Testimony | Unions and Labor Standards

Testimony prepared for the U.S. Senate Subcommittee on Economic Policy for a hearing titled “Banning Noncompete Agreements”

EPI’s Heidi Shierholz appeared as a witness and submitted the following written testimony before the U.S. Senate Banking, Housing, and Urban Affairs Subcommittee on Economic Policy, on Tuesday, July 30 at 2:30 PM, for a hearing titled “Banning Noncompete Agreements: Benefits for Workers, Businesses, and the Economy.”

Introduction

Chair Warren, Ranking Member Kennedy, and members of the subcommittee, thank you for the opportunity to testify today on the benefits to workers, businesses, and the economy of banning noncompete agreements.

My name is Heidi Shierholz, and I am an economist and the president of the Economic Policy Institute (EPI) in Washington, D.C. EPI is a nonprofit, nonpartisan think tank created in 1986 to include the needs of low- and middle-wage 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-wage workers, and assesses policies with respect to how well they further those goals. I previously served as Chief Economist at the U.S. Department of Labor during the Obama administration.

Noncompete agreements are clauses in employment contracts that prevent workers from going to work for, or starting, a competing business within a certain period of time after leaving a job. Under the Federal Trade Commission’s (FTC’s) noncompete ban, employers will be barred from asking any new workers to sign noncompetes, and existing noncompetes would be made unenforceable for the vast majority of workers. Today I will highlight the importance of the rule by discussing the ubiquity of noncompete agreements and describing the effects these agreements have on wages, business formation, economic dynamism, labor mobility, productivity, innovation, and prices. I will also discuss options firms have to protect trade secrets without noncompetes, and the FTC’s authority to ban noncompetes.

Noncompetes are widely used

Noncompete agreements are ubiquitous. Many studies of workers find that roughly one in five workers are subject to noncompete agreements (see for example Starr, Prescott, and Bishara (2020); Balasubramanian, Starr, and Yamaguchi (2023); and Rothstein and Evan Starr (2021)).1

My 2019 study with Alexander Colvin finds that more than one out of every four workers is subject to noncompete agreements. Using a 2017 survey of a random sample of private-sector businesses with 50 or more employees, Colvin and I found that almost half (49.4%) of businesses required at least some employees to sign a noncompete agreement and almost one-third (31.8%) indicated that all employees were required to accede to a noncompete as a condition of employment. Based on these data, we estimate that at least 27.8% are subject to noncompete agreements. 2 The higher share subject to a noncompete found in this survey compared with the other surveys cited above could be due to the fact that ours was a survey of business establishments, while the others are surveys of individual workers. While businesses know whether their workers are subject to noncompete agreements, workers are often asked to sign a noncompete on the first day of work when they are dealing with a great deal of administrative paperwork, and as a result often do not know or remember they are covered by a noncompete until they try to leave their job, and thus are likely to underreport being subject to them.

It is important to note that noncompete agreements are not limited to high-wage workers in knowledge-sensitive occupations and industries. My 2019 research with Alexander Colvin found that more than a quarter (29.0%) of private workplaces that had an average wage of less than $13.00 per hour used noncompete agreements for all their workers.3 Michael Lipsitz and Evan Starr (2021) note that “while [noncompetes] are frequently assumed to occur only in high-wage jobs…we find that the modal worker bound by a [noncompete] is paid by the hour, with median wages of $14.”4 And in their analysis of private-sector workers in 2017–2018, Rothstein and Starr (2021) reported that 14.4% of workers who earned less than the equivalent of $20 per hour and 14.7% of workers with less than a college degree had signed a noncompete agreement (compared with 21.7% for those earning more than $20 per hour and 24.3% of college graduates).5

As noted by Starr, Prescott, and Bishara (2020)—who found that 13.3% of labor force participants who earned less than $40,000 per year in 2017 reported that they were subject to a noncompete agreement—“the frequency of noncompetes among low-wage employees without access to trade secrets and the lack of negotiation in the contracting process hint at more anticompetitive rationales for the use of noncompetes by employers.”6

Noncompetes lower wages

To understand the impact of noncompetes on wages, it is useful to remember that essentially the only source of leverage an individual nonunionized worker has with respect to their employers is their ability to quit and take a job somewhere else. That ability means their employer has to provide a job that is competitive enough in terms of compensation, working conditions, and opportunities that the worker is not incentivized to leave to take another job or start their own business. Noncompetes cut that source of leverage off at the knees. When workers do not have the freedom to take another job, or start a business, in their line of work in their community, their employers simply don’t have to pay them as much or treat them as well.

