Bayh-Dole’s management of title

There is no need for “of the contractor” in the definition of subject invention if a subject invention is just any invention made within the scope of a funding agreement. 

The US Supreme Court looked at the definition of subject invention in Bayh-Dole and held that “of the contractor” did mean something–“of” meant “owned by,” so “of the contractor” restricted the definition to those patentable inventions made with federal support that the contractor owned. The definition did not presume that the contractor owns all patentable inventions made with federal support. Just the opposite: the definition divides inventions into those that are owned by the contractor and those that are not owned by the contractor. The definition also does not operate to assign or vest all patentable inventions made with federal support to the contractor. It simply distinguishes between some inventions (those that are patentable, made within scope of the funding agreement, and owned by the contractor) and other inventions. The definition, appearing in federal patent law, makes patents on subject inventions a new category in federal patent law with its own boundaries of property rights, different from those of other patents.

The IPA approach worked this same territory, but did so with greater clarity. In the IPA approach, a subject invention was any invention made with federal support–regardless of who owned the invention. The government could acquire title to any subject invention according to the terms of the funding agreement under which the invention was made. And the terms of the funding agreement were in turn based on the agencies’ regulations and contracting policies. The IPA, as a master agreement between a federal agency and a university, permitted the university to acquire title to any subject invention that the university decided to patent, subject to a public covenant on how the university could use the patent. The IPA required the university to have patent agreements with employees under which they promised to assign subject inventions to the university if the university chose to patent their inventions. The university had the right to acquire, under the IPA, only those subject inventions that the university had already chosen to patent. Everything else was between the inventors and the government.

To recap the IPA scheme: the federal government required the university to have an agreement with its employees under which they promised to assign subject inventions to the university IF the university decided to patent their invention. (For such an agreement to be binding, there would have to be some sort of consideration offered in return by the university, no?–for another time.) That is, the IPA was selective: first REVIEW the invention, then DETERMINE that a non-exclusive licensing program will meet the public interest or that an exclusive license will result in practical application within three years of the patent issuing and the exclusive licensee is willing for the term of exclusivity to be three years from date of commercial sale or eight years overall. If so, then DECIDE to patent and obtain assignment. If not, then don’t waste any time or money on it–LET IT GO to the federal government to patent for the federal patent commons or to release into the public domain.

The IPA-required patent agreement was between the university and its employees, unlike the (f)(2) agreement in Bayh-Dole’s standard patent rights clause, which is made by the employees for the benefit of the federal government as an extension of the funding agreement. In the IPA scheme, employees never become contractors. In the Bayh-Dole scheme employees must become contractors to make the definition of subject invention cover the situation in which the university does not acquire title.

Under the IPA, patents on subject inventions were regular old patents, but the IPA dictated how the university could exploit those patent rights. This is what I call the public covenant. It ran with the federal contract. The IPA required that

• the university license non-exclusively, royalty free or for a reasonable royalty; OR
• the university justify any exclusive licensing AND
• exclusive licenses may last only for three years from the date of first commercial sale or eight years from the date of the license unless the federal agency approved an extension AND in any case
• the university, assignees, or licensees achieve practical application within three years from the date of the patent issuing AND
• the government could intervene at any time if it determined that the public interest was better served by non-exclusive licensing or licensing to more companies.

Unlike the IPA, Bayh-Dole does not define a subject invention as simply a patentable invention made under a federal funding agreement. As an added nit, it inserts as well plant variety protections–which are not patentable inventions and an odd thing to add to patent law in this backhanded way. Bayh-Dole also adds “of the contractor,” making the definition apply only when a contractor has acquired an invention. Since the funding agreement is with a university and not with individual inventors, nothing the inventors invent is a subject invention until it is acquired by a contractor–but for implementation of the (f)(2) provision in the standard patent rights clause. This gap or gaffe or intentional lacuna in Bayh-Dole gets “fixed” in the standard patent rights clause, but in an odd–and perhaps an astute–way. University administrators and their mercenary legal advisors, however, try to “fix” the problem another way, by asserting that the university must own everything up front “to comply with federal law”–as if Stanford v Roche addressed merely a technicality of how the law forced or should force or was intended to force all invention rights to federal contractor-employers.

