Using IP to Create a Noncompete
Given the limited duration of a noncompete agreement, the geographical restrictions, and the practical nonenforceability of many of these agreements, they are clearly inferior to using Intellectual Property (IP) litigation in the field of finance. The IP mechanism is to assert that an employee agreement relating to confidential information entitles you the ability to vet ex-employee's proposed activities. If the employee reveals information, say it is derived from what he did at your company. Surely it will be 'related' if it involves finance, making the question of derivation plausible. If he does not reveal it, say it sounds like it was derived from what he did at your company, and his failure to fully reveal it is suspicious. Employers are granted large lattitude in their rights of discovery, and only allow damages if you can prove bad faith (as well as damages and unjustified allegations), which has an incredibly high standard of proof (eg, an email stating "let's screw John by filing frivolous allegations"). The employer can then credibly say to the ex-employee, either you do not compete, ever, or be prepared to not work for a few years and pay hundreds of thousands of dollars in litigation expenses. Anyone with less than $200k available to for such emergencies will find it an offer hard to refuse.
The key is a confidentiality agreement. Judges believe in protecting property rights, which are surely a key to a modern economy (see Hernando De Soto's work on the importance of property rights in the developing world). The problem is that any right is counterbalanced by other rights, an example famously being the right to swing my fist ends at your nose. In this case, people have a right to pursue business activities without fear of civil action on themselves or their business associates.
The beautiful thing about the confidentiality agreement is that one can use it to start discovery on a broad scope of information, and then generate a post hoc definition of what is covered. This is a long and costly process, certain to provide issues that need defending. The key is that overbreadth in technical matters is not obvious, for example, one can say, "he took the secret of mean-variance optimization", and by the time the court figures out such a claim is absurd, a more tenable claim can be made, such as 'a specific application of mean-variance opitimization that will be defined after discovery' (also note, to the defendant, both claims place all his activities as potentially poisoned, so the effect on the defendant's ability to work is unabated). Claims are allowed to change, or be replaced entirely, so one is not stuck with original, deficient claims that might surely stop the process.
Inconsistency, and ignorance of public domain or prior use in IP allegations are allowed as long as they are not obviously in bad faith. For example, pairs trading, while common among hedge funds, is something most people have never heard of. The onus is on the ex-employee to convince the judge that pairs trading is common, and so a program on pairs trading was not derived from work uniquely created at the direction of the ex-employee. Such an assertion may be easy to defeat ultimately, but that will be a question of fact only a jury can decide, only an expert witness can testify to, and the ex-employer will point to suspicious similarities between programs ran at his shop, and those he planned to run elsewhere. All the while, any potential employer or partner the ex-employee might work with faces undefined liability because no one knows what the tainted concepts or programs are, or more importantly, are not.
Many courts do not require a trade secret plaintiff to identify its trade secrets with specificity before discovery, which makes this tactic even more effective. The problem with ex poste definition of trade secrets was nicely argued by Graves and Range in a 2006 article. I read this article and figured this tactic had surely jumped the shark, because it is a patently calculated and abusive, and in effect makes the defendant have to prove himself innocent in order to work at great expense. The basic strategy is to define the IP at the broadest level possible, then, after discovery, and in response to cues from the court and learning about the issues (generally lawyers do not know financial issues at the level of trading strategies), change one's claims based on the new information. Given enough information, and a complex issue, something will appear sufficiently questionable to allow litigation to proceed. I presumed that the this backdoor of using intellectual property to enact a noncompete via post hoc discovery would have been shut. But it appears the Graves and Range piece merely articulated a playbook that is growing in popularity because it works. Hedge funds are good at arbitrage, in this case, arbitraging IP law to effect noncompete agreements.
Many people presume that if you have not done anything wrong, you have no worries from invasive search (see Solove), in that the only reasons to prevent showing everything to everyone is modesty (eg, letting people see you naked) or that you actually committed a crime. Many judges think they can distinguish between legitimate modesty issues and a crime, so what is the problem? The problem is that if the crime is something defined after discovery, that something will be found with sufficient motivation and data. Not everyone looks at data in good faith, even if they aren't on record as saying they have bad faith, and given the complexity of the law, the ambiguity of words, with enough data, and the right attitude, establishing sufficient grounds for litigation is simply a matter of time. For example, in Friedrich Durrenmatt’s novel Traps, which involves a seemingly innocent man put on trial by a group of retired lawyers for a mock trial game, the man inquires what his crime shall be. “An altogether minor matter,” the prosecutor says, “a crime can always be found.” An example of abusive open-ended discovery is when President Clinton was impeached for perjury, and this crime was the end result of an investigation into a real estate deal from the 1980's while he was governor of Arkansas. He was technically guilty, but people generally were sympathetic to the injustice of it given he was lying about a sexual indiscretion in 1997, which had absolutely nothing to do with the Whitewater real estate transaction of the 1980's.
The US Government's ability to listen in on overseas phone calls is trivial compared with the total bandwidth involved when having all of your computers given to a vindictive ex-employer, yet there is no movement to stop a wealthy employer's ability to punish ex-employees with this tactic. 10 years of daily writing emails, rants to colleagues, discussions with potential employers, websites I have visited, jokes to friends, downloaded files, etc. Have a law firm spend 18 months combing through those, and then tell you what you did wrong, is like drawing a bullseye around a dart thrown at a blank wall.
In July 2007, Renaissance won an estimated $20mm settlement with Millenium though the defendands admitted no guilt. What interested me most was that the suit commenced in December 2003, discovery was granted in December 2004, and the settlement was in July 2007. Renaissance's case against Pavel Volfbeyn and Alexander Belopolsky continues. One needs a considerable nest egg merely to fight such a suit.