[caskstrength]

> Quant firms at least are one of the few places where noncompetes can make sense. It's an extremely IP sensitive industry with stupendously high pay where the employee is going to someone probably competing very directly with you, for the same/similar opportunities.

Cry me a river. If knowledge of some particular employees worth so much to the quant firms, then they should pay them not to leave accordingly.

[bumby]

Employees don't hold ownership of that intellectual property, though. You're speaking almost in terms of a moral right; IP rights are legal rights of convention. An employee isn't entitled to them in the same way.

[caskstrength]

Right, and you don't need a non-compete to go after former employee stealing your IP.

[bumby]

Non-competes and NDAs are literally the mechanisms that companies try to protect their trade secrets. Patents, copyrights, etc cover publically disclosed IP.

The OP was about how non-competes make sense in an IP-intensive field, like quant finance. The reason is that these contracts help protect the IP by explicitly stating their case. Your comment goes against the very foundation of IP law: creating reasonably fair commercial opportunities. If I can extort you because you hired me and I learned your secrets, I think that pushes the scales beyond "reasonable."

[hadlock]

If you are making $10 million a year based on an employee's personal contribution to the company, and paying them $135,000, they are likely underpaid, and another company might gladly pay them $250,000 to add $10mm to their bottom line. But the non compete holds them in the job paying less. Their value to the company clearly allows them to pay $250k to that employee, but it's the non-compete that is allowing the company to profit an additional $120k. There's no case for non-compete beyond "excessive profit margins".

[usefulcat]

Merely having knowledge of a profitable trading strategy is not at all the same as having invented it, or even being capable of doing so.

[bumby]

The short answer is "inevitable disclosure doctrine" that prevents you from working for a competitor if it's inevitable that you will disclose trade secrets. It's a sticky wicket for engineers.

However, this might be confusing different issues. My comment was specific to using NDAs/non-competes to protect trade secrets. This is different from merely using them to prevent poaching by competitors. In cases were there isn't inevitable disclosure, I think it's much less likely that a non-compete would be enforced in court.

[LeifCarrotson]

I totally agree that if an employee adds $10M or $1M to the bottom line and you're paying him $100k, that's under compensation.

But there's a categorical difference between that situation and when an employee or dozens of employees who may be a break even or negative impact on profits have knowledge of a trade secret researched by a team of their predecessors that makes the company $100M.

I'm all about fair compensation and worker's rights, but a business shouldn't have to pay all those people $100M salaries.

[bobthepanda]

You can certainly pay them the duration of the noncompete at their prevailing wage.

In fact, this already exists. https://en.wikipedia.org/wiki/Garden_leave

It really should be required of all noncompetes.

[olliej]

No, NDAs are the tool to stop your employees leaking information. Non competes are the tools you use to avoid paying market price for your employee.

[bumby]

It depends. Some jurisdictions blur the lines between NDAs and non-competes. For example, in many areas NDAs can also prevent you from working for a competitor due to "inevitable disclosure."

[caskstrength]

You are arguing with straw man since no one in this thread argued to bad NDAs. To me personally NDA seems to be a reasonable concept while NCA is just a wage depression tool.

[bumby]

The parent comment of the one I replied to was specifically referencing IP as a use case for NCA. You allude to “knowledge” of IP warranting higher pay. Besides, elsewhere it’s discussed how NDAs can act as de facto NCAs due to the inevitable disclose doctrine.

[caskstrength]

> Besides, elsewhere it’s discussed how NDAs can act as de facto NCAs due to the inevitable disclose doctrine.

And that is much better case than NCA since it would only apply in specific narrow cases and wouldn't prevent a McDonalds employee from working in fast food industry for a year, for example.

[shermozle]

You're suggesting that if an employee wants to change jobs, they should work in a field for which they have less expertise and thus get paid less. How is that reasonable?

[bumby]

That is not what I’m implying. You can work in a similar industry without taking trade secrets, except in the case of inevitable disclosure. For example, if you quit writing for Coke you can still go work for Pepsi without disclosing the Coke recipe.

[hamandcheese]

Knowledge of a secret does not imply that you provide value.

[akira2501]

Are you speaking towards the employee or towards the quant firm? If the employee has no standing to claim value, then why does the underlying business get to?

[bumby]

Because they own the trade secret. For example, they have the legal right to license a trade secret; an employee does not. It's about legal ownership of intellectual property.

As a corollary, you may read a patent and now have the knowledge of a product. But you don't have the same legal right to create and sell that product. That right is protected by the patent owner.

[SOLAR_FIELDS]

Of course, that line of thinking involves the inevitable follow up: when is something a “copy” vs a “genuine invention”. If company has a patent on making widget A, how different does a previous employee who leaves and makes widget B have to be before it’s not considered a violation?

