 2019. “Construction Company Operator, Foreperson and Engineer Indicted for Manslaughter in Death of Laborer Buried in Debris Following Wall Collapse” (press release). November 21, 2019.

Bureau of Labor Statistics, Current Population Survey (BLS-CPS). 2021. “Union Members—2020” (news release). Accessed March 1, 2021.

Bureau of Labor Statistics, Injuries, Illnesses, and Fatalities (BLS-IFF). 2019a. “National Census of Fatal Occupational Injuries in 2019” (news release). Accessed December 16, 2020.

Bureau of Labor Statistics, Injuries, Illnesses, and Fatalities (BLS-IFF). 2019b. “Employer-Reported Workplace Injuries and Illnesses — 2019” (news release). Accessed November 4, 2020.

Byars, Mitchell. 2017. “Boulder Ice Cream Co-owner Facing New Sex Assault Charges.” Daily Camera, March 29, 2017.

Byars, Mitchell. 2019a. “Boulder Man Fails to Pay Undocumented Workers, Then Threatens Them, Police Say.” Daily Camera, January 8, 2019.

Byars, Mitchell. 2019b. “Boulder Man Sentenced For Threatening Undocumented Workers.” Daily Camera, November 12, 2019

Byfield, Erica. 2019. “6 Charged in 2018 NYC Wall Collapse that Killed Construction Worker.” NBC New York, November 21, 2019.

Byrne, Matt. 2019. “Contractor Indicted on Manslaughter Charges in Worker’s Fatal Fall from Roof.” Portland Press Herald, April 10, 2019.

California Attorney General’s Office (Calif. AG). 2018. “Attorney General Becerra Announces Criminal Charges in a Bay Area Labor Exploitation and Human Trafficking Case” (press release). September 7, 2018

California Department of Insurance (Calif. DOI). n.d.a. “Workers’ Compensation Fraud” (web page). Accessed April 10, 2021.

California Department of Insurance (Calif. DOI). n.d.b. “Fraud Assessment Commission.” (web page). Accessed April 12, 2021.

California Labor Commissioner’s Office (CA Lab. Com). 2021. “Wage Theft is a Crime Campaign Materials” (web page). Accessed April 11, 2021.

California Statewide Law Enforcement Association (CSLEA). 2016. “San Diego Restaurant Owner Sentenced to Jail for Wage Abuses.” December 12, 2016.

California Statewide Law Enforcement Association (CSLEA). 2019. “Former Owner of Temp Agency Convicted of Workers’ Comp. Insurance Premium Fraud.” December 10, 2019.

California Superior Court for the County of Los Angeles, “Protective Order,” Case No. 9CA01116, August 13, 2010.

Carré, Françoise. 2015. (In)dependent Contractor Misclassification. Economic Policy Institute, June 2015.

Cediel, Andrés, and Lowell Bergman. 2013. “Rape in the Fields.” Frontline PBS and Univision Documentary. Aired June 25, 2013.

Center for Progressive Reform (CPR). n.d. “CPR’s Crimes Against Workers Database” (database). Accessed, April 6, 2021.

Center for Court Innovation (CCI). n.d. Restorative Justice (webpage). Accessed April 19, 2021.

Centro de Trabajadores Unidos en La Lucha (CTUL). 2019. “Victory!” (web post). May 2019.

Chatterjee, Rhitu. 2018. “A New Survey Finds 81 Percent of Women Have Experienced Sexual Harassment.” The Two-Way, NPR. February 21, 2018.

Chen, David W. 2016. “Construction Company Guilty of Manslaughter in Immigrant Worker’s Death.” New York Times, June 10, 2016.

Cho, Eunice Hyunhye, Tia Koonse, and Anthony Mischel. 2013. Hollow Victories: The Crisis in Collecting Unpaid Wages for California’s Workers. UCLA Labor Center and National Employment Law Project, 2013.

Christian, Murray. “Queens DA Launches Bureau Targeting Predatory Lenders and Housing Scammers.” Astoria Post, June 3, 2020.

Christian, Peter. 2019. “Helena Man Pleads Guilty to Felony Employer Misconduct.” KGVO. October 3, 2019.

Colorado Fifth Judicial District Attorney’s Office. 2020. (Colo. 5th DA). “Local Contractor Charged with Wage Theft from His Employees, and Tax Evasion” (press release). June 25, 2020.

Colorado General Assembly. 2019. Bill Summary for HB19-1267; House Committee on Judiciary Report, Hearing Summary, April 2, 2019.

Colvin, Alexander J.S. 2018. The Growing Use of Mandatory Arbitration: Access to the Courts is Now Barred for More Than 60 Million American Workers. Economic Policy Institute, April 2018.

Cooper, David. 2016. Balancing Paychecks and Public Assistance: How Higher Wages Would Strengthen What Government Can Do. Economic Policy Institute, February 2016.

Cooper, David, and Teresa Kroeger. 2017. Employers Steal Billions from Workers’ Paychecks Each Year. Economic Policy Institute, May 2017.

Costa, Daniel, Philip Martin, and Zachariah Rutledge. 2020. Federal Labor Standards Enforcement in Agriculture: Data Reveal the Biggest Violators and Raise New Questions about How to Improve and Target Efforts to Protect Farmworkers. Economic Policy Institute, December 2020.

Covert, Bryce. 2020. “Sexual Harassment at McDonald’s Is Even Worse Than We Knew.” The Nation, May 26, 2020.

District Attorneys Association of the State of NY. (DAASNY). 2019. Joint Budget Hearing Testimony on Public Protection before New York Senate. January 29, 2019.

District of Columbia Attorney General’s Office (D.C. AG). 2019. “AG Racine Releases Report on Payroll Fraud in District Construction Industry” (press release). September 10, 2019.

Dobruck, Jeremiah. 2014. “Contractor Convicted of Stealing Workers’ Wages During Costa Mesa Project.” Los Angeles Times, September 26, 2014.

Economic Policy Institute. 2018. Economic Policy Institute Policy Agenda, last updated December 2018.

Estlund, Cynthia, 2018. The Black Hole of Mandatory Arbitration, 96 N.C. L. Rev. 679, 692–699 (Mar. 2018).

Feldblum, Chai R., and Victoria A. Lipnic. 2016. EEOC Select Task Force on the Study of Harassment in the Workplace. U.S. Equal Employment Opportunity Commission. Accessed March 1, 2021.

Feshir, Riham. 2019. “Hennepin County’s First Labor-Trafficking Case Ends in Guilty Plea.” MPR News, November 19, 2019.

Fine, Janice, Daniel Galvin, Jenn Round, and Hana Shepherd. 2020. Maintaining Effective U.S. Labor Standards Enforcement Through the Coronavirus Recession. Washington Center for Equitable Growth, September 3, 2020.

Flaherty, Nora. 2019. “Saco Contractor Indicted on Rare Charge of Workplace Manslaughter.” Maine Public Radio, April 11, 2019.

Flanagan, Jane, Terri Gerstein, and Patricia Smith. 2020. How States and Localities Can Protect Workplace Safety and Health. Harvard Labor and Worklife Program and National Employment Law Project, May 2020.

Gartrell, Nate. 2016. “Wage Theft Allegations Grow Against Bay Area Elder Care Homes.” Mercury News. Updated August 11, 2016.

Gerstein, Terri, and David Seligman. 2018. “A Response to ‘Rethinking Wage Theft Criminalization.’” OnLabor, April 20, 2018.

Gerstein, Terri. 2020. Workers’ Rights Protection and Enforcement By State Attorneys General: State AG Labor Rights Activities from 2018 to 2020. Economic Policy Institute and Harvard Labor and Worklife Program, August 27, 2020.

Goodell, Nathaniel, and Frank Manzo. 2021. The Costs of Wage Theft and Payroll Fraud in the Construction Industries of Wisconsin, Minnesota, and Illinois. Midwest Economic Policy Institute, January 2021.

Gordon, Colin, Matthew Glasson, Jennifer Sherer, and Robin Clark-Bennett. 2012. Wage Theft in Iowa. Iowa Policy Project, August 2012.

Graves, Chris. 2020. “Contractor Gets 9 Months in Hennepin Co. Labor Trafficking Case.” MPR News. January 15, 2020.

Green, Jason. 2021. “San Jose Flooring Company Owner Charged with Wage Theft.” The Mercury News. April 13, 2021.

Green, Sara Jean. 2016. “Seattle Contractor Charged with Felony for Employee’s Death in 2016 Trench Collapse.” Seattle Times, January 9, 2018.

Greenhouse, Steven. 2019. Beaten Down, Worked Up: The Past, Present, and Future of American Labor. New York: Alfred A. Knopf (2019).

Hamaji, Kate, Rachel Deutsch, Elizabeth Nicolas, Celine McNicholas, Heidi Shierholz, and Margaret Poydock. 2019. Unchecked Corporate Power: Forced Arbitration, the Enforcement Crisis, and How Workers Are Fighting Back. Economic Policy Institute, Center for Popular Democracy, and National Employment Law Project, May 2019.

Harris, Sarah. 2014. “Violi’s Restaurant Owners Violated Child Labor Laws.” NCPR, December 12, 2014.

Haynes, Madisson. 2019. “Owner of Construction Company Faces Charges After Teen Dies in Trench Collapse.” WUSA9, November 21, 2019.

Human Exploitation and Trafficking Watch (HEAT Watch) (website). n.d. accessed April 10, 2021.

Hennepin County Attorney’s Office (Hennepin CA). 2019. “Ricardo Batres Pleads Guilty to Labor Trafficking” (press release). November 22, 2019.

Hennepin County Attorney’s Office (Hennepin CA). 2020a. “Husband and Wife Face Three Felony Charges for Elaborate Workers’ Compensation Fraud Scheme” (press release). January, 2020.

Hennepin County Attorney’s Office (Hennepin CA). 2020b. “Former Owners of Merit Drywall Sentenced for Insurance Fraud” (press release). November 2020.

Hennepin County Attorney’s Office (Hennepin CA). 2020c. “Ricardo Batres Sentenced to 270 Days in Jail and Five Years Probation for Labor Trafficking” (press release). January 2020.

HPM Digital Team. 2018. “Prosecutors File First Wage Theft Case in Harris County.” Houston Public Media, September 5, 2018. (HPM Digital Team 2018)

Huizar, Laura. 2019. Exposing Wage Theft Without Fear: States Must Protect Workers from Retaliation. National Employment Law Project, June 2019.

Intarasuwan, Kiki. 2018. “Rainbow Bright Daycare Trafficked Workers, Stole $8.5 Million in Wages: Attorney General.” NBC Bay Area, September 7, 2018.

Johnson, Matthew S. 2020. “Regulation by Shaming: Deterrence Effects of Publicizing Violations of Workplace Safety and Health Laws.” American Economic Review, 110 6 1866-1904. June 2020. https://www.aeaweb.org/articles?id=10.1257/aer.20180501

Grittner, Amanda M. and Johnson, Matthew S. 2021. “When Labor Enforcement and Immigration Enforcement Collide: Deterring Worker Complaints Worsens Workplace Safety.” Working Paper. 2021.

Kashinsky, Lisa. 2019. “Drain Company Owner Convicted of Manslaughter in South End Trench Collapse.” Boston Herald. October 31, 2019.

Levin, Ben. 2018a. “Rethinking Wage Theft Criminalization.” OnLabor, April, 13, 2018.

Levin, Ben. 2018b. “Prosecutorial Power, Prisons, and the Problem with Wage Theft Criminalization: A Reply.” OnLabor, April 30, 2018.

Levine, Marianne. 2018. “Behind the Minimum Wage Fight, a Sweeping Failure to Enforce the Law.” Politico, February 18, 2018.

Littlefield, Dana. 2017. “Company Owners Plead Guilty in $7 Million San Diego Hotel-Worker Fraud Case.” San Diego Union-Tribune. April 9, 2017.

Lopez, Pablo. 2018a. “Prosecutors Rack Up Fresno County’s First Ever Human Labor Trafficking Conviction,” Fresno Bee. August 31, 2018.

Lopez, Pablo. 2018b “Man Says He’s Innocent of Labor Trafficking. Judge Says He Turned Three Women into Slaves.” Fresno Bee. October 29, 2018.

Los Angeles County Sheriff’s Department (LACSD). 2021a. “LA County Sheriff Villanueva Launches the ‘Wage Theft Task Force’” (press release). February 9, 2021.

Los Angeles County Sheriff Department (LACSD). 2021b. “Wage Theft Task Force” (website). Accessed April 10, 2021.

Los Angeles County District Attorney’s Office (L.A. DA). 2021. “District Attorney Gascón Announces Worker Protection Lawsuit Against Handy for Misclassifying Its Workers.” (press release). March 17, 2021.

Lotshaw, Tom. 2021. “Local Contractor Pleads Guilty in Tax Evasion, Wage Theft Case.” VailDaily. January 6, 2021.

Mahalia, Nooshin. 2008. Prevailing Wages and Government Contracting Costs, Economic Policy Institute, July 3, 2008

Massachusetts Attorney General’s Office (Mass. AG). 2018. “New Bedford Man and East Providence Cleaning Company Charged in Connection With Labor Trafficking” (press release). May 7, 2018.

Massachusetts Attorney General’s Office (Mass. AG). 2019. “Temp Company Owners Plead Guilty to Wage Theft, Intimidation, and Retaliation Against Warehouse Workers” (press release). December 12, 2019.

McCluskey, Martha T.; Thomas O. McGarity; Sidney Shapiro; Rena Steinzor; Ron Wright; and Katherine Weatherford. 2016. Preventing Death and Injury on the Job: The Criminal Justice Alternative in State Law. Center for Progressive Reform. March 2016.

McNicholas, Celine, Margaret Poydock, Julia Wolfe, Ben Zipperer, Gordon Lafer, and Lola Loustaunau. 2019. Unlawful: U.S. Employers are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns. Economic Policy Institute, December 2019.

McPherson, Elizabeth, “Texas Legislature Passes Groundbreaking Wage Theft Bill.” Construction Citizen, June 9, 2011.

McPherson, Elizabeth, “First Wage Theft Charges in Harris County History Brought Against Million-Dollar Homeowner for Refusing to Pay House Painter.” Construction Citizen, September 7, 2018.

Meixell, Brady, and Ross Eisenbrey. 2014. An Epidemic of Wage Theft Is Costing Workers Hundreds of Millions of Dollars a Year. Economic Policy Institute. September 2014.

Michigan Attorney General’s Office. (Mich. AG). 2019. “Michigan AG Nessel Issues 13 Felony Charges in First Payroll Fraud Case.” (press release). August 28, 2019

Mid Hudson News. 2020. “Hoovler Holds Labor Fraud Training for State DAs.” June 24, 2020.

Migiel-Schwartz, Michael. 2021. “Why District Attorneys Should Take Up Wage Theft Criminalization.” OnLabor, February 22, 2021.

Minkler, Meredith, Alicia L. Salvatore, Charlotte Chang, Megan Gaydos, Shaw San Liu, Pam Tau Lee, Alex Tom, Rajiv Bhatia, and Niklas Krause. 2014. “Wage Theft as a Neglected Public Health Problem: An Overview and Case Study from San Francisco’s Chinatown District.” American Journal of Public Health, vol. 104(6), 1010–1020. June, 2014. https://doi.org/10.2105/AJPH.2013.301813

Minnesota Attorney General’s Office (Minn. AG). 2019. “Attorney General Ellison Announces New Wage Theft Unit in AGO.” (press release). July 15, 2019.

Nassau District Attorney’s Office. n.d. “Nassau County District Attorney’s Office Bureaus.” (web page). Accessed April 19, 2021.

National Academies of Sciences, Engineering, and Medicine (NASEM). 2020. Sexual Harassment of Women: Climate, Culture, and Consequences in Academic Sciences, Engineering, and Medicine. National Academies Press, June 2018.

National Conference of State Legislatures (NCSL). 2018. “Prosecuting Human Traffickers” (web page). Accessed April 1, 2021.

National Human Trafficking Hotline (NHTH). 2019. U.S. National Human Trafficking Hotline 2019 Data Report. National Human Trafficking Hotline and Polaris Project. Accessed March 1, 2021.

National Immigration Law Center (NILC). 2010. The U Visa and How It Can Protect Workers. September 2010.

NBC10 Boston. 2019. “Atlantic Drain Owner Convicted in Boston Trench Collapse That Killed 2 Workers.” October 31, 2019.

New Jersey Attorney General’s Office (N.J. AG). 2019. “Contractor Pleads Guilty to Falsifying Records to Cheat Workers Out of $200,000 by Not Paying Prevailing Wages” (press release). March 27, 2019.

New York Attorney General’s Office (N.Y. AG). 2012a. “A.G. Schneiderman Announces Arrest of Garment Factory Owner for Witness Tampering” (press release). February 2, 2012.

New York Attorney General’s Office (N.Y.AG). 2012b. “AG Schneiderman Announces Guilty Plea of Home Health Agency Owner for Failing to Pay Workers $300,000 in Wages.” (press release). October 4, 2012.

New York Attorney General’s Office (N.Y. AG). 2013. “A.G. Schneiderman Announces Arrest of Construction Firm Owner for Underpaying Workers and Laundering Stolen Wages” (press release). February 4, 2013.

New York Attorney General’s Office (N.Y. AG). 2014a. “A.G. Schneiderman Obtains Jail Sentence for Owner of Movie Theater Cleaning Company That Underpaid Workers” (press release). January 17, 2014.

New York Attorney General’s Office (N.Y. AG). 2014b. “A.G. Schneiderman Announces Conviction of Construction Boss for Underpaying Workers on Project at JFK Airport” (press release). November 20, 2014.

New York Attorney General’s Office (N.Y. AG). 2014c. A.G. Schneiderman Secures Criminal Conviction of Employers in Child Labor Case, (press release). December 11, 2014.

New York Attorney General’s Office (N.Y. AG) 2015a. “A.G. Schneiderman and U.S. Department of Labor Announce Criminal Charges against, and Civil Settlement with, Papa John’s Franchisee for Wage Theft” (press release). July 15, 2015.

New York Attorney General’s Office (N.Y. AG) 2015b. “A.G. Schneiderman and U.S. Department of Labor Announce Jail Time For Bronx Papa John’s Franchisee Convicted of Wage Theft” (press release). November 16, 2015.

New York Attorney General’s Office (N.Y. AG). 2018. “A.G. Schneiderman Announces Guilty Pleas and Convictions of Three Queens Construction Companies for Failing to Pay 150 Workers Over $370,000 in Wages” (press release). February 7, 2018.

New York Attorney General’s Office (N.Y. AG). 2019. “Attorney General James Announces Sentencing of Cortland County Farmer Following Death of 14-Year-Old Employee” (press release). January 19, 2019.

New York Attorney General’s Office (N.Y. AG). n.d. “Criminal Enforcement of Labor Laws: Prevailing Wage (webpage). Accessed April 7, 2021.

New York County District Attorney (NY Cty. DA). 2017. “DA Vance Invests $11.8 Million in Services for Historically Underserved Victims of Crime” (press release). April 26, 2017.

New York County District Attorney (NY Cty. DA). 2018. “DA Vance, Partners Announce Assault Charges in Harlem Construction Site Incident; Warn Industry About Increasingly Popular ‘Mini Cranes’” (press release). November 8, 2018.

New York State Division of Criminal Justice Services (NYS DCJS). 2015. “Governor Cuomo Announces More than $14 Million to Help Prosecutors Target Tax and Public Benefits Fraud.” (press release). February 19, 2015.

New York State Division of Criminal Justice Services (NYS DCJS). 2020. Communications between author and agency. Notes on file with author. November 27, 2020.

New York State Division of Criminal Justice Services (NYS DCJS). 2021. “Crimes Against Revenue Program (CARP)” (website). Accessed April 10, 2021.

New York State Legislature. (NYS Leg.) 2019. Accepting and Appropriating 100% Grant Funds Received from the New York State Division of Criminal Justice Services to the Suffolk Country District Attorney’s Office, Under the Crimes Against Revenue Program. Resolution No. 1018 -2019. November 6, 2019

Oravetz, Janet. 2019. “Boulder Man Accused of Failing to Pay Workers, Threatening to Report them to ICE.” 9News. November 15, 2019.

Ormiston, Russell, Dale Belman, and Mark Erlich. 2020. An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry. Institute for Construction Economics Research, January 2020. page 3.

Pace, Eli. 2019a. “Employer Charged with Manslaughter in Worker’s Death at Colorado Construction Site.” Post Independent, August 23, 2019.

Pace, Eli. 2019b. “Avon-Based Employer Charged with Manslaughter in 2018 Worker’s Death at Granby Construction Site.” Greeley Tribune, August 23, 2019.

Pennsylvania Attorney General’s Office (Pa. AG). 2019. “Central Pennsylvania Contractor Charged for Theft of Workers’ Wages and Benefits” (press release). September 27, 2019.

Pennsylvania Attorney General’s Office (Pa. AG). 2021. “AG Shapiro Announces Multi-Million Dollar Theft Charges Against State College Contractor” (press release). April 8, 2021.

Philadelphia District Attorney’s Office (Phil. DA). 2021. “DAO Worker Protection Unit Charges Plumbing Contractor for Misappropriating Union Fees” (press release). January 25, 2021

Piore, Michael J. and Schrank, Andrew. 2018. Root Cause Regulation: Protecting Work and Workers in the Twenty-First Century, Harvard University Press, 2018.

Queens District Attorney (Queens DA). 2020. “Queens Contractor and His Business Plead Guilty to Violating Prevailing Wage Labor Laws and Stealing More than $1.5 Million from Workers” (press release). September 22, 2020.

Ransom, Jan. 2020a. “Harvey Weinstein is Found Guilty of Sex Crimes in #MeToo Watershed,” New York Times, February 24, 2020.

Ransom, Jan. 2020b. “Harvey Weinstein’s Stunning Downfall: 23 Years in Prison.” New York Times, March 11, 2020.

Ramirez, Mary. 2018. “How To Prosecute Wage Theft: A Training for Prosecutors & Investigators” (event invitation). December 13, 2018.

Restaurant Opportunities Centers (ROC) United and Forward Together. 2014. The Glass Floor: Sexual Harassment in the Restaurant Industry. October 6, 2014.

Reyes, Juliana Feliciano. 2019. “Philly DA’s Office Launches a Unit to Prosecute Employers for Crimes Against Workers.” Philadelphia Inquirer, October 8, 2019

Reyes, Juliana Feliciano. 2021. “Delaware County DA is Charging a Contractor with Misclassifying Its Workers.” Philadelphia Inquirer, January 11, 2021.

Rhinehart, Lynn, and Celine McNicholas. 2021. Shortchanged—Weak Anti-Retaliation Provisions in the National Labor Relations Act Cost Workers Billions. Economic Policy Institute, April 2021.

Rhinehart, Lynn, Lane Windham, and Lawrence Mishel. 2020. Explaining the Erosion of Private-Sector Unions: How Corporate Practices and Legal Changes Have Undercut the Ability of Workers to Organize and Bargain. Unequal Power, Economic Policy Institute, November 2020.

Rhode Island Attorney General’s Office (R.I. AG) 2019. “Lincoln Man Arrested on Wage Theft Charges After Joint Investigation by RIAG, RISP, and DLT” (press release). September 23, 2019.

Rhode Island Attorney General’s Office (R.I. AG). 2020. “Attorney General Charges Former Owner of Cleaning Company Wage Theft and Workers’ Compensation Insurance Fraud” (press release) July 9, 2020.

Rhode Island Attorney General’s Office (R.I. AG). 2021. “AG Neronha Seeks to Strengthen Penalties for Wage Theft, Labor Violations” (press release). February 24, 2021.

