The start of this article is here.
We have been working through what NIST’s introduction of an assignment requirement for subject inventions actually does. In one view, nothing. In another, a technical if not prissy requirement that inventions a contractor already owns must be assigned to the contractor. From that, we have worked through the idea of equitable ownership–what’s fair from the circumstances–as a possible explanation for NIST’s new rule requiring contractors to require inventors to make a written agreement to assign inventions to the contractors that the contractors already own–to protect the federal government’s interest.
D. C. Toedt III, an attorney, has a helpful diagram and discussion of when an employer owns an invention. If an inventor uses an employer’s resources, the employer generally can expect to have a “shop right” in any invention–much like a non-transferable, non-exclusive license to the invention. Otherwise, look to see whether the employee was “hired to invent,” or “set to experimenting,” or there were “special circumstances” (such as, the employee is a senior officer of the company with an equitable duty to assign inventions to the company). In any of these situations, a company may have equitable title in an invention made by an employee. The result can be characterized various ways, including (as Toedt points out in his helpful notes) courts can construe an “implied in fact” contract between the employer and employee with regard to assignment of any inventions within the scope of the claim of equitable title.
To underscore this line of development. An employer (in general, any employer) has no claim on an employee’s inventions merely because the employee is employed. No equitable title claim there. Similarly, there’s no equitable title claim based on the use of an employer’s “resources.” While an employer and employee can enter into an agreement that the use of an employer’s resources (made available so the employee can do whatever the employee has been hired to do, short of inventing or assigned to conduct experiments) in fact is the consideration for assignment of inventions made with those resources, the employer and employee have to actually enter into that agreement. The agreement is not implied by what is “equitable.” What is equitable, in situations of employment or use of an employer’s resources is that the employer have a “shop right” to use the invention in the employer’s business. Any other outcome depends on a voluntary agreement with its own offer, acceptance, and consideration.
Put it another way, the value of the “resources” provided may well be tiny compared to the value of either an invention or a patent on that invention. Only the stupid, desperate, and indifferent among inventors would give up ownership like that for next to nothing. They would publish openly before they would accept a deal that would doom their work to being excluded from broad availability. If an inventor gives up ownership of an invention, and the employer obtains a patent, then even the inventor cannot practice the invention independently without the employer’s approval. Graduate students cannot graduate and practice what they learned in the lab–even their own inventions–without a bureaucrat’s permission and that isn’t likely if the bureaucrat is unwilling to grant to those graduate students royalty-free non-exclusive licenses.
For inventions that are already closely tied to an employer’s proprietary products and related IP, perhaps that is not a big deal. But if the invention is something that stands alone from the employer’s assets, it can be a big deal that the employer takes ownership anyway.
For universities and their faculty involved in creative activities of their own choosing, universities rarely have a claim to equitable title in inventions. This is even more the case for federally supported inventions, where the federal government reimburses the university for its direct and indirect costs and faculty must request that the university “release” them from their “university duties” in order to participate in the federally sponsored work.
Now we have to avoid the fuzzy thinking of university administrators, who often carry fuzzy thoughts with them like Pigpen with his dust cloud in the Peanuts cartoon strip. If a university patent policy asserts that faculty have an obligation to assign all inventions they make during their employment or using university resources or related to an area in which they have professional expertise (these are the usual assertions), then there has to be some basis for the obligation. It cannot be that it is equitable that the university have ownership of a faculty member’s work. That’s the one thing that is *not* equitable!
Consider, for instance, university policy statements assuring faculty they have freedom in research and freedom of publication–neither assurance is true if a university can demand to own a faculty member’s inventions, in which case the faculty member cannot choose where and how to do research in the area of invention because the faculty member must obtain permission from the university to do so–not free; and the faculty member is compelled by the university to publish in the patent literature–again, not free.
Were it not for administrative confustion about Bayh-Dole (confustion meaning refusal to comply with Bayh-Dole and chronic misrepresentation of the law and use of that misrepresentation to ignore formal university policies), it would be crystal clear that universities have no claim on faculty inventive work unless the faculty member has voluntarily agreed to give up academic freedom (freedom of research and freedom to publish) and tenure in order to receive some offered benefit other than employment, use of resources made available with employment, and continued employment.
Bayh-Dole’s patent rights clause, then, has been drafted to limit the federal government’s interest in inventions that a contractor acquires to only those that the federal government itself, were it to contract directly with a contractor’s employees, would have a basis for claiming a right to at least a non-exclusive, royalty-free license. The federal government would have no claim to the inventive work of non-employees or employees who were not hired for their technical expertise. But Bayh-Dole goes even further: the federal government does not assert an interest in any such inventions–the ones made in projects receiving federal support–unless and until those technical employee inventors assign their inventions to a party to the funding agreement. The contractor-employer has no obligation or mandate or even encouragement under Bayh-Dole to obtain any such assignment and Bayh-Dole gives employers no special privilege to obtain any such assignment.
Think past university administrative confustion and fuzziness. If the federal government under Bayh-Dole has no interest in inventions made in projects receiving federal support until those inventions are owned by a contractor–and the federal government is providing financial support for the work (and not the university in some express and significant way), then the university also has no interest in those inventions. And that’s what Bayh-Dole’s standard patent rights clause packs down–entirely consistent with the Supreme Court’s ruling in Stanford v Roche.
