Q&A: Economist examines impact of noncompete agreements

Liyan Shi, an assistant professor of economics in the Tepper School of Business at Carnegie Mellon University, argued in a study published in Econometrica that noncompete agreements harm the economy. She suggested that a near ban of these noncompete agreements may be the best policy.

In April, the Federal Trade Commission issued a final rule aimed at promoting competition by banning noncompetes nationwide, to protect the fundamental freedom of workers to change jobs, increasing innovation and fostering new business formation.

In a recent Q&A, Shi provided insights into noncompete agreements and their influence on both the workforce and the economy.

Q: What are noncompete contracts and how common are they among U.S. workers?

Shi: Noncompete contracts are agreements between a company and an employee where the employee promises not to work for a competitor after leaving the company. These agreements are an elusive aspect of the labor market for researchers and policymakers because they are private contracts, which makes them difficult to measure and quantify.

Economists and researchers have gone to great lengths to conduct surveys and collect data to understand the prevalence of noncompetes and have found that they are surprisingly common. According to a nationwide survey by Evan Starr at the University of Maryland and his co-authors, about 30 million U.S. workers—roughly 18% of the entire workforce—are bound by noncompetes, and an even higher 38% have signed such contracts during their careers.

Q: What does your research reveal about the prevalence and duration of noncompete clauses among executives?

Shi: One segment of the labor market where we can observe the details of employment contracts is among executives and key employees at publicly listed firms, who must disclose these details for regulatory reasons. To understand the use and effects of noncompete clauses, I utilized machine learning and natural language processing tools to collect a large set of these contracts and analyze the details of the arrangements. My findings reveal that nearly two-thirds of executives are bound by noncompete clauses, which last about two years on average.

Q: What are the main concerns about noncompete agreements?

Shi: The main concern about noncompete contracts is their anticompetitive effects: restricted labor mobility prevents workers from moving to more productive jobs and inhibits the entry of new firms. With declining labor market fluidity, stagnant wage growth and decreasing business dynamism in the U.S., noncompete contracts have been identified as a likely contributing factor.

Noncompete contracts are justified as protecting employers' investments, especially in their employees. This debate has led to varied legal approaches across the country: many states allow noncompetes, while others, like California, ban them entirely.

Q: What policy recommendations do you have regarding noncompete agreements?

Shi: The effects of noncompete agreements present a policy trade-off. My research aimed to evaluate whether intervention is necessary. While noncompetes benefit firm investment, particularly in intangible assets like R&D, these benefits are outweighed by the costs of reduced labor reallocation and fewer new jobs. This indicates that a laissez-faire outcome is undesirable and that regulation is needed. On balance, I recommend a near ban on noncompetes.

Q: How does the noncompetes affect workers and employers?

Shi: Another issue of contention is how noncompete agreements affect wages—it's complicated. One reason noncompete clauses lower wages is because workers cannot take new, potentially more lucrative jobs. They are also less likely to get raises because their outside option is diminished. However, wages are often set through employer-worker bargaining, especially for high-skill workers, who may demand a higher starting wage to compensate for future restrictions. This is precisely what I found for key personnel such as executives in a firm.

While employment restrictions may seem unappealing, workers can be enticed with appropriate compensation. Nevertheless, noncompetes should not remain unregulated because they affect third parties—potential employers who could offer more productive jobs but cannot recruit workers bound by these agreements.

Q: What is next in the research world regarding noncompete contracts?

Shi: Researchers are examining the effects on different labor market segments, in particular low-skill workers who may have limited bargaining power against their employers, and why they agree to sign noncompetes. There is also exciting new research on how noncompete enforcement affects knowledge spillovers and the firm's innovativeness. More broadly, researchers are exploring how noncompete policies interact with other employment contracts and intellectual property policies, such as nondisclosure agreements, trade secret protection and patent protection.