(Note: this post continues from the previous one. I don’t take up the particulars of the Stanford v. Roche case, and I wish the disputants, the judges, the attorneys and inventors all the best. Here, I aim to show how a reading of Bayh-Dole as I’ve developed it makes things understandable, even tractable. There’s no huge problem, but actually some relatively straightforward, even if not fun, consequences. Certainly, there is no reason to go rebuilding the patent laws to single out university inventors as lacking the capacity to hold personal rights in inventions like most everyone else, including federal employees working on the same projects. Bayh-Dole was not built on the premise that university faculty, unlike everyone else in the world, cannot read a contract, understand the significance of what they do, and are babes in the woods to be duped by malicious companies. That’s the implication of the amicus brief filed by WARF, AUTM, and others. Honestly, they should withdraw it with apologies all around).
We established last time that a non-federally funded collaborator in a joint invention under Bayh-Dole may do what it will with its interest. It has no obligations to share licensing income with inventors, to license to small businesses, or to use the remaining balance for scientific research or education. It has no obligation to report on utilization. It has, by holding title, satisfied a great deal of Bayh-Dole already, by getting rights away from government and university alike and out into the field.
What does this mean for a situation like that in Stanford v. Roche? I am not concerned here with the details of the case or the particular merits of the various arguments one way or another between those parties. I empathize, folks, but not here. Rather, let’s look at how the Bayh-Dole component ought to play out in similar proceedings. The amicus brief submitted by WARF, AUTM, and others constructs a highly implausible, overtly self-interested restatement of Bayh-Dole, arguing for outright ownership of inventions (and subsequent patents) by universities, preempting any other obligations or interests that inventors of subject inventions may have. Some good attorneys with good heads on their shoulders have agreed with the brief. I have shown previously some sampling of how badly argued the brief is. It’s just assertions pinned on reasoning from one meaning among others given to “retain” in one place and not others. Or, as we might say, it’s not horses, just two coconuts being beat together.
Let’s look at a simplified situation for joint invention. A research team at a university invents under a federal grant. One of the members of the team is a company employee collaborating with the team. The company employee’s participation comes within the scope of the federally funded project (that is, it is not in the way of closely related project to apply discoveries made in the federally funded project). That team member has an obligation to her employer in the form of a present assignment of future inventions made within the scope of interest of her employer, and this invention meets that condition. The other members of the team have made written agreements to protect the government’s interest in subject inventions, and this surely is one. How does title to the invention shake out?
Title is joint. On one side, the company employee enjoyed, briefly, a personal, undivided interest in the invention, but her title to the invention passed upon the making of the invention to her employer by means of the present assignment. On the other side, the university operates under Bayh-Dole. It has the opportunity under Bayh-Dole to elect to retain title or to assign that option to another qualified organization. If it exercises the option to retain title, then it has the further option of determining how it will comply with the law. One way would be to require assignment of patent rights from its employee inventors. Another way would be to direct the employee inventors to assign their rights to an invention management organization. (Again, the law permits the university to choose this organization, or nominate the inventors to choose this organization—or, for emphasis, the law does not dictate who in the university is authorized to make these decisions). Upon election of title, the university with its share of the invention (also undivided), must embark on an effort to use the patent system to promote use of the invention, or cause that effort to be made. It does this by causing patent applications to be filed, granting the government its license, and the like.
An easy out at this point for the university would be to rely on the company collaborator, which also holds title in the invention, and may be in the best position of all to accomplish Bayh-Dole objectives. For this, the university could agree simply to a standstill in its own licensing efforts, providing the company collaborator with what amounts to an exclusive commercial position. In exchange, the university would ask for reports of utilization so it could comply with its Bayh-Dole reporting obligations to the funding agency. It might take some great restraint for the university not to ask to be paid for being such a good partner, but nothing in Bayh-Dole requires it.
The university could also license its interest to the collaborating company, and build such compliance matters into the relationship, but the company may see this approach as much less desirable, as it would create other obligations that would go with a license from the university—such as reporting, US manufacture (in some cases), and government march-in rights on at least the university side of the deal. However, this is not the only way things can go, and the university could offer rights to industry broadly at no charge, defeating the collaborating company’s potentially significant commercial interest in the invention. Alternatively, just to give a shape to the variety of possibilities, the university could offer an exclusive license (again, only in its undivided interest) to a competitor of the collaborating company, or both the university and collaborating company could license their rights to a joint venture or start up or other company willing to take on the work. Again, in a joint venture or common licensee, the collaborating company would license on whatever terms it chose, but the university would be constrained by Bayh-Dole, and wherever the terms between the two varied, the licensee would have accept the Bayh-Dole terms to gain the benefit of the university’s interest in the invention and/or patent(s).