All the convolutions here come from the rather foolish premise that Bayh-Dole has to establish an arbitrary (“uniform”) policy on invention ownership to displace all other statutes and executive branch policies. Rather than treating different contracting situations differently–and clearly, and openly, and expressly–Bayh-Dole requires an arbitrary default patent rights clause that conflates contracts and grants, procurement and subvention, work directed by employers with work in which employers have no right. Then within that clause we see a pile of conditionals, waivers, and inconsistent federal agency responses–to federal licenses, to reporting on invention utilization, to march-in.
Federal agencies may vary from the default standard patent rights clause, but there is a huge overhead in doing so–which in itself mocks the Bayh-Dole policy and objective to “minimize the costs of administering policies in this area.” Bayh-Dole does no such thing–Bayh-Dole changes where the costs of administrating the policies in this area will be huge–in making changes in the default patent rights clause for exceptional circumstances, in challenging whether an invention is a subject invention, in marching in to protect the public from nonuse or unreasonable use. Those are now the areas of administration with such high costs that federal agencies don’t bother to use them. Inspired? Hardly.
In software programming, there is a concept called the “antipattern.” Antipatterns are the design approaches to software development that historically have not worked. Here’s a brief synopsis:
While it is reasonable to assume that the principal reason we write software is to provide solutions to specific problems, it is also arguable that these solutions frequently leave us worse off before we started.
Think of Bayh-Dole as a software development project. Fine intentions, but what about the design architecture?
A survey of hundreds of corporate software development projects indicated that five out of six software projects are considered unsuccessful. About a third of software projects are canceled. The remaining projects delivered software that was typically twice the expected budget and took twice as long to developed as originally planned
Now think about how university administrators (and federal officials, and NIST to boot) represent the architecture (the “highlights”) of Bayh-Dole–federal funding! inventions! patents! licenses! commercialization! public benefit! This sequence happens perhaps one in 1,000 inventions. Add in last “on reasonable terms!” and the odds of success grow even longer. NIST’s contribution is to insert an overt assignment requirement between “inventions!” and “patents!” for inventions for which a contractor already owns equitable title. NIST does not disrupt the overall design concept–even though that design concept is nowhere to be found in Bayh-Dole.
If NIST performed properly its public role under Bayh-Dole, its rule changes would be directed at making clear that the prevailing design claims for Bayh-Dole are not supported by the law or the standard patent rights clause. NIST would make rule changes that lowered the cost for federal agencies to challenge failures to report subject inventions and to march-in to protect the public from nonuse and unreasonable use. NIST would make it administratively low cost to introduce new patent rights clauses to address circumstances outside the default patent rights clause. But NIST has done nothing like this.
Instead, NIST has packed even more conditionals into a single patent rights clause. The software analogy is one of creating a program that grows into a monster of non-operating code, buggy code, and more buggy code inserted to deal with the other buggy code. No one casually reading the law or the implementing regulations can make sense of what they see. So university administrators and Bayh-Dole gurus simply make up what they want and attribute it to Bayh-Dole. Even NIH and NASA and NIST officials can have fun doing this. Instead of public policy on the government’s interest in inventions made with federal support established by a uniform law, we get just the opposite–a law that doesn’t operate, backed by a patent rights clause that keeps getting worse and worse, and that backed by university officials and pundits making something else up entirely.
University officials refuse to comply with Bayh-Dole’s patent rights clause in nearly all of its substantive requirements. Universities don’t accurately report subject inventions; don’t use the patent system to promote the use of subject inventions; don’t comply with the written agreement requirement (so it’s funny that NIST amends that agreement with an assignment requirement); ignore the restrictions on nonprofit assignment of subject inventions; ignore the restrictions on nonprofit use of income earned with respect to subject inventions; ignore the small business preference–pretty much anything that might have to do with shaping the public covenant that should run with inventions made in projects receiving federal support, pretty much anything that would have to do with encouraging a diversity of approaches to research, to innovation, to public access, to industry use, to commercial ventures, to venture-backed development, to economic development.
Instead, we get a Moloch-class invention bureaucracy that would rather eat its own inventions than to allow others to decide how inventions arising in publicly supported inventions might be best used. The Moloch idea is that a university bureaucrat (or a federal one) is a way better judge of what might beneficially happen next than is the (arguably) world-class research team that came up with the invention. If more bureaucratic invention management is the inspiration of Bayh-Dole–just now university bureaucrats rather than federal ones, and fragmentation of new research across scores of private players all attempting exclusive licenses rather than a few federal agencies all generally releasing contracted research inventions for general public access (including access by the inventors) with only a few, special exceptions–then we might see why only Moloch-class bureaucrats worship their image of the law and ask others to bow down as well.
The NIST assignment requirement is just another tickle adjustment to make the bureaucratic restatement of Bayh-Dole appear to be all that closer to a vesting statute, now one with an assignment provision that can be misrepresented along with the rest of the law to make it appear that Bayh-Dole does indeed give contractors a special privilege to acquire inventions made with federal support. In fact, Bayh-Dole does just the opposite. But in practice, it would appear that NIST ensures that Bayh-Dole will continue to be ignored in the name of complying with, um, Bayh-Dole.
Someday, perhaps we will ask ourselves, with regard to the fake Bayh-Dole that now dominates public discourse on this arcane area, “Where does that highway lead to?” and we may say to ourselves, “My God, what have we done?” Why must Bayh-Dole end up being “same as it ever was”? NIST’s rule changes make implementation of Bayh-Dole as it expresses public policy that much further from ever happening. That’s what NIST has done, actually.