A legal document called a non-compete contract is constraining the potential of workers and the broader labor market. These contracts, which prohibit workers from taking a similar position at a new employer or starting a competing business, limit the ability of workers to negotiate wage increases or find the best job for them. And by doing so, non-compete contracts limit our economy’s potential.
Many workers will encounter non-compete contracts at some point during their careers, and they may find that non-competes can shape their career progress and affect their ability to negotiate with employers. To better understand the scope and effects of non-competes throughout the labor force, we used data from the Bureau of Labor Statistics to study the prevalence of non-compete contracts in recent years. Our findings show that overall, more than one in seven mid-career workers reported having a non-compete contract in their current or most recent job. And non-competes are not just for high-tech workers and executives. Among low- and moderate-income workers, more than one in ten reported having a non-compete contract.1
Many workers will encounter non-compete contracts at some point during their careers, and they may find that non-competes can shape their career progress and affect their ability to negotiate with employers.
Non-compete contracts signed by low-wage workers are especially concerning. First, non-competes reduce pay for lower-wage workers. When Oregon ended the enforcement of non-competes for hourly paid workers, wages for those workers rose by 2–3 percent, with larger effects in occupations where non-competes are more common (Lipsitz and Starr 2021). Second, low-wage workers have less access to legal advice than other workers have, making it more difficult for them to enter a fair, well-informed negotiation with employers over their non-compete contracts. Indeed, workers—many of whom receive non-competes only on or after the first day of work and without an opportunity to negotiate—report that they rarely negotiate over non-competes and frequently misunderstand whether and to what extent their non-competes are enforceable (Marx 2018; Prescott and Starr 2021; Starr, Prescott, and Bishara 2021).
For employers, non-competes can have mixed effects. On the one hand, employers can use non-competes as a tool for limiting wage growth (by reducing workers’ leverage to negotiate better wages) and deterring workers from leaving. On the other hand, some employers and entrepreneurs may find that non-compete contracts make recruiting new workers or starting new businesses more difficult.
To better understand these considerations, we now provide more detail on how non-compete contracts are used and discuss the characteristics of workers who have non-competes.
The purpose, enforcement, and impact of non-competes
A non-compete contract is a legal agreement between an employee and employer that outlines the circumstances under which the employee cannot seek new employment or start a business. Typically, the contract specifies the industry, geographic boundary, and time period during which the individual is prohibited from competing with the employer. Details of the non-compete and its enforceability can vary. Some non-competes restrict employment across very broad geographic areas, while others are more limited. In some states, non-competes may bind even if an employee is fired without cause, while some other states require that employers pay employees a fraction of their prior wage during the period in which certain non-competes are enforced.
Two common justifications for non-compete contracts are that they protect employers’ proprietary information (i.e., trade secrets) and client relationships or increase employers’ incentive to provide training to their workers. In the first instance, an employer might be unwilling to hire a worker without assurance that the worker won’t share trade secrets after leaving the company. In the second instance, an employer that assumes its workers will depart in order to work for a competitor or start a competing business might underinvest in training those workers, leading to lower productivity.
For workers with non-competes, the mere threat of enforcement can limit their negotiating power and career opportunities, even in states that do not enforce the contracts stringently.
While these justifications for non-compete contracts may have merit in some situations, non-compete contracts are blunt instruments. For example, many workers—especially low-wage workers—do not possess important trade secrets, and even for those who do, the protections provided by trade-secrets law can be sufficient (and better targeted) to address this specific concern. Though research suggests that workers with non-compete contracts may receive more training (Starr 2019), other, targeted options for promoting training (e.g., public subsidies for training, or employee repayment contracts) do not have the negative career effects associated with non-competes.
The enforcement of non-competes varies substantially across states. California, North Dakota, and Oklahoma essentially do not enforce non-compete contracts, meaning that employers cannot legally hold former employees to their non-competes. At the other end of the spectrum, some states (such as Florida) enforce them relatively stringently. (One example of stringent enforcement comes in the form of courts that will rewrite and enforce what would otherwise be an overly broad non-compete.) For workers with non-competes, the mere threat of enforcement can limit their negotiating power and career opportunities, even in states that do not enforce the contracts stringently (Marx 2018).
The pervasiveness of non-competes
Our analysis shows that more than one in seven workers (15 percent in our sample) have non-compete contracts with their current or most recent employer. The data we analyzed are from the 2017–2018 round of the 1997 cohort of the National Longitudinal Survey of Youth: a widely used, nationally representative, Bureau of Labor Statistics survey of a cohort of individuals born from 1980 through 1984 (ages 32 to 38 in this most recent round of the survey).2 Of workers earning $20 per hour or less, 12 percent reported having a non-compete contract in their current or most recent job, compared with 18 percent of workers earning more than $20 per hour. As hourly wages increase, so does the likelihood of a worker having a non-compete contract.
