The high-frequency trading business is nothing if not opaque. Noted author Michael Lewis shone light on it in the wonderful book Flash Boys. And the industry gave us Sergey Aleynikov, who has managed to contribute mightily to the area of trade secrets and corporate indemnification law over the past decade.
The HFT world, when it comes to non-competes, is predictably ahead of the curve. Having represented many quants and traders, something of an industry standard has developed. And I'm not sure it has been replicated elsewhere.
HFT non-competes often work like this. Employee signs an employment agreement that contains salary, bonus eligibility, and non-compete restrictions, along with the panoply of confidentiality clauses and invention assignment provisions you would expect in a business that thrives on opacity.
For starters, HFT firms don't really need customer-based clauses because they have no customers. So the non-compete is a true industry-wide restriction, justified largely on the basis that the firm's business model is proprietary and the employees are privy to a wide range of non-public trading strategies and confidential information.
The non-compete, though, shifts. It does so based on the company's election, at the time of termination, as to how long it will last. Typically, the HFT firm agrees to pay the employee his or her salary (or some percentage of total income) during the selected period. In my experience, the period can last 0, 3, 6, 9, or 12 months. And, according to a typical HFT contract, the company decides and notifies the employee of the non-compete term, paying the employee what amounts to garden-leave during that period of time.
This seems relatively uncontroversial (some might even say it's fair). But as shown by Reed v. Getco, LLC, a Chicago HFT firm, Getco, deviated from this model. Its contract contained a six-month non-compete clause, with consideration of $1 million payable to the employee during the term. When Reed quit, Getco advised him that his non-compete term was "zero months and/or is waived. You will not receive any noncompete payments."
Reed then sat out and did not compete for more than six months, seemingly abiding in full with his contractual covenant. He then sued for the noncompete payment. The appellate court found he was entitled to it. The court did not allow Getco to rely on boilerplate contract provisions concerning "waiver" and "modification," finding apparently that Reed had a contractual right to the $1 million - which he earned by complying with a non-compete Getco told him it was not going to enforce.
What is not clear from this opinion is whether Getco's agreement incorporated the industry standard non-compete that shifts, based on the employer's election. It appears that Reed negotiated something separate than that. So at first blush, I am not sure the Reed decision does much to upend the HFT business model for enforcing non-competes.
To that end, HFT firms still may be able to get away with their shifting non-compete term, provided that the contract makes it clear the payment is optional and dependent on the firm selecting an appropriate non-compete period. Getco may have been thinking that it would treat Reed like other employees - and simply forgot that his contract was much more custom.