The NIH’s View of Bayh-Dole Compliance, 3

We are working through NIH guidance on Bayh-Dole reporting requirements. In the process we are making note about how thoroughly NIH misrepresents Bayh-Dole. Sloppy? Indifferent? Does it matter?

The next bit is going to be a bother:

If it helps, just assume that NIH has things backwards. Then it almost makes sense. The Supreme Court made it clear in Stanford v Roche that inventors own their inventions, even those made with federal support. Bayh-Dole does not disturb that ownership. Bayh-Dole gives no special privilege for anyone to take ownership of inventions from inventors. Bayh-Dole manages the priority of claim between a grantee/contractor and the federal agency only once a grantee/contractor has taken ownership of an invention from an inventor.

Now the drafters of Bayh-Dole’s standard patent rights clause (including Latker, it appears, who also drafted Bayh-Dole) were clever folk. They left Bayh-Dole with a huge hole as a statute (yes, I know, the statute itself is something of a hole). The definition of subject invention means that only those inventions owned by a contractor come within the scope of the law. Since Bayh-Dole merely takes precedence over other “acts” with regard to Bayh-Dole’s subject matter, those other acts remain in force for all inventions that a grantee/contractor does not find a way to own. It’s as if there are now two invention management regimes–the former regime of laws and executive branch policies that operate if a grantee/contractor does not obtain ownership, and Bayh-Dole when a grantee/contractor does take ownership. If a grantee/contractor doesn’t take ownership of an invention, then Bayh-Dole doesn’t operate–the former set of special laws and executive branch policies apply.

You see the design? Everything is as it was, unless a contractor finds a way to obtain ownership of a given invention. Then Bayh-Dole’s standard patent rights clause kicks in and limits the conditions under which a federal agency may request ownership, the extent of the patent property rights in subject inventions, and federal agency actions to protect the public from nonuse and unreasonable use by an owner of a subject invention. Bayh-Dole does not create a “uniform” policy on invention ownership–it just adds another layer to the garble–a layer that’s no more uniform in its application and interpretation than any of the other efforts at government patent policy.

But that’s not the end of it. The drafters of the standard patent rights clause saw a way to expand the scope of Bayh-Dole. If inventors own their inventions, then make inventors parties to the funding agreement. Then the inventors will also be contractors. Then their inventions will always be subject inventions (if, of course, otherwise within scope). It’s just that these subject inventions aren’t ones owned by the grantee/contractor–they are owned by the inventor/contractor. But now Bayh-Dole has been used to preempt all those statutes and executive branch patent policies, because all inventions within scope become subject inventions, without the need for any assignments whatsoever. Genius cleverness, if greatly expanding Bayh-Dole after it has been signed into law is your idea of great public policy.

Thus, under Bayh-Dole’s standard patent rights clause, inventors own their inventions as always, but now inventors are parties to the funding agreement–contractors–and their inventions are subject inventions. The grantee/contractors have no special privilege under Bayh-Dole to acquire the inventors’ subject inventions. The inventors are subject to the inventor patent rights clause at 37 CFR 401.9, and all this has nothing to do with 35 USC 202(d)–it’s all under 35 USC 202(a):

Each nonprofit organization or small business firm may, within a reasonable time after disclosure as required by paragraph (c)(1) of this section, elect to retain title to any subject invention

Inventors are turned into small business firms by 37 CFR 401.9 by becoming parties to the funding agreement and owning their inventions. Cleverly done.

Inventors don’t have to assign to anyone. If inventors have timely disclosed their subject inventions and have timely notified the federal agency that they will retain title, then they don’t have to even file a patent application. This is hugely important. This, right here, this very thing–is the federal government’s deference to the academic freedom of faculty inventors working with the support of federal grants. In particular, faculty have the freedom to publish. A patent application is a publication. If faculty had to publish in the patent literature, then they would not have freedom to publish. The omission in 37 CFR 401.9 of the requirement at (d)(2) of the patent rights clause to file patent applications is precisely directed at faculty inventors. Yes, it’s buried. Yes, even faculty don’t know it’s there–thanks to the Bayh-Dole-know-it-all folks at the NIH and at universities. But it’s there.

But of course, we are not done. There’s another pathway. A grantee/contractor gets title somehow and then decides not to retain title.

The inventor may request assignment of invention rights.

Now the inventor does not have rights, which the grantee/contractor acquired but now doesn’t want. The inventor may beg back rights, but now Bayh-Dole’s disposition of priority between the grantee/contractor and federal agency comes into play, since the federal agency now has the right to request title. The inventor, then, begs from the federal agency, not from the grantee/contractor, who already has indicated that it does not want to own.