Romero, Dennis. 2010. “Brothers Benny and Nissan Pirian Get One Year in Jail for Underpaying, Overworking Car Wash Employees.” LA Weekly. August 17, 2010.

Rosado Marzán, César F. 2020. “Wage Theft as Crime: An Institutional View.” Journal of Law & Society 300 (2020), U Iowa Legal Studies Research Paper No. 2021-02. Accessed April 7, 2021.

Rosenthal, Ann. Death by Inequality: How Workers’ Lack of Power Harms Their Health and Safety. Economic Policy Institute, April 2021.

Rushton, Geoff. 2021. “State College Contractor Gets Jail Time After Pleading Guilty to Wage Theft Charges.” StateCollege.com. April 26, 2021.

Ryan, Kate. 2017. “Collateral Consequences: Brooklyn DA Seeks to Protect Immigrant Defendants from Deportation,” WNYC News. December 12, 2017.

San Diego District Attorney’s Office (S.D. DA). 2021. “New Workplace Justice Unit to Protect Workers’ Rights” (press release). February 22, 2021.

San Francisco District Attorney’s Office (S.F. DA). 2020. “District Attorney Launches New Economic Crimes Unit to Protect Workers’ Rights” (press release). April 21, 2020.

Schrank, Andrew, and Jessica Garrick. 2013. Mexican Immigrants and Wage Theft in New Mexico. Department of Sociology, University of New Mexico and Somos Un Pueblo Unido, August, 2013.

Sen, Aditi. 2016. By a Thousand Cuts: The Complex Face of Wage Theft in New York. Center for Popular Democracy, November 2016.

Sernoffsky, Evan. 2018. “2 Charged in Death of Forklift Operator at San Francisco Lumber Company.” SFGate, May 7, 2018.

Shaw, Elyse, Ariane Hegewisch, and Cynthia Hess. 2018. “Sexual Harassment and Assault at Work: Understanding the Costs,” Institute for Women’s Policy Research Briefing Paper. October 2018.

Shields, Michael. 2019. Wage Theft: A Quiet Crisis in Ohio. Policy Matters Ohio, April 10, 2019.

Shierholz, Heidi. 2020. “EPI Comments on Independent Contractor Status Under the Fair Labor Standards Act.” Comments submitted on behalf of the Economic Policy Institute to Amy DeBisschop, Division of Regulations, Legislation, and Interpretation, Wage and Hour Division, U.S. Department of Labor, October 2020.

Shierholz, Heidi, and Margaret Poydock. 2021. “EPI Comments on the Rescission of the Joint Employer Rule.” Public comments submitted to Amy DeBisschop, Division of Regulations, Legislation, and Interpretation, Wage and Hour Division, U.S. Department of Labor, April 12, 2021.

Smith, Annie. 2021. “The Underprosecution of Labor Trafficking” South Carolina Law Review, 72 S.C. L. Rev. 477, January 2021.

Steinzor, Rena. 2015. Why Not Jail? Industrial Catastrophes, Corporate Malfeasance, and Government Inaction. Cambridge University Press, 2015.

Stiffler, Chris. 2019. Stolen Labor: Wage Theft in Colorado. Colorado Fiscal Institute, September 2019.

Stone, Katherine V.W. and Alexander J.S. Colvin. 2015. The Arbitration Epidemic: Mandatory Arbitration Deprives Workers and Consumers of Their Rights. Economic Policy Institute, December 2015.

Suffolk County District Attorney’s Office (Suffolk Cty. DA). 2020a. “Gas Station Owner Arrested for Allegedly Exploiting Workers, Threatening Retaliation Against Immigrant Employees” (press release). December 15, 2020.

Suffolk County District Attorney’s Office (Suffolk Cty. DA). 2020b. “DA Sini & DA’s Labor Advisory Board Announce Arrest of 8 Individuals and 9 Corporations in Labor Crime Crackdown” (press release). March 9, 2020.

Svoboda, Haleigh. 2011. “Enforcement Is Next Task for Law on Wage Theft.” New York Times, October 29, 2011.

Travis County Sheriff’s Office (Travis CSO). 2021. “Online Citizen Report Filing” (website). Accessed April 10, 2021.

U.S. Department of Homeland Security (U.S. DHS). 2019. U Visa Law Enforcement Resource Guide, 2019.

U.S. Department of Justice (U.S. DOJ). 2020. “Justice Department Awards Nearly $101 Million to Combat Human Trafficking.” (press release). September 21, 2020.

U.S. Department of Justice (U.S. DOJ). N.d. “Partnerships to Address Labor Trafficking” (web page, Community Oriented Policing Services). Accessed May 10, 2021.

U.S. Department of Labor, Occupational Safety and Health Administration (U.S. DOL-OSHA). n.d.a. (U.S. federal government website). Accessed April 6, 2021.

U.S. Department of Labor, Occupational Safety and Health Administration (U.S. DOL-OSHA). n.d.b “State Plans” (web page). Accessed April 6, 2021.

U.S. Department of State. 2021. “Understanding Human Trafficking” (fact sheet). January 2021.

United States Government Accountability Office (U.S. GAO). 2020. “Workplace Sexual Harassment: Experts Suggest Expanding Data Collection to Improve Understanding of Prevalence and Costs,” GAO-20-564 (September 2020).

Villanueva, Alex. 2021. “Wage Theft Thrives in Secrecy. Here’s How My Office Is Working to Stop It.” Washington Post, March 23, 2021.

Vosseller, Bob. 2019. “Ocean County Contractor: 3 Years for Fraud.” Jersey Shore Online, September 6, 2019.

Warsmith, Stephanie. 2019. “Akron Construction Company Owner Sentenced to 3 Years in Prison for Employee’s Death in a Fall.” Akron Beacon Journal, updated September 9, 2019.

Washington State Attorney General’s Office (Wash. AG). 2018. “Contractor Sentenced for Criminal Wage Theft, False Reporting of Workers’ Comp Payments.” July 27, 2018.

Weil, David, and Amanda Pyles. 2007. “Exploring the Complaints and Compliance Gap Under U.S. Workplace Policies”, in LERA (ed.): Proceedings of the 59th Annual Meeting, January 5–7 2007, Chicago, Illinois. Labor and Employment Relations Association Series. Champaign, IL, LERA, pp. 168–181.

Weil, David. 2010. “Improving Workplace Conditions Through Strategic Enforcement: A Report to the Wage and Hour Division.” May 2010. Available at https://www.dol.gov/sites/dolgov/files/WHD/legacy/files/strategicEnforcement.pdf.

Weil, David. 2014. The Fissured Workplace: Why Work Became So Bad for So Many and What Can Be Done to Improve It. Cambridge, Mass.: Harvard University Press, 2014.

Westchester County District Attorney. 2021. “DA Rocah Announces Multilingual Hotline: 914-995-TIPS.” (press release). March 16, 2021.

Xu, Lisa, and Mark Erlich. 2019. Economic Consequences of Misclassification in the State of Washington. Harvard Law School Labor and Worklife Program, December 2019.

Yakin, Heather. 2019. “Crackdown on Labor Crimes in Region Intensifies.” Times Herald-Record, October 17, 2019.

Amy DeBisschop
Division of Regulations, Legislation, and Interpretation
Wage and Hour Division
U.S. Department of Labor
Room S-3502
200 Constitution Avenue NW
Washington, DC 20210

Re: Comments on Independent Contractor Status Under the Fair Labor Standards Act (RIN 1235-AA34)

Dear Ms. DeBisschop:

The Economic Policy Institute (EPI) is a nonprofit, nonpartisan think tank created in 1986 to include the needs of low- and middle-income workers in economic policy discussions. EPI conducts research and analysis on the economic status of working America, proposes public policies that protect and improve the economic conditions of low- and middle-income workers, and assesses policies with respect to how well they further those goals. EPI submits these comments on the Department of Labor’s (Department/DOL) Notice of Proposed Rulemaking regarding the standard for determining who is a covered employee and who is an independent contractor under the Fair Labor Standards Act (FLSA).1

The Department’s proposed interpretation is contrary to law because it ignores the plain language of the FLSA’s definition of “employ,” which “includes to suffer or permit to work,”2 and ignores U.S. Supreme Court and federal circuit court authority interpreting the Act. The Department is attempting to impermissibly narrow this very broad definition of “employ” by proposing a restrictive interpretation of the long-accepted “economic realities” test.3 This five-part test has always been interpreted by the Supreme Court in its totality, not weighing any one factor more than another. But now, DOL proposes amending it in a fashion that places undue weight on two factors and then narrows those two factors further—individual control over the work and opportunity for profit or loss. As a result, the Department is proposing to constrict the FLSA’s broad coverage in a way that will undermine its statutory intent. This proposal makes it easier for employers to classify workers as independent contractors, and if finalized, it would lead to an increase in the share of the workforce that are independent contractors.

In the proposed rule, the Department egregiously fails to estimate the transfers between employers, workers, and the social insurance system that would occur if this proposal were finalized. The requirements that agencies must follow as a part of the rulemaking process are very clear, and among them is the requirement that agencies must assess all quantifiable costs and benefits “to the fullest extent that these can be usefully estimated.”4 There is no question that DOL could have produced estimates; in what follows, I show that it is straightforward to produce estimates using data researchers routinely use and taking a methodological approach that is in the spirit of estimates the Department of Labor undertakes on a regular basis. One plausible explanation for why DOL left out the required estimate is that any good-faith estimate would have shown this rule would result in a substantial transfer from workers and the social insurance system to employers.

In this comment I will estimate these transfers. The basic structure of this analysis is to estimate (1) the change in the value of a job to a worker if the worker is classified as an independent contractor instead of an employee as a result of this rule, and (2) the change in payments to social insurance funds if a worker is classified as an independent contractor instead of an employee. Multiplying these figures by the estimated number of workers who would shift to independent contractor status if this rule were finalized will yield the aggregate impact of the rule on workers and on social insurance system coffers.

To estimate (1) and (2) above, I need to determine the earnings of workers whose classification would change from payroll employee to independent contractor as a result of this rule. The workers most likely to be affected by this rule are workers in lower-wage occupations in labor-intensive industries, such as delivery workers, transportation workers like taxi drivers and some truckers, logistics workers including warehouse workers, home care workers, housecleaners, construction laborers and carpenters, agricultural workers, janitors, call center workers, and staffing agency workers in lower-paid placements. Using Occupational Employment Statistics data from the Bureau of Labor Statistics where available, the median annual earnings for these jobs ranges from $24,850 for maids and housekeeping cleaners to $48,330 for carpenters. Given that I do not have a way to determine the precise earnings of those workers whose classification would change if the rule were finalized, I will assume that a worker whose status would change as a result of this rule is at the bottom of this range, $24,850, to be extremely conservative.

Take-home earnings represent only a portion of labor costs to an employer and the value of a job to an employee. In order to estimate the total compensation of a worker whose classification would change as a result of the rule, I use data from the Employer Costs for Employee Compensation (ECEC) program of the Bureau of Labor Statistics. The ECEC breaks down total worker compensation into regular pay (wages and salaries), supplemental pay (overtime, premium pay, shift differentials, and nonproduction bonuses), paid leave, insurance (health, life, long- and short-term disability), retirement benefits, and legally required benefits (Social Security, Medicare, federal and state unemployment insurance, and workers’ compensation). The ECEC has these compensation profiles by occupation and industry, but there obviously isn’t a compensation profile for the set of workers whose classification would change as a result of this rule. To be extremely conservative, I assume that the compensation profile of workers whose status would change as a result of this rule is the same as the compensation profile of workers in the occupation/industry breakdown with the lowest level of total compensation, service occupations in private service-providing industries. The cost per hour worked for various components of total compensation for service occupations in private service-providing industries is given in Table 1.5 Table 1 also calculates the ratio of each of these components to pay (wages, salaries, and supplemental pay), which I will use to estimate various compensation components of workers whose classification status would change as a result of the rule.

Table 1 shows that 2.6% of these workers’ pay is composed of supplemental pay (mostly overtime and other premium pay like holiday pay, but also shift differentials and nonproduction bonuses). Recall that I am conservatively assuming workers whose classification status would change as a result of the rule earn $24,850 annually. That means that $24,202 is earned as regular pay, and $648 was earned as supplemental pay. Table 1 also shows that paid leave benefits are equivalent to 5.5% of pay, which is equal to $1,357 for a worker who earns $24,850. Further, Table 1 shows that insurance (health, life, long- and short-term disability) and retirement benefits are equivalent to 11.2% of pay, which is equal to $2,771 for a worker who earns $24,850.

Table 1

The employee will have 7.65% deducted for Social Security and Medicare, which is equal to $1,901 for a worker who earns $24,850.6 The employer is also required to pay $1,901 to cover the employer share of Social Security and Medicare. Table 1 also provides the cost to the employer of other legally required benefits—unemployment insurance and workers’ compensation—showing that these costs are equivalent to 1.0% and 2.7% of pay, respectively, which is equal to $259 and $677, respectively, for a worker who earns $24,850.

These figures are in the first column in Table 2. Table 2 shows that the total value to an employee of a job where the pay is $24,850 is estimated to be $27,077, after adding in the value of paid leave, insurance, and retirement benefits, and subtracting off FICA (Social Security and Medicare) taxes.7 The first column in Table 2 also shows that the total payment to social insurance funds (Social Security, Medicare, unemployment insurance, and workers’ compensation) associated with this payroll job is $4,739 (summing the employer contribution to social insurance and the worker share of Social Security and Medicare).

Table 2

The second column of Table 2 estimates how these quantities would be different if the worker were classified as an independent contractor instead of as an employee. The first question that arises is what pay the worker would receive as an independent contractor. The Department states that that “in a competitive labor market, any reduction in benefits and increase in taxes is likely to be offset by higher base earnings—referred to as an ‘earnings premium,’” which would mean that “in theory, companies would likely have to pay more per hour to independent contractors than to employees because independent contractors generally do not receive employer-provided benefits and have higher tax liabilities,” and “any tax-related transfers from employers to workers are likely to be offset by higher wages employers pay to ensure workers’ take-home pay remains the same.”8 However, the Department goes on to note that “this expected wage premium may not always be observable at a statistically significant level” and in fact, the Department’s own analysis of 2017 Contingent Worker Supplement (CWS) data “did not show a statistically significant difference” between the wages of employees and independent contractors with the same demographic characteristics in the same occupation.

This is not surprising when considering the fact that the theory that businesses will not be able to pay less in total compensation to workers if their status shifts from employee to independent contractor—that their base pay will rise to make up for a reduction in benefits—is based on the assumption of perfectly competitive labor markets. There is broad and growing evidence that perfect competition is rare, and that most labor markets do not function competitively—particularly low-wage labor markets like those under consideration here, in which workers are more likely to lack the power to bargain for higher wages to compensate for their loss of benefits and increase in taxes when they become independent contractors.9,10 Further, very-low-wage employees whose wages are elevated by the minimum wage could easily see their wages drop when, as independent contractors, they no longer legally must be paid the minimum wage.

After review by the Office of Information and Regulatory Affairs (OIRA) at the White House, the following sentence was removed from DOL’s original economic analysis: “The Department anticipates a positive wage effect due to the expected increase in labor force activity, but did not attempt to quantify estimates of changes in earnings.” That removal following OIRA’s review is no surprise, given that a positive wage effect of the proposed rule can’t be supported by the evidence.

Given these findings, I assume that the low-earnings workers under consideration here get no wage premium when they shift from being an employee to being an independent contractor, and instead receive the same regular pay that they received when they were an employee. I further assume that this worker will receive no supplemental pay (such as overtime, holiday premium pay, shift differentials, or nonproduction bonuses), or paid leave. Overtime is not legally mandated for workers who are not payroll employees, and it and other kinds of supplemental pay or paid leave are unlikely to be received by an independent contractor. I also assume that a worker who changes status as a result of this proposed rule would receive no insurance or retirement benefits from their contract job.11

An additional cost that people who switch from being payroll employees to independent contractors have to face is greater paperwork costs. For example, the IRS estimates that business taxpayers spend 13 more hours than nonbusiness taxpayers doing their taxes.12 If we conservatively assume that independent contractors spend 30 minutes per week on other (nontax) paperwork costs that they wouldn’t have to spend if they were a payroll employee, that, plus the additional 13 hours spent on taxes, is an additional 39 hours of paperwork per year. This is equivalent to 1.8% of pay, or $445 annually for an independent contractor who earns $24,202 in regular pay annually. Further, independent contractors are likely to need software for doing their bookkeeping and taxes. A commonly used program for things like tracking expenses is FreshBooks, which costs $162 per year for the cheapest option.13 A commonly used tax preparation program for independent contractors, TurboTax Self-Employed, costs $120 plus $50 per state for the cheapest option, whereas an individual who is a payroll employee with a simple tax return can prepare their return using TurboTax at no cost.14 For an independent contractor who earns $24,202, this sums to a new annual paperwork cost of $777.15

Independent contractors are required to pay both the employee and the employer portions of taxes for Social Security and Medicare, which is 15.3% of pay—though they are able to deduct the employer-equivalent portion of the tax, along with business expenses. Taking these factors into account means an independent contractor who earns $24,202 and has $777 in business expenses will pay $3,310 in Social Security and Medicare taxes.16 Furthermore, when a worker is reclassified from an employee to an independent contractor, the former employer no longer pays for unemployment insurance or workers’ compensation. As Table 2 shows, the value of the job to the worker drops from $27,077 when they are an employee to $20,114 when they are an independent contractor, a drop of $6,963, or 25.7%.17 Social insurance funds get $1,429 less, or 30.2%, when the worker’s status changes.

What might be being left out in this analysis? The Department focuses on “flexibility and satisfaction” as important nonpecuniary attribute that workers may trade income to receive.18 However, it is difficult to imagine that there are a meaningful number of workers who would get more satisfaction from doing the same job for substantially less compensation as an independent contractor than for substantially more compensation as a payroll employee. Many workers indeed may value flexibility, but notably, employers are able to provide a huge amount of flexibility to payroll employees if they choose to; the “inherent” tradeoff between flexibility and payroll employment is greatly exaggerated. Workers also highly value other factors, like income stability, which are much less prevalent among independent contractors and are not taken into account here.

How will the share of workers who are payroll employees and the share of workers who are independent contractors change as a result of this rule? To begin to answer that question, we need to know how many independent contractors there currently are. There is a great deal of uncertainty around this number (the Department notes that “there are a variety of estimates of the number of independent contractors and these span a wide range based on methodologies and how the population is defined”).19 The 2017 Contingent Worker Supplement (CWS) estimated that there were 10.6 million workers who are independent contractors in their main job. This number, however, drastically underestimates the total number of independent contractors by not including workers who do independent contracting on the side, in addition to their payroll job. The Department makes a correction for this issue and estimates that there are 18.9 million individuals working as contractors at a given time. For the sake of my calculations, however, I will limit my analysis to the 10.6 million workers the CWS finds are independent contractors in their main job, since workers who do independent contracting as a side job may, in many cases, make far less than the $24,202 I am assuming workers whose status changes as a result of this rule earn. It should be noted that this means I am leaving out many millions of independent contractors and my estimates will, as a result, be extremely conservative in this way, in addition to other ways my estimate is conservative (some mentioned above and more to come below).

It should be noted that the Department emphasizes the difference between workers whose classifications would change to independent contractor status as a result of the rule and workers who are newly hired as independent contractors. Given the high degree of churn in the labor market—particularly in the low-wage labor market—the distinction between newly hired independent contractors and workers whose status changes is not a relevant distinction, and the Department should drop this emphasis—and should instead focus on the change in the share of the share of the workforce that is made up of independent contractors. Throughout this comment I have, for simplicity, referred to workers shifting from being payroll employees to independent contractors, but that should be understood to represent an increase in the share of the workforce that is made up of independent contractors. It is further worth noting that the Department provided no evidence that this rule would increase overall employment levels, merely citing a McKinsey Global Institute study that finds that 15% of those not working are interested in becoming an independent contractor, but providing no evidence that this rule would likely lead to those workers joining the workforce.20 In fact, by reducing the income of lower-income workers who have a high marginal propensity to consume, this rule would reduce aggregate demand in the economy and likely decrease the overall number of jobs.

How much will independent contracting increase as a result of this rule? The DOL proposal would potentially allow companies to legally argue that workers who are now misclassified as independent contractors or who are working “off the books” would be legitimately classified as independent contractors under the narrow terms of the proposal. As such, one approach would be to use the percentage of workers misclassified or working off the books under current law to estimate the number of workers who could be reclassified as independent contractors under the proposed rule. However, due to severe data constraints, estimates of the share of workers who are misclassified as independent contractors or working off the books are limited. A recent paper estimates that between 12.4% and 20.5% of workers in the construction industry are either misclassified as independent contractors or working off the books.21 Conservatively assuming that the bottom of this range applies more broadly to the lowest-paid quartile of the U.S. labor market, that is 4.9 million low wage workers who may be affected by this rule.22 Of course, these are workers who are already not getting the benefit of being a payroll employee, so the cost impacts described above would not apply without a substantial increase in enforcement. However, this exercise does provide a broad sense of the potential scope of workers affected. Further, even these workers lose something of value under this rule given the current enforcement regime, namely the legal right to the wages and benefits they would receive if they were properly classified. I do not attempt to quantify this effect.

To be exceedingly conservative, I will simply assume that there would be an increase as a result of this rule of 5% in the number of workers who are independent contractors in their main job.23 This translates into an increase of just 530,000 workers who are independent contractors at their main job, given the CWS estimate of 10.6 million workers who are independent contractors in their main job. Multiplying that by my conservative estimate that these workers would lose $6,963 per year yields an aggregate loss to workers of at least $3.7 billion annually. This loss to workers is composed of at least $400 million in new annual paperwork costs, and a transfer to employers of at least $3.3 billion in the form of reduced compensation. Further, social insurance funds would lose at least $750 million annually in the form of reduced employer contributions, meaning this rule also results in a transfer of at least $750 million annually from social insurance funds to employers.24 Table 3 provides the final estimates.

Table 3

It is important to note that these estimates are lower bounds for many reasons. In particular, I assumed the workers whose status would change are very low-wage workers with a very low level of benefits (meaning that to the extent the status of workers with higher wages or greater benefits changes, I have underestimated the impacts), I only took into account workers who are independent contractors in their main job (meaning that I have left out the impact of any status shift for workers who would be independent contractors for supplemental income, not their primary job), and I assumed that there would only be a 5% increase in the number of workers who are independent contractors in their main job as a result of this rule (meaning that to the extent the status of more workers changes, I have underestimated the impact). Taking into account these factors, the true impact could be many times my estimates. For example, if the true number of workers affected is equivalent to just half of the 4.9 million workers currently estimated above to be misclassified or working off the books, workers would lose on the order of $17 billion annually as a result of this rule.

EPI strongly opposes the Department of Labor’s proposed rulemaking regarding independent contractor status under the Fair Labor Standards Act, because this rule will cost workers billions of dollars annually and will cost the social insurance system hundreds of millions of dollars annually. Further, due to things like occupational segregation by race, discrimination, and other labor market disparities rooted in structural racism, Black and Latinx workers are more likely to work in the occupations affected by this rule. As a result, this rule would exacerbate existing racial disparities. It must not be finalized.

Thank you for the opportunity to submit comments. Please do not hesitate to contact me at hshierholz@epi.org if you have questions.

