I have been looking at laws regarding non-compete covenants. A non-compete agreement aims to prevent a worker from accepting other work that would compete with his or her employer or business partner. In employment situations, this might include non-solicitation of the business’s customers, or operating a competing business within a given range of miles from the business. In publishing, a publisher may ask for a non-compete from an author, so that the publisher is the only one publishing the author’s work on a given topic. Non-compete agreements may exclude competitive activity for a year or may go for much longer.
Many states have laws that permit these sorts of non-compete agreements, provided that they are “reasonable” to protect a “legitimate business interest” or the “goodwill” of the business and balancing that interest with the effect on the public of the restraint of activity. A common strategy of employers seeking non-compete clauses, therefore, is to demand a very extensive non-compete, but then qualify the claim with a statement that the demand is no broader than is required by law. The idea is to get the “full benefit” of the law by asking for whatever the maximum interpretation of “reasonableness” might be. A happy side attribute of such a strategy, from the employer’s perspective, is that absent a law case, the employer can claim pretty much anything is reasonable up to whatever limit an employee is willing to tolerate, short of filing suit. That is, “reasonableness” is not so much a matter of public policy as it is how far an employer can go before some employee chooses to fight over it. Some states won’t tolerate this sort of thing, while others will. California, for one, declines to re-write invalid arrangements to make them valid. Folks have to do their own heavy lifting in California. (Here is a nice summary by Christopher S. Drewry of the variations among the states.) We will come back to this strategy in a moment, so keep it in mind.
In this regard, California is something of an outlier, along with Texas and North Dakota. Since the 1860s California has simply banned non-compete agreements. The current statement of the law is California Business and Professions Code Section 16600:
Except as provided in this chapter, every contract by which anyone is restrained from engaging in a lawful profession, trade, or business of any kind is to that extent void.
The exceptions in the chapter pertain to non-compete clauses included in the sale of a business and the management of partnerships and S corporation memberships. Otherwise, in California, non-compete deals are pretty much void.
While employers might not like California’s approach at all, it is worth asking whether technology development in California has an advantage when it comes to innovation that other parts of the country can’t touch, and that an important part of Silicon Valley and Hollywood’s productivity comes from this simple prohibition on non-compete clauses.
Consider the “gig” culture of making movies. A production company is formed, works for three years or so and produces a movie, and then disbands and its members and others reform to create new companies. So much for “sustainable” and “growth”. The company exists to get a job done, does that, and then poof. The idea of working for a salary for 10 years seems downright stultifying compared to the idea that there is always a new company with a fresh idea looking for talent on a limited engagement basis.
Consider also the movement of technology expertise in Silicon Valley. I had one veteran of the Valley tell me that non-disclosure agreements were ignored in the Valley, though lawyers always demanded them. There were only two bits of information that weren’t shared: budgets and release dates. Technology ideas flow freely. The neat thing about no non-compete clauses is that if a current employer ignores your best ideas and your work product, you simply go next door. Again, plenty of available companies that are more than happy to compete, and there’s no territorial requirement that you have to move out of state to land a competing job.
If a state is fussing about how to create a world-class innovation environment–as the state of Washington has been doing recently–one might start by forbidding non-compete clauses and seeing if gig culture strategies start to emerge, leading to more spin outs, more production-style companies, greater freedom of movement for creative talent, to find environments suited to its imagination and expertise.
In all this, intellectual property plays a role as well. In Germany, limits were placed on what an employer could claim from employees by way of invention. “Service inventions” are those that arise from employment for the company regardless of where the invention might take place, while “free inventions” are everything else. The idea is, an employer ought to have what the employer expects to have from employing people, but should not get what those people otherwise might come up with, just because these other things might be valuable, too. If one wants to stimulate innovation, so the thinking goes, then the last thing one wants is to have employers with a lot of power laying claim to everything they can, using the power of employment and the threat of a big fuss to dissuade employees from doing anything with their inventions other than passively handing them over to company attorneys for disposition. Instead, one wants employee inventors to be able to pursue an open market for their work, outside the constraints of the narrow expectations of work produced for, or through the benefit of access to the information of, the employer. That is, shopped around to whomever will develop them. Gig culture. Opportunism. Choose your breakthrough network rather than conform to the demands of an employer more interested in the value of your invention for its future demand from others than to develop it as part of the employer’s business.
