We are working with an article by Sean O’Connor to get at an underlying problem with discussion of Bayh-Dole. O’Connor, a law professor, appears to be working diligently to find a way to “fix” Bayh-Dole so that universities end up owning faculty inventive work as a requirement of federal funding.
We might then argue that Bayh-Dole’s “failure” (in the eyes of university administrators and the legal professionals who advise them, badly) is actually quite an intelligent public policy expectation. If faculty inventors choose not to assign inventions to the university that hosts their work, then federal agencies must deal with them under the inventors patent rights clause–37 CFR 401.9–not the nonprofit patent rights clause–37 CFR 401.14. The mistaken assumption is that Bayh-Dole’s public policy was to include whole sections of the law that everyone should work diligently to ensure that they never operate.
Bayh-Dole provides for inventors to own their inventions and even to be preserved from predatory actions by the organizations that host federally supported research, but apparently the assumption is that the law here is bad and noncompliance–a kind of administrative disobedience–virtuously restores for the law what surely it must have intended but failed to provide–institutional ownership of all inventions made with federal support. That’s faux Bayh-Dole in a nutshell. But that is not Bayh-Dole, even if some folks continue to enjoy being confused on the matter, and even if some of them make good money going around the country giving talks as if their confusion was newly discovered clarity.
We can see then that Bayh-Dole was drafted by people with an expert awareness of university invention policies and practices. The drafters knew the IPA program and chose not to re-implement the IPA assignment requirement. Bayh-Dole preempts the authority of any standing federal claim to own inventions made in federally funded projects. Instead, federal agencies must derive their authority to claim ownership of inventions from Bayh-Dole. As the Supreme Court held, Bayh-Dole applies only to subject inventions–only to those inventions that a contractor has already come to own. The (f)(2) written agreement extends the scope of the law to all inventions made by inventors who are parties to the funding agreement, not just to the institutional contractors that initially accept federal funds. In all of this, there are no mistaken assumptions regarding university invention agreements and assignment requirements.
There are, however, deeply flawed representations of Bayh-Dole and its implementing regulations, and these go well beyond Sean O’Connors claim of a mistaken assumption. After the Supreme Court decision in Stanford v Roche, we might say these are misrepresentations, not merely mistaken assumptions. One of these flawed representations is that Bayh-Dole requires or endorses institutional ownership of inventions made with federal support. Another is that Bayh-Dole requires commercialization of these inventions. Yet another is the idea that Bayh-Dole implicitly expects exclusive licensing, or that the patent system was to be used “as it was originally intended.” Such a claim ignores the obvious, that Bayh-Dole creates a public covenant that runs with patents on subject inventions, so that the patent system is reconfigured to address publicly funded research and development, limiting the property rights of owners of patents on subject inventions, just as the Kennedy patent policy had done.
Faux Bayh-Dole–the exploitation of misrepresented Bayh-Dole–opens up a pipeline by which university patent administrators can supply patent monopolies to the biomedical industry (and to speculators hoping to sell companies or “monetize” patent rights within this industry). The patent administrators point to the money they receive (or, more often, hope to receive someday, some year, some decade) as the justification for faux Bayh-Dole, for their refusal to comply with the law in almost all of its substantive, non-paperwork requirements, including the (f)(2) written agreement, the restrictions on assignment of subject inventions, and the restrictions on the use of income earned with respect to subject inventions. They would not call it a scam, but we should.
What Sean O’Connor has identified as a mistaken assumption at the foundation of Bayh-Dole is in fact talking-point artifact of non-compliant university practice under Bayh-Dole.
O’Connor, having done a careful job tracing legislative and legal matters, then gives up and draws stick figures for university administrators:
Possibly in part because nothing in the Patent Rights Clauses expressly requires the contractor to have a full patent assignment agreement in place with all inventive employees, some universities began assuming that Bayh-Dole obligated contractor employees to assign subject inventions by act of law so long as the university timely elected to retain title.
“Possibly in part” is equivalent to “I am grossly speculating with no information, but wish to make things sound authoritative.” The adverb “expressly” goes well beyond qualification and suggests that somewhere in the patent rights clauses there is an implication that contractors must have assignment clauses. Nothing of the sort. Thus, O’Connor speculates that universities (abstraction for “university patent administrators and their legal advisors”) “began assuming” something. Such speculation defies belief.
