Not fixing a hole in Bayh-Dole

Bayh-Dole does not disturb federal common law with regard to inventions. Inventors supported by federal research funds own their inventions. The Supreme Court made this clear in Stanford v Roche.

Bayh-Dole applies to subject inventions only. Subject inventions are patentable inventions made with federal funds and owned by a contractor. A contractor is any party to a federal funding agreement for research.

Unless a contractor owns an invention made with federal funds, Bayh-Dole does not apply. Under Bayh-Dole, an inventor has no obligation to assign inventions and so turn them into subject inventions or even to report inventions unless they are subject inventions.

There is nothing, moreover, in Bayh-Dole that requires inventors to assign inventions made with federal funds to the federal government, or requires inventors to promise to assign such inventions to the federal government.

Pause with me a moment. If an invention is not owned by a contractor, then the invention is not a subject invention and Bayh-Dole’s apparatus for patent rights clauses in funding agreements does not apply.

You might see how this outcome consternates people convinced that Bayh-Dole was the government’s gift to university technology licensing programs, that somehow Bayh-Dole was intended to give universities outright ownership of inventions made with federal support, with a mandate to do anything they could to make money from patents on those inventions. How can any of these good things happen if Bayh-Dole only applies after a university has already acquired such an invention? Surely the law is supposed to make it easy for universities to acquire inventions–you know, by “electing title.” But that’s not how Bayh-Dole puts it.

Perhaps this absence of a demand that inventors report and assign inventions made with federal funding is a hole in Bayh-Dole, caused by the same crappy drafting that plagues the law throughout. Or perhaps this is a clever ruse to make it appear that Bayh-Dole vests rights automagically in contractors, and putting Bayh-Dole in federal patent law aimed to heighten the claim to this “presumption.” Or– just to make clear there are plenty of options, none exclusive of the others–maybe the drafters of Bayh-Dole (i.e., Norman Latker) reached the limit of federal constitutional authority and simply had no standing to force ownership of inventions to move by federal law from inventors to the organizations that happen to handle federal money provided to support research. The Constitution, after all, reserves to the federal government a right only with respect to inventors, not to employers, or contractors, or certainly not to institutional money-handlers. Or–and this seems more likely to my mind–Bayh-Dole would never have been passed, even as a parting gift to Senator Bayh, if the law had expressly stated that inventions made with federal support would be owned outright, not by the inventors, but by universities and their patent broker associates.

Call it another political expediency needed to get a law through Congress that looked as much as possible like the defunct Institutional Patent Agreement program. The aim was–I suggest–to make Bayh-Dole look as close to the IPA as possible, and then use the implementing regulations and extend the standard patent rights clause to make it appear that Bayh-Dole did whatever it was that universities and pharma wanted the law to do–namely to preserve monopolies in federally supported inventions, to obtain these monopolies with impunity, and to convey these monopolies to companies (or speculators) for profit. The cover for the claim has been that only by doing these things can inventions made with federal support benefit the public. Otherwise, you know, inventions just “sit on the shelf.” It’s a crazy nuts bonkers argument, but somehow it manages to persuade folks. The argument implies such nonsense as:

  • Only patented things get used.
  • People only make new things when there’s a monopoly to extract maximum profit.
  • Nothing new can be made unless it is first made as a commercial product.
  • The public is served by ensuring that any new product carries a monopoly price.
  • We must monopolize all research inventions so some can be exploited for profit.
  • If just anyone could use an invention made with federal support, then no one would have an incentive to make a product version–so people have to be forbidden to use inventions until someone is positioned to sell a product version to them.

Or, just to fill out the possibilities, perhaps this hole in Bayh-Dole was intentional and may have been set up to make it clear that neither the federal government or an organization that hosts federally supported research should have any claim whatsoever on inventions unless and until an inventor assigns such an invention to a contractor. Then and only then does Bayh-Dole kick in. Now, whether this last possibility was intentional does not so much matter: this is the law. The Supreme Court ruled as much in Stanford v Roche. Bayh-Dole concerns the disposition of rights between a research host and the federal government, but only after an invention has become a subject invention–that is, after the invention is owned by a contractor–a party to a funding agreement. “Nothing more”–quoth the Supreme Court on the matter.

Now some commentators will argue that this hole in Bayh-Dole has to be “fixed,” Beatles-like. That somehow, there is a free-standing, authoritative intention for the law that universities must own inventions made with federal support, and must be able to pass monopolies on these inventions to companies–especially those in pharma–so that companies, their investors, and university administrators could all share in the benefits of patent monopolies on federally supported research. A monopoly is a terrible thing to squander, especially on people who need life-saving medications and therefore will pay a pile for access. But there is no such authoritative intention in the law, and certainly not in the statement of policy and objective that fronts the law, at 35 USC 200.