Agencies support requests of this type to variously.

Garble. Ah, the hell with it. Just write anything you want, NIH.

In all cases, documentation is required when a grantee/contractor organization waives rights to the invention and the inventor(s) wishes to retain the invention

More garble. Nothing in Bayh-Dole requires “documentation” in the case where an inventor retains rights. Nothing in Bayh-Dole requires an inventor to beg rights. At best, 35 USC 202(d) requires a federal agency to consult with a contractor before permitting an inventor to retain rights. But that consultation is–I would presume–to ensure that there is not some agreement–nothing to do with Bayh-Dole–under which the inventor is obligated to assign to the grantee/contractor or (in the case of some universities, back in the less jaded days, a university faculty prohibited itself from holding patents on inventions, especially in areas of medicine). A courtesy consultation, then.

waives rights to the invention

Silly. The assignee/contractor has title. It cannot “waive” title. It can assign title. It can blow patent rights. It could waive the right to require assignment of title. Cannot merely “waive” title and leave title floating above the dead body of a subject invention, to be fought over by angels and demons and the everlasting universe.

and the inventor(s) wishes to retain the invention rights.


Here, “retain” must mean “reacquire” even though the Supreme Court was clear that “retain” in Bayh-Dole means possession: “you cannot retain something unless you already have it.” It’s the reacquisition of a subject invention by an inventor. Look at 35 USC 202(d) again:

If a contractor does not elect to retain title to a subject invention in cases subject to this section, the Federal agency may consider and after consultation with the contractor grant requests for retention of rights by the inventor subject to the provisions of this Act and regulations promulgated hereunder.

The Supreme Court had trouble with the wording here, imagining that an inventor might assign an invention but retain some rights in it, and thus could continue to hold those retained rights. That actually makes sense for university practice. Many universities had patent policies that required the university to take actions with regard to inventions that were assigned to them or give them back to the inventors–so, file a patent application, actively market the invention, get a deal done within a given time frame. Otherwise, assign it back to the inventors. In those cases, the inventors do retain substantial rights. If a university acquires an invention made with federal support–a subject invention–and then gives up on it, then under many university patent policies, the university must give up title to the inventors (now, universities are wont to put strings on the deal–demanding a broad license or a royalty or a requirement that the inventor commercialize–stupid stuff).

Having requested reassignment of a subject invention, the inventor must, according to the NIH,

agree to all terms associated with invention reporting as detailed in 37 CFR 401, and must pursue commercialization of the invention through patent filing or licensing as a
research tool.

At least the NIH is consistent in producing garbled nonsense. The inventor has agreed to invention reporting with the (f)(2) written agreement. There’s no further need for additional agreement on the matter. Further, the inventor is subject to 37 CFR 401.9, not 37 CFR 401.14 with regard to the inventor’s management of a subject invention.

The inventor does not have to pursue commercialization. We’ve been through that. Practical application. And 37 CFR 401.9 omits (d)(2)–an inventor as a small business contractor does not have to file a patent application nor does an inventor have to grant any license. The NIH wording makes it appear that the inventor has the option of either filing a patent application or “licensing as a research tool.” This is garbled. The NIH dearly would love contractors to release research tools on open science/open innovation principles, but to require that the NIH would have to declare exceptional circumstances, which they have never done. So they beg and encourage and bitch and moan. But here, they make it sound as if an inventor could “license” an invention without having a patent position. I suppose–but then we are screwing around with bailment or trade secret or copyright in a regulation pertaining to the use of the patent system. As I said, garble.

At best, the NIH has decided that if a contractor elects not to retain title, or retains title but doesn’t want to file a patent application, then if the invention is a research tool and the grantee/contractor intends to license the invention anyway, without those all-valuable-everything-sits-on-the-shelf-if-there-are-no-monopoly-patent-rights-to-call-forth-the-risk-capital-of-speculative-investors,  then the NIH won’t request title to the subject invention. All the hubbub about Bayh-Dole teeing up patent rights goes away. And the NIH doesn’t even insist that the licensing without patent rights for research tools is non-exclusive and favors small businesses if done by nonprofits. Because. Doesn’t matter.

Only a bit more and we are done. What about assignments of inventions? NIH: There is no restriction under Bayh-Dole for assignments of subject inventions by for-profit firms. No documentation is therefore required.