Anyone attending any AUTM conference from the get-go of Bayh-Dole was told by the “experts” that Bayh-Dole’s “elect to retain title” meant “elect to take title.” Anyone reading COGR’s guide to Bayh-Dole would see exactly this same claim. Here’s COGR:
The university is obligated to have written agreements
with its faculty and technical staff requiring disclosure
and assignment of inventions.
Misleading–it’s only true if by “assignment of inventions” is meant, but incompetently communicated, “assignment of inventions to the federal government as may be required under the inventor patent rights clause.” Here’s COGR misusing “election of title” to mean “take title” or “notify the government that title that vested with the university will stay there”:
Upon election of title, the university must file a patent application within one year
And there’s nonsense usage like this:
The retention of title to inventions by the institution is the only way of ensuring that the institution will be able to accept funding from interested research partners in the future
Not only is the claim that institutions must own inventions in order to ensure they can receive funding from multiple sponsors simply unsupportable, but that “retention of title” makes it appear that the action of Bayh-Dole is to permit an institution to keep title to inventions from inventors. No university administrator had to “begin to assume” anything. The patent administrators were told that faux Bayh-Dole was true, and these administrators in turn make it clear to senior university officials that Bayh-Dole gave universities ownership of federally supported inventions. It wasn’t “some” universities–it was nearly all of them, all the time, from the first day any administrator “learned” about Bayh-Dole from COGR or AUTM.
But O’Connor speculates that universities somehow spontaneously “assumed” something about Bayh-Dole. It just has not been the case. But if one ignores practice and constructs a history only on the proper legislative documents, one can easily come to belief speculation must be fact, because otherwise the whole history begins to look bogus.
Of course, non-compliant university administrator Bayh-Dole practice might be described as a “mistaken assumption” and one might even go chasing into history to find some support for the description. But that direction misses the fundamentals of both the law and the practice–folks need to read not only the law but the implementing regulations and the standard patent rights clauses with their waivers, no requirements for enforcement, and unworkable procedures. And folks need to look at practice to see how the patent rights clauses have been dealt with.
Let’s pinch this tighter. Here’s O’Connor’s assertion again:
The Bayh-Dole Act of 1980 was built on a mistaken assumption that “contractors”—recipients of federal funding—were securing assignments of inventions from their employees.
The patent policies and practices adopted by university administrators have been built on a set of political talking points that misrepresent Bayh-Dole multiple ways: that it gave special rights to the organizations that hosted federally supported research; that it requires commercialization; that commercialization requires patent monopolies; that federally owned inventions were not used because no private patent monopoly was available; that universities were doing a better job in commercializing inventions than the federal government was doing. All of it was and is bullshit, but it keeps coming out of the mouths of Bayh-Dole pundits. O’Connor bumbles into Latker and Bremer’s post-drafting politics and mistakes their claims for some mistaken assumption that found its way into their drafting.
Here’s what’s most discomforting about O’Connor’s argument. By accepting the talking points, O’Connor implies that Bayh-Dole was intended to vest title of federally supported inventions with institutions, and the folks doing the drafting just screwed up because they ignored a fundamental point in any research procurement. After all, look at all the university attorneys signing on to the amicus briefs filed in Stanford v Roche arguing that Bayh-Dole vested inventions with universities. They represented that Bayh-Dole intended vesting. O’Connor’s argument then runs–Bayh-Dole intended vesting, but there was a technical snafu and they botched the fundamental idea that first Bayh-Dole had to force inventors to assign their inventions to the organizations hosting the research, and then Bayh-Dole could vest title with those organizations. The obvious fix, then, for the law is to force inventions to be assigned, or force organizations to require assignment–the same thing, essentially.
The Supreme Court wrote that such a “fix” would require a clear intent by Congress to overturn a basic tenet of patent law:
It would be noteworthy enough for Congress to supplant
one of the fundamental precepts of patent law and deprive
inventors of rights in their own inventions. To do so under
such unusual terms would be truly surprising. We are
confident that if Congress had intended such a sea change
in intellectual property rights it would have said so
clearly—not obliquely through an ambiguous definition of
“subject invention” and an idiosyncratic use of the word
It is not even clear that for subvention funding, the federal government has the authority to compel inventors to assign their inventions to the organizations that host the research. In the pre-Bayh-Dole regime, the federal government joined the research investigators as parties to the funding agreement, and required assignment of their inventions to the federal government (when statute or executive branch patent policy required such assignment). That would be a matter of contracting. The IPA program circumvented even this protocol by requiring future federal contractors to require inventors to assign to the contractor rather than to the federal government. Perhaps the federal government had the authority to do so–though it does not appear the IPA program complied with executive branch regulations. But federal patent law is different, as the Supreme Court observed in Stanford v Roche. Whether the law aims to vest ownership of inventions with research hosts, or to deny inventors rights to their inventions (including the right not to file a patent application) in favor of some special privilege bestowed on government favored organizations, patent law would appear not to allow such a thing. The Constitution grants the federal government the right to reserve for inventors exclusive rights. There’s nothing there that the federal government has the right to reserve for institutions that host federal research exclusive rights in inventors’ inventions.