Sincerely,

Heidi Shierholz
Senior Economist and Director of Policy
Economic Policy Institute
Washington, DC

Endnotes

1. Independent Contractor Status Under the Fair Labor Standards Act, 85 Fed. Reg. 60600–60639 (September 25, 2020).

2. 29 U.S.C. §203(g)

3. See U.S. v. Silk, 331 U.S. 704 (1947).

4. Maeve P. Carey, Cost-Benefit and Other Analysis Requirements in the Rulemaking Process, Congressional Research Service, December 9, 2014.

5. For service workers in service industries, legally required benefits are not disaggregated into subcomponents (Social Security and Medicare, unemployment insurance, and workers’ compensation). This disaggregation exists for service workers in all industries, however, so I simply apply the share each of these subcomponents constitutes of legally required benefits for services workers in all industries to the total value of legally required benefits for service workers in service industries, to estimate the subcomponents for service workers in services industries.

6. The 7.65% used here is slightly smaller than the 8.3% found in the ECEC data presented in Table 1. I use 7.65% because it is the official tax rate.

7. I should note that “total value” is defined as a relatively near-term concept here, in the sense that paying Social Security taxes would likely increase a worker’s Social Security benefits in retirement, so those taxes are not a long-run negative.

8. 85 Fed. Reg. at 60626-60627, 60628

9. Alan Manning, Monopsony in Motion: Imperfect Competition in Labor Markets (Princeton, N.J.: Princeton University Press, 2003); Anna Sokolova and Todd Sorensen, Monopsony in Labor Markets: A Meta-Analysis, Washington Center for Equitable Growth, February 2020; Arindrajit Dube, Jeff Jacobs, Suresh Naidu, and Siddharth Suri, “Monopsony in Online Labor Markets,” American Economic Review: Insights 2, no. 1 (March 2020): 33-46, https://www.aeaweb.org/articles?id=10.1257/aeri.20180150.

10. It is worth noting that if the labor market were to get to full employment and stay there for an extended period, the economic leverage that this would provide workers might mean that even these workers could recoup some of the value of lost benefits and increased taxes in the form of higher wages. However, the instances of full employment in the labor market are the exception, not the norm.

11. An individual with total household earnings of $24,850 and a household size of three or more (for example, one individual with two children) would likely be eligible for Medicaid. I do not attempt to quantify this effect but it should be noted that to the extent workers who change status as a result of this rule are able to take up Medicaid, the transfer to employers related to health care that would result from this rule would be transfers from the social insurance system to employers, not from the worker to employers.

12. See page 101 in Internal Revenue Service, 1040 and 1040-SB Instructions: Including the instructions for Schedules 1 through 3, Cat. No. 24811V, January 2020.

13. FreshBooks, “FreshBooks Pricing, Lite, Plus and Premium Packages ” (web page), accessed October 26, 2020.

14. TurboTax (website), accessed on October 26, 2020.

15. $777 = $445 + $162 + $120 + $50.

16. $3,310 = ($24,202 – $777)*(1-0.0765)*0.153.

17. It is worth noting that this is a similar effect to what other studies find, for example, a paper titled “Pay, Passengers and Profits: Effects of Employee Status for California TNC Drivers” by Michael Reich of the University of California, Berkeley finds that employee status for Uber and Lyft drivers would increase total driver compensation by about 30 percent.

18. 85 Fed. Reg. at 60628.

19. 85 Fed. Reg. at 60623.

20. 85 Fed. Reg. at 60627.

21. Russell Ormiston, Dale Belman, and Mark Erlich, An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry, Institute for Construction Economics Research, January 2020.

22. Data from the Current Population Survey from the Bureau of Labor Statistics find that there were 157.5 million workers in the US in 2019; 4.9 million = 157.5 million * .25 * .124.

23. A 5% increase is a conservative assumption given the Department is proposing to amend the five-part economic realities test—which has always been interpreted by the Supreme Court in its totality, not weighing any one factor more than another—in a way that will place undue weight on two factors and then narrows those two factors further, making it more likely that workers will be classified as independent contractors and as a result likely leading to a substantial increase in the number of independent contractors.

24. Some might argue that social insurance funds wouldn’t be hurt by not having employers pay into unemployment insurance and workers’ compensation because independent contractors aren’t eligible for those benefits. However, independent contractors are currently eligible for unemployment insurance benefits through Pandemic Unemployment Assistance, and it is reasonable to assume that this will occur in future recessions, as well. Further, low-paid independent contractors who lose their contracts and are without work, or get hurt on the job, would be likely to need to depend on safety net programs to survive, so the social insurance system as a whole would still be depleted.

What this report finds: Corporate boards running America’s largest public firms are giving top executives outsize compensation packages that have grown much faster than the stock market and the pay of typical workers, college graduates, and even the top 0.1%. In 2019, a CEO at one of the top 350 firms in the U.S. was paid $21.3 million on average (using a “realized” measure of CEO pay that counts stock awards when vested and stock options when cashed in rather than when granted). This 14% increase from 2018 occurred because of rapid growth in vested stock awards and exercised stock options tied to stock market growth. Using a different “granted” measure of CEO pay, average top CEO compensation was $14.5 million in 2019. In 2019, the ratio of CEO-to-typical-worker compensation was 320-to-1 under the realized measure of CEO pay; that is up from 293-to-1 in 2018 and a big increase from 21-to-1 in 1965 and 61-to-1 in 1989. CEOs are even making a lot more—about six times as much—as other very high earners (wage earners in the top 0.1%). From 1978 to 2019, CEO pay based on realized compensation grew by 1,167%, far outstripping S&P stock market growth (741%) and top 0.1% earnings growth (which was 337% between 1978 and 2018, the latest data year available). In contrast, compensation of the typical worker grew by just 13.7% from 1978 to 2019.

Why it matters: Exorbitant CEO pay is a major contributor to rising inequality that we could safely do away with. CEOs are getting more because of their power to set pay—and because so much of their pay (about three-fourths) is stock-related, not because they are increasing productivity or possess specific, high-demand skills. This escalation of CEO compensation, and of executive compensation more generally, has fueled the growth of top 1.0% and top 0.1% incomes, leaving less of the fruits of economic growth for ordinary workers and widening the gap between very high earners and the bottom 90%. The economy would suffer no harm if CEOs were paid less (or were taxed more).

How we can solve the problem: We need to enact policy solutions that would both reduce incentives for CEOs to extract economic concessions and limit their ability to do so. Such policies could include reinstating higher marginal income tax rates at the very top; setting corporate tax rates higher for firms that have higher ratios of CEO-to-worker compensation; establishing a luxury tax on compensation such that for every dollar in compensation over a set cap, a firm must pay a dollar in taxes; reforming corporate governance to give other stakeholders better tools to exercise countervailing power against CEOs’ pay demands; and allowing greater use of “say on pay,” which allows a firm’s shareholders to vote on top executives’ compensation.

Introduction and key findings

Chief executive officers (CEOs) of the largest firms in the U.S. earn far more today than they did in the mid-1990s and many times what they earned in the 1960s or late 1970s. They also earn far more than the typical worker, and their pay—which relies heavily on stock-related compensation— has grown much more rapidly than typical worker pay. Importantly, rising CEO pay does not reflect rising value of skills, but rather CEOs’ use of their power to set their own pay. And this growing earning power at the top has been driving the growth of inequality in our country.

About the CEO pay series and this report

This report is part of an ongoing series of annual reports monitoring trends in CEO compensation. In this report, we examine current trends to determine how CEOs of the top 350 largest U.S. firms (by sales) are faring compared with typical workers through 2019. We also compare top CEO pay with earnings of workers in the top 0.1% (through 2018), and look at the relationship between CEO pay and the stock market.

For most of our analyses, we use two measures of CEO compensation, one based on compensation as “realized” and the other based on compensation as “granted.” Both measures include the same measures of salary, bonuses, and long-term incentive payouts. The difference is how each measure treats stock awards and stock options, major components of CEO compensation that change value from when they are first provided, or granted, to when they are realized. The realized measure of compensation includes the value of stock options as realized (i.e., exercised), capturing the change from when the options were granted to when the CEO invokes the options, usually after the stock price has risen and the options values have increased. The realized compensation measure also values stock awards at their value when vested (usually three years after being granted), capturing any change in the stock price as well as additional stock awards provided as part of a performance award. The granted measure of compensation values stock options and restricted stock awards by their “fair value” when granted.

We have changed our definition of CEO compensation in the realized measure from that employed in earlier reports. Previous reports used the value of stock awards as granted in both the realized and granted compensation measures, so that the measures differed in only their treatment of stock options. As noted in our previous report (Mishel and Wolfe 2019) the increased importance of stock awards in executive pay and the increased divergence between the value of stock awards when granted (measured as “fair value” when granted) versus when vested means that excluding the realized gains from stock awards increasingly understates total CEO compensation. We therefore have incorporated a realized measure of stock awards along with the realized measure of stock options in our realized compensation metric. This first metric can be compared with the second metric, compensation granted, whose measurement is the same as in prior reports.

CEO compensation growth in 2019 and recent years

Both measures of CEO compensation grew strongly in 2019. Realized CEO compensation grew to $21.3 million in 2019, which was $2.6 million or 14.0% higher than in 2018. The growth in realized CEO compensation was driven by a 19.5% growth in vested stock awards and a 17.5% growth in exercised stock options. Granted CEO compensation grew $1.1 million or by 8.6% to $14.5 million in 2019.

Long-term trends

Realized CEO compensation grew 105.1% from 2009 to 2019, the period capturing the recovery from the Great Recession; in that period granted CEO compensation grew 35.7%. In contrast, typical workers in these large firms saw their average annual compensation grow by just 7.6% over the last 10 years. (Typical workers in these firms are production and nonsupervisory workers in the industries that the top 350 firms operate in. Their compensation measure includes wages and benefits.)

CEO compensation attained its peak in 2000, at the height of the late 1990s tech stock bubble, at $21.9 million (in 2019 dollars) based on either measure. That same year the CEO-to-typical-worker compensation ratio was 366-to-1 (realized) or 386-to-1 (granted).1 CEO compensation fell in the early 2000s after the stock market bubble burst, but mostly recovered by 2007, at least for the realized compensation measure (the measure using compensation granted remained substantially below the 2000 level). Realized CEO compensation fell again during the financial crash of 2008–2009 and rose strongly after 2009 and with the strong growth in 2019 regained and exceeded its 2007 pre-financial crisis level but in 2019 still remained below the 2000 peak level. CEO compensation continues to be dramatically higher than it was in the decades before the turn of the millennium. Realized CEO compensation was 1,167% higher in 2019 than in 1978 and granted CEO compensation was 1,033% higher. Correspondingly, the CEO-to-average-worker pay ratio, using the realized compensation measure, was 320-to-1 in 2019, far higher than the ratios in earlier years: 118-to-1 in 1995, 61-to-1 in 1989, 31-to-1 in 1978, and 21-to-1 in 1965.

The relationship between CEO pay and the stock market

CEO pay has become closely associated with the growth of the stock market. The generally tight link between stock prices and CEO compensation indicates that CEO pay is not being established by a “market for talent,” as pay surged with the overall rise in profits and stocks, and not with the better performance of a CEO’s particular firm relative to the performance of that firm’s competitors.

The relationship between CEO pay and the pay of other top earners; the rise of inequality

Amid a healthy recovery on Wall Street following the Great Recession, CEOs enjoyed outsized gains in compensation even relative to other very-high-wage earners (those in the top 0.1%); CEOs of large firms earned 6.0 times as much as the average top 0.1% earner in 2018, up from 4.4 times as much in 2007 and 3.3 times as much in 1979. This is yet another indicator that CEO pay is more likely based on CEOs’ power to set their own pay, not on a market for talent.

To be clear, these other very-high-wage earners aren’t suffering: Their earnings grew 337% between 1978 and 2018. CEO pay growth has had spillover effects, pulling up the pay of other executives and managers, who constitute more than 40% of all top 1.0% and 0.1% earners.2 Consequently, the growth of CEO and executive compensation overall was a major factor driving the doubling of the income shares of the top 1% and top 0.1% of U.S. households from 1979 to 2007 (Bakija, Cole, and Heim 2012; Bivens and Mishel 2013). Income growth has remained unbalanced. As profits and stock market prices have reached record highs, the wages of most workers have grown very modestly, including in the recovery from the Great Recession (Bivens et al. 2014; Gould 2020b).

Key findings

The measures analyzed in the report and associated key findings include the following:

• CEO compensation in 2019 (realized compensation measure). Using the realized compensation measure, the average compensation of CEOs of the 350 largest U.S. firms was $21.3 million in 2019. Compensation grew 14.0% in 2019 following a 1.5% loss in 2018. Top CEO compensation doubled over the recovery from 2009 to 2019, growing 105.1%.
• CEO compensation in 2019 (granted compensation measure). Using the granted compensation measure, the average compensation of CEOs of the 350 largest U.S. firms was $14.5 million in 2019, up 8.6% from $13.3 million in 2018 and up 35.7% since the recovery from the Great Recession began in 2009.
• Growth of CEO compensation (1978–2019). Using the realized compensation measure, compensation of the top CEOs increased 1,167% from 1978 to 2019 (adjusting for inflation). Top CEO compensation growth was roughly 50% greater than stock market growth during this period and far eclipsed the painfully slow 13.7% growth in a typical worker’s annual compensation. CEO granted compensation rose 1,033% from 1978 to 2019.
• Changes in the CEO-to-worker compensation ratio (1965–2019). Using the realized compensation measure, the CEO-to-worker compensation ratio was 21-to-1 in 1965. It peaked at 366-to-1 in 2000. In 2019 the ratio was 320-to-1, up from 293-to-1 in 2018. Most important, the ratio was far higher than at any point in the 1960s, 1970s, 1980s, or 1990s. Using the CEO granted compensation measure, the CEO-to-worker compensation ratio rose to 223-to-1 in 2019 (up from 212-to-1 in 2018), significantly lower than its peak of 386-to-1 in 2000 but still many times higher than the 45-to-1 ratio of 1989 or the 15-to-1 ratio of 1965.
• Changes in the composition of CEO compensation. The composition of CEO compensation is shifting away from the use of stock options and toward the use of stock awards. Vested stock awards and exercised stock options totaled 16.7 million in 2019 and accounted for 78.6% of average realized CEO compensation.
• Changes in the CEO-to-top-0.1% compensation ratio. Over the last three decades, compensation grew far faster for CEOs than it did for other very highly paid workers (the top 0.1%, or those earning more than 99.9% of wage earners). CEO compensation in 2018 (the latest year for which data on top wage earners are available) was 6.04 times as high as wages of the top 0.1% of wage earners, a ratio 2.86 points greater than the 3.18-to-1 average CEO-to-top-0.1% ratio over the 1947–1979 period.
• Implications of the growth of CEO-to-top-0.1% compensation ratio. The fact that CEO compensation has grown far faster than the pay of the top 0.1% of wage earners indicates that CEO compensation growth does not simply reflect a competitive race for skills (the “market for talent”) that also increased the value of highly paid professionals: Rather, the growing pay differential between CEOs and top 0.1% earners suggests the growth of substantial economic rents (income not related to a corresponding growth of productivity) in CEO compensation. CEO compensation appears to reflect not greater productivity of executives but the power of CEOs to extract concessions. Consequently, if CEOs earned less or were taxed more, there would be no adverse impact on the economy’s output or on employment.
• Growth of top 0.1% compensation (1978–2018). Even though CEO compensation grew much faster than the earnings of the top 0.1% of wage earners, that doesn’t mean the top 0.1% did not fare well. Quite the contrary. The inflation-adjusted annual earnings of the top 0.1% grew 337% from 1978 to 2018. CEO compensation, however, grew three times as fast!
• CEO pay growth compared with growth in the college wage premium. Over the last three decades, CEO compensation increased more relative to the pay of other very-high-wage earners than did the wages of college graduates relative to the wages of high school graduates. This finding indicates that the escalation of CEO pay does not simply reflect a more general rise in the returns to education.

Analysis

This section provides detailed analysis of our findings. We examine several decades of available data to identify recent and historical trends in CEO compensation.

Trends in CEO compensation growth

Table 1 presents recent trends in CEO compensation and for the key underlying components over the 2016–2019 period. It shows the average compensation of CEOs at the 350 largest publicly owned U.S. firms (i.e., firms that sell stock on the open market) by revenue.3 To analyze current trends, we use two measures of compensation, one based on compensation “granted” and the other based on compensation as “realized.” Both measures include the same measures of salary, bonuses, and long-term incentive payouts (columns 3, 4 and 5). The difference is how each measure treats stock awards and stock options, major components of CEO compensation that change value from when they are first provided, or granted, to when they are exercised or realized. The first measure, realized compensation (column 1), includes the value of stock options as realized (buying stocks at a previously set price and reselling them at the current market price) shown in column 8. The realized compensation measure also values stock awards at their value when vested (usually three years after being granted), capturing any change in the stock price as well as additional stock awards provided as part of a performance award (column 6). The second measure, compensation granted, values stock options and restricted stock awards by their “fair value” when granted (columns 9 and 7).4 (For details on the construction of these measures and benchmarking to other studies, see Sabadish and Mishel 2013.)

Table 1

We have changed our definition of CEO compensation in the realized measure from that employed in earlier reports. Previous reports used the value of stock awards as granted in both our measures, so that the measures only differed in their treatment of stock options. As noted in our previous report (Mishel and Wolfe 2019) the increased importance of stock awards in executive pay and the increased divergence between the value of stock awards when granted versus when vested means that excluding the realized gains from stock awards increasingly understates total CEO compensation. We therefore have incorporated a realized measure of stock awards along with the realized measure of stock options in our realized compensation metric. This first metric can be compared with the second metric, compensation granted, whose measurement is the same as in prior reports. More explanation of this measurement change and the impact on measured trends is provided in the Appendix, “Revising the stock awards component of our CEO compensation measure.”

Note that Table 1 provides a projection for data for 2019. The data now available for 2019 are limited to the executive compensation disclosed by firms filing proxy statements through June of 2019. To provide data for CEO compensation in 2019 that are consistent with the historical data, we construct our estimates by looking at the growth of compensation from 2018 to 2019 using the first-half-year samples of data available each year and then applying that growth rate to the compensation for 2018 based on the full-year sample. This method corrects for the fact that full-year samples show higher average CEO compensation than samples for the first half of a year. It allows us to avoid artificially lowering the estimated change in CEO compensation in 2019 relative to last year and earlier years.5

Both measures of CEO compensation grew strongly in 2019. CEO realized compensation grew to $21,283,000 in 2019, $2,621,000 or 14.0% higher than in 2018. The compensation granted measure grew $1,148,000, or by 8.6%, to $14,487,000 in 2019.

This growth in CEO compensation in 2019 was entirely driven by stock-related components: salary, bonuses, and nonequity incentives remained stable throughout the 2016–2019 period while stock options and stock awards grew.6 Stock options granted (column 9) did not grow much (up only $8,000) in 2019 though realized stock options (column 8) increased by roughly a million dollars ($977,000). The bigger growth was in stock awards, $1,179,000 for stock awards granted (column 7) and a larger 19.5% boost of $1,659,000 for vested stock awards (column 6).

The stock-related components of CEO compensation constitute a large and increasing share of total compensation: realized stock awards and stock options (column 10) were 73.1% of total compensation in 2016 and 78.6% in 2019. Vested stock awards (the realized metric, column 6) alone were nearly half (47.7%) of all CEO compensation in 2019.

There is a simple logic behind companies’ decisions to shift from stock options to stock awards, as Clifford (2017) explains. With stock options, CEOs can only make gains: They realize a gain if the stock price rises beyond the price of the initial options granted and they lose nothing if the stock price falls. The fact that they have nothing to lose—but potentially a lot to gain—might lead options-holding CEOs to take excessive risks to bump up the stock price. Stock awards, on the other hand, promote better alignment of a CEO’s goals with shareholders’ goals. A stock award has the value when given, or vested, and can increase or decrease in value as the firm’s stock price changes. If stock awards have a lengthy vesting period, say three to five years, then the CEO has an interest in lifting the firm’s stock price over that period while being mindful to avoid any implosion in the stock price—to maintain the value of what they have.

The growth of these stock-related components from 2016 to 2019, up 34.5%, or $4,286,000, was the sole reason that CEO realized compensation grew $4,277,000, or 25.2%. The smaller growth of CEO granted compensation (up $1,765,000) in the same period, 2016–2019, reflects the smaller growth of stock awards granted ($1,861,000) and the failure of stock options granted to grow. This pattern, as explored further below, mirrors the strong growth of the stock market between 2016 and 2019, up 30.6% in the S&P 500.

Table 2 presents the longer-term trends in CEO compensation for selected years from 1965 to 2019 using the same two measures used in Table 1.7

For comparison, Table 2 also presents the average annual compensation (wages and benefits of a full-time, full-year worker) of private-sector production/nonsupervisory workers (a group covering more than 80% of payroll employment, see Gould 2020a), allowing us to compare CEO compensation with that of a typical worker. From 1995 onward, the table also identifies the average annual compensation of production/nonsupervisory workers in each of the industries of the firms included in the sample. We take this compensation as a proxy for the pay of typical workers in these particular firms and use it to calculate the CEO-to-worker compensation ratio for each firm.

Table 2

Finally, the table shows changes in the stock market, as measured by the Dow Jones Industrial Average and S&P 500 Index. Figure A uses data from all years since 1965 to show what happened to average annual CEO compensation and the S&P 500 Index over the last five and a half decades. It uses the realized CEO compensation measure.

Although the stock market fell by roughly half between 1965 and 1978, realized CEO compensation increased by 78.9%. Typical worker pay saw relatively strong growth over that period (relative to subsequent periods, not relative to CEO pay or the pay of other earners at the top of the wage distribution). Annual worker compensation grew by 19.9% from 1965 to 1978, only about a fourth as fast as CEO compensation growth.

Realized CEO compensation grew strongly throughout the 1980s but exploded in the 1990s. It peaked at the end of the stock market bubble, in 2000, at about $21.9 million, a 261% increase over just five years earlier in 1995 and a 1,204% increase over 1978. This latter increase exceeded even the growth of the booming stock market (513% for the S&P 500 and 439% for the Dow) between 1978 and 2000. In stark contrast to both the stock market and CEO compensation, private-sector worker compensation increased just 0.6% over the same period.

When the stock market bubble burst in the early 2000s there was a substantial paring back of CEO compensation. By 2007, however, when the stock market had mostly recovered, realized CEO compensation reached $19.4 million, just $2.5 million below its 2000 level. However, granted CEO compensation remained down, at $14.4 million in 2007, a substantial $7.6 million fall from the 2000 level.

The stock market decline during the 2008 financial crisis also sent CEO compensation tumbling, as it had in the early 2000s. After 2009, realized CEO compensation resumed an upward trajectory, as shown in Figure A. It stalled from 2014 to 2018. The strong growth in CEO compensation in 2019 raised it to $1.9 million above where is was in 2007, before the 2008 financial crisis. Although Figure A does not track the trajectory of the change in granted CEO compensation, we know from the data behind Tables 1 and 2 that it also shot up until 2013 and then leveled out over the 2013–2017 period before a $1.2 and 1.1 million growth, respectively, in 2018 and 2019, leaving granted CEO compensation in 2019 slightly ($134,000) above the pre-2008-financial crisis level.

Figure A

For the period from 1978 to 2019, realized CEO compensation increased 1,166.8%—roughly 50% as fast as stock market growth (depending on the market index used) and substantially faster than the painfully slow 13.7% growth in the typical worker’s compensation over the same period. CEO granted compensation grew 1,032.7% over this period. Realized CEO compensation in 2019 remained below its stock market bubble 2000 peak, but was only off the peak by $627,000, or 2.9%.

