What are some possible outcomes of Stanford v. Roche? We will start out simple. Then we will go into the darkness with a flickering candle.
1. AUTM is right and Bayh-Dole is a vesting statute. Urk, cruel world.
2. The SG is right, and Bayh-Dole voids scope of all outstanding contracts pertaining to transfer of title upon participation in subsequent federally funded research. End of private contracting for future inventions.
3. Stanford failed to require the proper written agreement under (f)(2) and therefore lacks standing to require assignment of title, but lucks out because government has all the rights and outcomes that it bargained for.
4. Everything is fine. There is no issue. Holodniy was free to assign future rights, and participates as a representative of Cetus with regard to his contributions to the federal research. Stanford had no obligation to require a written agreement from him. The government has all the rights it requires, there is product available to the public. Worries that haven’t happened are different worries.
5. Cetus knows that Holodniy is back at Stanford working in the same area but takes no action to protect its potential claims so its generic, broad present assignment re-scopes narrowly through laches or some similar argument, just as Stanford’s earlier promise to assign re-scoped when Holodniy committed future rights to Cetus. Holodniy’s assignment to Stanford is valid, but Bayh-Dole has nothing to do with it.
I don’t know which of these might win out. See here for yet another possibility. At this point, it’s a paid dispute, lots of politics in the wind, a quarter billion or so in money riding on the outcome, as well as 30 years of private contracting for invention rights. Will those be voided by a OMG! interpretation of Bayh-Dole?
My thought is that invention rights are what they are in the Patent Act and Bayh-Dole does not void them or rescope them. There are many reasons to believe that Bayh-Dole was not re-writing the Patent Act without notice. Assignment practice did not change. There is no note of it in the MPEP, for instance. And universities did not change their practices. Even Stanford asked its inventors in the patents in dispute for assignments. Bayh-Dole is implemented through written agreements, so the proper analysis is to follow those agreements, not reason from abstractions. For this, timing and rules of precedence matter.
Until Holodniy invents in the federally funded work at Stanford, it’s all pretty clear. Stanford’s policy/promise to assign claim on Holodniy inventions is entirely conditional, and his later commitment of certain future invention rights to Cetus is entirely proper, and Cetus provided good and valuable consideration for the rights they obtained.
When Holodniy invents, then, there are five competing obligations to assign title.
1) the Stanford conditional promise to assign, which has been displaced by
2) the later conditional (on scope) present assignment of future inventions to Cetus, which is potentially in conflict with the yet later
3) written agreement to protect the government’s interest required at (f)(2) which may or may not be the same as
4) a re-imposition of Stanford’s promise to assign, invoked to comply with Bayh-Dole’s requirements or invoked or reaffirmed by Stanford under its own IP policy without reference to research compliance, and
5) Holodniy’s subsequent assignment of title to Stanford after Stanford has filed a patent application and then notified the government of its election to retain title to the subject invention.
From this, we can get rid of 1). The conditions are not met for a Stanford claim when Holodniy signs 2). Stanford does not contest that 2) is valid, and apparently knows of 2). 2 is valid at least until Holodniy begins participation in the federal research.
The CAFC says in essence that 2) remains valid against all later claims.
Why? Because the invention was made in scope of 2), Stanford’s later claims came with knowledge of Cetus’s prior claims and therefore fail. That Stanford has obligations to the government which Stanford may not have properly met is not material to the standing of the assignments in question. It would appear that Stanford has put itself into the position that if it lacks title from Holodniy, then it is in breach of the standard patent rights clause. The elegance of the AUTM vesting argument is that Stanford could never be in breach of Bayh-Dole in this way, because the federal government imposes its claim directly on all inventors at the moment they a) come within scope of federal funding and b) are research employees.
The argument boils down to whether Stanford is in breach of its obligations, or whether to preserve Stanford’s position, the court should void a private contract otherwise in good standing. Restated: is it possible for a university to breach the standard patent rights clause with regard to title under (f)(2)? Why have (f)(2) at all if the university has title on notice?
The arguments for Bayh-Dole voiding or rescoping 2) are elegant in that they preserve Standford’s standing under its Bayh-Dole obligations, and hand title to Stanford without regard for any existing, otherwise valid private commitments. No one is wrong, but the federal law steps in and undoes any private dealings, hands title to the university, and imposes its protocols.
