Your Non-Compete Wasn't Protecting Secrets. It Was Cutting Your Pay.

The non-compete in your onboarding folder almost certainly protects no secret. 95% of companies that use one say it guards trade secrets, but only 6% ever actually enforce it in court. The gap between those two numbers is where your raise went. The agreement was never a legal shield. It was a way to keep you too scared to leave.

That fear is the entire product. Once you see that, you can read the clause honestly instead of obeying it on reflex.

Who actually signs non-competes, and is it really mostly executives?

It is not mostly executives. Around 30 million American workers, roughly 1 in 5, are bound by a non-compete. The median holder of one earns about $14 an hour. That is sandwich makers, warehouse staff, hair stylists, and nonprofit volunteers, not just the VP of engineering with the source code in her head.

The reach goes further down than most people assume. 14% of workers earning under $40,000 a year, and 15% of workers without a four-year degree, are subject to non-competes. On the employer side, 55% of surveyed private firms use them for at least some workers, 72% do not exempt hourly workers, and 53% apply them to all part-time employees.

So if you signed one and you are not a chief officer of anything, you are the norm. The instrument was built for you, not for the corner office.

Are non-competes really about protecting trade secrets?

For most of the people bound by them, no. This is the smoking gun, and it is two numbers from a single government report sitting side by side.

95% of employers who use non-competes cite trade secret protection as the justification. Only 6% frequently enforce them, and 73% rarely or never do. Read that again. Ninety-five percent claim a secret worth defending. Six percent ever defend it. The other eighty-nine percentage points are workers walking around believing a fence is real that the company has no intention of ever testing.

Look at the weak version, the one that made the news. Jimmy John's required sandwich employees to sign a non-compete barring them, for two years, from working at any shop within three miles that earned 10% or more of its revenue from sandwiches. There is no trade secret in a sandwich. There is no proprietary lettuce. The only thing that clause protected was the company's ability to keep wages low by making it harder for you to walk across the street for a better hourly rate, which is the same pressure behind what gets called a skills shortage but is usually just a wages-too-low shortage.

Now the strong version. A VP of Sales at a SaaS firm who personally manages the top 50 enterprise accounts, carries the renewal pricing model in her head, and is being recruited by the direct competitor. A time-limited, geographically specific clause there has a defensible business rationale. That is the 6%. Almost nobody reading this is the 6%.

How much is a non-compete actually costing you in pay?

Real money, and it compounds. The research is consistent across wage tiers.

Banning non-competes raises wages by 3 to 4% for both low-wage and high-tech workers and increases worker mobility by 11 to 17%. When the FTC modeled a national ban, it estimated earnings for the average worker would rise by $524 a year.

$524 a year sounds like a rounding error. It is not, over a working life. Compounded across 20 years, that suppression gap is a used car, a meaningful chunk of a down payment, or a year of retirement savings. The cage has a price tag, and you have been quietly paying it.

What the research finds	The number
Wage increase when non-competes are banned	3-4% for low-wage and high-tech workers
Mobility increase when banned	11-17%
FTC estimate of per-worker raise from a ban	$524/year
Reduction in patenting from higher enforceability	16-19% over 10 years

The mechanism is simple. You cannot get a raise from a competing offer if you believe you are not allowed to take one, and a competing offer is the single biggest lever you have on your pay. The clause does not have to win in court. It only has to win in your head.

What is the "in terrorem" effect, and why does a useless clause still work?

It works because you do not know it is useless. The Latin term lawyers use is in terrorem: by fear. The paper does not need teeth. It only needs you to believe it has them.

Here is the proof that the fear, not the law, is the product. 29.3% of California firms still hand out non-competes to all workers, even though they have been completely unenforceable in California since 1872. The clause is void on signing. It has no legal force whatsoever. Companies use it anyway, and workers sign it, and then workers quietly take themselves out of better jobs because the paper said so.

The Washington state experiment is the cleaner test. When Washington made non-competes unenforceable for the roughly 80% of workers earning under about $100,000, researchers expected employers to nudge near-threshold pay up to keep those workers covered. None did. And publicly traded company valuations did not drop.