Figure A shows how realized CEO compensation historically fluctuates in tandem with the stock market, as measured by the S&P 500 Index, confirming that CEOs tend to cash in their options when stock prices are high and accumulate unexercised options when stock prices are low. The growth of stock prices also increases the value of stock awards between when they are granted and when they vest, usually three years later. The financial crisis of 2008 and the accompanying stock market tumble knocked CEO compensation down 46.6% from 2007 to 2009. By 2014 the stock market had recouped more than all of the ground lost in the downturn. Not surprisingly, CEO compensation also made a strong recovery. The close connection between stock market growth and CEO compensation loosened a bit over the 2014–2017 period as realized CEO compensation did not follow the sharp upward trajectory of the stock market in those years. However, as shown in Figure A and Table 2, the growth of both realized and granted CEO compensation from 2017 to 2019 closely mirrors the growth of the stock market.

The normally tight relationship between overall stock prices and CEO compensation, as shown in Figure A, casts doubt on the theory that CEOs are enjoying high and rising pay because their individual productivity is increasing (e.g., because they head larger firms, have adopted new technology, or for other reasons). CEO compensation often grows strongly when the overall stock market rises and individual firms’ stock values rise along with it. This is a marketwide phenomenon, not one of improved performance of individual firms: Most CEO pay packages allow pay to rise whenever the firm’s stock value rises; that is, they permit CEOs to cash out stock options regardless of whether the rise in the firm’s stock value was exceptional relative to comparable firms in the same industry. Similarly, vested stock awards will increase in value when the firm’s stock price rises and simply corresponds to a marketwide escalation of stock prices.

Trends in the CEO-to-worker compensation ratio

Table 2 also presents historical and current trends in the ratio of CEO-to-worker compensation, using both measures of CEO compensation. This ratio, which illustrates the increased divergence between CEO and worker pay over time, is computed in two steps. The first step is to construct, for each of the 350 largest U.S. firms, the ratio of the CEO’s compensation to the annual average compensation of production and nonsupervisory workers in the key industry of the firm (data on the pay of workers at individual firms are not available).8 The second step is to average that ratio across all 350 firms. Note however that trends before 1995 are based on the changes in average top-company CEO and economywide private-sector production/nonsupervisory worker compensation.

The last two columns in Table 2 show the resulting ratio for both measures of CEO pay. We adjust the ratio for 2019 to reflect the percentage-point growth between the ratios in the first-half-year samples in 2018 and 2019 and add that growth to the ratio estimated for the full-year sample in 2018 to derive the 2019 ratio consistent with the historical data (this corresponds to how we project CEO compensation for 2019 based on first half data in 2018 and 2019). The trends are depicted in Figure B.

Figure B

The Securities and Exchange Commission (SEC) now requires publicly owned firms to provide a metric for the ratio of CEO compensation to that of the median worker in a firm, as mandated by the Dodd-Frank financial reform bill of 2010 (SEC 2015). Those ratios differ from those in this report in several ways. First, because of limitations in data availability, the measure of worker compensation in our ratios reflects workers in a firm’s key industry, not workers actually working for the firm. The ratios reported to the SEC will reflect compensation of workers in the specific firm. Second, our measure reflects an exclusively domestic workforce; it excludes the compensation of workers in other countries who work for the firm. The ratios reported to the SEC may include workers in other countries. Third, our metric is based on hourly compensation annualized to reflect a full-time, full-year worker (i.e., multiplying the hourly compensation rate by 2,080). In contrast, the measures firms provide to the SEC can be and are sometimes based on the actual annual (not annualized) wages of part-year (seasonal) or part-time workers. As a result, comparisons across firms may reflect not only hourly pay differences but also differences in annual or weekly hours worked. Fourth, our metric includes both wages and benefits, whereas the SEC metric solely focuses on wages. Finally, we use consistent data and methodology to construct our ratios; our ratios are thus comparable across firms and from year to year. The SEC allows firms flexibility in how they construct the CEO-to-median worker pay comparison; this means there is not comparability across firms—and ratios may not even be comparable from year to year for any given firm, if the firm changes the metrics it uses.

There is certainly value in the new metrics being provided to the SEC, but the measures we rely on allow us to make appropriate comparisons between firms and across time. The text box provides more information on the ratios firms are providing to the SEC.

As Figure B shows, using the realized measure of CEO compensation, CEOs of major U.S. companies earned 21 times as much as the typical worker in 1965. This ratio grew to 31-to-1 in 1978 and 61-to-1 by 1989. It surged in the 1990s, hitting 366-to-1 in 2000, at the end of the 1990s recovery and at the height of the stock market bubble.9 The fall in the stock market after 2000 reduced CEO stock-related pay such as realized stock options and caused CEO compensation to tumble in 2002 before beginning to rise again in 2003. Realized CEO compensation recovered to a level of 331 times worker pay by 2007, still below its 2000 level. The financial crisis of 2008 and accompanying stock market decline reduced CEO compensation between 2007 and 2009, as discussed above, and the CEO-to-worker compensation ratio fell in tandem. By 2014 the stock market had recouped all of the value it had lost following the financial crisis, and the CEO-to-worker compensation ratio in 2014 had recovered to 327-to-1. Because CEO compensation was relatively stable between 2014 and 2016 while worker compensation experienced moderate growth, the CEO-to-worker pay ratio fell. Over the 2016–2019 period CEO pay resumed its upward trajectory and the 14% surge in realized CEO compensation in 2019 brought the ratio to 320-to-1, not far from its 2007 level. Though the realized CEO-to-worker compensation ratio remains below the value achieved in 2000, at the peak of a stock market bubble, it is far higher than it was in the 1960s, 1970s, 1980s, and most of the 1990s.

The pattern using the granted measure of CEO compensation is similar. The CEO-to-worker pay ratio peaked in 2000, at 386-to-1, even higher than the ratio with the realized compensation measure. The fall from 2000 to 2007 was steeper than for the other measure, hitting 242-to-1 in 2007. The stock market decline during the financial crisis drove the ratio down to 178-to-1 in 2009. It recovered to 217-to-1 by 2014 and, after dipping a bit over the next three years, ended back up at 212-to-1 in 2018 before rising to 223-to-1 with the strong 8.6% growth of CEO granted compensation in 2019. This level is far lower than its peak in 2000 but still far greater than the 1989 ratio of 45-to-1 or the 1965 ratio of 15-to-1.

The exponential growth in the CEO-to-worker compensation ratio reflects the strikingly different trajectories of the pay of CEOs and that of the typical worker. On the one hand, there has been very little growth in the compensation of a typical worker since the late 1970s, growing just 15.1% over the 40 years from 1979 to 2019, despite a corresponding growth of economywide productivity of 70% (Bivens and Mishel 2015, updated at EPI 2019). The 1,167% growth in realized CEO compensation from 1978 (there are no data for 1979) to 2019 far exceeded the growth in productivity, profits, or stock market values in that period.

Dramatically high CEO pay does not simply reflect the market for skills

This section reviews competing explanations for the extraordinary rise in CEO compensation over the past several decades. CEO compensation has grown a great deal since 1965, but so has the pay of other high-wage earners. To some analysts, this suggests that the dramatic rise in CEO compensation has been driven largely by the demand for the skills of CEOs and other highly paid professionals. In this interpretation, CEO compensation is being set by the market for “skills” or “talent,” not by managerial power or rent-seeking behavior.10 This explanation lies in contrast to that offered by Bebchuk and Fried (2004) or Clifford (2017), who claim that the long-term increase in CEO pay is a result of managerial power.

The “market for talent” argument is based on the premise that “it is other professionals, too,” not just CEOs, who are seeing a generous rise in pay. One prominent example of this argument comes from Kaplan (2012a, 2012b). In the prestigious 2012 Martin Feldstein Lecture at the National Bureau of Economic Research, he claims:

Over the last 20 years, then, public company CEO pay relative to the top 0.1% has remained relatively constant or declined. These patterns are consistent with a competitive market for talent. They are less consistent with managerial power. Other top income groups, not subject to managerial power forces, have seen similar growth in pay. (Kaplan 2012a, 4)

In a follow-up paper for the Cato Institute, published as a National Bureau of Economic Research working paper, Kaplan expands this point:

The point of these comparisons is to confirm that while public company CEOs earn a great deal, they are not unique. Other groups with similar backgrounds—private company executives, corporate lawyers, hedge fund investors, private equity investors and others—have seen significant pay increases where there is a competitive market for talent and managerial power problems are absent. Again, if one uses evidence of higher CEO pay as evidence of managerial power or capture, one must also explain why these professional groups have had a similar or even higher growth in pay. It seems more likely that a meaningful portion of the increase in CEO pay has been driven by market forces as well. (Kaplan 2012b, 21)

However, the argument that CEO compensation is being set by the market for “skills” does not square with the available data corresponding to what Kaplan employed. Bivens and Mishel (2013) address the larger issue of the role of CEO compensation in generating income gains at the very top and conclude that substantial rents are embedded in executive pay. According to Bivens and Mishel, CEO pay gains are not the result of a competitive market for talent but rather reflect the power of CEOs to extract concessions.

Here we draw on and update the Bivens and Mishel (2013) analysis to show that the evidence does not support Kaplan’s claim that “professional groups have had a similar or even higher growth in pay” than CEOs (Kaplan 2012b). CEO compensation grew far faster than compensation of very highly paid workers over the last few decades, which suggests that the market for skills was not responsible for the rapid growth of CEO compensation. To reach this finding, we use Kaplan’s series on CEO compensation and compare it with the wages of top wage earners (reflecting W-2 annual earnings, which includes the value of exercised stock options and vested stock awards), rather than the household income of the top 0.1% as Kaplan did.11 The wage benchmark seems the most appropriate one because it avoids issues of changing household demographics (e.g., increases in the number of two-earner households over time) and limits the income to labor income (i.e., it excludes capital income, which is included in household income measures). We update Kaplan’s series (Kaplan 2012b) beyond 2010 using the growth of our measure of realized CEO compensation.

The data presented in Table 3 show the result of our analysis: It shows that, contrary to Kaplan’s findings, the compensation of CEOs has far outpaced that of the top 0.1% of earners. Specifically, it shows the ratio of the average compensation of CEOs of large firms (the series developed by Kaplan, incorporating stock options realized) to the average annual earnings of the top 0.1% of wage earners (based on a series developed by Kopczuk, Saez, and Song 2010 and updated by Mishel and Kassa 2019). The comparison is presented as a simple ratio and logged (to convert to a “premium,” defined as the relative pay differential between two groups). Both the simple ratios and the log ratios understate the relative pay of CEOs, because CEO pay is a nontrivial share of the denominator, a bias that has probably grown over time as CEO relative pay has grown. If we were able to remove top CEOs’ pay from the top 0.1% category, it would reduce the average for the broader group.12

Table 3

The very highest earners—those in the top 0.1% of all earners—saw their compensation grow fantastically though far less than the compensation of the CEOs of large firms (note that the gains from exercised stock options are taxed as W-2 wage income and so are reflected in measures of wages in the data we analyze).

CEO realized compensation was 6.04 times the pay of the top 0.1% of wage earners in 2018, a bit below the 6.10 ratio in 2017 and substantially higher than the 4.36 ratio in 2007. CEO compensation grew far faster than that of the top 0.1% of earners over the recovery from 2009 to 2018, as the ratio spiked from 4.61 to 6.04. CEO compensation relative to the wages of the top 0.1% of wage earners in 2018 far exceeded the ratio of 2.63 in 1989, a rise (3.41) equal to the pay of more than three very-high-wage earners.13 The log ratio of CEO relative pay grew 83 log points from 1989 to 2018 with respect to wage earners in the top 0.1%.

Is this increase large? As noted earlier, Kaplan (2012a, 4) concludes  in his prestigious Martin Feldstein Lecture that CEO relative pay “has remained relatively constant or declined.” In another paper, Kaplan (2012b, 21), claimed that high earning professional groups such as “private company executives, corporate lawyers, hedge fund investors, private equity investors, and others” had a  “similar or even higher growth in pay” as CEOs. Kaplan’s historical comparisons are inaccurate, however. Figure C compares the ratios of CEO compensation to top 0.1% earnings back to 1947. In 2018 this ratio was 6.04, 2.86 points higher than the historical average of 3.18 in the 1947–1979 period (a relative gain in wages earned by the equivalent of 2.9 very-high-wage earners).

Figure C

That CEO compensation grew much faster than the earnings of the top 0.1% of wage earners is not because the top 0.1% did not fare well. The inflation-adjusted annual earnings of the top 0.1% grew 337% from 1978 to 2018 (Mishel and Kassa 2019). CEO compensation, however, grew more than three times faster than that, up 1,167%!

If CEO pay growing far faster than that of other high earners is evidence of the presence of rents, as Kaplan suggests, one would conclude that today’s top executives are collecting substantial rents, meaning that if they were paid less there would be no loss of productivity or output in the economy. The large discrepancy between the pay of CEOs and other very-high-wage earners also casts doubt on the claim that CEOs are being paid these extraordinary amounts because of their special skills and the market for those skills. It is unlikely that the skills of CEOs of very large firms are so outsized and disconnected from the skills of other high earners that they propel CEOs past most of their cohort in the top one-tenth of 1%. For everyone else, the distribution of skills, as reflected in the overall wage distribution, tends to be much more continuous so this discontinuity is evidence that factors beyond skills drive the compensation levels of CEOs.

For comparison purposes, Table 3 also shows the changes in the gross (not regression-adjusted) college-to-high-school wage premium. This premium is simply how much higher are the hourly wages of workers with a (four-year) college degree (but not an advanced degree) relative to hourly wages of workers with just a high school diploma. This premium is useful because some commentators, such as Mankiw (2013), assert that the wage and income growth of the top 1% reflects the general rise in the return to skills, as reflected in higher college wage premiums. (The comparisons end in 2018 because 2019 data for top 0.1% wages are not yet available).

Since 1979, and particularly since 1989, the increase in the logged CEO pay premium relative to other high-wage earners far exceeded the rise in the college-to-high-school wage premium, which is widely and appropriately considered to have had substantial growth: The logged college wage premium grew from 0.46 in 1989 to 0.59 in 2018, a far smaller rise than the logged ratio of CEO-to-top-0.1% earnings, a rise from 0.97 to 1.80. Mankiw’s claim that top 1% pay or top executive pay simply corresponds to the rise in the college-to-high-school wage premium is unfounded (Mishel 2013a, 2013b). Moreover, the data we present here would show even faster growth of CEO relative pay if Kaplan’s historical CEO compensation series (which we use as the basis for the ratios in Table 3) had been built using the Frydman and Saks (2010) series for the 1980–1994 period rather than the Hall and Liebman (1997) data.14

Conclusion and the connection to overall inequality

Some observers argue that exorbitant CEO compensation is merely a symbolic issue, with no consequences for the vast majority of workers. However, the escalation of CEO compensation, and of executive compensation more generally, has fueled the growth of top 1.0% and top 0.1% incomes, generating widespread inequality.

In their study of tax returns from 1979 to 2005, Bakija, Cole, and Heim (2010) establish that the increases in income among the top 1% and top 0.1% of households were disproportionately driven by households headed by someone who was either a nonfinancial-sector “executive” (including managers and supervisors, hereafter referred to as “nonfinance executives”) or a financial-sector worker (executive or otherwise). Forty-four percent of the growth of the top 0.1%’s income share and 36% of the top 1%’s income share accrued to households headed by nonfinance executives; another 23% for each group accrued to households headed by financial-sector workers (some portion of which were executives).

Together, finance workers (including some share who are executives) and nonfinance executives accounted for 58% of the expansion of income for the top 1% of households and 67% of the income growth of the top 0.1%. Relative to others in the top 1%, households headed by nonfinance executives had roughly average income growth; those headed by someone in the financial sector had above-average income growth; and the remaining households (nonexecutive, nonfinance) had slower-than-average income growth. These shares may actually understate the role of nonfinance executives and the financial sector, because they do not account for increased spousal income from these sources in those cases where the head of household is not an executive or in finance.15

High CEO pay reflects economic rents—concessions CEOs can draw from the economy not by virtue of their contribution to economic output but by virtue of their position. Alluding to the fictional town in the radio program “A Prairie Home Companion,” Clifford (2017) describes the Lake Wobegon world of setting CEO compensation that fuels its growth: Every firm wants to believe its CEO is above average and therefore needs to be correspondingly remunerated. But, in fact, CEO compensation could be reduced across the board and the economy would not suffer any loss of output.

Another implication of rising pay for CEOs and other executives is that it reflects income that otherwise would have accrued to others: What these executives earned was not available for broader-based wage growth for other workers. (Bivens and Mishel 2013 explore this issue in depth.) It is useful, in this context, to note that wage growth for the bottom 90% would have been nearly twice as fast over the 1979–2018 period had wage inequality not grown.16 Most of the rise of inequality took the form of redistributing wages from the bottom 90% (whose share of wages fell from 69.8% to 61.0%) to the top 1.0% (whose wage share nearly doubled, rising from 7.3% to 13.3%, with most of the increase among the top 0.1% whose share of all wages grew from 1.6% to 5.1%) (Mishel and Kassa 2019).

Although the analyses in this report predate the economic shock of the coronavirus pandemic, there is a renewed focus on CEO pay because so many American workers are out of work or have seen their hours or wages cut. As our analyses show, CEOs who volunteer to take salary cuts aren’t giving up a lot given how much of their pay comes from stock awards and options. Moreover, the inflation-adjusted growth of the stock market, as reflected in the S&P 500, was about 8% higher in mid-2020 (last half of June and first half of July) than it was in 2019, indicating that CEO compensation in 2020 will very likely grow over its 2019 levels.

Several policy options could reverse the trend of excessive executive pay and broaden wage growth. Some involve taxes. Implementing higher marginal income tax rates at the very top would limit rent-seeking behavior and reduce the incentives for executives to push for such high pay.17 Another option is to set corporate tax rates higher for firms that have higher ratios of CEO-to-worker compensation. Clifford (2017) recommends setting a cap on compensation and taxing companies on any amount over the cap, similar to the way baseball team payrolls are taxed when salaries exceed a cap. Other policies that could potentially limit executive pay growth are changes in corporate governance, such as greater use of “say on pay,” which allows a firm’s shareholders to vote on top executives’ compensation. Baker, Bivens, and Schieder (2019) review policies to restrain CEO compensation and explain how tax policy and corporate governance reform can work in tandem: “Tax policy that penalizes corporations for excess CEO-to-worker pay ratios can boost incentives for shareholders to restrain excess pay,” but, “to boost the power of shareholders [to restrain pay], fundamental changes to corporate governance have to be made. One key example of such a fundamental change would be to provide worker representation on corporate boards.”

The CEOs examined in this report head large firms. These large firms, almost by definition, enjoy a degree of market power that has grown in recent decades. It seems that CEOs and other executives may have been prime beneficiaries of these firms’ greater market power. This suggests using the tools of anti-trust enforcement and regulation to restrain these firms’ market power. This not only promotes economic efficiency and competition, but might help restrain executive pay as well.

About the authors

Lawrence Mishel is a distinguished fellow and former president of the Economic Policy Institute. He is the co-author of all 12 editions of The State of Working America. His articles have appeared in a variety of academic and nonacademic journals. His areas of research include labor economics, wage and income distribution, industrial relations, productivity growth, and the economics of education. He holds a Ph.D. in economics from the University of Wisconsin at Madison.

Jori Kandra is a research assistant at the Economic Policy Institute.

Acknowledgments

The authors thank the Stephen Silberstein Foundation for its generous support of this research. Steven Balsam, an accounting professor at Temple University and author of Equity Compensation: Motivations and Implications (2013), has provided useful advice on data construction and interpretation over the years. Steven Clifford, author of The CEO Pay Machine: How It Trashes America and How to Stop It (2017), has also provided technical advice. Clifford served as CEO for King Broadcasting Company from 1987 to 1992 and National Mobile Television from 1992 to 2000 and has been a director of thirteen public and private companies.

Appendix: Revising the stock awards component of our CEO compensation measure

In this report we have revised our realized measure of CEO compensation to reflect the growth of the value of stock awards from the time they are granted to when they are vested—growth capturing both rising stock prices and the awarding of more stock based on meeting performance targets. With this change, our realized metric includes the realized value of stock awards as well as of stock options, as recommended in Hopkins and Lazonick (2016). Our other metric of CEO compensation—granted CEO compensation—captures the “fair value” of both stock options and stock awards.

The need for this change in measurement was described in last year’s report on CEO compensation (Mishel and Wolfe 2019),

Analyses of the underlying components of CEO compensation over the 2016–2018 period… showed a strong growth in stock awards, which are simply stocks granted to employees. Stock awards can increase or decrease in value depending on the trend in the firm’s stock price. Stock awards, which are included in both definitions of CEO compensation, rose to $7.5 million in 2018, a substantial amount of income alone. The composition of CEO compensation has been shifting toward stock awards and away from stock options since the end of the last cycle in 2006–2007. These two stock-related items—stock options and stock awards—together still make up the bulk of CEO compensation, at 74% and 68%, respectively, of options-exercised and options-granted CEO compensation measures in 2018….

As the share of CEO compensation represented by stock options declines, and the share represented by stock awards grows, CEO compensation levels and growth will possibly be increasingly understated in our measures as well as in other measures, including those used by companies to construct the CEO-to-worker ratios reported to the SEC. The reason is this: The exact compensation earned through stock options is measurable—the exercised-options measure of compensation captures any rise in the stock price from the time the options are granted. But for stock awards, the value is determined at the time stocks are granted; any future gains in the value of the stock that accrue to the CEO are not captured by data disclosed by the firms. Nor are they captured in the SEC measure. Because stock awards have become more important, and stock options less important, there is increased likelihood that measures of CEO compensation will not fully capture CEOs’ gains going forward. This increased understatement of CEO compensation in turn tamps down measures of CEO compensation growth.

The measure of stock awards used in both of our CEO compensation metrics was the fair value of stock awards: the number of shares granted times the stock price at the grant date. Now, in our realized CEO compensation measure, we are using a realized value measure of stock awards which reflects the value of stock awards when vested. This will capture both the rise and fall of the value of the stock awards between grant and vesting and any increase in the stock awards due to performance equity programs that award more shares for exceeding performance targets (Francis 2019, Hodak 2019). Hodak (2019) reports that executives are likely to receive at least half their awards after three years based on performance programs rather than time since award. So, using a fair market value measure of stock awards at the time awards were granted understates the compensation actually received by executives—and this understatement is increasingly acute in recent years as stock awards have become a greater share of compensation, necessitating a change in how we measure CEO compensation.

Appendix Figure A shows the trend in the fair market value of stock awards when granted and the vested value of stock awards that is now incorporated into our realized pay CEO compensation metric, both set in 2019 dollars. This allows us to assess how the change in the stock award measure affects CEO compensation trends since 2006, which is the first year for which the vested value of stock awards is available. We use data through 2018, which is the latest year for which we have a full year of data. Whereas the vested value of stock awards was $632,000 less than that of the fair value of stock awards in 2006 ($3,761,000 versus $4,393,000) the vested value measure grew faster so that by 2018 it was $1,214,000, or 16.7%, more than the fair market value of stock awards ($8,490,000 versus $7,276,000). This shows both the sizable growth of the value of stock awards and the even faster growth in the value of stock awards when the rising value of individual shares is combined with an increase in the number of shares awarded from performance programs and thus counted in the vested measure.

Appendix Figure A

Appendix Figure B shows how the use of the vested value rather than the fair value of stock awards affects the overall trend of CEO compensation. Figure B displays our new realized metric of CEO compensation, capturing both realized stock options and vested stock awards, and the prior measure of realized CEO pay used in earlier reports, which captures realized stock options but measures stock awards as granted not as vested.18 Again, we show the trends between 2006 and 2018 in $2019.