Any effort to push inventions made with federal support into the arms of a contractor comes from outside Bayh-Dole.

There are two paths people use to force inventions into those waiting arms. The first, and most used, is that university administrators demand that inventors assign their inventions to the university. Administrators use a variety of methods. Some simply bullshit and claim that Bayh-Dole requires assignment. Others put language in their patent policy that makes it appear that inventors have agreed to assign, or face termination (of employment) if they don’t assign. Still others make all new employees sign paperwork saying that they will and already have assigned. All of these methods are then backed by university attorneys paid well to prevent anyone from challenging the legality or sufficiency of these policies, contracts, and sham-contracts.

I’ve written plenty on the topic of policies, contracts, and patent agreements–especially for university faculty–so won’t belabor things here. Faculty aren’t employed for much of their research. They are assured of academic freedom. Faculty don’t generally agree to things buried in policy documents that repudiate that academic freedom. And even if faculty do agree, it’s not clear there is any consideration to make that agreement contractual with regard to their inventions. And even if the agreement is contractual, then the faculty member gets the benefit of stability of the contract and the university cannot change the terms of that contract at will, and moreover, the contract should be interpreted in favor of the non-drafting party (and thus, not by a senior university official looking out after university administrators). The Shaw court made much of this expressly clear.

But for the apparently unlimited university budgets for attorneys to defend defective policy claims, sham-contracts, and self-interested interpretation of adhesion contracts, this approach to forcing inventions from inventors to university administrators would fall apart quickly.

The second path to getting inventions into the hands of contractors is to insert that requirement into the standard patent rights clause authorized by Bayh-Dole. This is what NIST proposes to do. It’s just that there is no authority in Bayh-Dole for such a requirement, especially after Stanford v Roche. Bayh-Dole represents federal policy on the matter of inventions in contracted research, displacing all other legislation on the matter and dictating executive branch contracting terms. Bayh-Dole does not require assignment by inventors. So how could NIST create a standard patent rights clause that does? No, the impetus to make inventors assign to universities does not come from Bayh-Dole, and does not even come from Congress. It must come from someone who wants inventors to give up their inventions so that patent brokers can have a livelihood conveying monopolies to companies. Who might that be? Figure it out.

The standard patent rights clause authorized by Bayh-Dole (actually one of four–[NIST has changed these citations in its May 2018 changes–(b) stuff is rolled into (a) and (a) is eliminated] 37 CFR 401.14(a) (for nonprofits), 37 CFR 401.14(b) (for certain weapons related work), 37 CFR 401.14(a) less (k) (for small businesses), and 37 CFR 401.9 (for inventors to be treated a bit like small businesses)) does contain a gesture regarding inventors assigning–but not to the research host. 37 CFR 401.14(a)(f)(2) requires the research host (a “contractor”) to require each employee (other than clerical and nontechnical employees) to make a written agreement “to protect the government’s interest” (as the section heading has it). That is, universities are to flow down to inventors–delegate to–assign to–stand aside in favor of–substitute–certain obligations and rights relative to inventions made with federal support:

to report subject inventions to personnel designated for patent matters

to sign papers to permit patent applications to be filed

to sign papers to establish the government’s rights in subject inventions

Of course, an inventor could not sign any papers to establish the government’s rights if the inventor has no rights to convey–such as because the university’s administrators have created a policy statement that says the inventors do not own their inventions. But the standard patent rights clause requires the university to take an action–and so the university must waive its policy claims and instead require (and thus permit) each potential inventor to make a written agreement to establish the government’s rights in subject inventions.

But we still have the hole. The written agreement pertains only to subject inventions–to report them, to allow patent applications on them, to establish the government’s rights in them. But a subject invention must already be owned by a contractor–so how could an inventor ever have any rights to sign over to the government, if a contractor must already own those rights to make an invention “subject”? And even then, what rights in a subject inventions would an inventor would still have, once an invention has become subject, and thus has been assigned to a contractor?

Well, here’s the cleverness. The written agreement required by (f)(2) turns inventors into contractors. The definition of contractor is “any party to a funding agreement.” And the definition of funding agreement includes assignment, substitution, or subcontract–and thus, a party to any assignment, substitution, or subcontract is also a party to the funding agreement and hence a contractor. Follow that until you follow that.

When universities delegate research work to another organization, they subcontract, and the subcontract becomes part of the funding agreement, and thus the subcontractor becomes a contractor with the same rights and obligations under the standard patent rights clause as the primary research host. The same thing happens with the (f)(2) written agreement requirement, except now the recipient of the flow down is the inventor-to-be. When the inventor-to-be makes the written agreement, the written agreement–as an assignment of duties and rights–becomes part of the funding agreement, and the inventor-to-be becomes a party to the funding agreement–delegated by the university in the same manner as a subcontractor–and hence the inventor-to-be, too, becomes a contractor.