While higher-earning and more-educated workers are more likely to have a non-compete contract, the shares of lower-earning and less-educated workers that also have non-competes are still substantial. Around 8 percent of workers in the bottom quintile of wages have a non-compete. For workers in the top quintile, the figure is 21 percent. These results largely confirm those from the first survey of non-competes, conducted in 2014 by Evan Starr, J.J. Prescott, and Norm Bishara. Before that survey was conducted, non-competes had often been thought of as unique to high-tech workers, CEOs, and others likely to possess trade secrets—the most common justification for the use of non-competes. Our results, along with the 2014 survey, show that non-competes affect a much broader set of workers.
Non-competes are broadly distributed throughout the labor force
Loading chart 1...
Non-compete contracts are present in a wide variety of business types and across many groups of workers. The drop-down menu in Figure 1 displays how the prevalence of non-competes varies by occupation, educational attainment, hourly wage, race, industry, and geography. Key highlights from Figure 1 include:
Workers in sales and related occupations have the highest prevalence of non-competes—22 percent—while only 6 percent of workers in agricultural or food-preparation and -serving occupations have non-competes. However, after adjusting for worker and employer characteristics, variation across occupations is substantially diminished.
Around 32 percent of workers in the professional-, scientific-, and technical-services industry have signed non-competes, compared to 7 percent of educational-services and accommodation and food-services workers. Differences by industry remain after adjusting for worker and employer characteristics, although the magnitudes of the differences are diminished. (See the Appendix Figure for details of our calculations that adjust for worker and employer characteristics.)
The shares of urban and rural workers signing non-competes are similar, at 15 percent and 14 percent, respectively. (From a worker perspective, non-competes may have different effects in rural areas where there are fewer employment options.)
The share of workers with non-competes generally increases with education levels, though individuals with graduate degrees are less likely to have non-competes than those with four-year degrees. These differences largely remain after adjusting for other factors.
Workers with higher wages are more likely to have non-competes, but a substantial fraction of workers have non-competes at all wage-levels.
The prevalence of non-competes is relatively similar across racial and ethnic groups: 14 percent of Latino/a workers and 13 percent of non-Latino/a Black workers have non-competes, compared to 15 percent of workers generally.
Characteristics of employers may also predict how likely a worker is to have a non-compete. As Figure 2 shows, workers whose employers have multiple locations are more likely to have non-competes than those working at single-location employers, perhaps suggesting that larger employers are more likely to have the human resources and legal departments that could deploy non-competes to their workers.
Loading chart 2...
Government policies can limit or eliminate non-competes
Policymakers have recently begun to focus on the ways in which uncompetitive labor markets shortchange workers. Accordingly, non-competes are receiving more scrutiny. Policies that limit non-compete contracts for low- and moderate-income workers can raise those workers’ wages and employment opportunities while contributing to a more flexible labor market.
A growing number of federal, state, and municipal policymakers have proposed or enacted reforms to non-competes in recent years. These proposals include a federal bill that prohibits the use and enforcement of non-competes in general and another federal bill that prohibits use and enforcement for workers (largely lower-wage) who are eligible for overtime pay under the Fair Labor Standards Act.
Oregon has made non-compete contracts unenforceable for hourly paid workers and certain other groups. Similarly, reforms in Massachusetts made non-competes unenforceable for employees eligible for overtime pay, student interns, and workers fired without cause. And recently, Washington, D.C., banned non-competes (going beyond simply making them unenforceable) for essentially all employees.
As policymakers consider reforms to non-compete contracts, it’s important that they not lose sight of non-competes’ effects on workers, and on low-wage workers in particular. A growing body of evidence shows that non-competes lower these workers’ pay and are relatively common, while the justifications for non-competes are less likely to apply to them.
The authors thank Michael Lipsitz and Evan Starr for insightful feedback on an earlier draft.
1 Throughout this article, our data—the 1997 Cohort of the National Longitudinal Survey of Youth—limit us to examining workers ages 32 to 38. In this calculation, we further limit the sample to those workers earning $20 per hour or less.
2 The original survey sample included 8,984 respondents, who have been interviewed 18 times. We utilize data from the most recent interviews, Round 18, which took place in 2017 and 2018 when the respondents were in their mid-30s. In the most recent round of interviews, 6,734 respondents participated. Those outside of the universe for the non-compete question for their main job (i.e., those who were self-employed or members of the military) were excluded from our sample. Furthermore, we only include individuals who answered the non-compete question, either in regard to a current job or a most recent job that ended in 2015 or later. Our analysis focuses on the respondent’s main job, which the 1997 Cohort of the National Longitudinal Survey of Youth defines as their current or most recent job.
Data Scientist, Community Development and Engagement
Tyler Boesch analyzes data, develops visualizations, and creates statistical models to help the Community Development and Engagement team understand issues affecting low- and moderate-income communities. Before joining the Bank, he was a graduate research assistant with the University of Minnesota Center for Urban and Regional Affairs.
Assistant Vice President, Community Development and Engagement
Ryan Nunn is an assistant vice president in the Minneapolis Fed’s Community Development and Engagement Department. Leading the Bank’s applied research function, Ryan works to improve outcomes for low- and moderate-income communities with the help of better evidence and analysis.