A law firm gives an overview of the new NIST regulations for Bayh-Dole. It leads with fake history. Not a good sign.
First enacted in 1980, the Bayh-Dole Act (as amended, the “Act”) for the first time permitted research institutions and other recipients of federal research funding (e.g. from the NIH, DOD and NSF) to retain ownership of resulting inventions and be able to license them to companies and other entities in exchange for royalties.
With all due respect, total nonsense. The PHS IPA program started in 1953, lapsed in the late 1950s, was revived in 1968 and ran until it was shut down in 1978 for being ineffective. Under the IPA program, nonprofits could own inventions made with federal support and do exclusive licenses, even for royalties. There were over 70 nonprofits in the NIH’s IPA program by 1978. The NSF also ran its own IPA program. Norman Latker claimed Bayh-Dole was based on the IPA program (and he ought to know, since he drafted both). The Department of Defense had a long-standing policy of allowing contractors to own inventions made with DoD funding. The issue of ownership was *not* with these federal agencies in their contracting for research–the problems were with the Department of Energy, which apparently took a long time to make determinations to allow contractors to own inventions, and with the Department of Health, Education, and Welfare policy of not granting exclusive licenses to inventions that HEW owned.
In addition to the fake history, we also find the ambiguous use of “retain.” Research institutions were allowed to “retain” title to inventions that they otherwise had acquired title to. “Retain” does not mean “preclude inventors from holding title to the inventions they make,” but rather means “preclude the federal government from acquiring title, once a research institution has obtained title.” But anyway.
More fake history:
The profession called “Technology Transfer” emerged from the Act, with vastly more patents being filed and licensed out and a wave of financing of new enterprises and technological development that has continued, to the benefit of the US economy.
“Technology transfer,” as David Mowery et al. point out in Ivory Tower and Industrial Innovation, was used by Research Corporation to describe setting up offices at universities to assist inventors in presenting their inventions to Research Corporation for possible management. Yeah, the technology transfer was from inventors to Research Corporation–really. It may be that the “profession” of “technology transfer” “emerged” from Bayh-Dole practices, but the activity called technology transfer came well before. Bayh-Dole was predicated on the growth of that technology transfer activity at universities. The history here is backasswards.
It’s true, universities did file more patent applications–in part because they referred fewer inventions to Research Corporation and other invention management organizations. But here’s a graph of issued utility patents from universities and other nonprofit foundations. Quick, where’s the Bayh-Dole effect?
Yes, the huge increase in patenting comes about 1995–gene patents, anyone? race to get applications in before the 17-year term from date of issuance changed over to 20-years term from date of filing? We are a long time after Bayh-Dole’s passage. Besides, there was the IPA program before Bayh-Dole. Bayh-Dole changed mostly nothing for most universities already receiving NSF and NIH funding. There ought not be any Bayh-Dole effect for these institutions. But look–the weird thing is that universities are getting more patents in the last decade that don’t recite federal funding. There may be a Bayh-Dole effect is now reducing universities’ interest in obtaining patents on federally supported inventions. Who can say?
As far as I can tell, even the overhyped crazy-assed “economic” model of university invention licensing impact amounts to a rounding error on both startups and economic impact. The benefit to the American economy of the profession of technology transfer is a rounding error, even with the silliest of “models” to estimate that impact. And we haven’t even gotten to the idea that university technology transfer does more to drag down innovation opportunities than it supports. That is, university inventions appear to get used in many cases despite the profession of technology transfer. Despite university bureaucrats insisting on having a thumb in every innovation pie. Great picnic. Pity the bureaucrats have got to all the pie first.
With this fake history as a premise to establish the importance of Bayh-Dole, the article moves into a discussion of NIST’s changes. Here’s some interesting news with regard to NIST’s removal of the 60-day window in which a federal agency must act to request title to a subject invention that a contractor has failed to disclose or failed to elect to retain title:
It is believed that many funding recipients fail to make these required reports, but rely on the anticipated failure of the Federal agency to give the required notice to assert title within the 60-day limit.
