Why don’t people copy other funds?
Sounds like a stupid question, I know.
But let’s say John joins a quant fund like Two Sigma (or RenTech or Citadel) and does well there. He then starts his own fund, copying 2Sig’s entire setup, or at least the majority of their models and systems. John then raises money, makes returns, etc. Obviously he implements his own strategies and adds his own skill to the new fund, but by and large his great returns are based on his copying.
There are two cases: this is illegal or it isn’t. If not, why doesn’t everyone do it?
And if it is, how does any quant researcher or trader ever start their own fund? Wouldn’t they necessarily be using strategies or models from their former employer to get started?
This is just a random thought I had. I don’t know too much about this stuff so any help is appreciated. Thanks in advance.
Pre sure those mfs have one of the most aggressive non-competes in the world.
Right, but if there are non-competes how does anyone ever start their own fund? Like how did the founders of Jane Street start their prop firm after leaving SIG?
Bro these non-competes dont last a lifetime
A couple things:
1) non competes: usually won’t allow you to compete directly for a specified period of time
2) silo: many quant funds will be more siloed to prevent this type of thing. You might know one strategy or even just a piece of a strategy but not the full thing. Makes it much harder to go copy the fund somewhere else.
Right, but if there are non-competes how does anyone ever start their own fund? Like how did the founders of Jane Street start their prop firm after leaving SIG?
Short answer, because they're geniuses that represent the 0.001% of outcomes so there's 0 point in trying to compare your situation to theirs in any way meaningful enough to base decisions off of.
Non competes are temporary (1-2 years).
As for other questions you’ve been asking:
1) in the quant space, the data and infrastructure cost is high. You can’t just leave and have all that tech and data come with you. Even if you take “the models” the ingesting the data (and paying for it), having the tech to support it, etc is A LOT of work.
2) as I mentioned, the silos, even without clear silos, there is a lot of code at many of these places and knowing how it all works (not just at the concept level but at the actual code) is very difficult (and usually very guarded)
3) as for why places sue, it’s to enforce the contracts. If someone leaves and goes and tries to replicate your stuff, you need to decide what to do about the contract. If you let a bunch of people do that because you don’t think they are doing anything “good” then as soon as someone who does have “good” ideas does that, you have set a whole lot of precedent on not enforcing the contracts and that’s not good.
Non specific to those firms, but the short answer is this happens a lot. Many experienced investors have spun out their own firms using similar frameworks and methods to their old firms but with a twist. Algorithms aside, most intellectual property in investing is impossible to protect (frameworks, rubrics, mental models, industry work) but rather it's the application of these things which is hard to replicate. For quant hf probably clearer lines around not stealing actual code but I don't know that space.
And the non compete is often notional. It's a bad look for a PM to enforce that on an analyst who wants to be a PM but can't be in their setup. Most spin outs are done with the blessing of the PM.
Then why do firms like RenTech and 2 Sigma go crazy with the lawsuits whenever someone leaves and starts their own fund? What determines whether someone is allowed to start a fund and whether someone is blocked by their original firm?
The quant places definitely do sue people who try this, and whether they succeed or not depends on a lot of things, how the court settlements pan out, etc. Most of the successful spin-out quant funds were formed decades ago when the industry was much less mature. Some of the early hires were the founders' friends and had their backing, which would not be the case for a recent hire. The siloed codebases also make it harder today but there was a time when they were not siloed. Even if the employee does not take code, they can be sued for remembering the code in their head.
So then how does anyone start a quant fund these days? Do people start them?
They sue because they can, they give aggressive noncompete cause they do not care. As someone mentioned the margin of error is much smaller in the quant world and certain products so if someone goes out of their way to do what you mentioned they doomed.
So why is the little guy sued and the bigger name allowed to make a new fund? Cause the bigger names have more money behind them, its all about money and rep.
How does anyone start a quant fund today then (ex: Alameda Research from SBF)? Do people start them?
some probably use something similar to what they were using at the quant fund. Thing is, most people think they can make an improvement on something, so they try to make something better and a lot of times it ends up being worse for various reasons.
You could probably ask this same question in every industry, why don't people just copy what the best do. A lot of times that happens, and it makes the market more crowded. However, sometimes someone can get close but not exact. Look at the smart phone market, there's those pictures where they show phones before the iPhone and after; the before everyone is different, the after is everything looks the same, but Apple still kept a dominate share of the market.
