Contract to sign States that want to cultivate the technology clusters of tomorrow would do well to follow California’s example and ban noncompete agreements. (Photo: Shutterstock)

Despite sky-high housing prices, Silicon Valley remains the vital center of the U.S. tech industry. Once an industry takes root in a particular place, it’s very hard to pull it away. Which makes it all the more important to understand: Why did Silicon Valley become Silicon Valley in the first place?

One prominent theory is that California’s culture encouraged a free flow of workers and ideas between companies, combining and re-combining in ways that spurred continuous innovation, instead of being jealously hoarded in corporate offices.

Ideas flow from company to company in a few ways. They can be communicated in a casual conversation, or by showing a product to a friend. They can be conveyed in formal settings like an industry conference or a research paper. But most ideas are contained in people’s heads, and so the main way for ideas to get from one company to another is for workers to switch employers.

Related: The recruitment economy hinges on poaching workers

This can’t happen if workers sign noncompete agreements, which mandate that a departing employee not go to work for a competing company — even if she or he is fired. California, fortunately, doesn’t allow this sort of agreement to be legally binding. Stanford professor AnnaLee Saxenian has said that if noncompetes had been enforced, Silicon Valley “would probably not be what it is today.”

States that want to cultivate the technology clusters of tomorrow would do well to follow California’s example. I have suggested that Wisconsin could ban noncompetes in order to help turn Madison into a tech hub. Pennsylvania could do the same to solidify Pittsburgh’s place as an emerging robotics cluster, and Ohio could help nurture nascent startup booms in Cincinnati and Columbus.

But the benefit of banning noncompetes probably goes far beyond the boost it gives to technology clusters. Workers in general have been seeing very small raises over the last few decades:

Noncompetes are contributing to this problem. When workers can’t choose to quit and go work for another company in the same industry and city, it represents a dramatic increase in the power of the employer over the employee. When a worker is shackled to a company, the company has very little reason to give that worker a raise.

Nor are noncompetes limited to technology workers. A 2016 Treasury Department report found that 15 percent of workers without a college degree, and 14 percent of workers making less than $40,000, were bound by noncompete agreements. Sandwich maker Jimmy John’s was forced to abandon the practice only after a public scandal. A 2016 survey of over 11,000 workers found that almost 40 percent of workers had signed such an agreement in the past.

The harmful effect of noncompetes appears to extend far beyond those workers that are directly bound by them. A recent study by economists Evan Starr, Justin Frake and Rajshee Agarwal found that workers in states and industries with lots of noncompetes suffer all sorts of ill effects — lower wages, less job mobility and lower levels of job satisfaction — even when they themselves aren’t bound by any such agreements. The economists found that the more enforceable noncompetes are according to local law, the more severe the effects become.

Why would workers who hadn’t signed noncompetes be hurt by them? For one thing, they’re competing against workers who have signed them — who can’t remain in their field if they change jobs, and are therefore willing to accept lower wages when they do switch jobs. Also, when noncompetes are prevalent, workers who don’t want to be bound by such agreements will be shut out of some jobs, decreasing the number of companies that can bid up their wages. Finally, noncompetes reduce entrepreneurship, by preventing workers from quitting and starting their own rival companies in a space; that also tends to reduce wages and hiring.

Noncompetes aren’t just hurting workers; they’re hurting economic dynamism.

There is at least one potential upside of noncompetes: increased business investment. Economist Jessica Jeffers recently found that although noncompetes do lead to fewer startups and less entrepreneurship, they also cause existing companies in knowledge-intensive industries to invest more. Secure in the knowledge that their employees can’t easily abscond with their ideas, companies are more willing to commit to new projects. More business investment means more economic activity, so that’s a good thing, but it’s tempered by the lack of investment that comes from having fewer new companies. Ultimately, new fast-growing companies provide more labor demand than confident established employers, so noncompetes are still a net minus for workers.

Given the public interest in helping American workers and supporting entrepreneurship, every state should take action to curb the use of noncompetes. Some are doing so. Montana, Oklahoma and North Dakota ban the practice, and Illinois severely limits it. Hawaii bans noncompetes for technology jobs, and New Mexico bans them for health-care workers. Utah and Oregon limit how long such agreements can last. Massachusetts, which famously lost out to Silicon Valley in the competition to become the country’s technology center, is considering reform as well.

This movement needs to spread. In a time of stagnant wages and falling dynamism, the country’s workers and startups need all the help they can get. States that move fast on this issue are likely to reap outsize rewards.

Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion. This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Copyright 2018 Bloomberg. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.