Limiting the scope of an employer’s claims to employee inventions is a form of restriction on non-compete clauses. The employer should not be permitted to restrain employees who create new ideas, where those ideas are not the result of the employer’s commissioning and “reasonable” expectations. A number of states–but by no means a majority–have implemented such laws, Washington (RCW 49.44.140) and California (California Labor Code Section 2870), among them.
Consider Washington’s version (California’s is nearly the same language):
(1) A provision in an employment agreement which provides that an employee shall assign or offer to assign any of the employee’s rights in an invention to the employer does not apply to an invention for which no equipment, supplies, facilities, or trade secret information of the employer was used and which was developed entirely on the employee’s own time, unless (a) the invention relates (i) directly to the business of the employer, or (ii) to the employer’s actual or demonstrably anticipated research or development, or (b) the invention results from any work performed by the employee for the employer. Any provision which purports to apply to such an invention is to that extent against the public policy of this state and is to that extent void and unenforceable.
If we break this down, we have a stack that goes something like this:
The Washington law does not require employers to claim everything in the stack that they can, but rather sets an outer limit to what is reasonably what an employer can bargain for in requiring employee inventions. But it is a tempting taste treat for company attorneys–if the law says that such claims are acceptable, then why not take everything one can? The effort, then, for the employer who wants everything, is to expand the scope of service inventions as much as possible.
Work done for an employer is pretty clear. Identifying the employer’s equipment, supplies, and facilities is also clear. However, trade secrets can be a challenge. Typically there’s no list in a safe labeled “Our Trade Secrets”. One can imagine an employer claiming as a trade secret “anything you think of that might have value to us.” And then a lot of stuff would be a service invention. The law works hard at the next level to limit claims that related to the employer’s business and business directions to those that are “directly related” and “actual or demonstrably anticipated” directions–that is, documentable things that precede the invention and can show that the employer was already headed in that direction. The thing to do, obviously, is for the employer to write down as many things as possible that the employer might ever want to do and put that in a safe labeled “Our Demonstrably Anticipated Business.”
The last box looks pretty secure, too–“one’s own time”. The effort of the complicated statement of scope is to attempt to separate one’s own time from taking the stuff and experiences of work “home” to be done on “one’s own time” out ahead of the employer.
The actual claims an employer might make depend not merely on this outer limits of reasonableness, but also by how the scope of employment is set out in an employment agreement, the assignment and direction of tasks, and any understandings or patterns of dealing that might arise in the course of those tasks, such as an employer offer to loan equipment or rent facilities to employees for their own work “on the side”. Any of these sorts of things can limit the scope of claim that can be made within the limits of “reasonableness.”
Now consider the problem facing a university faculty member. What is the scope of his or her employment? Is it just meeting assigned courses and assigning grades while serving on various administrative committees? Or does it include all research activities, even if not assigned or directed or reviewed by the university? Does it include any ideas one has that might be construed as either i) within the general area of professional work that the faculty member is hired to engage in or ii) might be patentable and therefore worth trying to sell to venture capitalists or to industry? What does one do with the idea of the “business of the university”? Is that the education of students and the provision of resources for the conduct of research, or does the university’s business include all the various areas in which research is conducted, so that an electrical engineering professor who invents a new video game architecture has to deal with the communications and computer science departments’ areas of research–no matter what you do, there’s someone in some department already working on that as part of their professional work. And what about resources provided in sponsored research, where it is a donor or sponsor that pays for the resources? Are those still the “equipment and supplies of the employer” even though the employer is only serving as a trustee of convenience for these assets and does not provide any control over the research work–which is clearly not for the employer? Even “own time” is called into question. I have heard frequently administrators say there is no such thing as faculty being “on their own time”. That would make it difficult for any faculty member to expect to have the benefit of the employee-inventor protections of the law. It is as if the law was not intended to protect faculty from administrative overreach. Or so it would seem.