Gosh, how could that be? Bayh-Dole is a do WTF you want law. There’s no obligation for a federal agency to enforce the standard patent rights clause or to act on the federal government’s rights. Even so, why is compliance a matter of “belief”? Well, that’s apparently because NIST doesn’t bother to compile any data to determine the degree of compliance. And really, how is it that anyone can determine what inventions a university owns that were within the scope of a federal funding agreement if those inventions are not disclosed?
Under the old regulations, if a federal agency didn’t act timely to request title, then title acquiesced where it was. No big deal–if an invention was a subject invention, it still was, paper formalities or not. A subject invention does not become a subject invention by being disclosed to the federal government. The big consequence is that the federal government does not receive its royalty-free license to practice and have practiced because that license is triggered by a contractor electing to retain title, not by a contractor acquiring title. It’s just–uh–have you heard of any government agency practicing or having practiced a subject invention? Yeah, doesn’t much happen. No one cares. It’s a paperwork fuss.
The result is that if a funding recipient fails to timely comply with the invention reporting or title claiming requirements, the Federal agency can at any time during the life of a reported invention, a resulting patent application, or resulting issued patent demand a transfer of title to it. This, in effect, creates a cloud on the invention/patent, which would apparently put any licensee at risk of losing its rights. The funding agency could, of course, elect to recognize the license, but at the least it would have a chance to re-negotiate it.
And pigs may grow wings. The cloud on the title to the invention is created by the failure of a contractor to disclose the invention to the federal agency. There’s a protocol for disputes. A contractor with a concern might think to trigger that protocol rather than wring its poor institutional hands about it. A licensee is hardly at risk for losing rights–yeah, perhaps a renegotiation–but that would be about losing money. We might ask why it is that a federal agency would disrupt a license (if in good standing) for more money. Back in the day, when federal government patenting was discussed, the argument was that the federal government should not be in the patent business for money. That’s why, in fact, that the patent clause in the Constitution is directed at inventors only, and that federal patent law grants patents based on the merits of inventions and not on the favoritism of government. But oh well. Who cares about history? If federal agencies aren’t requesting title in the face of defective disclosure and election to retain title, why would anything change with the removal of a time limitation to act? Create more uncertainty where no one cares? How is that a regulatory improvement we all need?
All Federal funding recipients must now take the regulations seriously and establish procedures to ensure that deadlines are not missed.
Parties negotiating for licenses to Federally-funded inventions or patents must now ensure that the licensing organization has fully complied with all requirement of the Act.
Silliness. The issue is whether a federal contractor has properly disclosed an invention and elected to retain title to it–not all requirements of “the Act.” The contractor’s requirements are those of the standard patent rights clause authorized by Bayh-Dole, not in Bayh-Dole the statute. All Bayh-Dole does is to require the use of a standard patent rights clause if a federal agency cannot justify an exception. No contractor I know of complies with the (f)(2) written agreement requirement of the standard patent rights clause. Universities routinely disregard the patent rights clause requirements on assignment of inventions and use of income earned with respect to subject inventions. They also ignore the requirement to attract small business licensees. So what?
Nonprofit institutions are notoriously reluctant to make any representations or warranties, but they must be prepared to warrant that they own, without qualification, the inventions and patents which they are licensing.
But most universities I know of do represent that they have the right to grant the licenses they grant. Here, for instance, is text from a template patent license from the University of California:
The Regents warrants to Licensee that The Regents has the lawful right to grant this license.
Universities often aim to pass on the “Bayh-Dole requirements.” If the license is an exclusive one–to make, use, and sell–in the United States, then the license is an assignment of the invention. The standard patent rights clause requires that nonprofits require the nonprofit patent rights clause to apply to any such assignee. I don’t see that universities have much to do to change their practice.
Licensees can consider seeking copies of the necessary and timely filings required under the Act.
Bayh-Dole and its patent rights clause make the filings exempt from FOIA disclosure. The licensees will have to get the copies from the university licensors. But even then, it will take some heroic work to determine that the disclosure is complete (in addition to being timely)–one needs to compare the disclosure to the government with the patent application, and one needs to confirm the date on which personnel responsible for patent matters received a compliant disclosure from the inventors, in addition to the date on which the federal government received the disclosure. Election to retain title is simpler, as it depends only on the date of disclosure to the federal agency–for that, I suppose one needs a confirmation from the federal agency of that date. Lotsa luck with that!