Rather than requiring each university to demand that inventors assign to the university all inventions made with federal support, and thus turn them all into subject inventions, the standard patent rights clause authorized by Bayh-Dole (but not in Bayh-Dole) requires each university, for each funding agreement, to require employees to make a written agreement to protect the government’s interest. When a university requires this agreement, each employee making the agreement becomes a party to the funding agreement for the limited purpose of reporting inventions, signing papers to permit patent applications to be filed, and signing papers to establish the government’s rights in these inventions. As a party to the funding agreement, each inventor is therefore a contractor. Since the inventor owns his or her invention, the invention is a subject invention and the rest of the standard patent rights clause can work as necessary. Obscure? Yes, of course. Clever? Apparently so. Necessary? Only if the fundamental architecture of Bayh-Dole is a garbled mess–which it is.

We can see that inventors, having made the (f)(2) written agreement, are contractors. In 37 CFR 401.9, when inventors are allowed by a federal agency to retain ownership of their inventions (that is, the agency decides not to request assignment under its rights in the (f)(2) agreement), then the agency is to treat the inventors as a “small business firm contractor“–but with fewer constraints:

Agencies which allow an employee/inventor of the contractor to retain rights to a subject invention made under a funding agreement with a small business firm or nonprofit organization contractor, as authorized by 35 U.S.C. 202(d), will impose upon the inventor at least those conditions that would apply to a small business firm contractor under paragraphs (d)(1) and (3); (f)(4); (h); (i); and (j) of the clause at § 401.14(a).

For the march-in provisions in 37 CFR 401.6, the definition of “contractor” expressly includes inventors when they own their inventions:

The procedures of this part shall also apply to the exercise of march-in rights against inventors receiving title to subject inventions under 35 U.S.C. 202(d) and, for that purpose, the term “contractor” as used in this section shall be deemed to include the inventor.

But why doesn’t the standard patent rights clause just require universities to obtain promises to assign from all employees? Why the strange apparatus relying on contractors? Keep in mind that Latker, the HEW patent counsel that was one of the key people, along with Howard Bremer, in reviving the IPA system and in drafting Bayh-Dole, moved from HEW to the federal agency charged with drafting the implementing regulations for Bayh-Dole. Whatever Bremer and Latker wanted to achieve, it was going to be an inside job–if they could get away with it.

The management of title to inventions is one of Bayh-Dole’s least apt pieces of work, which is all the worse since title is the one thing that Bayh-Dole was put up to addressing. In 401.9, inventors “retain” rights. In 401.6, they “receive” rights. In the standard patent rights clause, at 401.14(a)(c) and (f) a contractor “elects to retain title” but in 401.14(a)(d) a contractor “elects title.” Title is never assigned–rather it is elected or conveyed (401.14(a)(d) and (f)) or established (f) or received. It’s no better in Bayh-Dole proper–a contractor may “elect to retain title” (35 USC 202(a) and (d)) or “elect title” (35 USC 202(a)) or “elect to retain rights” (202(c)) or “elect rights” (202(c)). Sloppy? Incompetent? Clever? Precise? Dunno.

The Supreme Court in Stanford v Roche ruled that all such uses refer to the disposition of rights in the funding agreement as to whether the claims of a contractor or federal agency take precedence, and these uses do not refer to the vesting of title in the first place–which is with the inventors. An inventor may then “receive” title simply by inventing, or by obtaining title from a university that has obtained title but decided not to “elect to retain title” and has assigned title back to the inventor. Or an inventor may apparently also receive title from the federal agency, if a university obtains title and then decides not to retain that title and assigns title to the federal government.

“of the contractor” matters even if the management of title under Bayh-Dole is screwy. Franken-sausage legal drafting. Perhaps folks were trying to be too clever. Perhaps folks were trying to snooker Congress and then university faculty into believing that Congress really had changed 200 years of federal patent law with an idiosyncratic use of “retain” (as the Supreme Court wryly put it). In the IPA, universities are required to obtain assignments. In Bayh-Dole, assignment doesn’t appear, as if by eliminating the word somehow title could float freely among whatever claimants might make a stretch for it. “Elect” appears to come from the IPA–where universities had an option on a first right to obtain title. They “elected” the option offered by the IPA contract to intervene in what otherwise would be the federal agency’s decision. In Bayh-Dole, “elect” is paired with “retain” and “rights”–echoes of the option in the IPA. One “elects to retain” under an option; one “elects rights” of the option. To “elect title” is to choose the option under which one has the right to acquire title.