It's different, because in say the tech industry, everyone is competing for the same quantity - customers - and it's a zero-sum game. Whereas tomorrow if I copy every trade that Citadel makes (impossible but demonstrates my point), I can make the same percentage returns that they do.
Not sure if you been following but no quant dude on earth needs to start their own fund today. Ken will steal Izzy's guy give them a crazy guarantee and let them do their thing. Izzy will nab David's guy and let them var up like mad and go to war. Dmitry will intersect Ken's guy as he is going to the airport on family vacay and offer an out ready to go.
Izzy just changed his fee structure to match Ken...."steady profits" is all they ask and if you a good quant they will give you all you need. Non competes are getting extended all the over the place with bold letters "if you work for us, you cannot and will not talk to Ken ever"
Also for many quant algos, they tend not to last after a couple of years as these inefficiencies either disappear or become exploited by more funds, making it harder for the algo to extract alpha. This is why quant HFs are constantly innovating on their strategies to find inefficiencies that no once else knows about yet.
This is also why a non-compete is a decent barrier to preventing ex-employees from copying code and starting their own fund using that code.
Not to mention, as others have previously mentioned, that the start up costs for a quant hf is immense due to the massive tech infrastructure required. Finally, not everyone can raise money to set up their own fund. It takes a lot of trust and due diligence before an institutional investor is willing to trust you with their money.
If they don't last a while, why do firms like RenTech pull out the lawsuits every time someone tries to leave? What separates someone who successfully exits and starts their own fund from someone who is sued to oblivion? What is the boundary between preventing copying of stuff and letting people start their own funds?
Well it's better to be safe than to be sorry. You also sort of answered your own question in that lawsuits make the hassle of ex-employees stealing IP in order to set up their own firm not worth it.But also the people who left to start their own quant funds have either done it with the blessings of their previous fund, finished their non compete period, or did not have any non-competes to bind them (possible way back when quant investing was very much Wild West)
Also in one’s contract, there’s typically a clause that says any work/invention/idea created by the employee during employment will be considered as part of the firm’s property. So it’s literally theft to copy code from a quant firm once the employee leaves the firm.
How do you think Two Sigma was founded?
2 Sigma was founded by the people who straight up started the quant stuff, if I understand correctly. Not many people used ML or similar for trading before Siegel and Overdeck.
lol. Where do you think Siegel and Overdeck worked before 2Sigma
have seen many quants who are sued/even some sentenced to jail lol because of this IP issue.
Yeah, that's why I asked. Never thought you could go to jail for something like this.
Quant heavy shops silo things out a ton structurally. You can’t just upload something to a USB drive and have the firms magic in a box lol
If this is true, why do RenTech and 2 Sigma go crazy with the lawsuits? What is the problem if everything is siloed?
It is true and they do that as a deterrent.
Same reason you litigate aggressively with any IP you own...
Dude not everything is black and white. Even if teams/info is silo'd, a former employee still has knowledge of strategies belonging to their silo, which is still alpha and they're not going to want that to leak. It's like asking why people care about losing 10% of their net worth when they still have the other 90%???
While there are certainly some investors who may try to replicate the strategies of successful funds, there are several reasons why this is not always a viable or desirable approach:
Access to information: Many successful funds have proprietary research, data, and investment strategies that are not publicly available. Without access to this information, it can be difficult to accurately replicate their approach.
Different risk tolerances: Even if an investor is able to replicate a successful fund's strategy, they may not have the same risk tolerance. A successful fund may be taking on more risk than an individual investor is comfortable with.
Fees and expenses: Successful funds often charge management fees and other expenses that can eat into returns. An individual investor may not be willing or able to pay these fees.
Voluptatem et id odit ut. Ab ipsum aut a est. Cumque eveniet et eveniet. Dolore fugit nulla qui voluptatum vitae. Distinctio rerum enim soluta est. Et necessitatibus rem et facilis vitae quisquam quae quis. Molestiae repudiandae quaerat eos illum quo molestias maxime.
Consectetur vitae ratione assumenda. Architecto numquam rerum laborum deserunt vero provident atque. Cupiditate voluptates quos blanditiis quis officiis dolores aliquam.
Architecto qui qui cupiditate unde voluptas officiis. Consequatur consectetur est omnis reprehenderit quaerat corporis molestias porro. Non quasi voluptates maxime dolorum autem deserunt. Repellendus voluptas quisquam accusantium rem.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...
Totam dolores odit ipsum numquam. Natus et temporibus et hic ut fuga eos. Ea saepe dolor et aut.