Appendix Figure B

Changing measures of stock awards means a lower value of CEO compensation in 2006 by $632,000 million from $19,560,000 to $18,929,000. By 2018, however, the revised CEO compensation measure is $18,663,000, $1,214,000, or 7.0%, greater than the unrevised measure, $17,448,000. Revising the measurement of CEO compensation to include the realized value of stock awards increases the level of CEO compensation in the latest full year, 2018, and also shows greater growth since 2006.

The change in measurement of CEO compensation in this report creates a discontinuity between 2005 and 2006, the year that the data for both the fair value and vested value of stock awards are first available. Before 2006 the value of stock awards was a “restricted stock grant” measure capturing the “value of restricted stock awarded during fiscal year, determined at grant date as share price times the number of shares granted” (Hopkins and Lazonick 2016). The pre-2006 measure is similar conceptually to the fair market value but not exactly the same. In 2006, two stock award measures were tracked: the “fair value” and the “vested value.” Restricted stock grants were valued at $2,988,000 ($2019) in 2005. The fair market value of stock awards in 2006 was $4,393,000. It is not possible to know how much of the jump between 2005 and 2006 is due to a change in definition rather than part of the trend toward increased use of stock awards. This discontinuity, and associated possible measurement error, does not matter a great deal for our analysis since our focus is on longer-term trends, analyzing CEO compensation trends since 2007, the year before the financial crisis that sparked the Great Recession, or since 2009, the beginning of the recovery from the Great Recession. We also make comparisons of CEO pay in recent years to CEO pay in 2000 and 1978. Since stock awards were far less popular in the earlier years—in 2000 and certainly in 1978—our judgment is that the discontinuity over 2005–2006 is not a major concern, especially since the discontinuity already necessarily was embedded in our prior metric (as the data switched from restricted grants to fair market value of stock awards).

Appendix Table 1 provides an assessment of how the growth of CEO compensation over key periods is affected by our change of metrics.

Compared with the old measure, the revised measure shows a smaller decline of CEO compensation from 2007 to 2018, 3.9% versus 14.2%, and a much larger growth of CEO compensation over the 2009–2018 recovery period (79.9% versus 54.2%). Because the revised measure is greater in 2018 than our prior metric, the growth measured over the longer term will be greater: specifically, the growth between r 1978 or 2000 and now is greater (less of a fall since 2000, which was the stock bubble–related peak of CEO compensation, and much more since 1978) because of the change in measurement.

Appendix Table 1

Endnotes

1. It may seem counterintuitive that the granted and realized CEO-to-worker pay ratios for 2000 are different from each other when the average CEO compensation is the same. As we describe later in this report, we do not create the ratio from the averages; rather we construct a ratio for each firm and then average the ratios across firms.

2. There were 38,824 executives in publicly held firms and 9,692 people in the top 0.1% of wage earners in 2007, according to the Capital IQ database (tabulations provided by Temple University professor Steve Balsam).

3. Each year’s sample includes the largest 350 firms for which ExecuComp provides data.

4. We use Compustat estimates of the fair value of options and stock awards as granted. These estimates are determined using the Black Scholes model. See Sabadish and Mishel 2013 for more information about our data sources and methodology.

5. Most Fortune 500 companies release annual financial data in early spring; the data are included in samples limited to the first half of the year. However, the data we present for previous years include all of the data that were released during each calendar year. This creates a bias in comparing data for the first half of the year relative to the full year’s data in the prior or earlier years: Compensation levels for the full year’s data are higher than compensation in the data limited to the first half. A comparison of data available in June thus shows a smaller increase when compared with the previous year’s full data than a comparison with the data that were available at the same time a year earlier. We analyze the impact of this bias and find that the vast majority of top firms remain unchanged between the samples for the first half and the full year. However, there is churn among the smaller firms in the sample. Among firms with lower net annual sales, average CEO compensation tends to be higher in the full-year sample. Additionally, in recent years firms reporting later in the year have tended to be firms with lower worker compensation levels and therefore higher CEO-to-worker compensation ratios.

6. In order to calculate the projected full-year 2019 value of the vested stock awards we assume that the vested stock awards as a share of CEO realized compensation for first-half-year 2019 remains consistent for the full-year 2019. We then multiply the share of vested stock awards by the projected full-year 2019 CEO realized compensation. We use the projected full-year 2019 value to calculate the growth rate of the vested stock awards from 2018 to 2019. A similar process is used to calculate the projected full-year 2019 value of exercised stock options.

7. We chose which years to present in the table in part based on data availability. Where possible, we chose cyclical peaks (years of low unemployment).

8. There are a limited number of firms, which existed only for certain years between 1992 and 1996, for which a North American Industry Classification System (NAICS) value is unassigned. This makes it impossible to identify the pay of the workers in the firm’s key industry. These firms are therefore not included in the calculation of the CEO-to-worker compensation ratio.

9. As noted earlier, it may seem counterintuitive that the two ratios for 2000 are different from each other when the average CEO compensation is the same. It is important to understand that (as we describe later in this report) we do not create the ratio from the averages; rather we construct a ratio for each firm and then average the ratios across firms.

10. The managerial power view asserts that CEOs have excessive, noncompetitive influence over the compensation packages they receive. Rent-seeking behavior is the practice of manipulating systems to obtain more than one’s fair share of wealth—that is, finding ways to increase one’s own gains without actually increasing the productive value one contributes to an organization or to the economy.

11. We thank Steve Kaplan for sharing his CEO compensation series with us (Kaplan 2012b). The series on the income of the top 0.1% of households that Kaplan used is no longer available. Moreover, as we discuss, the appropriate comparison is to other earners, not to households, which could have multiple earners and shifts in the number of earners over time.

12. Temple University professor Steve Balsam provided tabulations from the Capital IQ database of annual wages of executives exceeding the wage thresholds (provided to him) that place them in the top 0.1% of wage earners. There were 38,824 executives in publicly held firms and 9,692 executives in the top 0.1% of wage earners in 2007. The 9,692 executives in publicly held firms who were in the top 0.1% of wage earners had average annual earnings of $4.4 million. Using Mishel et al.’s (2012) estimates of top 0.1% wages, we find that executive wages make up 13.3% of total top 0.1% wages. One can gauge the bias of including executive wages in the denominator by noting that the ratio of executive wages to all top 0.1% wages in 2007 was 2.14 but the ratio of executive wages to nonexecutive wages was 2.32. We do not have data that would permit an assessment of the bias in 1979 or 1989. We also lack information on the number and wages of executives in privately held firms; to the extent that their CEO compensation exceeds that of publicly traded firms, their inclusion would indicate an even larger bias. The Internal Revenue Service Statistics of Income (SOI) Bulletin reports that there were nearly 15,000 corporate tax returns in 2007 of firms with assets exceeding $250 million, indicating that there are many more executives of large firms than just those in publicly held firms (IRS 2019).

13. A one-point rise in the ratio is the equivalent of the average CEO earning an additional amount equal to that of the average earnings of someone in the top 0.1%.

14. Kaplan (2012b, 14) notes that the Frydman and Saks series grew 289% whereas the Hall and Liebman series grew 209%. He also notes that the Frydman and Saks series grows faster than the series reported by Murphy (2012).

15. The tax data analyzed categorizes a household’s income according to the occupation and industry of the head of household. It is possible that a “secondary earner,” or spouse, has income as an executive or in finance. If the household is in the top 1.0% or top 0.1%, but the head of household is not an executive or in finance, then the spouse’s contribution to income growth will not be identified as being connected to executive pay or finance sector pay. The discussion in this paragraph draws on Bivens and Mishel 2013.

16. This follows from the fact that over 1979–2017 annual earnings rose by 22.2% for the bottom 90%, while the average growth across all earners was 40.1% (Mishel and Wolfe 2018). That means that the bottom 90% would have seen their earnings grow 17.9 percentage points more over the 1979–2017 period if they had enjoyed average growth (i.e., no increase in equality, 40.1 less 22.2).

17. Exercised stock options are considered W-2 wages so taxed as “income.” Stock awards are also taxed as income when vested.

18. We also remove a small amount of restricted grant awards ($494,000 in $2019) for 2006 that was included in our measure.

References

Baker, Dean, Josh Bivens, and Jessica Schieder. 2019. Reining in CEO Compensation and Curbing the Rise of Inequality. Economic Policy Institute, June 2019.

Bakija, Jon, Adam Cole, and Bradley Heim. 2010. “Job and Income Growth of Top Earners and the Causes of Changing Income Inequality: Evidence from U.S. Tax Return Data.” Department of Economics Working Paper 2010-24, Williams College, November 2010.

Bakija, Jon, Adam Cole, and Bradley Heim. 2012. “Job and Income Growth of Top Earners and the Causes of Changing Income Inequality: Evidence from U.S. Tax Return Data.” Department of Economics Working Paper, Williams College, April 2012.

Balsam, Steven. 2007. Executive Compensation: An Introduction to Practice and Theory. Washington, D.C.: WorldatWork Press.

Balsam, Steven. 2013. Equity Compensation: Motivations and Implications. Washington, D.C.: WorldatWork Press.

Bebchuk, Lucian, and Jesse Fried. 2004. Pay Without Performance: The Unfulfilled Promise of Executive Remuneration. Cambridge, Mass.: Harvard Univ. Press.

Bivens, Josh, Elise Gould, Lawrence Mishel, and Heidi Shierholz. 2014. Raising America’s Pay: Why It’s Our Central Economic Policy Challenge. Economic Policy Institute Briefing Paper no. 378, June 2014.

Bivens, Josh, and Lawrence Mishel. 2013. “The Pay of Corporate Executives and Financial Professionals as Evidence of Rents in Top 1 Percent Incomes.” Economic Policy Institute Working Paper no. 296, June 2013.

Bivens, Josh, and Lawrence Mishel 2015. Understanding the Historic Divergence Between Productivity and a Typical Worker’s Pay: Why It Matters and Why It’s Real 2015. Economic Policy Institute Briefing Paper no. 406, September 2015.

Bureau of Economic Analysis (BEA). Various years. National Income and Product Accounts (NIPA) Tables [online data tables]. Tables 6.2C, 6.2D, 6.3C, and 6.3D.

Bureau of Labor Statistics (BLS). Various years. Employment, Hours, and Earnings—National [database]. In Current Employment Statistics [public data series].

Clifford, Steven. 2017. The CEO Pay Machine: How It Trashes America and How to Stop It. New York: Penguin Random House.

Compustat. Various years. ExecuComp [commercial database].

Economic Policy Institute (EPI). 2019. “The Productivity–Pay Gap” (web page), updated July 2019.

Federal Reserve Bank of St. Louis. Various years. Federal Reserve Economic Data (FRED) [database].

Francis, Theo. 2019. “The New Pay Gap: What Firms Report Paying CEOs Versus What They Take Home.” Wall Street Journal, August 25, 2019.

Frydman, Carola, and Raven E. Saks. 2010. “Executive Compensation: A New View from a Long-Term Perspective, 1936–2005.” Review of Financial Studies 23, no. 5: 2099–2138.

Gould, Elise, 2020a. “The Labor Market Continues to Improve in 2019 as Women Surpass Men in Payroll Employment, but Wage Growth Slows.” Working Economics Blog, Economic Policy Institute, January 10, 2020.

Gould, Elise. 2020b. State of Working America Wages 2019: A Story of Slow, Uneven, and Unequal Wage Growth Over the Last 40 Years. Economic Policy Institute, February 2020.

Hall, Brian J., and Jeffrey B. Liebman. 1997. “Are CEOs Really Paid Like Bureaucrats?” National Bureau of Economic Research Working Paper no. 6213, October 1997.

Hodak, Marc. 2019. “Are Performance Shares Shareholder Friendly?” Journal of Applied Corporate Finance 31, no. 3: 126–130.

Hopkins, Matt, and William Lazonick. 2016. “The Mismeasure of Mammon: Uses and Abuses of Executive Pay Data.” Institute for New Economic Thinking Working Paper no. 49, October 12, 2016.

Internal Revenue Service (IRS). 2019. “SOI Bulletin Historical Table 12: Number of Business Income Tax Returns, by Size of Business for Income Years, Tax Years 1990–2016, Expanded Version” (data table). Excel file downloadable at https://www.irs.gov/statistics/soi-tax-stats-historical-table-12 (web page when updated December 13, 2018).

Kaplan, Steven N. 2012a. “Executive Compensation and Corporate Governance in the U.S.: Perceptions, Facts, and Challenges.” Martin Feldstein Lecture, National Bureau of Economic Research, Washington, D.C., July 10, 2012.

Kaplan, Steven N. 2012b. “Executive Compensation and Corporate Governance in the U.S.: Perceptions, Facts, and Challenges.” National Bureau of Economic Research Working Paper no. 18395, September 2012.

Kopczuk, Wojciech, Emmanuel Saez, and Jae Song. 2010. “Earnings Inequality and Mobility in the United States: Evidence from Social Security Data Since 1937.” Quarterly Journal of Economics 125, no. 1: 91–128.

Mankiw, N. Gregory. 2013. “Defending the One Percent.” Journal of Economic Perspectives 27, no. 3: 21–24.

Mishel, Lawrence. 2013a. “Greg Mankiw Forgets to Offer Data for His Biggest Claim.” Working Economics Blog (Economic Policy Institute), June 25, 2013.

Mishel, Lawrence. 2013b. “Working as Designed: High Profits and Stagnant Wages.” Working Economics Blog (Economic Policy Institute), March 28, 2013.

Mishel, Lawrence, Josh Bivens, Elise Gould, and Heidi Shierholz. 2012. The State of Working America, 12th Edition. An Economic Policy Institute book. Ithaca, N.Y.: Cornell Univ. Press.

Mishel, Lawrence, and Melat Kassa. 2019. “Top 1.0% of Earners See Wages Up 157.8% Since 1979.” Working Economics Blog (Economic Policy Institute), December 18, 2019.

Mishel, Lawrence, and Jessica Schieder. 2018. CEO Compensation Surged in 2017. Economic Policy Institute, August 16, 2018.

Mishel, Lawrence, and Julia Wolfe. 2018. “Top 1.0 Percent Reaches Highest Wages Ever—Up 157 Percent Since 1979,” Working Economics Blog, Economic Policy Institute, October 18, 2018.

Mishel, Lawrence, and Julia Wolfe. 2019. CEO Compensation Has Grown 940% Since 1978: Typical Worker Compensation Has Risen Only 12% During That Time. Economic Policy Institute, August 2019.

Murphy, Kevin. 2012. “The Politics of Pay: A Legislative History of Executive Compensation.” University of Southern California Marshall School of Business Working Paper no. FBE 01.11.

Sabadish, Natalie, and Lawrence Mishel. 2013. “Methodology for Measuring CEO Compensation and the Ratio of CEO-to-Worker Compensation, 2012 Data Update.” Economic Policy Institute Working Paper no. 298, June 2013.

Securities and Exchange Commission (SEC). 2015. “SEC Adopts Rule for Pay Ratio Disclosure: Rule Implements Dodd-Frank Mandate While Providing Companies with Flexibility to Calculate Pay Ratio.” Press release no. 2015-160, August 5, 2015.

Staff of Congressman Keith Ellison. 2018. Rewarding or Hoarding? An Examination of Pay Ratios Revealed by Dodd-Frank. May 2018.

Introduction

The central fact of our economy is the long-term decline of employment conditions over the past 40 years. Since the late 1970s, corporate profits, executive salaries, and shareholder returns have grown handsomely while wages of workers creating this prosperity have stagnated.1 Chief executive officer compensation grew 940% from 1978 to 2018, while typical worker compensation rose only 12% in that period.2 Even the low unemployment rate reached by 2018 had not been enough to spur truly significant wage growth, leading one economic analyst to declare that “the competitive supply-and-demand model of labor markets is fundamentally broken.”3 Workers have responded to falling wages by working longer hours.4 Thus, American workers find themselves working harder, running faster, and still sliding slowly backwards.

One of the primary causes of this growing economic inequality is the shrinking share of American workers who have a union in their workplace.5 When workers have the ability to bargain collectively with their employers, the division of corporate profits is more equally shared. On average, if one compares a union employee with a nonunion employee of the same gender, race, ethnicity, education, and years of experience, working in the same occupation, same industry, and same geographic area, the unionized worker’s wages are 13.2% higher than the nonunion counterpart. When the value of health and pension benefits are added in, the union pay advantage is greater still.6

Unsurprisingly, many nonunion workers wish that they too could earn union wages and benefits and access the other protections that come with unions. In a 2017 survey, 49% of nonmanagerial nonunion employees—who in the population at large represent roughly 58 million workers—told pollsters they would vote for having a union if given the opportunity to do so.7 Yet that same year, only 50,000 employees were able to establish a new union through National Labor Relations Board (NLRB) elections, or less than 1% of the number who want a union.8 What makes unions so rare despite being so popular? The fact that federal labor law is profoundly broken. Instead of serving as a neutral expression for workers’ preferences, the NLRB election system forces workers to run a gauntlet of fear, threats, intimidation, forced propaganda, and stifled speech. This is what must change for American workers to have a meaningful right to collective bargaining and for our country to find our way out of the crisis of economic inequality.

Current events make the need to reform labor laws even more urgent. The COVID-19 pandemic and the national protests around racial justice have heightened the importance of unions. As the pandemic swept across the country, thousands of nonunion workers walked off their jobs demanding personal protective equipment, hazard pay, and access to sick leave. The concrete realization that these things could only be won through collective action has also led many of these workers to seek to unionize in order to protect themselves and their families.9 At the same time, the importance of the power of collective bargaining for essential workers and Black workers has become clearer. Unionization has helped bring living wages to once low-wage jobs in industries such as health care and is a key tool for closing racial wage gaps.10 In recent years the Black Lives Matter movement has joined with the fight for a $15 minimum wage and other union efforts in order to win economic dignity for African American workers.11

Elections without democracy

As the world’s first modern democracy, the United States has long served as the global standard-bearer for defining what constitutes “free and fair” elections, including equal access to the voters for all political parties, equal access to the media, freedom of speech for both candidates and voters, and a guarantee that voters will not be financially bribed or coerced to support one candidate or another. People who first hear of union “elections” may assume these elections are conducted according to the same standards. However, the standard practice of anti-union employers makes NLRB-supervised elections look more like the discredited customs of rogue regimes abroad than anything we would call American. First, because there is no meaningful enforcement for violating voters’ rights, these rights are often violated. And those rights themselves are limited. There is, for instance, no right of free speech for voters in union elections. There is no equal access to media. Indeed, there is not even equal access to the names and contact information of eligible voters. And there is no protection against economic coercion of voters. Anti-union employers take advantage of the lack of rights in many ways, as the following sections show.

Finally, even when workers vote to unionize and that vote is legally certified by the NLRB, employers often continue to deny these employees the right to collective bargaining by refusing to negotiate a contract. As illustrated in the second of the case studies below, this can be accomplished through both illegal and legal means, including legal tactics that create multiyear delays, causing workers to lose faith in their own power and often leading activists to quit the employer. Again, the norms of American democracy require that winning candidates assume their positions at the appointed time; if there are challenges about the election, these are addressed at a later time, but legal delaying tactics cannot be used to perpetuate an incumbent’s rule after voters have elected to replace the incumbent with a challenger. But under the National Labor Relations Act (NLRA), even when employees vote for collective bargaining, the outcome of this vote may not be implemented for years, if at all.

Lawlessness at work: How employers undermine workers’ legal right to organize

The National Labor Relations Act of 1935 established the right to a union and collective bargaining for all private-sector workers. However, in the 85 years since the law was enacted, those rights have become increasingly unattainable. In 2018, only 6.4% of private-sector workers had unions.12

Workers’ inability to secure union representation is in large part a product of the rampant lawlessness that characterizes NLRB elections, made possible by the absence of meaningful penalties under the law. In elections for Congress, those who violate elections law may face fines, imprisonment, or loss of commercial licenses. But in NLRB elections, even employers who willfully and repeatedly break the law by threatening employees, bribing employees, destroying union literature, firing union supporters, or lying to federal officials in an effort to cover up these deeds can never be fined a single cent, have any license or other commercial privilege revoked, or serve a day in prison. As a result, it is not merely rogue employers who violate workers’ rights under law, but many mainstream employers who decide it is worth breaking the law in order to intimidate employees out of organizing a union.

A December 2019 EPI report highlighted the rampant lawlessness that characterizes workplace elections under the NLRB.13 In 2016–2017:

• Employers were charged with violating workers’ legal rights in 41.5% of all NLRB-supervised union elections.14
• Employers were charged with illegally firing workers in at least one-fifth (19.9%) of elections.
• In nearly a third (29.2%) of all elections, employers were charged with illegally coercing, threatening, or retaliating against workers for union support.
• Larger employers are even more likely than others to break the law: in elections involving more than 60 voters, more than half (54.4%) of employers were charged with at least one illegal act.

To put these findings in the context of what we normally expect from democratic elections, the Federal Elections Commission reports a total of 372 charges of illegal activity related to federal election campaigns in 2016–2017, or one charge for every 367,000 voters.15 In comparison, NLRB-supervised elections saw one charge for every 161 eligible voters.16 By this math, illegalities are more than 2,000 times more common in NLRB elections than in elections for the U.S. Congress or president. Such widespread intimidation recalls the worst of authoritarian regimes abroad; but these are the conditions that govern unionization elections in workplaces across the country.

Lawful but exploitive coercion: Employers spend $340 million per year on “union avoidance” consultants to deny workers the right to organize

Even when employers obey the law, they rely on a set of tactics that are legal under the NLRA but illegal in elections for Congress, city council, or any other public office. A $340 million industry of “union avoidance” consultants helps employers exploit the weaknesses of federal labor law to deny workers the right to collective bargaining.17

Over the past five years, employers using union avoidance consultants have included FedEx, Bed Bath & Beyond, and LabCorp, among others. Table 1, reproduced from an EPI report published in late 2019, lists just a few of these employers, along with the reported financial investments they made to thwart union organizing during the specified years.18

Table 1

These firms’ tactics lie at the core of explaining why so few American workers who want a union actually get one, and their success in blocking unionization efforts represents a significant contribution to the country’s ongoing crisis of economic inequality.

The lack of a right of free speech enables coercion

NLRB elections are fundamentally framed by one-sided control over communication, with no free-speech rights for workers. Under current law, employers may require workers to attend mass anti-union meetings as often as once a day (mandatory meetings at which the employer delivers anti-union messaging are dubbed “captive audience meetings” in labor law). Not only is the union not granted equal time, but pro-union employees may be required to attend on condition that they not ask questions; those who speak up despite this condition can be legally fired on the spot.19 The most recent data show that nearly 90% of employers force employees to attend such anti-union campaign rallies, with the average employer holding 10 such mandatory meetings during the course of an election campaign.20

In addition to group meetings, employers typically have supervisors talk one-on-one with each of their direct subordinates.21 In these conversations, the same person who controls one’s schedule, assigns job duties, approves vacation requests, grants raises, and has the power to terminate employees “at will” conveys how important it is that their underlings oppose unionization. As one longtime consultant explained, a supervisor’s message is especially powerful because “the warnings…come from…the people counted on for that good review and that weekly paycheck.”22

Within this lopsided campaign environment, the employer’s message typically focuses on a few key themes: unions will drive employers out of business, unions only care about extorting dues payments from workers, and unionization is futile because employees can’t make management do something it doesn’t want to do.23 Many of these arguments are highly deceptive or even mutually contradictory. For instance, the dues message stands in direct contradiction to management’s warnings that unions inevitably lead to strikes and unemployment. If a union were primarily interested in extracting dues money from workers, it would never risk a strike or bankruptcy, because no one pays dues when they are on strike or out of work. But in an atmosphere in which pro-union employees have little effective right of reply, these messages may prove extremely powerful.

