Bayh-Dole the Monster

The Bayh-Dole Act makes a great deal about public interest. Throughout the law are gestures toward worthy objectives–use of inventions, manufacturing in the United States, government licenses, and the right of federal agencies to step if they need to. But all of this is for show. None of it operates. Bayh-Dole can be reduced to two propositions:

  1. Federal agencies must use by default a standard patent rights clause that forbids them from having an ownership interest in inventions made with federal support unless the contractor fails to obtain an ownership interest or subsequently waives that interest or fails to maintain that interest.
  2. Federal agencies, when they acquire inventions, may license those inventions exclusively.

Everything else in the law is for show and otherwise inoperative, ineffective, unimplemented, unpracticed, prohibitively difficult, or designed not to work. Utter waste. Think about it.

How does putting a government rights legend in a patent application promote innovation? Does anyone go, “Hey, this patent has federal funding–that’s awesome, let’s use this invention!”? No, of course not. What about notifying the government about electing to retain ownership? Why? What does it matter? Why not just use the normal investigator grant reporting pathway–the investigator reports the invention to the funding agency as a matter of course. If the inventor publishes and blows rights, so what? The federal government has never prosecuted a university investigator for publishing an invention and therefore blowing patent rights. Special notice to the government doesn’t matter. The government never practices under its licenses. There’s no point in requesting reports on use. The government never acts on non-use, never audits reports for truth, and has to keep reports a state secret. Nothing here that benefits innovation. We could go through a long list this way. Bayh-Dole is almost but not quite entirely a total waste of effort when it comes to anything having to do with innovation, national competitiveness, or public benefit.

The public interest elements of Bayh-Dole are for show. Their role is to cover for what the law does–prevent federal agencies from requiring inventions as deliverables in federally supported research conducted by universities, other nonprofits, and small businesses. The Supreme Court in Stanford v Roche commented on the lack of protections for inventors and third parties alike in Bayh-Dole. If Bayh-Dole did have public interest elements that operated, one would expect such protections. But no, where it counts, Bayh-Dole is silent or disconnects the public interest from its actual requirements or makes the public interest so difficult that something else happens instead–and what does happen is, unless challenged, what the law intends to happen.

The effect of Bayh-Dole is to institutionalize control of inventions made with federal support. This works to circumvent the U.S. patent system, which is based on the federal government having the right to reserve to inventors for limited times the exclusive rights to their inventions. Those rights are reflected in the grant of a federal patent, which is based on the merit of the invention alone–not on the government’s playing favorites, not on the government’s interest in profiting from the patent. In contracting situations, the commissioning party is free to negotiate for patentable inventions as deliverables. An inventor may agree to give up ownership of such inventions, then, as a condition of participating in the commissioned work. That’s what the basic executive branch patent policy expected for university inventors. When the government obtained ownership of such inventions, that same executive branch patent policy expected federal agencies to dedicate the invention to the public or if the federal agency obtained a patent to license the invention non-exclusively and royalty-free to all qualified persons.

Think about this government policy for a minute. If the government acquires an invention and then obtains a patent, it is in effect issuing a patent to itself. If the federal patent system is set up to preclude the government from having a profit interest in patents, and to preclude the government from granting patent franchises based on favoritism–both activities subject to past abuse–then when the government issues a patent to itself, these same prohibitions should operate. If a federal agency grants an exclusive license to all the substantial rights in an invention, the agency in effect grants the patent to another person–not the inventor. If a federal agency charges a royalty for this grant, the government has a profit motive tied to the monopoly exploitation of the invention by the new owner. In effect, the federal agency runs a shadow patent office that deals in patents on just those patterns of dealing that the official federal patent system has been designed to prevent.

Bayh-Dole, then, enables an end-run around the federal patent system while making a show of using the federal patent system to promote the use of inventions. Bayh-Dole is the anti-inventor, anti-federal patent system Act.

