Non-competes” are clauses in employment contracts that forbid people from quitting their jobs and going to work for a competitor, or sometimes from working in their field at all, for a period of time. They are far more common than most people realize. Non-competes show up in contracts across nearly all walks of life, from retail and food-service workers to professionals, executives, and creatives. Many employees do not read their contracts carefully, or do not understand the implications when they do. The result is often genuine shock when workers discover that quitting for a better job may expose them to the threat of a lawsuit.

I once knew a literary agent who, thanks to a particularly nasty non-compete, was effectively forced to spend an entire year doing nothing at all. Her story is not that unusual. For many workers, non-competes function less like a narrow contractual term and more like a hidden restraint on their power to seek a better life.

Economic intuition and basic theory have long suggested that non-competes are bad for workers. What has changed in recent years is that a growing body of empirical evidence now shows that they are even worse than many critics previously believed—and may be costing employees, in the aggregate, hundreds of billions of dollars each year. A nationwide ban of some kind is one of the most straightforward pro-worker policies available. Yet existing efforts remain trapped in a quagmire that has little to do with the merits. The next administration needs to charge hard and ban most or all non-compete clauses nationwide, once and for all.

The basic case against the non-compete is both economic and moral. Quitting for a higher-paying job—or even credibly threatening to do so—is one of the primary ways workers increase their earnings over time. Job switching is also how undervalued employees find better matches for their talents, skills, and interests. It is how new firms get staffed, and how startups are born. Economists describe this process as “labor mobility,” but the underlying idea is simple and intuitive.

You do not need a deep grasp of economic principles to see that there is something fundamental about the right to quit and seek better work in one’s field. This is especially so in a country that supposedly prizes freedom—the ability to walk away from a job and pursue a better opportunity in your field seems like a core economic liberty.

Employers, for their part, like non-competes for an obvious reason: they keep labor costs down and reduce the risk of losing valuable employees. A firm can get away with lower pay, fewer benefits, or worse working conditions if its workers are contractually barred from leaving for a better job. In that sense, non-competes operate as a kind of private regulation of the labor market, one that tilts bargaining power sharply toward employers.

Unsurprisingly, employers rarely defend non-competes as mere cost-cutting devices. Instead, they emphasize better-sounding justifications. The most common is the fear that employees who leave will take secret information with them and share it with competitors. Another is that non-competes are necessary to protect investments in employee training. The idea is that if a firm spends time and money training workers, it deserves contractual protection against those workers taking their new skills elsewhere. Others suggest that employees who dislike non-competes should simply negotiate them away.

The problem with most of those defenses is that there are ways to protect legitimate employer interests that don’t involve stripping away the right to find a better job. The real issue is the reduction in pay. And over the past half decade, researchers have examined these claims closely, and the case against non-competes has become increasingly devastating.

An extensive 2024 study by three economists associated with the National Bureau of Economic Research attempted to measure the raw effects of noncompetes on wages. The study compared states where non-competes are broadly enforceable with the four states that ban them outright. The authors concluded that “rendering NCAs unenforceable nationwide would increase average earnings among all workers by 3.5% to 13.7%.”

If that estimate is even approximately right, the implications are enormous. With total U.S. wages last year at roughly $11.71 trillion, a nationwide ban would translate into between $400 billion and $1.6 trillion in additional earnings for workers each year. Those are not marginal gains.

Another study, particularly ingenious in its design, helps explain how these effects arise in practice. Columbia University economist Bo Cowgill and two co-authors conducted a large-scale field experiment involving roughly 14,000 job offers for recruiter positions. The study was conducted in collaboration with real companies, and the job offers were genuine. The contracts were randomly assigned into three groups: some included no non-compete clause at all; some included a non-compete that was clearly disclosed early in the contract; and others included the same clause but buried deep in the document, on page seven.

Because the contracts were signed online, the researchers were able to observe reading behavior directly. Almost no one read the buried provision. Even more striking, although the firms made little effort to enforce the noncompetes, the effects on worker outcomes were dramatic. The researchers found that “removing a noncompete increases mobility between competing employers by 30–57% and raises workers’ total earnings from the two firms by 12–16%.” Scaled across the labor market, those figures once again translate into massive amounts of lost income.

Defenders of non-competes sometimes respond that these wage and mobility effects are the price society pays for encouraging training and innovation. If firms cannot lock in employees, the argument goes, they will underinvest in training and knowledge creation. Yet here too, the empirical literature points strongly in the opposite direction. Studies consistently find that non-compete enforcement reduces entrepreneurship and startup activity along multiple dimensions.

One influential study found that non-compete enforceability “substantially reduces the rate of patenting: an average-sized increase in NCA enforceability leads a state to have 16–19% fewer citation-weighted patents over the following ten years.” A 2025 paper in the Journal of Corporate Finance similarly concluded that non-competes compromise the value of innovation investments themselves. By examining stock market reactions to patent announcements, the authors found that non-competes “compromise value creation from real investments by reducing worker incentives and hindering knowledge diffusion.” In other words, non-competes appear to undermine precisely the innovation they are supposed to protect.

Given the growing pile of evidence and the American taste for economic liberty—at least in the post-Civil War era—it is reasonable to ask why noncompetes have not already been banned nationwide. To be sure, a growing number of states have enacted full bans, and others have adopted partial bans, often limited to workers earning under roughly $120,000 per year. These state-level efforts are commendable. But the resulting patchwork of rules creates uncertainty, uneven enforcement, and incentives for regulatory arbitrage. From a worker’s perspective, a clear nationwide rule would be far superior.

As in so many areas of American policy, the effort to ban non-competes has become sidetracked into a legal sideshow. In 2021, the Biden White House issued an executive order encouraging the abolition of non-competes. The Federal Trade Commission, then chaired by Lina Khan, promulgated a nationwide ban. Employers quickly challenged the rule, suing in Texas, and the issue became entangled in a broader fight over the scope of the administrative state and the FTC’s authority.

The result is that what should be a straightforward national policy debate about labor mobility and economic fairness has instead morphed into an arcane legal argument about the FTC’s statute. Specifically, the controversy now turns on what Congress meant when it granted rulemaking powers to the FTC in the late 1970s. As interesting as that question may be to a small number of academics, it is a poor substitute for an open and democratic debate over national labor policy.

Non-competes are low-hanging fruit. They clearly hurt workers. To the extent employers have legitimate interests in protecting trade secrets or confidential information, the law already provides less restrictive tools for doing so, including trade secret laws and narrowly tailored confidentiality agreements. A ban on non-competes is not meaningfully partisan, nor does it advantage one region of the country over another.

One need not be obsessed with artificial intelligence to observe that the labor market is changing rapidly. A dynamic, mobile workforce is likely to be essential to future economic prosperity. Locking workers in place has a distinctly eighteenth-century feel to it. It is easy to forget that the Thirteenth Amendment banned not only slavery, but also indentured servitude. Yet in subtle ways, we have allowed contractual practices that move uncomfortably back in that direction.

In the end, the calculus is relatively simple. Non-competes keep labor costs down, which benefits employers. For precisely the same reason, they harm workers. In an era when the balance of power between labor and business is already skewed, banning non-competes is an unusually direct way for a new administration to improve the lives of millions of workers. The justifications for non-competes increasingly look like distractions from that basic fact.