Table 2 list the most common legal but anti-democratic tactics used to defeat union organizing.

Table 2

The outcome of elections without democracy

It is common for unionization drives to start with two-thirds of employees supporting unionization and still end in a “no” vote. This reversal points to the anti-democratic dynamics of NLRB elections: voters are not being convinced of the merits of remaining without representation—they are being intimidated into the belief that unionization is at best futile and at worst dangerous. When a large national survey asked workers who had been through an election to name “the most important reason people voted against union representation,” the single most common response was management pressure, including fear of job loss.24 Those who vote on this basis are not expressing a preference to remain unrepresented. Indeed, many might still prefer unionization if they believed it could work. Where fear is the motivator, what is captured in the snapshot of the ballot is not preference but despair.

To understand what union elections look like in reality, we have profiled two cases in which workers sought to create a union and met with a harsh (and typical) employer backlash. In both cases—a tire plant in Georgia and a satellite TV company in Texas—the employer response ranges from illegally firing union activists to engaging in acts of coercion and intimidation that are illegal in any normal election to public office but are allowed under the NLRA.

Kumho Tire defeated an organizing campaign even when 80% of workers wanted a union

In Macon, Georgia, one of the world’s largest tire manufacturers opened a new plant in 2016 that promised to bring hundreds of middle-class jobs to the economically hard-hit city. When promised wages failed to materialize, the workers began organizing a union; within six weeks 80% had signed cards calling for an election.25 In response, the company hired “union avoidance” consultants who ran a traditional—and very aggressive—anti-union campaign. In an NLRB election less than a month later, employees narrowly voted against unionization.26 What follows is an account of what initially led employees to organize and how management’s campaign drove so many of them to abandon that effort so soon after they’d begun.

Working at Kumho: Promise and reality

Macon has the most concentrated poverty of any city in Georgia.27 In 2016, nearly 40% of the city was living below the poverty line. Half of this population—or one in five Macon residents—was living on incomes that were less than 50% of the federal poverty threshold, or less than $12,150 per year for a family of four. Macon’s African American residents were especially likely to be living in poverty, making new job creation particularly essential for the city’s Black community.28

Thus, when Kumho Tires—one the world’s largest producers of car and truck tires—announced plans to build its first U.S. factory in Macon, the news was celebrated as a critical lifeline for the city. Kumho promised to create more than 450 new jobs, with the local press reporting that jobs would pay $45,000 per year.29 “We have extreme poverty in Macon,” one worker explained. “When Kumho came in, I thought ‘this is our chance to make a good living.’… Everybody’s hopes were up.”30

Unfortunately, as Kumho’s majority-Black workforce soon discovered, the Kumho jobs did not turn out as advertised. On their first day at the company, employees were told they’d start at $15 per hour, one-third less than what they’d been led to believe. However, employees were assured that this was a temporary wage while the company was setting up and that everyone would get pay increases once production got rolling.31 Yet months went by with no wage increase. When employees asked about their promised raises, managers explained that first they needed to sign up customers. After customer contracts were signed, managers declared there couldn’t be any raises until the plant was running at 80% capacity. “Every time we hit one of those goals, they changed the goalpost,” one employee reported.32 Ultimately, employees went a year and a half with no raise, and have never come close to the wage level promised when Kumho opened.33

Beyond wages, employees soon discovered that Kumho operated with few known rules, with many decisions seemingly based solely on favoritism. For instance, some jobs in the plant are louder, dirtier, harder, or more complicated than others but all pay the same.34 When workers in the harder jobs sought to switch positions, they discovered that there was no process for moving from one job to another: no seniority system, no application process, no action they could take other than appealing to the personal whim of supervisors.35 Employees likewise complained of arbitrary discipline. One woman reported that she was fired for being on break too long—an offense often overlooked with others. “They were just making up things,” explained a co-worker. “If they like you, you can miss as many days as you want. But if they don’t like you, as soon as you go over, you’re out the door.”36 So too, employees frequently worked around safety hazards without proper training or equipment. Employees in Kumho’s mixing department spent 12 hours a day working without respirators, covered in carbon black—a powdery substance classified as a chemical hazard and potential carcinogen.37 In 2019, OSHA issued a $523,000 fine against the facility, citing Kumho for 22 separate violations including unsafe use of carbon black.38 To make their jobs safer, better paying, and more reliable, employees turned to organizing a union.

In early August 2017, United Steelworkers Local 572 President Alex Perkins received a message from a Kumho employee asking to talk about creating a union at the new plant. They met at a restaurant, where Perkins explained the union couldn’t take action based on one employee’s interest and that the next test would be whether a larger group of workers was interested in organizing. The employee then asked Perkins to step outside, where 17 Kumho workers waited to talk with him.39 The following week, the Steelworkers gave out 60 union cards to a group of four workers, thinking they might sign up 60 supporters in a month, or perhaps never; instead, they were all signed by the next day. Organizing proceeded rapidly, and on September 18, the workers petitioned for an NLRB election with 250 signed cards, representing nearly 80% of the plant’s workforce.

Management’s anti-union campaign

Almost immediately after the employees’ petition was filed, Kumho hired a consultant associated with Labor Relations Institute, one of the country’s most prolific union avoidance firms. The company instituted daily compulsory anti-union meetings running up to 90 minutes.40 Production was shut down and all workers were required to listen to managers and consultants delivering anti-union speeches, videos, and powerpoint presentations.41

Outside the meeting rooms, the workplace was saturated with one-sided propaganda. “The whole place was covered in anti-union posters,” one employee recalled. “’Vote No.’ ‘USW loves strikes.’ ‘USW plants closed down.’”42 In addition, “anti-union videos played 24/7 on flat screens that management put up in the employee entrance to the plant, at the security gates, in the cafeteria and in break rooms.… Any time you went on break or to the bathroom, they were in your face.”43

The company had hundreds of “Vote No” hats made, with supervisors encouraging their subordinates to don them. Kumho employs a large number of temporary employees who do the same work as, and are generally indistinguishable from, regular employees. Because temporary employees could be terminated at will and lacked the same legal protections afforded regular employees, they followed management’s wishes.44 The hats were meaningless for the temp workers, who couldn’t vote in the election. But as temporary employees donned the hats, they flooded the workplace with a sea of apparently anti-union sentiment, and most workers were unable to distinguish between hat-wearers who were temporary employees and those who were permanent employees. Thus the hats created an atmosphere of momentum for the anti-union cause and intimidation for pro-union employees.

Finally, in addition to the mandatory group meetings, workers report that each employee was personally and repeatedly addressed by their direct supervisor.45 One employee recounts that “my supervisor would stop me every day and say, ‘How you gonna vote, man? You gotta vote no!’ No question I was going to vote yes, but I couldn’t let them know that, ’cause I was in fear for my job security.”46 In extreme cases, employees were double-teamed, with two managers simultaneously talking against the union from either side of one’s workstation. One employee subject to this treatment “got to the point where he would tell them whatever they wanted to hear,” just to make the treatment stop.47

The role of fear in Kumho’s anti-union campaign

At the heart of management’s campaign was the threat that workers would lose their jobs if they voted to unionize. Under the NLRA, it is legal for employers to “predict” that they will shut down if workers organize, but illegal to “threaten” closure. Insofar as they scare workers out of organizing, there is no significant difference between these, and employers often issue a combination of illegal threats and technically legal predictions. In Kumho’s case, an administrative law judge of the NLRB ultimately determined that 12 different managers (including the company’s CEO) issued illegal threats to close the plant or lay off employees.48

On the day employees filed their election petition, three different managers told their subordinates that Kumho would shut the plant down if they voted the union in.49 Over the course of the next month, such threats were repeated daily. In mandatory whole-plant meetings, in departmental pre-shift meetings, and in supervisors’ one-on-one conversations with subordinates, workers were told that if they voted to organize, Kumho’s contractors would cut them off, the parent company would close the plant, the equipment would be shipped back to Korea, and everyone at the Macon facility would be out of work.

The election period was described by one Kumho employee simply as “Hell. Intimidation. Every day managers asking you, ‘What are you going to do.’ Some people trying to stay strong, other people trying to hide whatever they thought, and others trying to get brownie points by running up to management to tell them everything that was said in union meetings.”50

A lot of employees shut down in the face of intensive supervisory campaigning. “As soon as you come in, every day they were in your face,” one worker recalled. “A lot of people couldn’t handle it.”51 Even one of the core union leaders found the pressure unbearable: “I kept having them come up to me, demanding, ‘Can I count on you for a no vote?’ One day he kept at it for 10 minutes straight, so I told him ‘yes,’ just to get him off my back.”52 Likewise, many employees who had identified themselves to co-workers as union supporters nevertheless wore Vote No hats out of fear that refusing to do so would alienate their supervisor.53

In union elections, everyone is looking around to see who else is in; to judge whether they believe that their co-workers are capable of coming together to force management to pay more than it wants.54 A large collective action—such as everyone wearing union buttons or t-shirts—gives employees confidence that they have the collective power to stand up to management. For the same reason, management strategies focus on intimidating union supporters into silence and quiescence—or into open betrayal of their cause—in order to convince others that even union supporters lack the fortitude to unite in opposition to management. At Kumho, such incidents were plentiful.

One union leader recalls a friend, known to co-workers as a strong union supporter, who wore a Vote No hat in order to get his employer to stop harassing him. “Other people who looked to this guy as a strong union supporter got the message that management can cow anyone.”55 Another Kumho worker recalled a pro-union employee who ended up making an anti-union speech in a departmental meeting. This employee later apologized and swore that he was still a union supporter, but felt he had to take this step to secure his job.56

At one point, the employees considered organizing a day when supporters would all wear union buttons to work, but people were too fearful to carry this out.57 One of the core union leaders explained how even he came to believe wearing buttons was dangerous.

At first, I didn’t care who knew I was for the union. But then, once the pressure got on me, I also heard from another employee that in HR they had a list of people who were for the union. I thought, ‘I’m gonna end up getting fired.’ So I started laying low. I tried not to talk about it…. I felt my job was at risk. We weren’t about to wear a button supporting the union.58

Kumho’s active campaign of disinformation

When Kumho employees petitioned for a union election, 80% had signed cards supporting unionization; less than a month later, only 43% voted to organize, with 52% opposed.59 When asked how it is possible for political opinion to shift so dramatically in such a short period, management consultants typically assert that this reflects a process of objective education: in the pre-petition period workers only heard the union side of the story; once presented with all the facts, they turned against unionization. However, the messages conveyed by Kumho’s managers and consultants were not objective facts. Rather, at the heart of the anti-union effort was an active campaign of disinformation. Specifically, the company’s messaging focused on four demonstrably false assertions: (1) that the plant would shut down if workers organized; (2) that Kumho’s customers wouldn’t buy from a union supplier; (3) that Kumho was already paying above-average wages for the area; and (4) that workers would be forced to pay exorbitant union dues if they voted to organize. Each of these is briefly explained below.

False threats of plant closing

Kumho managers repeatedly hammered workers with the assertion that, if they voted to organize, all their jobs could be shipped overseas.60 But these threats fly in the face of the company’s $450 million investment in this state-of-the-art facility and its own declarations regarding the strategic value of its Macon facility.61 North America accounts for 22% of the global tire market, and the factory site—just off an interstate highway—makes for easy transportation to auto manufacturers and dealers in the South and throughout the U.S.62 When the facility opened, a company spokesperson said, “It helps us dramatically to be here” because “the U.S. tire market is the biggest in the world” and “shipping from China or Vietnam is an expensive process.”63 The chairman of Kumho’s parent corporation went further, declaring that the Macon plant was “a must if we are to significantly expand our presence in the market.”64 The idea that the company would walk away from this critical investment rather than negotiate with its employees is simply not plausible. But it was nevertheless asserted, repeatedly and strenuously, throughout the campaign.

Threats that Kumho’s customers wouldn’t work with unionized suppliers

Consultants and supervisors repeatedly told employees that the auto manufacturers that Kumho sells to don’t want to do business with unionized suppliers, specifically naming Hyundai and Kia as South Korean car manufacturers that would not want to contract with a unionized tire manufacturer. But this is false: both Hyundai and Kia contract with Goodyear and Firestone—both unionized companies—for their tires. When one worker challenged a consultant on this point, the consultant backpedaled, insisting that “I’m just saying it’s a possibility.”65 But the assertion was repeatedly made without any such qualification—and taken to heart by many workers who had no source of alternative information with which to evaluate management’s claims.

Threats that wages and benefits could go down with a union, and assertion that wages average only $11 in Macon area

It is a standard theme of anti-union consultants to tell employees that unionization might result in lower wages and benefits. Employers typically stress that the process of collective bargaining starts “from scratch,” with no guarantee of maintaining current wage and benefit levels.66 It is true, of course, that bargaining begins without preconditions. But in a nonunion workplace, wage and benefit levels always start “from scratch.” Presumably, wages in a nonunion firm represent the lowest management thinks it can pay and still attract the quality of labor it needs. It is nonsensical to think the addition of a union could result in lower wages. Indeed, in advertising its union avoidance services to employers, LRI notes that unions typically increase compensation by 25%.67

At Kumho, this misleading argument was reinforced with falsified data. In whole-plant meetings, the company’s consultant told employees that Bureau of Labor Statistics (BLS) reports showed that the average wage for Macon-area production workers was $11.35 per hour, and that therefore if workers voted in a union, the company would enter negotiations proposing to pay everyone $11 rather than the $15 most employees were then earning.68 “She told us we could lose a lot if we vote in the union,” one employee recalled, “because she said the company is already paying us a lot more than we’re supposed to be paid.”69 Not only was the claim illogical; it was false. In later questioning before an administrative judge, Kumho representatives acknowledged that BLS reported average production wages of $17.17 per hour in the Macon area.70

False claims about mandatory dues in a right-to-work state

A standard message of anti-union campaigns is that unions are businesses that exist only to coerce dues out of workers, and that the dues workers pay will amount to more than the benefits they derive from a union contract.71 This argument doesn’t make sense anyplace in the country—a union contract only goes into effect if workers vote to ratify it, and workers would not vote to ratify a contract whose terms will make them worse off. But the argument makes least sense in a state like Georgia where a state law requires that employees be granted the full benefits of a union contract even if they don’t contribute dues to cover the cost of maintaining that contract (such laws, misleading labeled right-to-work or RTW laws, exist in many states).72 RTW laws mean that no worker can ever be required to pay union dues. If there were a union at Kumho, dues would be entirely voluntary.

Nevertheless, communications from management sought to convince workers that they would be forced to pay dues against their will. One management poster informed workers, “Even if you do not participate in the Union activity, you will be required to pay dues which amounts to as much as…3% of your pay.”73 Similarly, the company’s website declared that

If the union wins the election, it will be speaking for all employees in the bargaining unit as a group, regardless of how an individual employee voted. When they can, unions want a ‘dues check-off’ provision in a contract, which means that dues are automatically deducted from employees’ paychecks.74

Why voters believe false information: Free speech stifled

The previous section detailed Kumho’s campaign of disinformation. Why would Kumho employees believe such demonstrably false claims from management? Above all, because they are bombarded by one-sided information, with little or no opportunity to hear from the union.

Democracy relies above all on free speech to create an informed citizenry. But under current law, it is impossible for most workers to have balanced information. Pro-union employees are only permitted to talk about their cause while on break time and in break areas. But at Kumho, supervisors were often present in both the cafeteria and break rooms, making these spaces feel unsafe. And given the company’s open hostility, most employees were scared to be associated with known union activists.

Indeed, even home visits by union organizers or pro-union workers proved difficult to arrange. Many Kumho employees live far outside of town. With employees working 12-hour shifts, it’s difficult to find a time when people are home and available to talk.75 The union office is just five miles from the Kumho plant, and organizers invited workers to meet there before or after their shifts. But, according to multiple employees, workers’ fear of retaliation from Kumho management soon led them to be wary of being seen meeting with union organizers—even away from company property. At first, between 20 and 25 employees came to meetings at the union office at the end of each shift. Over the next two weeks this number fell by 75%, and those who showed up were noticeably more fearful.76

USW Local 572 President Alex Perkins said he “started to see more people backing into parking spots, and couldn’t figure out why” until he realized that employees were shielding the back of their trucks from view out of fear that someone might drive by, record license plates, and report their presence to Kumho management.77

As a result of all these actions to stifle speech, most employees had little or no opportunity to hear the union’s arguments. The union estimates that 35% of Kumho’s employees never spoke with a union representative outside the workplace; an additional 30% had just one conversation; and only 35% had more than one such conversation.78 Above all, it appears that it was this one-sided control of speech that made so many employees vulnerable to disinformation.

The ‘no’ vote and its aftermath

In September 2017, employees at Kumho petitioned the NLRB for a union election, with 80% of workers having signed cards indicating their support for unionization. Less than one month later, employees voted 164–136 against unionization. Almost immediately following the election, the union filed charges with the NLRB, alleging that Kumho had broken the law by threatening and intimidating workers into voting no.79

Why did Kumho employees vote no?

The central question of the Kumho experience is how pro-union employees were so quickly led to vote against organizing. Some were likely driven to vote no simply out of desperation to end the state of heightened tension in the workplace. To the extent that management’s behavior convinces workers that a vote to unionize is a vote to permanently turn the workplace into a psychological war zone, the only way to return things to normalcy is to vote “no.”80

Above all, employees reported, people voted no because management convinced them that their jobs were at risk.81 Although Kumho’s wages were below the company’s competitors and below what was promised, they were still higher than many of the jobs otherwise available in this high-poverty region. “There are a lot of people around here making $9.50 an hour,” explained one employee. “A lot of people had never even made what they were making here. So they feel like, ‘They tell me they’re going to shut this job down. I can’t mess this up.’”82 “People on my shift voted no because they were afraid,” agreed a former employee. “Team leaders were saying if a union comes in Kumho is going to shut down.”83 “They hammered at people,” another added. “It’s that every day on your job [they say] ‘We really need you to vote no. We’re gonna go out of business if you vote yes.’ People believed them. People were afraid to lose their job.”84

One employee who signed a union card and then voted no was convinced by their supervisor that unionization might drive Kumho out of business. Despite its being a $2.5 billion multinational corporation that has been making tires for more than 50 years, this employee’s manager insisted that Kumho was a “startup” company and might not survive unionization.85 Following this conversation, “I decided I wanted to give the company a chance.”86 Eighteen months later, the employee had become disillusioned. “Every promise they gave us turned out to be a lie,” this employee explained.87 A co-worker reported going through a similar process—signing a union card and then voting “No” because their supervisor claimed Kumho would shut the plant down if a union was voted in, and convinced the employee to give the company “another chance.” Both these employees came to regret their choice, and subsequently signed new union cards.88

The challenge of a second election

In May 2019, a judge ruled that Kumho had committed “numerous, severe” violations of federal labor law, and ordered a new election to be held.89 Yet it is very difficult for employees to come together in a renewed unionization effort after experiencing such widespread intimidation. Many of the strongest union supporters were either fired or quit following the 2017 election.90 And the experience of watching activists punished has left a deep mark on remaining employees. “If you wear a button,” one current employee asks, “what’s to say they won’t fire you over it?”91

Indeed, there is little reason for employers facing rerun elections not to engage in the same illegal intimidation a second time, since the likely remedy for that would simply be a third election.92 This logic was long ago spelled out by a prominent “union avoidance” consultant:

What happens if you violate the law? The probability is you will never get caught. If you do get caught, the worst thing that can happen to you is you get a second election and the employer wins 96% of second elections. So the odds are with you.93

“Right now,” a Kumho worker reports, “everybody is too scared to be openly pro-union. People are scared to be seen meeting at the union hall, or even to meet at McDonald’s.”94 As in authoritarian regimes abroad, when the fear of retribution makes voters too scared to publicly demonstrate what they believe, even a secret ballot cannot guarantee democracy. In the second election in September 2019, Kumho workers voted 141‒137 in favor of unionization, with 13 challenged ballots.95 Both sides are awaiting an NLRB ruling to determine the election’s outcome.

What’s at stake: Union versus nonunion wages in Macon

This case study follows the initial defeat of a union organizing campaign in Macon followed by second election for which the outcome is uncertain. The outcome has real consequences. A union could make a dramatic difference in the lives of Macon workers. At Kumho, employees report a typical wage of $15 per hour; at this rate, a worker with three dependents still qualifies for food stamps.96 By contrast, if employees were able to raise their pay even just to the standard of entry-level employees at a nearby unionized packaging plant, they’d be making nearly $46,000 per year—just about what the Kumho jobs were promised to pay when the plant first opened, and a wage at which no one would need public assistance.97 There is good reason to believe that Kumho could afford to pay significantly better wages. Many of Kumho’s competitors are unionized, and pay well above Kumho’s rates.98

One union supporter who was terminated following the election laments having left, despite having landed in a better paying position. “I didn’t want to leave,” he explains. “I wanted to make Kumho a better place. This was 400 jobs that could help out 400 families in poverty in our community…. I’m in a great situation now but that’s not helping those people still there.”99

How DISH TV denied employees’ right to collective bargaining even after they won their union election

In 2009, 100 technicians and warehouse workers in two Dallas-area branches of the DISH TV corporation decided to organize a union. They affiliated with the Communications Workers of America (CWA), won their election, and had their union certified by the NLRB. But a decade later, they still have no contract. The employer—relying in part on advice from the anti-union Jackson Lewis law firm100—engaged in a series of tactics including legal delays, management intimidation, and economic retaliation. Some of these acts were legal and some were not; together, they illustrate how the weakness of current labor law allows employers to effectively block workers’ right to collective bargaining even after they have won an election.

DISH TV, headquartered in Englewood, Colorado, is one of the nation’s leading providers of satellite pay TV, with 12 million subscribers and annual revenues of approximately $13 billion.101 In 2009, the company announced that it was cutting technicians’ hourly pay and switching to an incentive pay system based on employee productivity. Employees felt they had already put up with “too many changes,” and viewed the pay cut as the last straw.102 In response to the company’s announcement, employees in DISH’s North Richland Hills and Farmers Branch offices in Texas began organizing. Workers signed cards saying they wanted a union election, the union filed the petition for an election, and then the NLRB scheduled a vote.103 Support for the union was widespread, and in 2010 a majority of employees in both locations voted to unionize. In theory, this is the point at which management is required to honor employees’ choice and sit down to begin good-faith negotiations toward a contract. Instead, it marked the beginning of a decade-long war of attrition aimed at undermining the union and thwarting employees’ ability to negotiate a fair contract.104

Immediately following the election, DISH filed complaints with the NLRB asserting that the CWA staff person representing employees at the North Richland Hills location was not an appropriate authority to file an election petition (despite his being the director of the local union and the fact that the NLRA does not require a specific type of person to file a petition) and contending that actions by a pro-union employee had tainted the election, even though the actions in question took place after voting was completed.105 Under the NLRB’s rules in effect at the time, the Labor Board was required to engage in a detailed hearing and lengthy delay in order to consider management’s objections. Ultimately, both complaints were dismissed, but it took 18 months to legally verify these facts, delaying certification of the election outcome.106 It wasn’t until November 2011 that negotiations for both units were ready to begin.