The University of Washington exploits its advantage in its patent and invention policy.
As a condition of employment, and even if a specific patent agreement is not signed, University employees agree to assign all inventions in which the University has an interest to the University, to an invention management agency designated by the University, or to the sponsor if required under agreements governing the research.
. . .
The University will not require assignment of interests for any invention for which no equipment, supplies, facilities, or trade secret information of the University was used and which was developed entirely on the employee’s own time, unless:
1) The inventions related:
a) Directly to the business of the University, or
b) The University’s actual or demonstrably anticipated research or development, or
2) The invention results from any work performed by the employee for the University.
Essentially the structure of the University’s claim is to claim everything, and then back off on anything that would be illegal to claim. There is no guidance as to what it would be reasonable for the University to claim under its employment arrangements with faculty, nor what other conditions, such as academic freedom and the provision of resources for the purposes of public service might play in what might make for a reasonable claim to ownership (as distinct, say, from seeking a financial interest, to recover actual costs).
The odd thing about this policy statement is that it never sets out just what inventions there are that the University “has an interest” in. It is “anything less what is illegal.” That is likely the result of reading the next subsection of the Washington law:
(2) An employer shall not require a provision made void and unenforceable by subsection (1) of this section as a condition of employment or continuing employment.
So rather than setting out a scope of interest, the University merely claims it is not requiring anything that would be illegal to claim. The uncertainty is set up in favor of the University: anything we say we want is ours, and the employee-inventor has to prove otherwise, on threat of breaching his or her employment agreement. The uncertainty goes further by the substitution of “University” for “employer” in the text derived from RCW 49.44.140. We have discussed at length elsewhere how it is possible for a faculty member to be “employed” by a university for a wide range of activities, but that the university acts as an “employer” for a small subset of these activities. The faculty generally do not work *for* the university, but rather *at* it, when they do their scholarly work, their teaching, and their research. By substituting in “University” for “employer” the policy effaces the difference in meaning and makes it appear that the business of the University is broader than the narrow terms of employment. It is hardly an innocent substitution.
In California, courts will not re-write a non-compete agreement so that it becomes valid, even if the agreement contains a provision for such a rewrite:
The court reasoned that permitting employers to include broadly written non-compete clauses in employment agreements and allowing them to rewrite the clauses if challenged would violate Section 16600. The court foresaw employees feeling obligated to honor the broad clauses without seeking the advice of counsel. To prevent this practice, the court refused to permit the revision of illegal covenants not to compete. Kolani, 64 Cal. App. at 407.
Of course, we are dealing here with Washington, not California, and with a patent policy not a non-compete agreement, but the pattern of attack and the public policy issues would appear to be highly parallel. The employer claims everything and then leaves it to interpretation to sort out just what the employer has legally claimed. One problem that arises in this sort of strategy is that it amounts to an agreement to agree: the employee by policy is told that he or she must agree with whatever the University may later decide is claimable. There is no meaningful scope provided by the “business of the University” when it comes to the service of the faculty. In contracts, an agreement to agree is not enforceable. One might then argue that a policy that claims everything except what is not legal to claim is an agreement to agree.
Interestingly, the University of California uses this same strategy. The UC patent policy makes the same general assertion with regard to agreement to assign.
An agreement to assign inventions and patents to the University, except those resulting from permissible consulting activities without use of University facilities, shall be mandatory for all employees, for persons not employed by the University but who use University research facilities, and for those who receive gift, grant, or contract funds through the University.