A large and growing body of research bears this out. For example, using data from the Survey of Income and Program Participation, Evan Starr (2019) finds that an increase in enforcement of state noncompete laws (from nonenforcement to an average state’s level of enforceability) is associated with a 4% decrease in hourly wages.7 Further, an analysis by Michael Lipsitz and Evan Starr (2021) of wage trends in Oregon after the state banned noncompete agreements found that wages for all hourly workers increased by 2–3% on average.8

It is worth noting that these studies show that noncompetes don’t just lower the wages of those subject to noncompetes, they also lower wages of workers who have not signed such agreements. As described below, noncompetes decrease entrepreneurship and new firm entry, which increases local labor market concentration and depresses wages. Further, noncompetes make labor markets “thinner,” reducing the likelihood of a successful employer-employee match and driving down equilibrium wages.

An analysis of data of the Noncompete Survey Project by Evan Starr, Justin Frake, and Rajshree Agarwal (2019) presents evidence that confirms this logic. Starr, Frake, and Agarwal “find that in state-industry combinations with a higher incidence and enforceability of noncompetes, workers–including those unconstrained by noncompetes–receive relatively fewer job offers, have reduced mobility, and experience lower wages” (emphasis added).9

Noncompetes also appear to exacerbate racial and gender wage gaps by exerting larger wage effects on women and black men than on white men.10

Long-run wages and employee training

Some have suggested that without noncompetes, firms will abandon investments in employee training and as a result, workers’ wages will be lower in the long run. While evidence is mixed on whether noncompetes are associated with more training, it is worth noting that one reason noncompetes might lead to more training is that they shrink the pool of available workers with relevant experience, so employers are forced to hire less experienced workers that require more training. Evidence also suggests that on net, any gains in long-term wages workers may receive from additional training in a noncompete regime is more than offset by the wage suppressing effects of the noncompetes themselves. Balasubramanian et al (2020) find that individuals who start a job in a state that enforces noncompetes see lower earning lasting over at least 8 years, compared to those who start a job in a state that doesn’t enforce noncompetes.11

It is also worth noting that employers who provide jobs with competitive wages, working conditions, and opportunities do not have to worry that their employees will defect to competitors en masse shortly after they receive valuable training.

Noncompetes reduce business formation, dynamism, labor mobility, productivity, and innovation

Some claim that though noncompetes reduce wages, they are still necessary because they boost innovation by incentivizing firms to invest in developing important advancements by reducing the ability of their workers to take valuable information about those advancements to a competitor. However, the best evidence consistently finds that on net, the effect of noncompetes on innovation is in the opposite direction.

By preventing workers from leaving their employers to create new businesses, noncompete agreements reduce business formation, reducing dynamism in the economy. An analysis based on the findings in Jeffers (2024)12 shows that banning noncompetes will increase the rate of new business formation by 2.7%, which, in the U.S. economy, would translate into an additional roughly 8,500 new businesses annually.

Noncompete agreements also reduce job mobility—i.e., they keep workers locked in jobs—and, as a result, they reduce productivity growth by blocking the efficient reallocation of labor from less productive to more productive job matches. For example, Johnson et al (2023) finds that noncompete agreements cause a significant decrease in job-to-job mobility.13 Eliminating noncompete agreements will allow workers to find firms, and firms to hire workers, that yield the most productive matches.