If the grantee/contractor is a non-profit, the government must approve the assignment

Not entirely true. Here’s the requirement at (k)(1). Read thou:

Rights to a subject invention in the United States may not be assigned without the approval of the Federal agency, except where such assignment is made to an organization which has as one of its primary functions the management of inventions, provided that such assignee will be subject to the same provisions as the contractor;

So a nonprofit may assign a subject invention without federal agency approval (or documentation) to any “organization which has as one of its primary functions the management of inventions.” Okay, an invention management firm, or a private equity firm, or a patent troll. Otherwise, yes, the federal agency must approve the assignment–that part the NIH gets right. Whew!

If the rights are assigned, new rights holder assumes the same reporting responsibilities as the grantee/contractor organization.

Yes, but it is much broader than that–the assignee is subject to the same provisions–of the patent rights clause–as the grantee/contractor. That means, for nonprofit contractors, that any assignee of a subject invention, whether approved by the federal government or not, must comply with the nonprofit patent rights clause, including paragraph (k). Obviously, if paragraph (k) didn’t flow along with assignment, then every university would assign to a for-profit patent management firm, and invention management would be free thereafter for any assignment one wanted to do, and there would be no requirement to share royalties or restriction on use of money earned with respect to a subject invention or requirement to attract small business licensees. It’s clear that (k) must flow along with any assignment of a subject invention. The NIH here is concerned only with reporting, so the NIH doesn’t bother with the big picture. We can give them that.

But in the big picture, the NIH doesn’t enforce this requirement. Nonprofits routinely assign subject inventions to companies that aren’t invention management firms. They do so by masking the assignment as an exclusive patent license–but in practice, and in legal form, they make an assignment of the underlying invention by granting exclusively all substantial rights in the invention–to make, use, and sell, and to enforce the patent. Assignment. NIH doesn’t care. Bayh-Dole, the do WTF you want law–but with gobs bureaucratic paperwork to ensure American innovation is unleashed! Well now.

And in that big picture, when a nonprofit assigns a subject invention to a company, that company is then restricted in its assignment of the invention (including exclusive sublicenses for all substantial rights), must share royalties with the inventors (this is independent of the university’s obligation to do so), must make an effort to attract small business licensees (sounds like a pro-competitive environment or a standard), and most of all, must use the balance of income after allowable costs for “scientific research or education”–that is, for a public interest purpose, not pocketed as profit. Holy cow, Batman–Bayh-Dole does have the public interest at heart. Too bad no one enforces the law.

If assignment approved, third party must pursue commercialization of the invention through patent filing or (for NIH) licensing of the invention as a research tool.

We have been through this once. No requirement to commercialize. No option not to file a patent application for institutional invention owners just because the invention is a research tool. At best, what the NIH is saying is that if an invention owner chooses not to file a patent application on a research tool, then the NIH will not request title to the invention if the grantee/contractor who has declined to retain title still retains title so long as it either goes ahead and files a patent application or without an ownership position in the invention goes ahead and licenses it anyway, as a bailment or copyright or trade secret or whatever.

All about right. The bigger point is that none of this is in Bayh-Dole. It’s a creature of the implementing regulations. Couldn’t help themselves once they started cooking up bureaucracy stew. If you think about it, there’s no meaningful logic here. If the contractor drops patenting, then the government has no need of a patent license–everyone can practice the invention. So what does the government gain by demanding the patent rights that a contractor has dropped in prosecution or after a patent has issued? Ah, the ability to license the invention exclusively–because to license non-exclusively when the invention would otherwise be available to all anyway would require some pretty special insight. Maybe for a standard. Maybe to disrupt companies in foreign countries. Heh, heh. Otherwise, pretty much the deal is that the federal government takes over ownership to trade in patent monopolies, as authorized by Bayh-Dole. Sorta sucks.

This is just something that’s true. Bayh-Dole requires federal agencies to insert an invention utilization reporting clause. But Bayh-Dole does not require federal agencies to obtain those reports. The NIH does. Bayh-Dole purports to make those reports government secrets, excluded from FOIA. Actually, FOIA doesn’t allow what Bayh-Dole claims–and especially what the standard patent rights clause claims–but no matter, Bayh-Dole gets to make things up any way it wants, and so do federal agencies, apparently. Freedom from public oversight and accountability–true bliss for patent speculators!

That’s enough. You get the picture. The NIH does a terrible job of presenting Bayh-Dole in trying to be helpful about reporting requirements.

This entry was posted in Bayh-Dole and tagged , , . Bookmark the permalink.