UC then follows up with a “Patent Acknowledgement” that is required at employment, along with an oath of allegiance to the state of California:
I acknowledge my obligation to assign inventions and patents that I conceive or develop while employed by University or during the course of my utilization of any University research facilities or any connection with my use of gift, grant, or contract research funds received through the University.
We skip over the details of the substitution of “develop” for “reduce to practice” and the use of “or” instead of “and”: “conceive or develop” rather than “conceive and reduce to practice” and the use of “while employed” rather than “within the course and scope of my employment” (which is what UC is trying to deal with in its ill-conceived amendment to the patent acknowledgement). We skip also the distinction between the narrower “research facility” in the acknowledgement and the use of the broader “facility” in the policy statement proper. We notice rather the addition of the phrase “any connection….” This is a remarkably broad construction, even over the wording of the patent policy. Compare:
who receive gift, grant, or contract funds through the University
any connection with my use of gift, grant, or contract research funds received through the University
The first construction has to do with those who are required to make an agreement to assign–and does not itself identify what those individuals are to assign. The second construction expands on the idea that if one is required to make an agreement to assign and is not an employee, then it must be for whatever reason that one has been singled out. Thus, we get “any connection with” use of certain funds “received through the University.” It is difficult to understand exactly how far one can take “any connection with” or “received through the University”. It would appear the work of an employer hot to claim everything it can, up to the limit of the law.
UC then includes the notice required by California Labor Law:
NOTICE: This acknowledgement does not apply to an invention which qualifies under the provision of Labor Code section 2870 of the State of California which provides that (a) Any provision in an employment agreement which provides that an employee shall assign, or offer to assign, any of his or her rights in an invention to his or her employer shall not apply to an invention that the employee developed entirely on his or her own time without using the employer’s equipment, supplies, facilities, or trade secret information except for those inventions that either: (1) Relate at the time of conception or reduction to practice of the invention to the employer’s business, or actual or demonstrably anticipated research or development of the employer; or (2) Result from any work performed by the employee for the employer. (b) To the extent a provision in an employment agreement purports to require an employee to assign an invention otherwise excluded from being required to be assigned under subdivision (a), the provision is against the public policy of this state and is unenforceable.
There is no mention of “any connection with” use of funds, nor of “conception” only of inventions. (An invention is not patentable until it has been made–that is, conceived and reduced to practice, both; UC aims to claim inventions when they are conceived even if they are not “developed” at UC–picking up on the use of “developed” in the Labor Code, where it might mean any of “made” or “reduced to practice” or “constructed and improved” but does not mean “conceived”.) Either of these claims to inventions is outside the scope of what is legally claimable per the Labor Code. UC relies on the notice of the Labor Code’s language to implicitly do the rewriting of their acknowledgement to make it valid. There’s an argument that this strategy makes the whole agreement invalid, as against public policy. One would think a public university, of all places, ought to be among the most careful with public policy.
Here we have folks in the midst of signing an employment agreement in which the agreement itself overstates the claims the employer makes on assignment of inventions, and then the employer qualifies those claims by saying they really only mean what is legal to require. It is essentially the same strategy as is used, and fails, in California non-compete agreements.