Declines in business formation and labor mobility as a result of noncompetes contribute to an overall decline in innovation. A study by Johnson, Lipsitz, and Pei (2023), finds that an average-sized increase in the enforceability of noncompetes leads to an 11%-19% reduction in patenting—including reductions in “breakthrough” inventions—over the following 10 years.14 Based on this research, it is estimated that banning noncompetes will lead to more than 17,000 patents each year. Other research, for example Reinmuth and Rockall (2024), also find that noncompetes have a significant negative impact on patenting and innovation by “reducing labor mobility as a channel of idea diffusion that increases overall innovation.”15

Banning noncompetes will reduce inflation

While eliminating noncompete agreements will raise wages, the net effect will likely be to reduce prices paid by consumers. Noncompete agreements increase concentration in the markets for goods and services by preventing workers from leaving their employers to create new businesses or join other firms that can increase the market supply and intensify competition. Banning noncompete agreements will reduce market concentration, thereby reducing market prices.

Noncompete agreements also reduce productivity growth by blocking the efficient reallocation of labor from less productive to more productive job matches. Eliminating noncompete agreements will allow workers to find firms, and firms to hire workers, that yield the most productive matches. The increased firm and economywide productivity will reduce consumer prices.

One area where direct evidence is available on the price effects of noncompetes is around the effect of noncompetes on prices in health care. Research shows that noncompetes significantly increase the cost of physician services (in particular, a 10% increase in the enforceability of noncompetes causes 4.3% higher physician prices).16 Based on that research, the FTC estimates that banning noncompetes will reduce health care costs by at least $74 billion over the next decade.

Noncompetes are often bundled with other restrictive contracts

Employers who require their workers to sign noncompete agreements are more likely to require their workers to sign additional restrictive contract provisions. My research with Alexander Colvin finds that over half of firms that require noncompetes for at least some of their employees also require at least some employees to agree to mandatory arbitration.17 Further, Balasubramanian, Starr, and Yamaguchi (2023, Table 2) analyzed data from the 2017 version of Payscale.com’s annual firm-level survey of publicly traded in the United States and found that almost one in four firms used noncompete agreements together with nondisclosure agreements, nonsolicitation agreements, and nonrecruitment agreements for all employees, while more than half used all four types of agreements for at least some of their employees.18

The reflexive bundling of noncompetes with mandatory arbitration, nondisclosure, nonsolicitation, and nonpoaching agreements provides further evidence that the primary purpose of noncompetes is often to restrict employee options rather than being a tailored strategy for protecting beneficial investments in and information held by employees.

Relatedly, and as John Lettieri highlighted while employed at the American Enterprise Institute, workers often enter into noncompete agreements after little or no bargaining with their employer and without full information about the rights they are signing away:

The vast majority of noncompete agreements are not subject to any negotiation between the employer and employee, suggesting that the employee is unlikely to receive any benefits in return for their signature. A large share of these agreements are presented for signature only after the employee has already accepted the job offer—often on the first day of work. Employers frequently exploit workers’ lack of knowledge and resources when crafting noncompetes. For example, employers commonly request that workers sign noncompetes even in states where they are completely unenforceable—and workers nevertheless sign the agreements assuming they are valid. Likewise, employers often craft extremely broad provisions knowing that employees generally lack both an understanding of what is enforceable and the wherewithal to challenge the terms in court. 19

Employers who need to protect trade secrets have other options

Employers do not need noncompete agreements in order to protect trade secrets. California, for example, made noncompete agreements unenforceable in 1872, but the state has still become a global technology hub—something that would be difficult to imagine if businesses in California faced misappropriation of trade secrets and other confidential information at a meaningfully higher rate than in other states.

That is because there are intellectual property laws that provide businesses with significant legal protections for trade secrets. Further, employers are still able to use tailored nondisclosure and nonsolicitation agreements. Policies like these, that directly address what employees may and may not do with company secrets, allow businesses to protect trade secrets while not harming competition by taking away workers’ freedom to seek other work or to start a business after they leave their job.

And of course, employers concerned about worker departure and the associated loss of firm knowhow and knowledge can retain staff by paying fair wages and salaries and treating them with respect. Many employers do not use noncompete clauses and have had no difficulty keeping employees using these salutary methods.

The broad-based nature of the FTC’s rule is important

The FTC’s rule is a complete ban on new noncompetes for nearly all workers, including independent contractors and senior executives, while existing noncompetes for the vast majority of workers will no longer be enforceable (a key exception is senior executives, for whom already-existing noncompetes will be allowed to remain in force—this group represents less than 1% of all workers).