To reserve exclusive rights for institutions puts us right back into the situation that the drafters of the Constitution chose to avoid–government playing favorites with companies and individuals by dealing in patents as private monopoly franchises. Set aside the (failed) argument that patents are the only way that the public might benefit from federally supported research results. We are left with the argument that inventors cannot possibly know how to manage their own affairs with regard to inventions (an assumption, apparently, of federal patent law is that inventors can manage their affairs), and therefore the federal government must direct that nonprofit organizations to step in and take over those inventions. First, that argument repudiates the fundamental federal authority to reserve exclusive rights for the inventors. Second, it plays favorites without merit. Nothing about a university hosting a faculty-proposed project has much at all to do with the merit of the university-as-administration. Bayh-Dole, unlike the IPA program, does not even consider whether a university has an invention management capability, whether its licensing practice is shoddy or capable, nor whether it has a track record of producing useful results. Third, it produces horrible results. Fourth, it is just a more complicated pipeline of patent monopolies on publicly supported inventions to private companies, now obscured by the use of universities.
Let’s look at the assumptions at work. Choose any of these:
Inventors cannot manage their inventions. Or, inventors cannot manage their inventions when they have benefited from public funds. Or, inventors should not manage their inventions when they have benefited from public funds, even if they could manage their inventions–even when they would be the best of anyone in managing their inventions. Or, inventors should not manage their inventions when they do research at a university, even if they would do a better job than university administrators.
Whatever one thinks of these assumptions, they run against the fundamental premise of the Constitution and federal patent law, that inventors are the only ones identified who should have the benefit of exclusive rights for limited times. Whatever version of these assumption one adopts, one also has to accept that the assumption also includes: use the patent system in a manner inconsistent with the premises of the patent system.
If inventors cannot manage their affairs (and the federal government should not take ownership of those inventions and make them available to all, with or without the patent system), then what is the difference between the federal government choosing companies to receive the inventions made with federal support and passing that choice over to the random competence of university administrators?
Let me try again. If it is good public policy for the federal government to force inventors to give up their rights in inventions to organizations that happen to host federally supported research, so that those organizations can pass the patent monopolies along to their favorite companies in the hopes of sharing in monopoly rents made possible by those patent monopolies, then why isn’t it good public policy for the federal government to do this directly? Why should Bayh-Dole have *any* contracting provisions for nonprofits or small companies other than that inventors are parties to the funding agreements and must assign any inventions to the federal government? Then use Bayh-Dole’s authorization for the federal government to deal in patent monopolies to choose the favorite companies to take those monopolies.
Why should a small company, just because it qualifies for a federal grant, be allowed to attempt to commercialize an invention its employees happen to make in that grant work? Why shouldn’t the company be happy with the grant income and come back for more? Wait–that’s what the small companies do, anyway, for the most part! After all, that grant work isn’t procurement–it’s a subsidy, a form of company welfare, a public offset to relieve investors of some of the financial burden of their investment in the small company. And given that many speculator-backed technology companies hope to be acquired by big companies anyway, why should it be good public policy to try to lower the private cost of setting up such acquisitions, just so speculators–er, investors–can have a crack at making a lot of money. It’s like saying that technology uptake is good, but first it has to go to the horse track and be run around a few times so that people can place bets and racetrack owners (university administrators) can attempt to make some money from that betting. Then, if there’s anything left of the technology, some company can use it to work the public over with monopoly patent rights. All this is declared to be virtuous, a wild success, better than anything else we could imagine. Bayh-Dole must be “fixed” to make this virtuous, wild success a matter of law, not merely bureaucratic impulse.
However, most of the time university administrators have no one to license to, and therefore cannot choose anyone to receive their new technology, and unless the administrators are forced to offer non-exclusive licenses (even royalty-free), they insist on attempting to find an exclusive licensee (especially also an assignee) for the invention. That is, the university administrators aim to play favorites, too–it’s just indiscriminately and desperately and unthinkingly playing favorites with anyone willing to take an exclusive license while saying pleasing things about commercialization.