But these arguments if they prevail also would void a whole lot of private commitments, including, I point out, all university transactions in which a university policy requires assignment of invention rights to a related foundation. If Bayh-Dole vests title with the university contractor, then it is the contractor, not the inventors, who must assign that title. Bayh-Dole might also pre-empt all written university IP policies that do not comply with its requirements, including the requirement for the substance of the written agreement at (f)(2). Lots and lots of things blow around in the wind if these arguments prevail.
So we look to the status of the written agreement required under the standard patent rights clause, 3). Does such a written agreement exist? I doubt there is a paper expressly referencing the funding agreement in which Holodniy evidences his agreement to 1) report subject inventions and execute papers to 2) permit patent applications to be filed on subject inventions and 3) to establish the government’s interest in subject inventions. I don’t know if any universities actually produce such paper for each grant.
Perhaps one reason the Supreme Court is taking this case is for the chance to lecture universities that if they want the benefit of a law, they really need to comply with it, and ubiquitous non-compliance is not a good argument for doing things differently. That might make a good message for banks, too.
If there’s no 3) agreement, then we are done with the title issue. Holodniy’s title passes to Cetus and Roche has co-ownership. The question for the courts is whether they can construct 3) and if so, does it displace obligations in 2)?
I expect that Stanford intends for its IP policy to implement 3) without any additional paperwork. That is, the agency imposes the standard patent rights clause at 37 CFR 401.14(a) via Circular A-110 _.36 (now 2 CFR 215.36) on Stanford as a condition of the grant, and Stanford then has the obligation for that award to require the (f)(2) written agreement of its research employees, arguably including Holodniy, and does this by the application of its conditional policy of 1), which it has previously waived for this particular scope of interest in Holodniy’s future work, which waiver Cetus and Holodniy have relied upon to do their deal that provides Holodniy with nine months access to Cetus’s facilities, researchers, technology, and data.
Now, however, to comply with the federal award, Stanford re-invokes at the time of the federal award its policy 1) as the later 4) and wants this re-invoked policy 4) to stand for the required written agreement 3).
There is a written policy statement, and Holodniy is required by the policy statement to agree to it, so it’s arguably a “written agreement”, though it is not signed nor specific to the grant nor ever evaluated by Stanford for whether the university has the power to invoke the policy unchanged for a scope of interest previously waived and relied upon.
This is the crux of the problem. Does Stanford have the power to change the scope of its policy, voiding 2) or at least excluding the federally funded work from 2) and without any notice to either Holodniy or Cetus? Put another way, does Stanford have privity in 2)? And if so, on what basis could Stanford terminate or change 2)? Or put yet another way, can Stanford permit Holodniy to participate in the research, and if it does, is the consequence to Cetus (loss of rights) or Stanford (failure to comply)?
It is not that Bayh-Dole pre-empts 2). Bayh-Dole requires Stanford to take on a contractual commitment. That is a federal commitment, yes. But that Stanford takes on a commitment to the government has nothing to do with how Stanford then complies with and implements that commitment. Stanford must *act* to comply. It must make a requirement of its research employees specific to the grant to protect the government’s interest. A general prior requirement won’t cut it, because that requirement has already been carved out for work that may be relevant to the grant. Whatever rights Stanford gets come about as a result of this agreement 3) that it must have, and not any other.
If Stanford cannot apply its policy 4) as a proxy to displace 2), then that 4) cannot be 3). If Holodniy cannot agree to 4) in light of 2) then he has not agreed to any 3) and i) he cannot participate OR ii) he is granted a waiver by the government OR iii) he is participating as a Cetus representative for the purpose of his invention rights.
The CACF decided, in essence, that Stanford having waived its conditional future interest via 1), allowing 2) to come into existence, cannot rely on 1) to establish 3), and cannot impose 4) to establish 3), and therefore lacks a contractual mechanism when the patent application is filed to have undivided ownership, and at that moment, at least, beyond any doubt, Cetus has a joint ownership position in the invention as a consequence of 2). 2) exists, is valid, is known to Stanford, is proper under Stanford’s policy, properly displaces 1) and Holodniy and Cetus have relied on Stanford’s position and actions with regard to 2). What else can the court do? The public is not harmed–there is product on the market. The government is not harmed–it has the notice and license it bargained for. Stanford through operation of its policies could not by contract law impose 1) or 4) in place of 3) because of 2), and therefore 2) prevails.