To my mind, this all appears straightforward, as far as anything in this business is. Invention rights are personal and disposed of by each inventor according to the agreements he or she has made with employers, sponsors, or others.
Now let’s change the scenario so that all the team members who invent are employees of the university, but one of them has been allowed to enter into a consulting arrangement with a collaborating company. That relationship is current, and it provides the same present assignment to future inventions that the company employee had agreed to. Now what happens under Bayh-Dole? It turns out Bayh-Dole doesn’t care, and doesn’t need to care. The place to look of course is the potential for collision between the written agreement (which the consulting member has accepted in compliance with Bayh-Dole) and the consulting contract (which is permitted by the university but the university has not inspected or approved as to invention obligations). Certainly there may be a “battle of forms” here, but it need not be not a fight to the death and it is not one Bayh-Dole cares about. If the Bayh-Dole written agreement supersedes the claims of the consulting agreement, then the consulting employee assigns (as requested) her rights to the university (or its designee, or the government, or not, depending on agency actions). If the consulting agreement trumps the written agreement, then the consulting employee’s rights are assigned to the company, and we are back to the situation already discussed, which is straightforward. Bayh-Dole does not care how this shakes out, because it is concerned with the role universities are to have in the promoting use of subject inventions. It is up to the university to construct written agreements that establish the interest the university may wish to have in an invention, but the university has no obligation, and indeed no mandate, under Bayh-Dole to have an exclusive ownership position in each subject invention. The university may waive, may assign, may subcontract, and may jointly invent to transfer, assign, convey, or reduce its interest in the inventions made with federal funds. The law leaves it to the university. The law is not a dictator of that interest, nor a nanny to assist the university in its desires.
This variation, too, appears to be resolved relatively directly. The answer jumps out, especially, if one drops the idea that Bayh-Dole is a special entitlement for universities to make money licensing exclusive patent rights to industry which otherwise seeks to frustrate this effort by enticing university research personnel into clever backroom deals. From the perspective of the stated objectives of Bayh-Dole, those arrangements by which university research personnel are of sufficient interest to industry to invite company support through consulting might be seen as one class of success stories for the law—only to have them portrayed by university patent licensing outfits as bad behavior! Bad, perhaps, for the licensing outfits, but not necessarily bad at all under Bayh-Dole, and in fact, by and large, very good.
We may add now additional complications at will. What if the university has an IP policy that requires assignment to the university, and the consulting agreement is considered, by the university, to be in breach of this policy? That is another conflict, but it does not have to do with Bayh-Dole. It is a matter of conflicting obligation of the same right. It is a matter of contract law, with regard to enforceable promises, and it is a matter of administrative discipline or tort law with regard to potential damage caused by the employee making the conflicting promises and by the company that may have induced or participated in creating the conflict. The dispute will involve chronology. It will follow the wording of the promises, such as promise to assign vs. present assignment, but also the scope of interest of the competing contracts. The dispute will involve what the university knew of the consulting relationship and what action it took or did not take to protect its interest, which now surfaces as so very important. The dispute will look at the behavior of the employee and the company. Did the employee report the relationship? Was there a working conflict of interest review? Did the university take any action when the consulting employee was identified to work with federal funds? Did the company know of the prior obligation and seek to circumvent it? Did the company know of the federal research and intend to obtain an interest in it through this means, despite other communication with the university?
Ah, it’s a lot to consider. But for Bayh-Dole, it doesn’t matter. It’s the kind of dispute that comes up between folks when they fail to manage their affairs. Bayh-Dole is not a law that demands order of the world–just of agency behaviors pertaining to invention rights in federal funding agreements with universities (and others). It is a law that sets out how a university is to behave if it is to take title. It does not set out that the university has to behave in only one way, and indeed, there are remarkably few penalties in Bayh-Dole for the university if it messes up—the government can take ownership of the contractor’s interest if the university waives or defaults on its obligations regarding invention title and patent work, and can march-in to force licensing if the university or its licensees fail to timely develop the invention to the point of practical application in compliance with the objectives of the law. In part, this may be because if there are no patent rights, or the patent rights fall non-exclusively into the hands of industry, or those rights remain with inventors personally, it’s all okay in terms of government support for university research. It’s not about how universities make licensing money. It’s about what happens when there are patents on federally funded inventions. The goal is use of those inventions, however it comes about. Really, the university doesn’t have to be there for every birth, expecting a fee.