Sit with that. If the clause genuinely protected something valuable, companies would have paid to keep it. They did not pay a cent. The agreement was, for most workers, a piece of paper the employer never needed and never intended to enforce. It works the same way salary history bans quietly reset how pay gets set: take away one lever the employer leaned on, and wages move, which tells you the lever was holding them down all along. They just never mentioned that part.

Are non-competes enforceable where you actually live?

It depends entirely on your state and your salary, and that is the most important sentence in this post. There is no uniform national answer.

Six states now fully ban non-competes: California, Minnesota, Montana, North Dakota, Oklahoma, and Wyoming. If you work in those, the clause in your drawer is void on arrival, full stop. Several other states set a salary floor below which the clause cannot be enforced at all. Washington makes them void for anyone earning below about $123,000 a year.

Then there is the other direction. Florida moved the opposite way in 2025, with its CHOICE Act strengthening employer protections as of July 1. So a marketing manager in Sacramento and a marketing manager in Tampa face completely different realities for the identical clause. Your zip code is doing more work here than your job title.

This is the part you cannot outsource to a blog post. You have to know your state and your salary tier before you decide what the clause means for you.

Didn't the FTC ban these? Where does that stand now?

It tried, and it lost. Do not assume a federal ban is protecting you, because there isn't one.

The FTC issued a national rule banning most non-competes in 2024. A Texas federal court struck it down in August 2024, and in September 2025 the Trump FTC formally dropped its appeal, leaving the rule dead.

That matters for one practical reason. The federal floor everyone expected is gone. What governs you is your state, and the state map is fragmenting, with some places loosening enforcement and others tightening it. The enforcement lottery continues, and your address is your ticket. Nobody is coming to rescue you with a single national rule, so the move is to understand the specific rules where you are.

When does a non-compete actually have a legitimate purpose?

Sometimes. It would be dishonest to pretend otherwise, and the honest read is what lets you tell your situation apart from boilerplate.

A senior person who genuinely carries client lists, pricing models, algorithms, or proprietary methods does represent a real business interest. An NDA alone may not deter a competitor who hires that person specifically to extract what they know. For that narrow band, a clause that is limited in time, tight in geography, and clear in scope can be defensible. That is the VP of Sales from earlier.

There is also a real trade-off worth naming. Some studies find non-compete enforcement correlates with more employer-sponsored training, because firms feel safer investing in workers they can keep. If you are in a specialized field getting genuinely valuable training, that is a real consideration. But a training-repayment clause does that job with far less collateral damage than locking you out of your whole industry.

A weak clause and a strong one look different on the page. The weak one is broad: any competitor, vague geography, long duration, signed by everyone on the team regardless of role. The strong one is narrow, named, and reserved for the handful of people who could actually do damage. If yours covers the whole industry and was handed to the entire headcount, it is a wage tool wearing a legal costume.

What should you do now?

Stop letting an unread clause make your career decisions for you. Run these four questions before you let a non-compete talk you out of a better job.

1. What does my state actually do? Full-ban state, salary-threshold state, or high-enforcement state. This answers most of it before you read a single word of the clause.
2. Am I above or below the salary floor? If your state has one and you earn under it, the clause is likely void on you regardless of what it says.
3. Am I genuinely the 6%, or the 89%? Do I carry a real secret a competitor would hire me to steal, or is this boilerplate that every hire signed? Be honest. Most people are the boilerplate.
4. Has the company ever actually enforced one? 73% of firms rarely or never do. Quiet research, a recruiter, or an employment lawyer can usually tell you whether yours is a paper tiger.

The worldview underneath all of this: agency over fatalism. The clause is designed to make you feel like the choice was made for you. It wasn't. The circumstance is real, the state map is real, and so is your ability to read your own situation and act on it. The fear was the whole mechanism. Take the fear away and you can evaluate the trade honestly, which is all a real career decision ever needs.

Want to pressure-test whether the clause in your offer is a real fence or just paper, before you turn down a better job over it? Message Praxy on WhatsApp and we'll walk through your state, your salary tier, and your actual exposure together.