I am no expert here but my understanding is that the case law around this is much more well trodden in patent land than it is for noncompetes

[bumby]

You're right, but there are some nuances that I would expect a good law to address. My presumption is that trade secrets would have to be covered by confidentiality agreements, which are distinct from noncompetes. The threshold for infringement of intellectual property is a "preponderance of evidence"; i.e., it's "more likely than not" or "greater than 50%" so it's not a terribly high threshold to prove compared to other sorts of law.

If it's truly patented (different from a trade secret), you can't produce it, even if your invention is slightly different. For example, if I hold a patent on a "car" and you make a "car with a radio," you still can't produce it because it infringes on my patent. You can't make your product without covering the totality of my claim. That's why people try to make patent claims as broad as possible.

[master_crab]

That’s not a very good example. A patent is available to read specifically because the discoverer has entered an agreement with the government to share the relevant information in return for exclusive use for a set period of time.

If they had not patented whatever it is they had, anyone could replicate the information/item in question with no penalty.

[bumby]

>anyone could replicate the information/item in question with no penalty

Sure, I suppose someone could develop/copy something in parallel with no knowledge. But that's not really the case in the discussion here as it comes to former employees.

If you worked for Company A which uses a proprietary algorithm for trading and somehow created the same for Company B later, would you really expect a jury to think the two are unrelated? As stated above, the threshold is "more likely than not" that your work for Company B is related to knowing the trade secrets of Company A. If you had never worked for Company A, maybe, but again that's not the case here because a noncompete would never enter the picture.

[master_crab]

> Sure, I suppose someone could develop/copy something in parallel with no knowledge.

I mean that’s a tad disingenuous as to how it worked before patents. Patents were meant to dissuade others from copying inventions for a certain set period. It was much rarer to see independent development of the same technology (not that it didn’t happen).

[bobthepanda]

clearly there's value in the secret if you're making them sign a noncompete.

The proposal is just to ask firms to put money where their mouths are.

[tester756]

He says that just because someone knows a secret, then it doesnt mean that he brings value/$$$ to the company

They should pay him 500k just because he knows a secret? even if he's making just 50k to the company?

[standardUser]

Obviously companies think it is worth a lot for certain employees to be quiet, so they pay accordingly. There is no "should" going on here, there is only what the two parties agree to.

[Clubber]

They should pay him whatever he would make going somewhere else. Expecting someone to not work in their career field for 1-5 years and not get compensated for that is silly.

[mistrial9]

LOL - you have stumbled upon the pay grades for security and executive management directly ! they absolutely are paid more to participate but stay quiet. It is a daily requirement.

[bobthepanda]

A noncompete has value. If its worth 50k then pay for a noncompete equivalent to 50k wages worth of hours. Simple as that.

[jjk166]

If someone else would pay them 500k to reveal the secret, then yes. Maybe don't give 500k secrets to people producing only 50k of value.

[davorak]

> Knowledge of a secret does not imply that you provide value.

If it did not provide value to tell the employee the secret then it follows that company would/should not tell the employee the secret.

[caskstrength]

"Secret" should be covered by NDA and/or patented.

[bumby]

That's the point. If it's covered by an NDA and the state declares NDAs unenforceable, there are no NDA-covered "secrets." I can get hired, learn all your secrets, and then sell my employment to your competitors. That system doesn't really work.

In practice, trade secrets are protected by other mechanisms. Patents are one of them because, by definition, patents are public knowledge so they are no longer secret.

[vinceguidry]

To play devil's advocate here, this law would literally make that illegal. What keeps me from taking the money then leaving anyway? A signed contract is the only way, and banning non-competes means making that contract unenforceable.

[helicalmix]

…sorry, how much exactly do you think quants make?

[KRAKRISMOTT]

Mid 6 to low 7 figs usually. They are usually one "tier" above SWEs i.e. an average new quant would make the same amount as an average mid-level NYC software engineer, in absolute monetary terms of total compensation. The high end firms like Jane Street and Two Sigma pay the equivalent, scaled to FAANG levels. Overall performance of the firm is a key factor in the compensation too.

[sesuximo]

People sometimes leave and take less money. Plenty of other reasons to change jobs.

[creer]

For a large team, and finance, there is no amount of pay that could compare to the potential gain of creating your own firm to reimplement some specific idea you learned on the job.

The more likely response then would be to move the team somewhere that does allow non-competes?

[charcircuit]

You can work at 2 companies without leaving either.

[bobthepanda]

presumably the noncompete wouldn't apply then, if the two companies were not competing, and then you wouldn't need to pay them at all.

[juped]

to the struggle SWEs who exaggerate your salary on HN and are upvoting this comment to the top: please stop and think for a second