For 2) not to prevail, it would have to be terminated or restructured by the parties to it, Holodniy and Cetus. AUTM argues, Bayh-Dole does this automatically by voiding 2). SG argues Bayh-Dole does this by automatically re-scoping 2). I don’t see how Bayh-Dole does anything of the sort. If any agent has to act, it is Stanford. It is Stanford that accepts the federal agreement, not Cetus, not Holodniy. It is Stanford that is under that agreement required to require written agreements of its research employees to protect the government’s interest in this particular grant. It is the status of how Stanford can cause 3) to be formed that is at issue. Rather than a Bayh-Doleus ex machina to save the day, the analysis must stay rooted in the circumstances that present, especially from the moment Holodniy joined the research to the moment that Stanford filed a patent application.
If one looks at this time period, 2) is the only agreement that Holodniy has that’s valid. 3) doesn’t exist because Stanford waived 1) and cannot impose 4) on Holodniy because 4) is just 1) with scoped locked to permit 2). Stanford does not have to accept the award, has the power to prevent Holodniy from participating, and has the opportunity to ask Holodniy to renegotiate 2) with Cetus or seek a waiver from the agency for Holodniy’s participation with 2) still in place. Yet Stanford does accept the award, relies on 4) to establish 3) regardless of 2), and requires Holodniy to execute 5) an assignment to Stanford as if 4) has done its job of establishing 3) and displacing 2).
The court looks at this and says, but 4) is void, so 3) is void, so 2) is not displaced, and Holodniy’s title goes to Cetus, and 5) is void because 2) happened no later than when the patent application was filed, so there was nothing left for Holodniy to assign. Makes some sense, got to give it that. There’s just not a lot of way around it, if one is following the contracting, and recognizes that Bayh-Dole is a law of contracting, not a law of taking personal rights.
Whew. This isn’t easy stuff. One can see how folks could have an argument about it, especially with a quarter billion in play.
What about the election to retain title? That mattered for the other two inventors. That does get Stanford a share of the title, but how? As a law of agreements, Bayh-Dole requires 3). Since the other two inventors have no intervening 2) commitments, 1) establishes 3). So when Stanford elects to retain title, it then has standing relative to any prior agency claim to require under 3) (–not 1)–) that these inventors assign to Stanford, which they do. Stanford thereby obtains title. But Holodniy has no 3) because Stanford lacks power to impose 4) as 3). Stanford is required to require 3), but 3) is not compelled on Holodniy except for Stanford’s action to do it, which Stanford cannot, having waived the power to do that back when 2) was established.
If the technology transfer community wants to worry something, it is not that Holodniy does a deal like 2). It is not that Stanford has a promise to assign rather than a present assignment at 1)–because it is the conditional scope at this point, not the form of assignment commitment, that is in play. It is furthermore not that Bayh-Dole is damaged by not rushing in to undo 2) automagically, at some huge loss to Cetus/Roche (and to a bunch of other agreements made over the years)–Bayh-Dole is fine, not the worse for wear. It is not even how to set up competing present assignments, so that somehow a present assignment at 4) can stand in for 3) and trump 2).
First, a present assignment cannot operate within the standard patent rights clause and the written agreement at (f)(2) until the university has standing to invoke 3). It is 3) by which the university the university has the right to request title, not 1) and not 4). 3) does not say present assignment to future inventions. 3) says agreement to sign papers to permit patent applications to be filed and to establish the government’s interest in subject inventions.
What papers are necessary for patent applications to be filed? Declaration and oath. Yes, that would appear needed. A present assignment of future inventions? No. That’s not necessary for an application to be filed. It’s the papers to establish the government’s interest that matter for the transfer of title.
The government has an interest in title to subject inventions until the university properly elects to retain that title. At the point of election, the university then has standing to request title from the employee / inventors under 3) with exactly the rights the government would have under 3). Since the government can request title as a condition of the research award, once the university has elected to retain title, it can request that title *as a condition of the research award*, not as a condition of employment, not as a condition of some IP policy, but as a condition of the same terms of award imposed on the funding agency by Bayh-Dole. Again, this is how federal agreements operate. It is not quite how state contracts operate. And Bayh-Dole is a law of federal agreements, imposed on funding agencies, transmitted to universities, with requirements for the university to obtain agreements, federal in nature, from certain research employees, to protect the government’s interest.
If you have read this through. Wow. It’s raining here. I’m watching A&M hold on against Oklahoma. Check my work. See what you think. I have been working on this for months. I agree that the case illustrates how Bayh-Dole operates, and for that it’s been a great incentive to work things through. I continue to believe Bayh-Dole is a great piece of work. It’s my hope that the Supreme Court will take the opportunity to work it through and show folks just how it works.
Oh, and I have at least one other line of development in this case, for another time.