Hacker News Reader

Comment by chaoz_ | Hacker News Reader

Would be interesting to see how contracts in HFT are going to change. From what I heard, when some specialists (e.g. hardware folks) are changing jobs, then the firm is likely to lose an "edge" as competitors will have access to similar technology/knowledge. Garden leave helps to mitigate this effect.

If non-competes are banned, then TCs are (probably) going to rise even more.

View on HN

> when some specialists (e.g. hardware folks) are changing jobs, then the firm is likely to lose an "edge" as competitors will have access to similar technology/knowledge.

So without the employee the firm can't compete? If that's really true, then that employee should be able to extract as much blood as they want. They are literally the keystone of the business.

The good news is that the FTC rule would allow the firm to enforce a non-compete by giving the employee a 25% equity stake. That seems like a fair trade in exchange for with-holding such incredibly valuable specialized knowledge from the market.

It's less that and more that once you see the inside here you can pass that on to the other place. Knowledge is everything. And it turns out you don't have to be the one to make something in order to memorize it.

I work at an HFT. I am expecting most of the major firms, especially Citadel Securities, lobby hard against this.

While non-competes on our side are paid, they're a pain in the ass to navigate if one is a visa worker (my case). So, I am praying for this to pass. If it does, I am starting interviewing with competitors the very next day.

Is it harder to come up a new idea for a trading strategy or is it harder to actually implement the system to perform those trades?

Not being in that industry, to me it seems like the easiest part is coming up with a new trading idea. The hardest part would seem to be actually implementing it without bugs. Therefore, I'm not sure if noncompete clauses really provide a super-significant amount of value. Even if you left a firm with a super good trading strategy today and wanted to implement it elsewhere, you're still going to have to perfectly re-build it, which is probably not trivial.

Both. If one is on a competitive exchange or a latency-sensitive asset class, need to be both smart and fast.

Also, strategies change all the time and are mostly the domain of quants and traders. What they did a year ago, could very well be history at the firm they worked at.

I'd say an idea that generates alpha is tougher. I have seen firms like XR have great technology, but sucky ideas. They were fast, not smart and hence they didn't make a killing like their competitors did, due to all the volatility in the last three years.

For quant finance and HFT noncompetes usually pay out pretty generously

Interestingly, the FTC rule does not include an exception for garden pay.

Probably just as well. Garden leave can be a pretty unsatisfactory mitigation, especially if it's not 100% of salary or if a lot of total comp is in bonus or equity. And even money aside, some people may not want to put their career on hold for a year.

Essentially garden leave is a compromise that's better than nothing and will make companies think twice about trying to enforce a non-compete. However, non-competes still raise flags for potential employers as well. (I used to work for a very small company and we wouldn't touch anyone who had a non-compete in place.)

Reply on news.ycombinator.com