Consider now the thought that UW and UC are using their patent policies as non-compete agreements. In California, there’s a case that has to do with a faculty member and a publisher that moves along these lines–Campbell vs Stanford. In this case, Campbell, a faculty member at University of Minnesota, and Stanford University, for its press, enter into a publishing deal for a psychological test. Part of that deal includes a covenant that Campbell will not prepare or publish “any similar work or anything that may injure the sale” of the test. Stanford’s press subsequently licenses the test to a commercial publisher. Campbell objects and argues the non-compete clause is invalid. The 9th Circuit hears the case on appeal and grapples with the core problem:
The problem in this case, then, is defining Dr. Campbell’s profession and determining whether he is precluded from pursuing that profession. (paragraph 17)
In this case, Stanford would have us say that Campbell’s “profession” is psychology generally, and that the competing works provision at issue does not entirely restrain him from pursuing his occupation as a psychologist. The California Court of Appeal precedents discussed above, however, compel us to define “profession” more narrowly. A person’s “profession” under section 16600 is not so expansive to include all work for which he is qualified. One may, by devoting all of his energy to a specialty within a traditional profession, limit his “profession, trade, or business” under section 16600 to that specialty. Thus, even though Campbell may hold a license to practice psychology generally, which includes work outside the field of “preparation of vocational interest exams,” Dr. Campbell alleged that these tests have become his life’s work, that he has attained a position of worldwide prominence in the preparation of the tests, and that the competing works provision would completely preclude him from preparing those tests.
The court orders a new trial in which Dr. Campbell can establish what his “profession” is for the purposes of non-compete. The issue for us, however, is whether a patent policy can come to have the same properties as a non-compete clause, when university faculty are concerned. There’s a general rule in contracting that one looks to how an instrument functions rather than reasoning from any title that it happens to have. A document labeled “Exclusive License” may well be an assignment if substantially all of the patent or copyright owner’s rights are transferred by the instrument. Here, we have a document labeled “Patent Policy” but as to function it behaves remarkably like a non-compete clause.
For faculty, what they invent or discover may well be their life’s work, their specialty, the very expertise for which they would be employed at any other university or company. A patent on their work held under a compulsory requirement of assignment by the university as employer makes it impossible for that faculty member to practice his or her expertise for any other organization without the permission of the university. Furthermore, the faculty inventor cannot even teach the invention to others–by publication, instruction, or consulting–without those others obtaining permission from the university-employer. Anything else is infringement. But it is also a “restraint” in “engaging in a lawful profession.” Faculty inventors caught in such a patent policy tangle cannot practice the very elements of their work that reflect the expertise for which they are known as scholars.
Under California law, it would appear that the UC patent policy, as implemented, is simply void because it functions as a non-compete clause. The faculty member cannot leave the university and practice his or her speciality in competition with the money-making aspirations of the university’s patent licensing office. Under Washington law, there is still the question of “reasonableness.” In a case before the Washington state Supreme Court involving a non-compete clause, the court found that:
The agreement at issue here is unreasonable because it bars Labriola from working in his field of expertise even where he takes no unfair advantage of his former employer.
The question might therefore be put: is a faculty member in practicing his or her invention for another organization taking unfair advantage of the university-as-employer, given that the university did not assign, direct, review, approve, or expect the invention, but simply claims it because it seeks to profit from licensing it? One might of course argue that such a thing is entirely proper and happens all the time in the corporate world. But the university world is not the corporate world, and a public university even more so. It is worrisome to see folks reasoning by analogy to corporate doings to arrive at what makes sense for a public university’s treatment of faculty. In the university world, faculty are employed to do a lot of things, but are employees for very little of what they do. A patent policy that seeks to prevent them from doing the key, important things that they accomplish as scholars–inventions, discoveries, works of authorship–merely because they appear to involve technology and could be licensed for profit–would appear to be a forthright and unreasonable effort to prevent faculty from competing with their employer outside of the narrow confines of the portion of their employment in which the employer directs their work and expects the benefit of their work product.
University patent policies that fail to establish a clear statement of the scope of university interest, and restrain faculty from practicing what they discover, invent, and author as scholars, are against public policy, work against a “gig” economy, and suppress rather than advance an innovation-favorable environment. Patent policies should be considered in light of the reasonableness standard of non-compete laws as well for particulars of employee-inventor protection laws. States that intend to foster innovation would do well to follow California’s lead in non-compete clauses–at least for their public universities–as well as demand robust protections for faculty and student inventors against administrators’ desires to claim everything and give up only what they are sued over. Policies and agreements written in this way should be void as against public policy.