The problem with “assign” has to do, perhaps, with placing Bayh-Dole in federal patent law. Here’s the definition of “assignment” in 37 CFR 3.1–

Assignment means a transfer by a party of all or part of its right, title and interest in a patent, patent application, registered mark or a mark for which an application to register has been filed.

In patent law, an assignment involves a patent application or patent. Inventions are outside the definition. Since Bayh-Dole introduces a new class of invention into federal patent law–the subject invention–and wants to deal with title to the invention rather than to a patent on the invention, it can’t use “assign” without disturbing the definition of “assignment” already in patent law. If the drafters of Bayh-Dole had used “assign,” then everyone would have had to wait until inventors had filed a patent application for any of the electing provisions of the standard patent rights clause to kick in. Maybe that would have been a good thing. If the inventors don’t file, then stuff hits the public domain and that’s the end of it. If inventors do file, then federal bureaucrats and university bureaucrats get to squabble over who gets to play the middleman–or, rather, under the standard patent rights clause, university bureaucrats get to keep title if they can get title, and then federal bureaucrats have zero motivation to pay attention to whatever the university bureaucrats do after that other than the petty paperwork compliance bits.

If the drafters of Bayh-Dole had used “assign,” then they would have made it clear that inventors owned the right to patent their inventions made with federal support and that a university had to obtain assignment to get title; electing to retain title was purely a matter of federal contract, not of patent law’s disposition of title. The Supreme Court rejected this scheme to separate title and inventor.

The U.S. Constitution gives Congress the right to secure for limited times to inventors exclusive rights in their discoveries. A patent is one way to secure such exclusive rights. The IPA is purely a creature of the executive branch–subject to executive orders and agency regulations. The IPA can manage the disposition of ownership once there is ownership, but it could not change that initial ownership–because that’s a matter for Congress, and a matter of Congressional authority. To create a federal patent law that declares then that some inventors do not have exclusive rights at all–their employers do, and not even as a result of employment, but only for having received federal money generally provided expressly to support the work of the inventors–that’s beyond the scope of what Congress has the authority to do.

The faux Bayh-Dole scheme is not even “invent for hire”–it’s “invent for profiteers posing as trustees”: first, trustees of federal money designated to support the work of individuals; second trustees of patent rights in inventions made by those individuals, to serve the public interest in the deployment of those patent rights. The middlemen violate the first trust by taking for their own institutional profit what was never theirs to take; they violate the second trust by abusing the reputation and public standing of the university as cover for any sort of patent exploitation they choose to do–it must all be in the public interest because the university can’t act otherwise. Or so the scam goes. Moral scams are the worst scams, because they draw in folks who otherwise would not participate but who are told it is okay to do this because it’s federal law.

“of the contractor” could not give university officials an institutional “equitable ownership of title” or a “presumption of ownership” (even if not equitable) in patentable inventions made with federal support. But it could create the illusion and support a misreading that if repeated often enough started to look true, and as a matter of federal law. Even if running against university values, policies, and practices, the illusion looked pretty darned enticing as a “new source of revenue”–and acting on the illusion was declared moral, too. Success meant merely acting on the illusion. Everything after became a matter of confirmation bias and halo effects. Even when the prophecy failed, believers have nowhere to go other than to admit their culpability–and so instead they deepen their commitment to the scam. They will not be persuaded otherwise.

Sean O’Connor posits that Bayh-Dole was based on a misconception arising from an obscure report on patents from the 1940s. Bremer and Latker, the architects of Bayh-Dole and of the IPA, had no need to rely on the Biddle report. They had detailed, technical knowledge of how universities operated. They were had no illusions. They knew what they were doing–it’s just that they could not quite pull it off, so they declared victory anyway and set about making university and federal practice conform to their happy misreading of the law rather than to what the law actually could deliver. If there has been a misconception involving Bayh-Dole, it is not that Congress assumed that universities always obtained promises to assign from their employees but rather it is that people continue to think that Bayh-Dole is anything other than scam to create work for patent brokers–with lots of moralizing surface detail about serving the public interest. The only folks to consistently make money over the 35 years of Bayh-Dole are the university administrators and their patent attorneys. In that regard, Bayh-Dole has been a huge success. Look at who drafts those accounts of Bayh-Dole’s success. It’s always someone who has got a pay-out as a result or is uncritically repeating the claims of those folks.