And if an inventor is a contractor, then when the inventor invents and federal common law provides that the inventor owns the invention, well then a contractor owns the invention and the invention is immediately a subject invention. And once the invention is a subject invention, then the inventor’s obligations under the (f)(2) written agreement kick in, the invention must be disclosed and the inventor has standing to establish the government’s rights. So far, so good, albeit only clever good.

The problem is, no university (at least I have not found one yet) complies with the (f)(2) written agreement requirement, and federal agencies do not appear interested in enforcing the requirement. Instead both the NIH and NSF advise universities to follow the first path, and make inventors assign inventions to the university. This, the federal agencies argue, is necessary for the universities to “comply” with Bayh-Dole. But it is nonsense–other than that it is the federal government telling university administrators how to ignore the implementation of Bayh-Dole but makes the law work anyway (sort of) to implement  Bayh-Dole without following the regulatory requirements of the law.

Even the (f)(2) written agreement requirement comes from outside Bayh-Dole. Bayh-Dole authorizes standard patent rights clauses, and at 202(c) states a number of provisions that must be in a standard clause, but (f)(2) is not one of those provisions. (f)(2) came from somewhere else. From Mars, maybe, or the planet Clarion in fulfillment of prophecy. But (f)(2) has just enough standing to require universities to make inventors into contractors for three tasks only–reporting subject inventions, assisting in patent applications on subject inventions, and establishing the government’s interest in subject inventions. And that makes inventors into contractors under the definitions in Bayh-Dole. Every invention they make with federal support then is a subject invention, and the whole lumbering beast called Bayh-Dole coughs and sputters into effect.

Bayh-Dole requires that patents on subject inventions  involve special handling. Such patents are not ordinary patents. Such patents do not have the same scope of rights of ordinary patents. Such patents cannot be assigned (if the contractor is a nonprofit) or licensed in certain ways. The income from such patents can be used only in specified ways (if the contractor is a nonprofit). But no where in Bayh-Dole is a contractor required to acquire rights, and no where in Bayh-Dole is an inventor required to assign rights. There is no authority in Bayh-Dole for any implementing regulation to require inventors to assign rights or to require contractors to acquire rights. If the executive branch creates such a requirement, it violates Bayh-Dole. The point of Bayh-Dole is that Congress controls, through patent law, executive branch research contracting for inventions. If the executive branch goes off making regulations on its own, then it is acting outside of Bayh-Dole and we are back to the old days, except now without a working idea of what constitutes the public interest–just patent monopolies based on public funding but exploited without public protections.

If Bayh-Dole were implemented and enforced as it is written, inventors using federal funds would be free to do what they will with those inventions–ordinary inventions, to be published, used, assigned, licensed, litigated, or ignored, just as the patent system provides. We might argue, then, that the patent system is in the public interest, but insisting on the bureaucratic use of the patent system is not necessarily in the public interest. That would be the fundamental lesson of Bayh-Dole: that federal funding does not cause an inventor to give up rights in inventions, not to federal bureaucrats, not to university bureaucrats, not to corporate bureaucrats. (Note, not bureaucats). That would be an interesting, if not inspiring, lesson. Too bad it’s not.

Instead, from outside the law we find interpretations, implementations, and especially a lack of compliance and disinterest in enforcement that turns Bayh-Dole into a law for middlemen, for preserving monopolies, for claiming that public benefit from federally supported research arises from monopoly pricing, monopoly litigation, and monopoly keeping everything else from public use unless someone pays.

Bayh-Dole is set up to be weak in this way. The law stipulates the requirements in federal contracts regarding inventions, but does not require the federal government to enforce any of these requirements. Title can be waived, even if the contractor messes up. Assignments can be tolerated, even if the nonprofit contractor does not get agency approval. The US manufacturing requirement can be waived. The small business preference can be waived or ignored. March-in doesn’t have to happen. Nothing material with regard to contractor behavior in Bayh-Dole has to be enforced.

It’s even worse for the general public. As the Supreme Court in Stanford v Roche pointed out, there are no protections whatsoever for those not involved in the research contracting (citations removed, my emphasis):

As Roche correctly notes, however, “the Act contains not a single procedural protection for third parties that have neither sought nor received federal funds,” such as cooperating private research institutions. Nor does the Bayh-Dole Act allow inventors employed by federal contractors to contest their employer’s claim to a subject invention. The Act, for example, does not expressly permit an interested third party or an inventor to challenge a claim that a particular invention was supported by federal funding.

In a world in which there is frequent collaboration between private entities, inventors, and federal contractors . . .that absence would be deeply troubling.