Let’s beat on it. Bayh-Dole sets up the proposition that the proper owner of an invention made with federal funds is not the inventor but the university that receives federal funds on behalf of the inventor (and other investigators). The university does not propose the work–the investigators do. The university does not put money into the work–the federal government pays the direct expenses for the research plus pays the university for the facilities and administration for the work. The university, for purposes of federal grant funding, is a host, not an employer. The university does not assign the work, does not direct the work, has no right to control the work, and the work is not done by investigators for the purpose of benefiting the university. The investigators have no intention to become servants of the university, taking direction from university administrators, with regard to their federally supported research.

Despite all this, however, university administrators insist that the intent of Bayh-Dole is to make universities the owners of inventions made by investigators supported with federal funds. The university has no merit in the matter–it did not invent, it does not employ, it is fully compensated for its involvement, it does not have a basis to own inventions. Except, of course, that administrators have represented that Bayh-Dole requires them to own, or sets it out as a virtue for them to own, or gives them the right to own. All of this is untrue, but no matter–that is how Bayh-Dole has been used, and unchallenged, that is Bayh-Dole’s purpose and effect.

And Bayh-Dole is placed, of all things, in federal patent law. There it operates like a tumor. If the federal government does not want inventors to have a right to their inventions (normal federal patent law), the government can provide federal funding and thus change the default in federal patent law, so that now an organization that merely hosts federally supported research–handles the money designated for investigators–has the right to own any inventions arising in that work. How utterly strange! The Supreme Court in Stanford v Roche was having none of this with regard to the idea that Bayh-Dole outright “vested” ownership of these inventions in the host universities.

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

But this is exactly what the authors of Bayh-Dole intended to do. That’s why Bayh-Dole is in federal patent law. That’s why there such ambiguity around the use of “elect to retain title” and even slipping in an “elect title.” The point of Bayh-Dole was to make a sea change in federal patent law without announcing it. The Supreme Court looked only at ownership. They were not asked to review as well whether patent law had also been changed to allow the federal government to re-issue patents but now not to the inventors but to whomever was the government’s favorite, and with a profit motive shared with the interests of the new owner of the patent monopoly. The government, in such cases, is reserving for limited times rights in inventions not to the inventors, but having set up the law to permit the government to strip inventors of the rights granted to them by official federal patent law, the government can then deal in the patent rights as it pleases. In the hands of the federal government, an exclusive license granting all substantial rights in an invention is not merely an assignment of ownership–it is a re-issuing of the patent, but now to a favorite and with a profit motive, rather than to the inventor.

Bayh-Dole is a shadow patent law for federally supported inventions. In the Bayh-Dole version of patent law, inventions are to be owned by institutions or the federal government, and the purpose of that ownership is to trade in monopolies on those inventions, choosing favorites and sharing in the profits those monopolies might bring.

This is the public purpose of Bayh-Dole. The argument is that institutional monopolies on inventions (Bayh-Dole) are superior to inventor-managed monopolies on inventions (the official patent system); that institutions should decide when to pursue patent monopolies (Bayh-Dole) rather than inventors (the official patent system); that institutions that merely host work should have a priority right to own any inventions arising in that work (Bayh-Dole) rather than either the inventors (the official system) or the federal government (via normal contracting requirements for deliverables).

The argument behind Bayh-Dole is that innovation best takes place when institutions create monopolies on inventions and exclude all use of those inventions pending a revenue sharing agreement with a speculative favorite willing to exploit such a monopoly on the premise of creating a commercial product.