However, just prior to the point at which negotiations would have begun, a petition was filed by some employees at the Farmers Branch location—supported by management— calling to decertify the union.107 Due to ongoing charges of illegal management activity, this vote was put on hold for nearly three years.108 The vote on whether to decertify the union finally took place in May 2014; the employees reaffirmed their desire for unionization, and the CWA was recertified as the workers’ union the following month. Management’s legal tactics thus succeeded in forestalling negotiations for nearly half a decade. Five years after first organizing, these workers were back at the beginning, ready to begin the collective bargaining process in mid-2014.

Just six months later, however, DISH put a halt to negotiations. The company announced that it was switching labor attorneys, and that the new attorney would contact the union in 2015. Instead, more than a year went by with no communication from the company. In early 2016, the union repeatedly asked the company to schedule bargaining dates, but was rebuffed. Instead, in April 2016 the company declared that the two sides were at an impasse—despite its new attorney never once having met with the union.109 By law, “impasse” can only exist when both parties have concluded there is no more compromising to be made. It’s not possible to reach that point without extensive negotiations, and indeed the NLRB subsequently found that DISH’s declaration of an impasse was illegal. But that process took two years. In the meantime, DISH invoked the bogus impasse declaration as a pretext to dramatically slash workers’ wages—an act significant both for its own sake and for its impact on demoralizing workers and undermining belief in the union.

As in many unionization drives, DISH’s pay system constituted a central issue in negotiations. After the company unilaterally imposed its incentive pay system in 2009, DISH employees gradually adapted to the new productivity measures, earning significant wage increases.110 In response, DISH notified employees in 2016 that it was now abolishing incentive pay and cutting compensation to a new and lower flat rate—constituting a paycut of up to 50% for current employees.111 At the same time, employees’ health insurance deductibles were doubled.112

Unsurprisingly, the impact of these cuts was severe. Multiple employees reported having to work second jobs, borrow from their relatives, or take out high-interest payday loans to keep their families afloat.113 “We are at the point we may have to put the house up for sale,” explained one employee. “My parents have been donating money to me because my wife…was pregnant…now she is applying for jobs as well.”114 Another described simultaneously working full-time at DISH, going to school full-time to qualify for a higher-paying job, and taking on a graveyard-shift job to make ends meet. Employees described co-workers who moved in with their in-laws, had their cars repossessed, put their 401(k) retirement funds up as loan collateral, or stopped answering their phone in hopes of avoiding bill collectors.115

Beyond their direct economic impact, these wage cuts also served to undermine employees’ bargaining power. By the end of 2016, 19 DISH employees had been forced to quit, including most of those who had served as members of the negotiating committee.116 In one location, workers were left with no union steward and no representative on the bargaining committee, as veterans left and other employees were too scared to step up as replacements. One long-term union supporter explained simply that he didn’t volunteer for a leadership role because “I didn’t want to get fired.”117 And as new employees came on to replace those driven out, management threatened veteran employees with termination if they discussed union issues with the newcomers.118

The ability of any group of workers to win a fair contract rests primarily on their ability to disrupt production or otherwise pressure the employer through collective action. In a typical campaign, workers engage in acts of solidarity within the workplace—wearing buttons or t-shirts, signing petitions, or holding group meetings. Often their actions escalate to public outreach activities including leafleting, picketing, staging boycotts, and making appeals to elected officials; ultimately building toward a potential strike. All of these actions depend on employees being united in commitment to their cause, and having confidence in the power of their own collective action.119 But each of the steps taken by DISH management served to make this type of collective action impossible. The experience of multiyear delays saps the momentum of an organizing drive and undercuts employees’ belief in the power of their own action. Unilateral wage and benefit cuts function as an object lesson in the omnipotence of management and the futility of organizing. Forcing union activists to quit deprives workers of their natural leaders and makes all other employees witness the union’s strongest supporters resignation in the face of management retaliation, and the same retaliation makes others scared to play leadership roles. Finally, the dramatic level of turnover means that many current employees have no knowledge of the union, and with veteran employees banned from educating them, there is little possibility of current employees ever gaining such knowledge. Under such conditions, it is virtually impossible for workers to take the sorts of collective action needed to convince management to sign a fair contract.

By early 2017, a CWA staff organizer reported that the union was “on the brink of losing all support…. We’ve seen people quit and people just not supporting us as they’ve already lost…trust.”120 The union’s regional director reported that “our members…feel defeated…. They feel like they were lied to, demoralized, betrayed.”121

In June 2018—after multiple hearings in federal court—the NLRB issued an order finding that DISH had committed multiple and egregious illegal acts, and required the company to reinstate incentive pay for technicians at its two unionized facilities, to reinstate the preexisting health insurance plan, to provide back pay for those who lost wages or were forced to pay higher health insurance premiums, to offer reinstatement to those who’d been forced to quit, to read aloud a statement pledging to respect the law, and to return to the bargaining table.122 Following this ruling, the company resumed negotiations in mid-2018, but appears to have continued to seek at every turn to undermine employees’ strength and unity within the workplace.

In late 2018, for instance, DISH reinstated three employees who had been forced to quit, but announced that it would lay off current employees to do so, sending letters notifying employees that they were losing their jobs due to “an order of the National Labor Relations Board.”123 Similarly, when the company offered merit raises to warehouse workers at its other locations, it withheld such raises from its unionized staff, with a manager falsely telling employees that they were being denied raises “because of your union.”124 As the drafting of this report wrapped up, the union and management were once again in negotiations; but it was unclear whether management had become more interested in signing a contract.

The NLRB’s order was designed, in theory, to rectify DISH’s illegal actions and restore an even playing field for negotiations; but this goal was unrealistic. The NLRB reasoned that requiring management to read its pledge to honor the law out loud to employees “will counteract the coercive impact of the [labor law] violations.”125 But when workers know that even if the company breaks its promise, no manager nor the corporation itself can ever be fined, jailed, lose its license or suffer any other penalty for illegal behavior, there is no reason for employees to trust that it is safe to join the bargaining committee, identify themselves as a union steward, or go on strike.

Employees’ fear of retaliation and their experience of management intransigence are not changed by reading a statement. Nor will reading a statement bring back the union supporters and leaders who quit when their wages were cut or give the newly hired the experience of organizing a union. What is needed is not an empty promise read out loud, but a legal requirement to settle a fair contract. Labor advocates for decades have proposed a system of neutral arbitration for settling union contracts—particularly first contracts—when the parties cannot reach agreement through traditional means.126 In the DISH case, this is sorely needed: the company has ignored the law for so long, and has so severely undermined workers’ ability to create an effective balance of bargaining power, that it is impossible to believe in the promise of a do-over election backed by a toothless pledge of good faith.

From the outside, it may seem puzzling that DISH has devoted so much time, energy, and money to fighting this small group of unionized employees. The company has never offered evidence that it can’t afford union wages; nor have its bargaining representatives ever claimed an inability to pay.127 Most tellingly, at the Dallas-area offices of AT&T DirecTV—a competitor company providing the same services as DISH but where employees have a union contract—technicians’ compensation is 50% higher than at DISH, and warehouse workers’ pay is almost twice that of equivalent DISH employees.128 If competitors operating in the same industry and same geographic market are able to remain profitable while paying higher wages, there is no reason to believe DISH could not do likewise.

Given that that union wages are affordable, why has DISH run such a concerted campaign—over so many years—to avoid signing a collective bargaining agreement with its employees? It may be that DISH’s corporate management was less concerned about wages at these two locations than they were about the potential for the example set by union-led wage increases to spread to others of the corporation’s locations. DISH employs 16,000 people across the United States.129 It’s hard to imagine that many of these workers wouldn’t respond to the knowledge that people were working the same jobs at significantly higher wages by demanding similar compensation for themselves.130

While it may be hard to justify the time and money spent on fighting the union in terms of the costs of this group of 100 employees, this expense may be more logical when understood as an investment in preventing higher wages from spreading to the rest of its workforce.

For the public, the spread of union wages throughout the DISH network would be welcome news indeed. The difference between DISH’s current wage scale in these facilities and the compensation earned by Dallas-area AT&T DirecTV employees doing the same work under a union contract is the difference between wages that may force one to rely on food stamps and wages providing a modest middle-class income.131 Nationwide, 16,000 DISH employees could be living in greater security and providing for their families—and, as evidenced by AT&T, the corporation would still be profitable. By devoting such intensive resources and dedication to stomping out any potential beachhead of unionization, DISH is focused instead on keeping those 16,000 employees locked into financial hardship and insecurity.

Conclusion: We must restore an effective right to organize to solve the crisis of inequality

We must restore the ability of American workers to negotiate with their employers if we are to reverse the growing crisis of inequality. Through a combination of legal intimidation and rampant lawbreaking, employers have made a sham of the theoretical right to collective bargaining.132

Employers’ ability to effectively subvert federal labor law is one of the primary barriers that stands in the way of creating an economy in which American workers can support their families in dignity. To once again make that right a reality, Congress must act to ensure that workers have a right to vote to unionize in an atmosphere defined by free speech and open communication, and without fear of retaliation for one’s beliefs.

The U.S. House of Representatives took an important step in the direction of restoring workers’ collective bargaining rights when it passed the Protecting the Right to Organize (PRO) Act in February 2020.133 Some of the most damaging tactics used by employers to oppose union organizing efforts would be restricted under the legislation, and meaningful penalties would be imposed when employers violate the law. Recently, a group of 70 economists, academics, and labor leaders led by Harvard Law School’s Labor and Worklife Program have called for even more reforms to U.S. labor law. The package of proposals, called the Clean Slate for Worker Power, includes extending labor rights to farmworkers, domestic workers, and independent contractors who are now excluded from federal union rights; mandating a national requirement that employees may only be fired for just cause rather than arbitrarily; prohibiting employers from permanently firing workers who engage in a strike; and enabling workers to engage in sector-wide negotiations rather than single-employer bargaining.134 All of these measures would be welcome steps toward restoring balance and effective democratic standards in federal labor law.

The evisceration of the American middle class, the increasing hardship that most nonprofessional workers face, and the escalating inequality that has come to define our country are not facts of nature over which we lack control. Congress has the power to reverse this decline and to restore the promise of the American economy by insisting that employers respect the right of workers to organize free from fear or intimidation. Ensuring the right of workers to organize and bargain collectively to secure a fair share of the prosperity that their work creates is the most important step that federal lawmakers can take to restore fairness in the American workplace and broad prosperity across our country.

About the authors

Gordon Lafer is an EPI research associate and a political economist and professor at the Labor Education and Research Center at the University of Oregon. Lola Loustaunau is a Ph.D. candidate and research fellow at the Labor Education and Research Center.

Acknowledgments

The authors thank Communications Workers of America for its generous support of this research, and Celine McNicholas, Heidi Shierholz, Julia Wolfe, Ben Zipperer, and John Lund for their contributions to this research.

Endnotes

1. Elise Gould, State of Working America Wages 2019, Economic Policy Institute, February 2020.

2. Lawrence Mishel and Julia Wolfe, CEO Compensation Has Grown 940% Since 1978, Economic Policy Institute, August 2019.

3. Noah Smith, “Econ 101 No Longer Explains the Job Market,” Bloomberg, April 5, 2018.

4. For example, the average woman now works 310 hours more per year—the equivalent of almost eight weeks of full-time work—than she would have in 1979. See Valerie Wilson and Janelle Jones, Working Harder or Finding It Harder to Work, Economic Policy Institute, February 2018.

5. See, for example, T. Shin, “Explaining Pay Disparities Between Top Executives and Nonexecutive Employees: A Relative Bargaining Power Approach,” Social Forces 92 (2014), pp. 1339–1372.

6. The data refers to the regression-based gap between workers covered by a contract (who are mostly but not exclusively union members) and workers not covered by such a contract. Josh Bivens et al., How Today’s Unions Help Working People: Giving Workers the Power to Improve Their Jobs and Unrig the Economy, Economic Policy Institute, August 2017.

7. Thomas Kochan, William Kimball, Duanyi Yang, and Erin Kelly, “Voice Gaps at Work, Options for Closing Them, and Challenges for Future Actions and Research,” MIT Sloan School of Management Institute for Work and Employment Research working paper, June 2018. In the survey, unionized workers were those who said that they were “represented by a union or a professional association” on their job.

8. In fiscal 2017, new unions were certified in 868 NLRB-supervised elections, with a combined total of 47,278 eligible voters. Election filings data for 2016–2017 were obtained from the NLRB through Freedom of Information Act (FOIA) requests NLRB-2018-001366 (submitted 9/26/2018, completed 10/26/2018) and NLRB-2019-000178 (submitted 11/28/2018, completed 12/27/2018). NLRB records sometimes make it difficult to identify duplicate records; thus it is possible that some of the elections we counted were duplicates, and the actual total of newly organized workers is less than that reported, but for purposes of this report we use this number as an upper-boundary estimate.

9. See Steven Greenhouse, “Stop Throwing Us Bones: Union Activism Surges amid Coronavirus,” The Guardian, May 6, 2020; Celine McNicholas, “In Midst of a Pandemic, Trump’s NLRB Makes It Nearly Impossible for Workers to Organize a Union,” Working Economics Blog, March 31, 2020; and Celine McNicholas and Margaret Poydock, “Workers Are Striking During the Coronavirus,” Working Economics Blog, June 22, 2020.

10. Josh Bivens et al., How Today’s Unions Help Working People: Giving Workers the Power to Improve Their Jobs and Unrig the Economy, Economic Policy Institute, 2017.

11. Mike Elk, “Justice in the Factory: How Black Lives Matter Breathed New Life into Unions,” The Guardian, February 10, 2018.

12. Bureau of Labor Statistics, “Union Members Summary” (news release), January 18, 2019.

13. Celine McNicholas, Margaret Poydock, Julia Wolfe, Ben Zipperer, Gordon Lafer, and Lola Loustaunau, Unlawful: U.S. Employers Are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns, Economic Policy Institute, December 2019.

14. It is important to note that our finding almost certainly underestimates the true frequency of illegal employer threats and terminations. Much of employers’ illegal behavior goes unrecorded, because workers are too scared to press charges. Indeed, Kate Bronfenbrenner’s 2009 survey of union organizers for EPI and the American Rights at Work Education Fund found that more than 75% of employers were believed to have broken the law at least once during the course of the campaign, almost double the number of cases in which charges were filed. See Kate Bronfenbrenner, No Holds Barred—The Intensification of Employer Opposition to Organizing, Economic Policy Institute and American Rights at Work Education Fund, May 2009, pp. 5–6.

15. Data on Federal Election Commission enforcement comes from U.S. Federal Election Commission, “FEC Enforcement Statistics 1977–2019,” last updated October 17, 2019. The estimate of charges of illegal activity per voter is based on the fact that more 136 million voters participated in the fall 2016 election; see Federal Elections Commission, “Official 2016 Presidential General Election Results,” January 30, 2017.

16. In fiscal 2016 there were just 1,193 representation elections conducted under the NLRB, with just under 80,000 eligible voters. Election data for 2016–2017 was obtained from the NLRB through FOIA requests NLRB 2018-001366 and NLRB 2019-00178.

17. Celine McNicholas, Margaret Poydock, Julia Wolfe, Ben Zipperer, Gordon Lafer, and Lola Loustaunau, Unlawful: U.S. Employers Are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns, Economic Policy Institute, December 2019.

18. EPI analysis of the union avoidance industry is built on an examination of publicly available forms LM-20 and LM-21 filed with the U.S. Department of Labor’s Office of Labor-Management Standards (OLMS). It is worth noting that, due to loopholes in reporting requirements, these amounts may represent only a fraction of these companies’ anti-union investments; for more information, see the methodological appendix in McNicholas et al., Unlawful: U.S. Employers Are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns, Economic Policy Institute, 2019.

19. The NLRB has ruled that employers have “no statutory obligation to accord the employees the opportunity to speak” at such meetings. Hicks-Ponder Co., 168 NLRB 806, 814 (1967). In Litton Systems, Inc., 173 NLRB 1024, 1030 (1968), the NLRB supported an employer who fired an employee for discreetly leaving a captive audience meeting, affirming that employees have “no statutorily protected right to leave a meeting which the employees were required by management to attend on company time and property to listen to management’s noncoercive antiuion [sic] speech designed to influence the outcome of a union election.” In one of the landmark cases concerning captive audience meetings, the NLRB acknowledged that an employer “did its best to inhibit the free play of discussion,” but nevertheless ruled the behavior legal. Luxuray of New York, 185 NLRB 100 (1970).

20. Kate Bronfenbrenner, No Holds Barred—The Intensification of Employer Opposition to Organizing, Economic Policy Institute and American Rights at Work Education Fund, May 2009.

21. Consultants universally stress the power of using supervisors to convey anti-union messages to their subordinates. Gene Levine notes, “The most effective method for gaining the support of employees is one-on-one, eyeball-to-eyeball conversations between supervisors and employees.” See Gene Levine, Complete Union Avoidance (Delray Beach, Fla.: Gene Levine Associates, 2005), ch. 8, p. 5. Robert Lewis and William Krupman note that “face to face communications between supervisors and employees” are key to management’s efforts: “If instructed properly, trained supervisors can be the most effective means of lawfully influencing employee attitudes.” See Robert Lewis and William A. Krupman, Winning NLRB Elections: Management’s Strategy and Preventive Programs Second Edition (New York: Practising Law Institute, 1979), p. 95. Alfred DeMaria states that “the most important factor influencing the individual’s choice of ‘Union’ or ‘Non-Union’ is his supervision—how well his supervisor communicates the company’s views during the organizing campaign.” See Alfred T. DeMaria, The Supervisor’s Handbook on Maintaining Non-Union Status (New York: Executive Enterprises, 1986), p. 37. DeMaria’s Management Report for Nonunion Organizations newsletter states that “the success of union prevention depends greatly on the ability of supervisors to influence their employees.” See Management Report for Nonunion Organizations 27, no. 11 (November 2004), p. 7. Bruce Kaufman and Paula Stephan report that the management attorneys they interviewed believed that “effectively marshalling the cooperation and support of the supervisors was the single most critical ingredient to defeating the union.” As one of the management attorneys concluded, “without [supervisors’] support, the employer’s chance of victory is substantially reduced.” See Bruce E. Kaufman and Paula E. Stephan, “The Role of Management Attorneys in Union Organizing Campaigns,” Journal of Labor Research 16 (1995), https://doi.org/10.1007/BF02685719, pp. 443, 447.

22. Marty Jay Levitt, Confessions of a Union Buster (New York: Crown, 1993), p. 10.

23. There is a remarkable degree of consistency in the themes of employer campaigns over the past 40 years. In his Management Report for Nonunion Organizations newsletter, Alfred DeMaria describes the “themes commonly used by employers” as including “threat to remove jobs,” “disparaging the moral character of union supporters,” “inevitability of strikes,” and “threat to reduce wages.” See “From the Editor: Learning Lessons (Good and Bad) from a Real-Life Campaign,” Management Report for Nonunion Organizations 24, no. 4 (April 2001), pp. 3–5. Another Management Report article in the same issue (p. 5), titled “Campaign Threat of Plant Closure,” notes that “predicting the future of a business if it becomes subject to an obligation to bargain with the union, is a recurring campaign theme.”

24. Other researchers likewise found that, among nonunion workers who wished they had one, 55% believed that “management opposition” was the central reason why they had been unable to organize. See Phil Comstock and Maier B. Fox, “Employer Tactics and Labor Law Reform,” in Sheldon Friedman et al., eds., Restoring the Promise of American Labor Law, ed. Sheldon Friedman et al. (Ithaca, N.Y.: Cornell Univ. Press, 1994), p. 98; and Richard Freeman and Joel Rogers, What Workers Want (Ithaca, N.Y.: Cornell Univ. Press, 1999), p. 62.

25. Under federal law, 30% of workers in a given workplace must sign cards requesting a union election in order for the NLRB to schedule an election; thus, the first step of any organizing campaign is collecting signed cards from at least 30% of the workforce.

26. The final vote was 164–136, with 16 employees not voting. United Steelworkers, Kumho Tires, Representation Case 10-RC-206308, National Labor Relations Board, September 2017. https://www.nlrb.gov/case/10-RC-206308.

27. WMAZ, “Georgia Ranked as One of Worst States to Live, How Does Macon-Bibb Stack Up?” December 5, 2016; Evan Comen and Samuel Stibbins, “What City Is Hit Hardest by Extreme Poverty in Your State?” USA Today, July 13, 2018.

28. Advameg, Inc., City-Data.com, “Macon (GA) Poverty Rate Data,” 2016. Federal poverty thresholds for 2016 are at Georgia Department of Community Health, “2016 Federal Poverty Guidelines (FPG) Annual & Monthly Income Levels from 100% to 250%,” January 2016. The federal poverty threshold for a family of four was $24,300 in 2016.

29. Randy Southerland, “Macon/Bibb County: It’s Better in Bibb,” Georgia Trend, March 1, 2015.

30. Employee #1, interview with G. Lafer, May 2019. Employee #2 similarly reported that newspaper accounts when Kumho opened stated that jobs at the plant would pay between $40,000–$45,000 per year (Employee #2 interview with G. Lafer, May 2019). The company’s decision was further cemented by an extensive package of incentives provided by both state and local economic authorities—including free land, taxpayer-funded job training, site development, and water infrastructure, along with a 20-year tax abatement—totaling more than $35 million in value. See “Kumho Tire–Macon, Georgia,” Trade & Industry Development, accessed June 2020, and Vera Linsalata, “Kumho Picks Georgia as Location for Its First U.S. Tire Plant,” Rubber & Plastic News, February 11, 2008. Georgia governor Sonny Perdue said the incentives were deemed well worth the price because Kumho would be “creating hundreds of new jobs for our citizens.” See Georgia Department of Economic Development, “Kumho Tire to Locate First U.S. Manufacturing Facility in Macon,” January 29, 2008.

31. “We’re starting you at $15 but don’t worry—we’re gonna make up for it,” one employee was told on the first day of work (Employee #1, interview with G. Lafer, May 2019).

32. Employee #3, interview with G. Lafer, May 2019.

33. Employee #1, interview with G. Lafer, May 2019.

34. “There was no connection between wages and the kind of job you were doing,” one employee explained. “If we’re all making $15, I’d rather move to a more relaxed department” (Employee #3, interview with G. Lafer, May 2019).

35. Employee #3, interview with G. Lafer, May 2019. Indeed, most employees had no job descriptions, making it all the more difficult to control what type of work one might be directed to do. Favoritism likewise shaped the assignment of hours and overtime. One employee, for instance, had been given as much overtime as he could work, enabling him to significantly boost his weekly earnings, until he lost favor with his supervisor and was subsequently completely shut out of opportunities for overtime work. (Employees #1 and #2, interview with G. Lafer, May 2019.) The practice of favoritism also secured the allegiance of those who benefited from the system. When Kumho employees approached their election, there was a block of nearly 40 employees—all white men—described as receiving preferential treatment from management. Where all other employees rotated between day and night shifts, these workers were on permanent day shift, and they were each paid $28 per hour, far more than other employees with similar skills. These employees knew that in a unionized workplace, these positions would be open to bid by seniority and skill rather than reserved for management favorites. As a result, they served as a bloc of anti-union voters, promoting the wearing of “Vote No” hats to other workers. Reported by Alex Perkins, USW Local 572 president and organizer, in an interview with G. Lafer, May 14, 2019.