Some have questioned whether it is reasonable to ban noncompetes for senior executives or other highly paid workers, who may be particularly likely to have confidential information. But as mentioned above, employers do not need noncompete agreements to protect trade secrets, because they have other options. And while senior executives or other highly paid workers are likely in a much better position than other workers to negotiate over their full employment package—including restrictive agreements like noncompetes—and are therefore less likely to themselves be made worse off by noncompetes, the broader economic harms caused by noncompetes mentioned above are still present, including reduced business formation, economic dynamism, labor mobility, and productivity, and increased inflation. Many of these harms may in fact be stronger as a result of noncompetes among this group than among other workers, as, for example, workers in this group may be more likely than other workers, in the absence of noncompetes, to be in the position to start a new firm, and to innovate and grow it.

It is also important that new noncompetes are banned under the rule, not just made unenforceable. The very existence of a noncompete, even if it could not ultimately be enforced, can deter workers from going to work for, or starting, a competing business. The high costs—financial and otherwise—of noncompete-related litigation means noncompetes create a chilling effect on new firm formation and worker mobility that exists whether or not the noncompete would be enforced by a court. Notably, the chilling effects of unenforceable noncompetes do not just deter workers from going to work for, or starting, a competing business, they can also deter other businesses from hiring a worker who had been working for a competitor and had signed a noncompete, due to fear of legal complications.

FTC has the authority to ban noncompetes

Business groups suing to block the FTC’s noncompete rule have claimed that the FTC does not have the authority to ban noncompetes. But as a federal judge in Pennsylvania stated last Tuesday in a decision declining to issue a preliminary injunction against the rule, “the FTC is empowered to make both procedural and substantive rules as is necessary to prevent unfair methods of competition.”20

The FTC Act states that unfair methods of competitions are illegal, and it authorizes the FTC to issue rules and regulations to prevent their use. Further, the Supreme Court in 1986 confirmed that the FTC can not only challenge traditional antitrust violations but can also outlaw practices “the Commission determines are against public policy for other reasons.”21 With the noncompete rule, the FTC is doing exactly what it was empowered (and directed) to do by the FTC Act—take action, based on the best evidence available, to protect fair competition.

Noncompetes are not trying to hide that they are contrary to open and fair competition—it’s in their name. They bar workers from leaving to accept a position in their line of work in their community, stifle the creation of new businesses, and depress employer competition for workers’ services. There is a large body of evidence, described above, that noncompetes hinder competition, disadvantaging consumers, workers, and competing businesses. Banning these coercive and unfair agreements is fundamental to the FTC’s mandate.

Conclusion

In closing, it’s worth considering who wins from having noncompetes. One might believe it is businesses who want to protect trade secrets—but, as described above, noncompetes are not needed for that; firms have other options for protecting their intellectual property.

A core group that does indeed benefit from noncompetes is existing firms who want to stop new firms from forming and competing with them. Noncompetes strongly benefit those incumbent firms who want to protect their advantage, but they make it harder for new firms to start, grow, and innovate, and, as a result, they are bad for consumers and the overall economy, both of which benefit from competition and innovation.

Another group that benefits from noncompetes is existing firms who want to prevent other firms from being able to hire away their workers by offering them higher wages, better working conditions, and better opportunities. Noncompetes clearly benefit those incumbent firms, but are harmful to workers and the overall economy, both of which benefit from employer competition for workers’ services.

Given all this, it is not a surprise that groups representing existing businesses are fighting tooth-and-nail to keep their ability to require workers to sign noncompetes. That, however, does not mean noncompetes are good for the nation. In fact, noncompetes are bad for economic dynamism and innovation, they depress business formation and labor mobility, they hurt productivity and growth, they raise prices, they shrink for workers’ wages, and they restrict workers’ freedom. Banning these coercive and unfair agreements is fundamental to the FTC’s mandate, and the workforce, consumers, and the broader economy will be better off for it.

Thank you for holding this important hearing and I look forward to your questions.