Bayh-Dole also does not set the table for the university. The university must be engaged and do its piece. It must establish the models it will use to meet its obligations should it choose to play. Bayh-Dole is no convenience law. The university must have written agreements with research personnel; it must designate who is responsible for receiving invention disclosures and reporting them to the funding agencies; it must designate who will decide whether or not to elect title, and how assignment or subcontracting or collaboration will take place, and how any of these actions may distribute personal invention rights as well as the university’s own rights under Bayh-Dole in title to inventions.
A final variation. Consider the case in which there is a state law requiring that inventions made with the use of state resources belong to the state. The research is conducted at a public university in the state. The university’s IP policy recites the state law and employees agree to the IP policy in writing as a condition of employment. Now comes the federally funded project and an invention. As before, the potential for dispute. The consulting employee may have entered into a consulting agreement in violation of state law. It may be that the agreement, to be valid and enforceable under state law, can scope only to subject matter permitted by state law. In this case, the collision will first be between state invention law, which requires state ownership of all inventions made with state resources, and state labor law (if it comes into play), which sets limits on the claims an employer may make on an employee’s inventions. From there, one can look at whether the university as a state agency can grant a waiver to the state’s invention law, and if so whether its actions in the case of the consulting agreement might constitute that waiver. If not then the university may have a basis to obtain a judgment that the consulting employee’s interest in title to the invention must be assigned to the university as agent of the state.
But as you may have figured out, Bayh-Dole doesn’t care. It doesn’t provide either party with a leg up on the deal. The argument the university may have is: we expected to hold interest in inventions under Bayh-Dole, and you (consultant, company) have defeated this expectation. The university can say: we relied on state invention law and our IP policy to obligate assignment. The university can even say; we relied on the written agreements in Bayh-Dole, or even just on the text of Bayh-Dole itself to get what we expected later. And the answer is: to get what you want, you have to be engaged in the practice with a level of diligence commensurate with the importance of the work, the sophistication of your research teams and their collaborations, and the particular requirements and leverage points of your local invention regulatory environment, however it may be established. It’s a complex game. The law doesn’t let you opt out of playing it with some skill.
This is a good thing for the law. Innovation is complex, and as social as it is business or legal or technical. Federal preemption of university inventors’ personal rights would not be good for public science, university research, or national innovation policy. And it would crush the thoughtful apparatus deployed by Bayh-Dole to address these areas of public policy. Bayh-Dole is an important law not just because it provides universities the opportunity to make money exploiting patent positions, but much more so because it allows universities to reveal the particular character of their administrators’ understanding of innovation in the public interest by offering choices, many of which, in nearly 30 years, most university administrators have failed to make—or even cared to consider. That’s the pity of Bayh-Dole, that it likely will get changed because university administrators have not been sufficiently informed or engaged to explore the potential it has offered to encompass and stimulate a wide range of science, research, and innovation models. Instead, we have got state-subsidized deal flow for needy venture capitalists and a few hit licenses for some biotech tools and drugs. And lots of the rest of the good inventive stuff, that would build relationships, fuel small companies, and support US manufacturing, doesn’t get the attention it should, and could, have–if only universities had developed diverse models to engage diverse circumstances. What worked for biotech in the speculative window from 1980 to 1995, didn’t mean much of anything for the rise of the internet during the same period, is not working for nanotechnology now, and likely won’t work for green tech or synthetic biology or anything else that might arise by way of new industry. Yet universities are staffed up for biotech like there’s no tomorrow, which of course, very likely, in terms of licensing on the biotech model, there isn’t, or at least not enough of it to make a full day of it.
The greatest threat to Bayh-Dole these days is not from special interests in industry, but rather from special interests in university administration, which seek short cuts to support disengaged accumulation of patents to feed a corporate-style licensing program with a fixation on productivity served up by corporate style metrics. It’s like churches competing on the size of their offering income, or orchestras reporting their value to the public as a function of the differential between the size of their grants and the successful reduction in payroll. Such a licensing program is permitted under Bayh-Dole. But that’s no way to argue that such an approach is an unqualified public good; serves science, research, or innovation objectives of the government; or is even close to what may be desired by way of local initiative under Bayh-Dole.
There’s yet a more particular discussion of why the pathway of ownership under Bayh-Dole matters. I’ll work on that if anyone is interested.