36. Employee #1, interview with G. Lafer, May 2019. Pay and promotion were also subject to unwritten and seemingly arbitrary procedures, employees reported. One employee, for example, described being asked to serve as a trainer in return for a pay increase of 75 cents per hour. One month into the new job he still hadn’t gotten a pay increase, but when he raised the issue with his supervisor, he was told that even though he had completed his probationary period and was a regular employee, he had to serve an additional probation period in the new position before he’d get his raise. None of this was ever put in writing, nor was it clear if other trainers were treated the same. “They just make up everything as they go,” this employee said. “They didn’t even tell me how long the probation would be.” Assuming this probation period would be set for the same 90-day length as the general employee probation, the employee approached his supervisor one week before completing his 90 days to ask about the raise. At that point, his supervisor announced that it wasn’t working out for him to be a trainer—giving no reason or evaluation whatsoever—and sent him back to his previous job, having never received a wage increase. (Employee #4, interview with G. Lafer, May 2019.)

37. Cancarb Limited, “Safety Data Sheet: Carbon Black,” revised May 25, 2015. One night, after there was an accident with the mixing machines, resulting in carbon black being dispersed through the vents and coating all areas of the building, the entire facility was evacuated in response. But employees in the mixing department were working in carbon black fumes all day long, every day, without respirators. (Employees #2 and #3, interview with G. Lafer, May 2019.)

38. U.S. Department of Labor, Occupational Safety and Health Administration, “U.S. Department of Labor Cites Georgia Tire Manufacturer and Contractors for 22 Safety and Health Violations” (news release), May 29, 2019.

39. USW Local 572 President Alex Perkins, interview with G. Lafer, May 15, 2019.

40. Employee #3, interview with G. Lafer, May 2019.

41. Employee #3, interview with G. Lafer, May 2019. While union supporters sometimes questioned management’s assertions, it took great fortitude to do so, they said. One employee who challenged the consultant’s facts said he found his blood pressure spiking so extremely that he had to be rushed to the hospital. But after the first week, some of the strongest union supporters were weeded out of meetings, enabling consultants to work on the less committed voters without having to worry about being challenged. (Employees #3 and #4, interview with G. Lafer, May 2019.)

42. Employee #3, interview with G. Lafer, May 2019.

43. Employee #3, interview with G. Lafer, May 2019.

44. “If you’re a temp, they can cut you right then and there. So they tell you wear this hat, and you have to wear it,” explained one employee (Employee #1, interview with G. Lafer, May 2019).

45. Kumho managers reportedly graded and ranked employees, tracking each employee’s political leanings, and adjusting strategy accordingly. In a department where workers were particularly vocal in their support for unionization, an additional manager was assigned who succeeded in quashing the department’s demonstrations of pro-union sentiment. (Employee #3, interview with G. Lafer, May 2019.) This manager was later found to have submitted a timesheet to receive overtime pay for “non-union support” (Bonus Request Form, October 7, 2017, Exhibit U-2, Kumho Tires and United Steelworkers Union, Cases 10-CA-208255, 10-CA-208414 and 10-RC-206308, March 22, 2019). In May 2019, an administrative law judge found that nine different Kumho managers had illegally interrogated employees on 15 separate occasions. While this number is large, it likely represents a small sample of what was standard practice during the campaign. (Arthur J. Amchan, Decision, Kumho Tires and United Steel Workers International Union, Cases 10-CA-208255, 10-CA-208414, and 10-RC-206308, National Labor Relations Board, May 14, 2019.) One union supporter was told by his supervisor, “I was told you were pro-union so I don’t have to talk to you ’cause your mind is made up” (Employee #3, interview with G. Lafer, May 2019).

46. Employee #5, interview with G. Lafer, May 2019.

47. Employee #1, interview with G. Lafer, May 2019.

48. Arthur J. Amchan, Decision, Kumho Tires and United Steel Workers International Union, Cases 10-CA-208255, 10-CA-208414, and 10-RC-206308, National Labor Relations Board, May 14, 2019.

49. Consolidated Complaint, Kumho Tires and United Steel Workers International Union, National Labor Relations Board, Cases 10-CA-208255 and 10-CA-208414, July 31, 2018.

50. Employee #1, interview with G. Lafer, May 2019.

51. Employee #1, interview with G. Lafer, May 2019.

52. Employee #4, interview with G. Lafer, May 2019.

53. Employee #3, interview with G. Lafer, May 2019.

54. This dynamic is described in detail in Gordon Lafer, “What’s More Democratic Than a Secret Ballot? The Case for Majority Sign-Up,” Journal of Labor and Society 11, no. 1 (March 2008), https://doi.org/10.1111/j.1743-4580.2008.00187.x. The dynamic is also inherent in the logic of collective action, as described in Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge, Mass.: Harvard Economic Studies, 1971; twentieth printing 2002).

55. Employee #3, interview with G. Lafer, May 2019.

56. Employee #1, interview with G. Lafer, May 2019.

57. After witnessing the company’s daily disregard for the law—in interrogations and threats, if not yet in terminations—the union concluded that a button day would entail unacceptable risks for employees (USW Local 572 President and organizer Alex Perkins, interview with G. Lafer, May 2019).

58. Employee #4, interview with G. Lafer, May 2019.

59. Five percent of employees did not vote.

60. In a company-wide meeting, Kumho’s human resources director announced that the Macon plant accounted for only 5% of the company’s worldwide production, and therefore could easily be shipped back to Korea: “Five percent of production could just quickly go away…. I’m asking you to consider not having our work taken abroad, and cast your vote as a clear No. There’s too much at stake…. We just cannot have this place shut down…. The union…if they strike, you could see the molds or tires being produced somewhere else…. I don’t want any of you to look at me later and say…why didn’t you tell me that if we don’t get along…what could happen?… I want to…have you know pretty much that we’ll be looking at tires being shipped somewhere else.” (Kumho Chief People Officer Jerome Miller, speech to Kumho employees, October 11, 2017. Recording shared with the author by USW. This speech is also documented in Arthur J. Amchan, Decision, Kumho Tires and United Steel Workers International Union, Cases 10-CA-208255, 10-CA-208414, and 10-RC-206308, National Labor Relations Board, May 14, 2019.)

61. William Schertz, “Kumho Opens Its First N.A. Tire Plant,” Rubber & Plastics News, June 1, 2016; Vera Linsalata, “Kumho Picks Georgia as Location for Its First U.S. Tire Plant,” Rubber & Plastics News, February 11, 2008. The company’s human resources director lamented, “I would love to show everybody the inside of the plant, but it’s all proprietary stuff we pretty much built ourselves” (quoted in Lori Johnson, “Kumho Tire Plant Brings Economic Boost to Central Georgia,” WMAZ, March 3, 2016).

62. Kumho cited “proximity to transportation and customers” as one of the reasons it chose this location for its plant (Vera Linsalata, “Kumho Picks Georgia as Location for Its First U.S. Tire Plant,” Rubber & Plastics News, February 11, 2008).

63. Kumho spokesperson Jim Frentheway, quoted in William Schertz, “Kumho Opens Its First N.A. Tire Plant,” Rubber & Plastics News, June 1, 2016.

64. Quoted in William Schertz, “Kumho Opens Its First N.A. Tire Plant,” Rubber & Plastics News, June 1, 2016.

65. Employee #3, interview with G. Lafer, May 2019.

66. See, for instance, Gene Levine, Guide to Union Avoidance (Gene Levine Associates, 2005), ch. 2, pp. 2–16.

67. Form LM-20, File C-00525-683681, filed September 27, 2018, by LRI President Phillip Wilson, including boilerplate contract terms, states that the “cost of unionization is estimated at 25% for most organizations.”

68. The BLS $11.35 figure was shown on a powerpoint slide with the question, “If the average hourly wage of manufacturing in Macon-Bibb County is 11.35, then why would Kumho Tire want to agree to anything more than that in negotiations?” (Official Report of Proceedings before the NLRB, Region 10, Kumho Tires versus United Steelworkers Union, Cases 10-CA-208255; 10-CA-208414; 10-RC-206308. March 21, 2019, volume 4.)

69. Employee #3, interview with G. Lafer, May 2019.

70. Official Report of Proceedings before the NLRB, Region 10, Kumho Tires versus United Steelworkers Union, Cases 10-CA-208255; 10-CA-208414; 10-RC-206308. March 21, 2019, volume 4.

71. See, for instance, National Association of Manufacturers, Remaining Union-Free: A Supervisor’s Guide (2004), p. 3; Alfred DeMaria, Management Report for Nonunion Organizations 26, no. 6 (2003), p. 7; Donald P. Wilson, Total Victory! The Complete Management Guide to a Successful NLRB Representation Election, 2nd edition (Broken Arrow, Okla.: Labor Relations Institute, 1997), pp. 77, 221.

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

73. “Did You Know This About Unions,” poster distributed at Kumho Tire plant, shared with G. Lafer by United Steel Workers, April 2019.

74. Kumhounionfacts.com, Frequently Asked Questions, screenshot, accessed September 2017. Shared with Gordon Lafer by USW attorney. Emphasis in the original.

75. At Kumho, the difficulty of home visits was compounded by the company’s providing the union a list of employee contact information—required by federal law—in which every single address was wrong. (Kumho mailed anti-union letters to employees’ correct home addresses, but provided the union with a different list of entirely false addresses.) The union considered filing charges with the NLRB to compel the company to provide accurate addresses. However, filing a charge would delay the election date and, given that workers were being subject to such intensive daily anti-union campaigning, the union determined that it was better to proceed toward the election than to file a complaint. (USW Local 572 President and organizer Alex Perkins, interview with G. Lafer, May 2019.)

76. USW Local 572 President and organizer Alex Perkins, interview with G. Lafer, May 2019. Unions’ ability to talk with workers outside the workplace has grown even more daunting since the COVID-19 crisis.

77. USW Local 572 President and organizer Alex Perkins, interview with G. Lafer, May 2019.

78. USW Local 572 President and organizer Alex Perkins, interview with G. Lafer, May 15, 2019.

79. United Steelworkers, Kumho Tires, Representation Case 10-RC-206308, National Labor Relations Board, September 2017. https://www.nlrb.gov/case/10-RC-206308.

80. Indeed, after two weeks of management’s anti-union messaging, a growing number of employees stopped reading the union leaflets. “I don’t want to hear anymore from the union or the company,” one woman told an organizer. “I just want it to be over.” Reported by USW Local 572 President and organizer Alex Perkins, interview with G. Lafer, May 14, 2019.

81. Employee #3, interview with G. Lafer, May 2019.

82. Employee #1, interview with G. Lafer, May 2019.

83. Employee #2, interview with G. Lafer, May 2019.

84. Employee #5, interview with G. Lafer, May 2019.

85. Kumho’s 2016 revenue reported in “Kumho Tire Co, Inc. GDR 144A,” MarketWatch, accessed June 8, 2020. G. Lafer interview with USW Local 572 President and organizer Alex Perkins, May 2019. In the interview, Perkins shared his post-election communication with Employee #6.

86. Employee #6, interview with G. Lafer, May 2019.

87. Employee #6, interview with G. Lafer, May 2019.

88. Employee #6, interview with G. Lafer, May 2019; and USW Local 572 President and organizer Alex Perkins, interview with G. Lafer, May 16, 2019. In the interview, Perkins shared his communication with Employee #7.

89. Arthur J. Amchan, Decision, Kumho Tires and United Steel Workers International Union, Cases 10-CA-208255, 10-CA-208414, and 10-RC-206308, National Labor Relations Board, May 14, 2019. Because such an order can take years to take effect, the union decided to withdraw its objections to the first election, let workers sign new cards, and petition for a new vote, which would happen much more quickly.

90. Employees #1 and #3, interview with G. Lafer, May 2019.

91. Employee #4, interview with G. Lafer, May 2019.

92. When a union has been certified after winning an election, employers are legally required to negotiate a contract in good faith. However, if an employer refuses to bargain in good faith, the legal remedy is simply to order the employer, once again, to negotiate in good faith. At the Dayton Hudson Corporation in Michigan, for instance, an election was overturned after the employer was found guilty of illegal threats, coercion, discrimination against union activists, videotaping workers talking to organizers, following employees into bathrooms, and monitoring employee phone calls. A second election was scheduled for 15 months after the first, and the employer was required to post notices acknowledging the law and pledging to respect it in the future. Yet even while these posters hung on the company’s walls, the employer repeated some of the exact behaviors it was pledging to rectify. The NLRB canceled the second election and charged the employer with more than 100 separate violations of the law. Yet all these workers could look forward to was, again, more signs posted, more promises voiced, and yet another delayed and rescheduled election. See Richard W. Hurd and Joseph B. Uehlein, Patterned Responses to Organizing: Case Studies of the Union-Busting Convention, Cornell University ILR School, 1994, p. 69. One of the most extreme such examples is the case of a casino in Sparks, Nevada, called the Sparks Nugget. In 1977, the NLRB found that the Sparks Nugget had been guilty of bargaining in bad faith for the three previous years, and instructed the employer to return to the negotiating table in good faith. In 1980, the Court of Appeals enforced the NLRB’s order, but the employer continued in its refusal to negotiate. In 1984, an administrative law judge once again found the employer was illegally bargaining in bad faith. In 1990, the NLRB upheld this decision, ordering the employer back to the table. Again, the employer appealed to the Ninth Circuit Court of Appeals, and in 1992, more than 17 years after the employer began disregarding the law, the court enforced another NLRB order requiring the company to return to the negotiating table. Sparks Nugget, Inc. v. NLRB, 968 F. 2d 991 (9th Cir. 1992) and related discussion in Andrew Strom, “Rethinking the NLRB’s Approach to Union Recognition Agreements,” Berkeley Journal of Employment and Labor Law 15, no. 1 (1994), pp. 50–86.

93. Fred R. Long, president of West Coast Industrial Relations Association, gave a presentation at Century Plaza Hotel, Los Angeles, July 28, 1976; Joel D. Smith attended the presentation and recorded it. The transcript was reproduced in John Logan, “Consultants, Lawyers, and the ‘Union Free’ Movement in the USA Since the 1970s,” Industrial Relations Journal 33, no. 3 (2002), pp. 197–214. In an issue of his Management Report for Nonunion Organizations newsletter, union-avoidance consultant Alfred DeMaria likewise advises readers, “Companies that win the first election are more successful in a second election. One might think that an employer that committed unfair labor practices and had to face the union again in the second election would be less likely to succeed because of its previous unfair labor practices. Yet NLRB statistics show that, overwhelmingly, the party that wins the first election (whether it be the union or the employer) wins the second election handily, often by a greater margin.” See “Companies Win More Second Elections—But Not Always,” Management Report for Nonunion Organizations 27, no. 11 (November 2004), p. 1.

94. Employee #1, interview with G. Lafer, May 2019.

95. Wayne Crenshaw, “Kumho Tire Union Vote Came Down to Four Votes. Now the Ballots Are Being Challenged,” Macon Telegraph, September 9, 2019.

96. Kumho does not publish its wage rates, but as of June 2019, Glassdoor was still reporting $15 per hour as the average wage for machine operators in Kumho’s Macon plant. Food stamps are available in Georgia for any family of four earning less than $33,475 per year. See Benefits.gov, “Georgia Food Stamps” (web page), accessed May 2020.

97. Macon Georgia Agreement, Graphic Packaging International, Inc., and United Steelworkers Union, Local 572, January 17, 2017–December 31, 2021; hard copy shared with G. Lafer by USW Local 572, Macon, Ga., May 2019.

98. Based on United Steelworkers’ contracts at Bridgestone/Firestone, Michelin, B.F. Goodrich/Uniroyal, Titan Tire, Goodyear, and Sumitomo Tire facilities in Alabama, Ohio, Iowa, Indiana, Tennessee, Virginia, North Carolina, and Kansas. Data provided to G. Lafer by USW via email, July 2019.

99. Employee #1, interview with G. Lafer, May 2019.

100. The Jackson Lewis website states that the firm has “developed specialized corporate campaign preventive advice and strategy/action plan recommendations…so that organizations can remain employers of choice.” Jackson Lewis P.C., “Labor and Preventive Practices” (web page), accessed June 2020. The same site advertises a JacksonLewis seminar titled “War is Hel…pful: Union Avoidance Training.” For a fuller description of the firm’s work, see John Logan, “The Union Avoidance Industry in the United States,” British Journal of Industrial Relations 44, no. 4 (December 2006), pp. 651–675.

101. DISH Network Corp., “Quick Facts” (web page), accessed May 22, 2020.

102. Martha Kinard, Regional Director, 16th Region, NLRB v. Dish Network Company, Evidentiary Hearing before Hon. Reed C. O’Connor, U.S. District Judge, Ft. Worth, Texas, January 6, 2017. Case 4:16-CV-952. Hearing Transcript. Witness #2, p. 24.

103. Under the NLRA, if at least 30% of workers sign cards saying they want a union election, the NLRB schedules a vote.

104. Gordon Lafer, interview with staff and attorneys of Communications Workers of America; and Martha Kinard, Regional Director, 16th Region, NLRB v. Dish Network Company, Evidentiary Hearing before Hon. Reed C. O’Connor, U.S. District Judge, Ft. Worth, Texas, January 6, 2017. Case 4:16-CV-952.

105. Email communication to G. Lafer from CWA attorney, August 2019.

106. National Labor Relations Board, Dish Network Corporation and Communications Workers of America Local 6171: Decision, Order and Order Remanding, April 11, 2012.

107. There is no evidence that the petition was originated by management, but management employees campaigned in support of decertification (CWA attorney, communication with G. Lafer, August 2019).

108. These charges were not related to the decertification petition itself, but because there were outstanding charges of illegal management activity, the NLRB put the vote on hold until those charges could be remedied, and this took almost three years (CWA attorney, communication with Gordon Lafer, August 2019).

109. Martha Kinard, Regional Director, 16th Region, NLRB v. Dish Network Company, Evidentiary Hearing before Hon. Reed C. O’Connor, U.S. District Judge, Ft. Worth, Texas, January 6, 2017. Case 4:16-CV-952; testimony of CWA Regional Assistant Vice President Sylvia Ramos.

110. DISH Regional Director Monty Beckham confirmed that technicians at the two unionized locations had higher average efficiency than similar employees in DISH’s other locations. Kinard, NLRB v. Dish Network Company, January 6, 2017, pp. 156–157.

111. The most skilled employees saw their pay cut from $70,000 to $35,000; lower-ranked employees experienced cuts that brought their wages down from over $50,000 per year to under $30,000. Kinard, NLRB v. Dish Network Company, January 6, 2017. Witnesses 2–8.

112. Dish Network Corporation and Communications Workers of America, Decision and Order, 366 NLRB no. 119, June 28, 2018, p. 8.

113. Kinard, NLRB v. Dish Network Company, January 6, 2017. Witnesses 5–8.

114. Kinard, NLRB v. Dish Network Company, January 6, 2017. Witness 6.

115. Kinard, NLRB v. Dish Network Company, January 6, 2017. Witness 7. The struggle to stay afloat also taxed family relationships. “A lot of the guys I work with,” this witness explained, “know that it’s pretty much a done deal that we’re going to wind up getting divorced.”

116. Kinard, NLRB v. Dish Network Company, January 6, 2017, pp. 6–7, 25, 202.

117. Kinard, NLRB v. Dish Network Company, January 6, 2017. Witness 2. Witness 1 likewise testified that employees did not volunteer to join the bargaining committee because “they were fearful for the terminations that they have seen.”

118. Kinard, NLRB v. Dish Network Company, January 6, 2017. Witnesses 2, 3, 11. DISH manager Waeland Thomas confirmed in federal testimony that he told employees, “I wouldn’t talk about the union while you are with these guys” because the newcomers would be surveyed by management about their experience and if they reported being made to feel uncomfortable by union talk, employees could suffer retaliation.

119. See, for example, Alexandra Bradbury, Mark Brenner, and Jane Slaughter, Secrets of a Successful Organizer (Detroit, Mich.: Labor Notes, 2016).

120. Kinard, NLRB v. Dish Network Company, January 6, 2017. Witness 1.

121. Kinard, NLRB v. Dish Network Company, January 6, 2017. CWA Assistant Vice President Sylvia Ramos.

122. Dish Network Corporation and Communications Workers of America, Decision and Order, 366 NLRB no. 119, June 28, 2018.

123. Letter from DISH Human Resources Manager Melissa Boillot to various employees, October 10, 2018.

124. Allegation made in unfair labor practice charge filed by Communications Workers of America against DISH TV, April 22, 2019; CWA attorney communication with G. Lafer, August 2019.

125. Dish Network Corporation and Communications Workers of America, Decision and Order, 366 NLRB no. 119, June 28, 2018.

126. The proposal for first contract arbitration was a central component of the Employee Free Choice Act of 2007, which was promoted by the AFL-CIO and adopted by the U.S. House of Representatives but not by the Senate. See https://www.congress.gov/bill/110th-congress/house-bill/800. The AFL-CIO continues to advocate for this proposal in a proposed Protecting the Right to Organize Act. See AFL-CIO, “Letter in Support of Legislation to Modernize the National Labor Relations Act,” April 29, 2019.

127. Further, the union offered to compromise on wages, maintaining the 2009 incentive wages only for the most veteran employees. Dish Network Corporation and Communications Workers of America, Decision and Order, 366 NLRB no. 119, June 28, 2018, p. 7.

128. AT&T Southwest and Communications Workers of America, 2017 Labor Agreements. As of 2019, warehouse employees at DISH earn between $10 and $11 per hour, or between $20,800 and $22,880 per year based on a 52-week, 40-hour-per-week schedule. Warehouse assistants with five years’ experience at AT&T DirecTV earn a base wage of $39,182 per year, not including pay premiums for evening and graveyard shifts or Sundays. DISH technicians post-2016 earn between $13 and $17 per hour, or between $27,000 and $35,000 per year, compared with AT&T technicians who, with five years’ seniority earn a base wage of $53,000 per year before accounting for shift premiums.

129. DISH Network Corp., “Quick Facts” (web page), accessed May 22, 2020.

130. Indeed, it appears that DISH executives sought to close the company’s two unionized locations shortly after workers organized. The U.S. District Court noted that DISH cut wages in the two unionized facilities to a level “lower than all other non-union technicians in the region” and, in addition, denied these employees the additional incentive pay provided to nonunion technicians. DISH offered no justification for this disparate treatment, and the court concluded that “improper retaliation seems the most likely answer.” Kinard v. Dish Network Co., United States District Court for the Northern District of Texas, Forth Worth Division, January 14, 2017, Civil Action no. 4:16-cv-00952-O. But a leaked text suggests that the company had more ambitious aims: closing the branch offices and eliminating any trace of unionization within DISH’s national network. When wage cuts were announced in 2016, a DISH manager unintentionally shared a message announcing that “the union is gone. Techs will be affixed hourly rates, no [incentives]…. The two offices are gradually closing…. They would rather have the techs quit en masse.” (Message from DISH Field Service Manager Hanns Obere, mistakenly sent to an employee. Reproduced in Dish Network Corporation and Communications Workers of America, Decision and Order, 366 NLRB no. 119, June 28, 2018.)

131. In Texas, a family of four is eligible for food stamps in 2019 if its income is below $33,480. See Texas Health and Human Services, “C-100, Income Limits and Proration Charts” (web page), accessed June 8, 2020.

132. Researcher Kate Bronfenbrenner shows that, by the early 2000s, 96% of all employers involved in union elections with 50 or more voters mounted campaigns against unionization and three-quarters of all employers involved in union elections with 50 or more voters hired union avoidance consultants. See Kate Bronfenbrenner, No Holds Barred: The Intensification of Employer Opposition to Organizing, Economic Policy Institute and American Rights at Work Education Fund, May 2009.

133. For a detailed description of the provisions of the PRO Act, see McNicholas et al., Unlawful: U.S. Employers Are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns, Economic Policy Institute, December 11, 2019.

134. Sharon Block and Benjamin Sachs, Clean Slate for Worker Power: Building a Just Economy and Democracy, Labor and Worklife Program, Harvard Law School, January 2020.