1. Evan P. Starr, J.J. Prescott, and Norman D. Bishara, “Noncompete Agreements in the U.S. Labor Force,” Journal of Law and Economics 64, no. 1 (October 2020); Natarajan Balasubramanian, Evan Starr, and Shotaro Yamaguchi, “Employment Restrictions on Resource Transferability and Value Appropriation from Employees,” January 2023; Donna S. Rothstein and Evan Starr, “Mobility Restrictions, Bargaining, and Wages: Evidence from the National Longitudinal Survey of Youth 1997,” November 2021.

2. Alexander J.S. Colvin and Heidi Shierholz, Noncompete Agreements: Ubiquitous, Harmful to Wages and to Competition, and Part of a Growing Trend of Employers Requiring Workers to Sign Away Their Rights, Economic Policy Institute, December 2019.

3. Ibid.

4. Michael Lipsitz and Evan Starr, “Low-Wage Workers and the Enforceability of Noncompete Agreements,” Management Science 68, no.1, April 2021.

5. Donna S. Rothstein and Evan Starr, “Mobility Restrictions, Bargaining, and Wages: Evidence from the National Longitudinal Survey of Youth 1997,” November 2021.

6. Evan P. Starr, J.J. Prescott, and Norman D. Bishara, “Noncompete Agreements in the U.S. Labor Force,” Journal of Law and Economics 64, no. 1, October 2020.

7. Evan Starr, “Consider This: Training, Wages, and the Enforceability of Covenants Not to Compete,” ILR Review 72, no.4, August 2019.

8. Michael Lipsitz and Evan Starr, “Low-Wage Workers and the Enforceability of Noncompete Agreements,” Management Science 68, no.1, April 2021.

9. Evan Starr, Justin Frake, and Rajshree Agarwal, “Mobility Constraint Externalities,” Organization Science 30, no. 5, July 2019.

10. Matthew S. Johnson, Kurt Lavetti, and Michael Lipsitz, “The Labor Market Effects of Legal Restrictions on Worker Mobility,” National Bureau of Economic Research Working Paper no. 31929, December 2023.

11. Balasubramanian, Natarajan, Jin Woo Chang, Mariko Sakakibara, Jagadeesh Sivadasan, and Evan Starr. “Locked in? The enforceability of covenants not to compete and the careers of high-tech workers.” Journal of Human Resources, April 2020.

12. Jessica Jeffers. “The Impact of Restricting Labor Mobility on Corporate Investment and Entrepreneurship.” The Review of Financial Studies, July 2024.

13. Matthew S. Johnson, Kurt Lavetti, and Michael Lipsitz, “The Labor Market Effects of Legal Restrictions on Worker Mobility,” National Bureau of Economic Research Working Paper no. 31929, December 2023.

14. Matthew S. Johnson, Michael Lipsitz, and Alison Pei, “Innovation and the Enforceability of Noncompete Agreements,” National Bureau of Economic Research Working Paper no. 31487, July 2023.

15. Kate Reinmuth and Emma Rockall, “Innovation through Labor Mobility: Evidence from Noncompete Agreements,” May 2024.

16. Naomi Hausman and Kurt Lavetti, “Physician Practice Organization and Negotiated Prices: Evidence from State Law Changes,” American Economic Journal: Applied Economics, Vol. 13, No. 2, April 2021.

17. Alexander J.S. Colvin and Heidi Shierholz, Noncompete Agreements: Ubiquitous, Harmful to Wages and to Competition, and Part of a Growing Trend of Employers Requiring Workers to Sign Away Their Rights, Economic Policy Institute, December 2019.

18. Natarajan Balasubramanian, Evan Starr, and Shotaro Yamaguchi, “Employment Restrictions on Resource Transferability and Value Appropriation from Employees,” January 2023.

19. John W. Lettieri, A Better Bargain: How Noncompete Reform Can Benefit Workers and Boost Economic Dynamism, American Enterprise Institute, December 2020.

20. Tomasz Mielniczuk and Craig Minerva, “FTC’s Noncompete Ban Survives Preliminary Challenge in Pennsylvania Federal Court,” JD Supra, July 26, 2024.

21. Federal Trade Commission (FTC), “Section 5 of the FTC Act as a Competition Statute” (web page), October 17, 2008.


See related work on Unions and Labor Standards | Wages

See more work by Heidi Shierholz