When these administrators work for public universities, we are right back to states offering to hand out patent franchises in exchange for money (and good publicity). State governments become complicit in speculation around patent monopolies–the license deals are set up so that the state makes money when the licensee-assignee makes money, and the state makes more money when the licensee-assignee uses its patent monopoly to suppress competition and charge the maximum price the resulting “market” will bear.
In this effort, of course, Sean O’Connor’s argument is that if you really want to do this as a matter of public policy, you need to force universities to force inventors to give up their rights in their inventions–and gosh people plum forgot that step in Bayh-Dole with the mistaken assumption that universities already did force inventors to give up their rights, even when the inventors weren’t working on behalf of the university. At best, under any argument for equitable interest, a university would have a “shop right” in such inventions unless the university had hired individuals expressly to invent for a problem that a resulting invention addressed.
The actual Bayh-Dole has no such mistaken assumption. Federal laws regarding inventions made in federal funding agreements were not repealed by Bayh-Dole, nor was executive branch patent policy made void, nor were federal agency regulations on inventions made with federal support set aside–other than when a contractor acquires ownership of any such invention. Those laws, policies, and regulations may be forgotten, but they are not gone.
Now here’s the garble.
(1) Bayh-Dole requires no inventor to assign any invention to an initial contractor.
(2) But Bayh-Dole’s standard patent rights clause requires initial contractors to join their potentially inventing employees as parties to each funding agreement, and that makes inventor-employees also contractors, and their inventions become subject inventions when they–the inventors–own them.
(3) The only way Bayh-Dole supports such a thing is if the inventors hold their obligation direct with the federal agency–that is, the inventors must protect the government’s interest, just as must the initial contractor, just as must any assignee or subcontractor. That’s the purpose of 37 CFR 401.9–the inventor-as-small-business-contractor patent rights clause.
(4) Not only does the requirement to make inventors parties to the funding agreement make those inventors contractors, but because the initial contractor is required to take this action itself, the initial contractor cannot at the same time comply with this requirement and also demand conflicting obligations from those same inventors. The federal requirement takes precedence.
(5) But so what? Universities don’t comply with the standard patent rights clause and federal agencies don’t enforce it and even NIST appears clueless about it by inserting new wording that requires initial contractors to require inventors to assign inventions that the initial contractors somehow already own.
(6) So none of this operates.
Follow all that? Bayh-Dole applies only to inventions acquired by a contractor. Bayh-Dole’s patent rights clause requires initial contractors to turn all their inventing employees into contractors. Thus any invention would be a subject invention. Except those subject inventions would be owned by the inventing employees under the inventor patent rights clause, with no obligation to assign their inventions to the initial contractors. Except that no one complies with Bayh-Dole’s patent rights clause. Instead, bureaucrats do WTF they want, and what they want more than anything is to own everything and try to do patent monopoly deals for profit using the findings of research conducted in the public interest. What crap.
These bureaucrats and their speculative buddies–and this is their argument, not mine–would only get involved if they can have outright ownership of inventions and the freedom to deal in patent monopolies. Think about it–do you want to drive these people away from public supported research? Take away their right to compel inventors to give up their inventions, and take away their right to deal in patent monopolies. That is–bluntly, enforce Bayh-Dole’s patent rights clause.
You can’t read the law and know practice, and you can’t review practice and think you know the law. You can’t even listen to the Bayh-Dole pundits at AUTM or any of a number of lobbying groups and think you have any idea at all about Bayh-Dole. You get fed a political ideology about what sounds good–at least sounds good if you are infatuated with bureaucratic, monopolistic control over the findings from research conducted in the public interest.
It is the debris of crappy drafting, clever compensation, bad faith, non-compliance, indifference, and bullshit. The mistaken assumption is that university patent administrators and their legal advisors are right that federal policy intended for inventions to vest (in however which way) in institutions, which were given a mandate to deal in patent monopolies or otherwise be free not to deal at all and withhold inventions from public access. And that’s truly a mistaken assumption. Bayh-Dole provides no support for the assumption, and in practice the university-led attempt to forestall all inventions reaching the public but for exclusive patent licenses is a dismal failure of technology transfer. Even if the mistaken assumption was accurate, that public policy should borgify the findings of research proposed and conducted in the public interest, it just doesn’t work. “Fixing” Bayh-Dole to enable administrative crap merely further entrenches the crap.