But the lack of procedures protecting inventor and third-party rights makes perfect sense if the Act applies only when a federal contractor has already acquired title to an inventor’s interest. In that case, there is no need to protect inventor or third-party rights, because the only rights at issue are those of the contractor and the Government.

In other words, the Supreme Court ruled that Bayh-Dole has no hole. Its scope and purpose is to deal with inventions made with federal support once a contractor has acquired rights. If it were otherwise, the lack of protection for inventors “would be deeply troubling.” So we should be troubled. The law has enabled a predatory environment, so that inventors, companies, and the general public all lack any standing to prevent the monopolization of inventions made with federal support. It’s not that the patent system itself is the problem–though there are plenty of issues there. If inventors owned their inventions undisturbed, at least we would be dealing with the patent system in its ordinary way, and would see the effects of inventors’ own judgment (and character, and luck).

And if we had back the Kennedy/Nixon executive branch patent policy, at least university inventors would have a choice between assigning to the federal government (and have access to their inventions via government dedication or non-exclusive release) and to a patent management organization (subject to agreement with the inventor over how to manage), where the federal government could step in easily if the patent management organization screwed up by not developing an invention to the point of practical application or not breaking up the monopoly represented by a patent to create competition and provide access to the benefits of the invention on reasonable terms. There were problems with the Kennedy/Nixon patent policy, too. It wasn’t perfect, but at least it was flexible, and the public interest took precedence over “middlemen” dealing monopolies.

Instead, we have got a version of Bayh-Dole that has a different sort of hole–implemented without protections for the public, as executive branch contracting that the executive branch has incentives not to enforce (such as express waiver or approval provisions in the law, and by convoluted implementing procedures). The problem with the Bayh-Dole we have–not the law, but the implementation that has exploited the law–is that the law doesn’t defend the public interest, whether that public interest would be expressed through the patent system or through federal policy regarding subvention funding in the public interest or through requirements that the federal government enforce the policy stated by the law and the requirements on the middlemen that gain control over monopolies on federally supported inventions.

Maybe there is some pragmatist justification–that university administrators should be above the law, should be able to make the law do whatever they want because they, apparently by definition, work only in the public interest–it’s just that the public is given no means of expressing any opinion on the matter, let alone enjoying any procedural protections. And worse, there’s no evidence that Bayh-Dole is that wild success claimed for it. If anything, Bayh-Dole has stifled the use of inventions and the very few inventions that have made it into use have done so largely despite university involvement, despite Bayh-Dole.

This, then, is the hole: that Bayh-Dole is written without procedural protections for anyone except patent brokers and their monopoly partners. Everything else is WTF. Bayh-Dole’s hole, as it were, is the absence of procedural protections for inventors, for companies, for the public. Freedom for patent brokers to withhold inventions from use, to sue companies caught using, to create monopolies and benefit from monopoly pricing. It is through this hole that university administrators suck inventions made with federal support and create monopolies that then prevent public use of these inventions in the hope that some few patents will be licensed for huge amounts of money–amounts made even greater by the prospect of preserving the monopoly for the entire term of the patent without competition.

The Supreme Court ruled that Bayh-Dole did not offer procedural protections to inventors and others because it concerned only the disposition of inventions as between a contractor and the federal government after a contractor had acquired rights. But for that ruling to have effect, contractors cannot place claims on inventors as a condition of the federal funding. Otherwise, it’s deeply troubling that there are no protections for inventors.

The hole, then, in Bayh-Dole is in fact the freedom of the inventor to decide what to do with inventions made with federal support. It’s not a hole at all. It is, in fact, a feature, but one patent brokers have worked diligently to close. The defect in Bayh-Dole is that the law permits the patent brokers to prey on inventors and to exploit monopolies at the expense of the public. Bayh-Dole has hand-waving about protections, but the law does not require the executive branch to enforce those protections. That’s the hole that needs fixing. Enforce the law–especially the statement of policy at 35 USC 200. Enforce the standard patent rights clause, especially the written agreement requirement at (f)(2), the requirements for nonprofits regarding assignment at (k)(1) and the requirements for nonprofits regarding the use of income at (k)(2) and (k)(3).  The result will be to make clear that Bayh-Dole does have a hole, and that it’s intended: inventors don’t have to assign any invention made with federal support, but if they do assign to a contractor, then they create a subject invention, and Bayh-Dole applies. If they assign otherwise, or don’t assign at all, then they never create a subject invention and Bayh-Dole and its standard patent rights clause does not apply. Even if universities require inventors make the written agreement in compliance with (f)(2), and so all inventions made with federal support become subject inventions immediately, there’s still nothing in Bayh-Dole under which inventors promise to assign inventions to a contractor or are required by law to assign.


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