I’ve written a bunch on this topic. Here’s a shorter restatement of the previous posts.
Patent rights under Bayh-Dole do not vest outright with the university, but rather with inventors as rights in personal property.
Title to inventions in Bayh-Dole is managed in the way of options in a contract. To “retain title” means to “exercise one’s prerogative to stand in place of the agency to direct disposition of title, subject to the requirements of the law’. “Retain title” means “to act so as to secure title” not “keep what you already have”. Think “retain attorneys” not “retain water”.
This reading recognizes that Bayh-Dole had to address three existing approaches: national patent management firms (such as Research Corp), affiliated foundations (such as WARF, but also others), and university-based programs (such as Stanford’s). Forcing title to vest in the contractor would not be consistent with established practices. It makes sense for Congress to draft the law so that any of these could continue to operate. That’s what the apparatus for “protecting the government interest” with “written agreements” is all about in 37 CFR 401.14(a)(f).
The reading “retain title means we must already have it” is reasoning from the *word* “retain” not from the *function* indicated. That’s a bad sign right there. It’s not the only way to consider the word or the function. In 37 CFR 401.9, contractor employees are referred to as “retaining title”. This couldn’t happen if title vested initially and outright with the institution when an invention was made, for then the institution would have to pass title to the agency or to the inventors by assignment. If “retain title” in 37 CFR 401.14(a) means “already have it”, then why doesn’t it mean that in 401.9? I suppose words can mean whatever you want them to mean, but that’s an odd way of reading the law–I can’t tell if it’s a will to power or just an egg talking on a wall. In 37 CFR 401.10, federal employees have the possibility of personal rights in subject inventions made jointly with university personnel–the government may “transfer or re-assign” those rights to the university. Can’t rightly re-assign if one hasn’t had an assignment first. That’s a good use of reasoning from a word that does unambiguously reflect its function.
Expecting Bayh-Dole to treat title in subject inventions in the usual manner makes sense of the requirement that contractors act to protect the government interest in written agreements with employees, which would not be necessary if the institution owned outright. If institutions owned outright, they would not be looking to their IP policies for punch in complying with Bayh-Dole. If they were claiming they had to have assignments (promises to assign, or present assignments of future inventions), it would not be true: first, because they do not need assignments for what they already have (so it would go); and second, because Bayh-Dole already provides apparatus for this and it is *not* a function of a university’s employment agreements or IP policy. If anything, it is a function of a university’s research policies requiring compliance with the terms of extramural research, for which the written agreements required in 37 CFR 401.14(a)(f) are an instance. Bayh-Dole is about uniform federal procurement behaviors with regard to inventions, not a revision of patent law to hand invention ownership to universities on a platter. A university with no IP policy at all has a better chance of implementing Bayh-Dole than one thinking Bayh-Dole needs the help of an IP policy!
If there really had been an intended change in vesting, so as to remove inventors’ personal rights from the proceedings if they happened to be university employees working with federal research funds, I would expect 35 USC 261 to be amended, or the MPEP to make note of a change in practice as substantial as this. I don’t see anything there at all. Have I missed this bit? Clearly the amicus brief folks did to, because they hung their hat on “retained”. Either it’s very cleverly done to slip something by everyone (including, it would appear, a lot of universities and the PTO), or it is legislation by implication (what we really mean is so manifestly obvious that we choose not to write it down directly, as that would, you know, sort of spoil the fun of seeing who gets it), or the big change never happened (because, obviously, it did not need to happen, was not intended to happen, would not do anything useful to meet the stated objectives of the law, and would be all around very bad).
In all, it’s vastly superior that title vest in inventors as a personal right, to be disposed of according to the processes already available and well established in patent law. That allows all things involving assignment and title to go on as before. There is no need to note any changes because there were none. This is a different kind of argument than the title-retentive bureauklepts must make–that there was a big change but it all got packed into “retain” in one place and not another, and then only discoverable by implication.
If a university elects title, then under Bayh-Dole and the written agreements it has in place, the university may require assignment to itself, to a related foundation, or any organization that has a primary function in managing inventions. Or of course, the university may waive its right to retain title, or give up at any point on the patenting, or, be totally disengaged and do nothing at all with its right to elect to retain title–and inform its research employees of this choice in the written agreements. The university may assign rights it holds under Bayh-Dole, not just title to a subject invention–that is, a university may transfer the right to elect to retain title, and to thereafter obtain title and file patent applications. See 37 CFR 401.14(a)(k)(1). This is similar to (g), involving flow down of Bayh-Dole in subcontracts. It’s the award conditions that may be assigned. Again, it is easy to see where confusion might arise, but one reading makes sense of the use of the language of the law and implementing regulations, and the other ignores whole swaths of stuff. In the reading that holds up the better, universities have choice not only regarding the disposition of title, but also in their approach to how that title is then managed. In the other reading, all this is gone. In its place, one approach only, dictated by law. When again did the creative class start advocating for getting less freedom from the law? How can AUTM think to represent anyone dealing in innovation, given their role in the amicus brief? What a disappointment, if not a self-disenfranchisement from a public leadership position on research innovation. Even if the law might allow title-retention, why would you advocate for it, representing innovation? No, it’s something else. It’s very bad. I don’t see this as a little OMG accident.
Finally, why did Stanford seek an assignment from its inventor if it already had title? If the reading favoring institutional vesting is correct, then there was nothing to assign to Stanford because *they already had title*. At best, it makes the assignment document a scavenger hunt item to trick the PTO into changing the assignment information on the face of the patent, since even the PTO apparently hasn’t got the word about Bayh-Dole. Gosh, that all seems so unlikely. Don’t you think the PTO a few decades ago would have issued a rule to stop universities from submitting void assignments and instead make them submit a copy of the notice to the agency of election to retain title (that they already had under Bayh-Dole, that clever, implicatious, title-retentive law)? But no, we don’t have that. It’s all a big surprise 29 years later. Does that make any sense?
Even sophisticated university practice points to the expectation that subject invention patent rights are personal and transferred by a writing executed by the inventor or the inventor’s legal representative.
For all that, inventor ownership is a good thing. There should be some respect in practice, even deference, especially at the institutional level. Innovation is more than patent rights. It’s tacit knowledge, breakthrough networks, and opportunity referrals. You need more than tech transfer professionals for that. Corporations, even those built on the strength of way cool inventions, tend to have a lousy track record the bigger and more stable they get, and the further removed from those early, heady days. Call it the PARC dilemma. Even better: universities should designate principal investigators to be a part of the process when an invention is made with federal funding, and should ensure that PIs have a voice in all matters pertaining to election to retain title and subsequent licensing efforts of subject inventions made on their watch. The deal that really matters is the one between PIs and inventors over the best course to advance the purposes of the federal funding, as set forth in the funding arrangement, and as set forth in Bayh-Dole. That would be consistent not only with Bayh-Dole but also with Circular A-110. So start with the collaboration that binds a research team to a public purpose beyond getting the next award, something worth doing, like inspiring uptake of something new in the world, discovered with federal funds at a university.