That’s a premise worth discussing. While it may be that there are some instances in which a monopoly might be necessary to justify private investment to create a commercial product–it’s possible, let’s say, in the abstract–in general this is nonsense. Many research inventions are tools developed in research for research. They do not need to take the form of a commercial product. Similarly for processes that may be adopted by industry directly: companies do not have buy a commercial product version, and do not have to sit on their hands waiting for a bureaucratically clumsy university to get around to picking a speculatively motivated monopolist to finance the creation of a commercial version. And finally, any amount of development cost that a private company acting alone could undertake, so also could the federal government. Furthermore, maintaining a patent monopoly all but requires a single company to provide the “risk capital.” But in open innovation environments, ones in which patent monopolies are broken up into the privileges of a commons or standard or non-discriminatory licensing scheme, the costs of development are distributed over all participants–some providing cash, others technology, others expertise, others test environments, others advocacy.
The determination to exploit the value of a patent as monopoly, and to do so in partnership with a single, favored company, and to do so on the premise that a commercial product must be made to the exclusion of all other work–that determination aims to destroy all other pathways by which research inventions might be used and developed. Bayh-Dole aims to crush government work to develop inventions to the point of practical application–something the government did well in service agencies such as the Department of Agriculture and even was doing well in public health, such as government projects targeting polio, leukemia, and malaria. Bayh-Dole aims to crush open innovation–allowing cumulative technology and networked, non-market development. This, too, is excluded by Bayh-Dole. Bayh-Dole, finally, aims to eradicate the inventor. Under prior federal contracting, inventors supported by federal funds delivered rights to their inventions to the federal government upon request. But the inventors always had access to their inventions because the federal government dedicated the inventions to the public or licensed non-exclusively, royalty free. Not under Bayh-Dole, which makes a virtue of dealing in monopolies. Under Bayh-Dole, inventors have no right to their work, no access to their work–unless institutions decide (whether with competence or otherwise) that their work is valueless as a monopoly right. It’s just that in the Bayh-Dole era universities virtually never waive ownership of patentable inventions made with federal support.

In all, Bayh-Dole is a monster. It undermines federal patent law with a shadow patent system that operates when federal funds support a patentable invention. That shadow patent system does the things that the federal patent system was set up to preclude–institutional control of inventions rather than inventor control; government issuing of patents based on favoritism and profit rather than the merit of the invention. And that shadow patent system was also set up to destroy federal government protections for inventors and the public in the special case where the government sponsors research as a form of public subvention–a donation for a public purpose–rather than as procurement for a programmatic agency purpose. In these situations, Bayh-Dole demands that such publicly spirited research must be treated as ordinary research to produce inventions for the financial benefit of institutions. Bayh-Dole, so this twisted logic asserts, demands that inventions become monopolies and that those monopolies are to be preserved in exclusive deals with favorites, for a share of the profits enhanced by monopoly. This, so goes Bayh-Dole, is in the public interest, this is proper federal contracting policy, this is proper patent law, the patent law as it should have been in the first place.

As a monster, Bayh-Dole attacks the patent system with its revisions. As a monster, Bayh-Dole attacks the “free play of free intellects” in university research. As a monster, Bayh-Dole attacks federal programs that would develop inventions to the point of practical application before release to commercial exploitation. As a monster, Bayh-Dole attacks open innovation, standards formation, industry collaboration, and competition based on the discoveries of federally supported university research.

There is nothing benign about Bayh-Dole. There are elements in Bayh-Dole that, were the operational, would change everything. But those elements are designed not to operate. They are there to make Bayh-Dole look like a cute little rabbit, when it is a dangerous killer. We may as well look at a bomb and go–but the parts can be assembled into a cute sculpture of a teddy bear, so it can’t be really dangerous.

Repeal Bayh-Dole the monster. Rip off its arm and send it back to its steaming lake to die. Reinstate the Kennedy/Nixon executive branch patent policy as a stop-gap. Then work to undo the damage in university patent policies and federal agency regulations.  Require, as a condition of federal grant funding, that a university can have no interest in inventions made with federal support. Suspend federal grants until a university submits a revised patent policy (and any ancillary contracting) that demonstrates compliance with this requirement. Reinstate default federal agency dedication or royalty-free non-exclusive licensing of inventions. Then, and only then, can we take on the special problem of medicinal chemistry created by an industry determined to deal in monopolies directed at matters of public health, and the psychological disorder of a federal agency–the NIH–that for nearly fifty years has made love to the monster.

This entry was posted in Bayh-Dole, Stanford v Roche. Bookmark the permalink.

Leave a Reply