The Dole-Bayh Act: Patentability

We are looking at a draft of a bill proposed by Senator Dole, with Bayh along for the ride. It–and other drafts, such as the Thornton bill (HR 8596), and the Institutional Patent Agreement master, and the Federal Procurement Regulation codification of the Nixon version of executive branch patent policy–eventually became Bayh-Dole. So we will call this draft bill the Dole-Bayh Act, though it never was.

The draft of Dole-Bayh starts with a premise that’s not in Bayh-Dole: “a consistent policy and procedure concerning patentability of inventions made with federal assistance.”

Oh wow. This is not just an idle reference. Here’s the title of the proposed new chapter of federal patent law:

The foundation of the bill is how inventions made with federal assistance are patentable. If an invention is not patentable, then it falls outside federal patent law. The implication is that if an invention made with federal assistance is not within the scope of this bill, then perhaps it is not patentable. That’s interesting.

We might push this idea all the way back to the 1947 Attorney General’s report, which argued that the federal government should own inventions made with federal assistance and should not enforce patent rights in those inventions, even if, perhaps, it could. There was some discussion, as well, whether absent an act of Congress, the government could grant an exclusive license to any patent it held (i.e., issued to itself). Bayh-Dole aims to fix that by asserting that federal agencies may grant exclusive licenses (citing conditions that aren’t enforced), and that the terms of those licenses should be kept secret. Perhaps it is still an open question whether Bayh-Dole at this point is Constitutional. Congress has the power to reserve exclusive rights for inventors. But does Congress have the right as well to force those inventors to assign those reserved exclusive rights back to the government? There was an argument that such rights were “exhausted” when so assigned, but the Attorney General’s report scoffed at that argument–but didn’t provide a substantive argument against it. One might vary the argument by agreeing that yes, the federal government might hold a patent, but that the attributes of that patent, and in particular the remedies for infringement, are not those available to individual inventors.

Before you scoff at this variation, it might be worth pointing out that Bayh-Dole uses something of the same approach. Inventions made with federal assistance are not ordinary inventions, and patents on these inventions are subject to un-ordinary requirements. The attributes of these patents are not those of ordinary patents. That goes, too, for inventions held by the federal government. What’s left out is whether the federal government has any right of enforcement in the patents it holds. While Bayh-Dole lists just about everything else a federal agency might do with a patent, it does not include suing the socks off citizens and companies for infringement–not to suppress use, not to demand compensation for the damaged suffered by the government when a citizen or company uses an invention made with federal assistance. In dealing with money made from licensing, there’s no mention either of awards of damages from infringement or settlements of infringement litigation. Just a vague reference to “receive funds from fees, royalties, or other management of federally owned inventions authorized under this Chapter.” Just that suing from infringement is not authorized under Chapter 18. Funny that.

The Attorney General’s report did not consider whether government-owned inventions were patentable, but if the government was not going to enforce patent rights by suing citizens and companies that used those inventions, then what is a patent but a publication with lots of technical fussiness? The Supreme Court in Dubilier was even more strident about it–tossing out the idea of the “dead hand” of government taking patents that, if left under private control, might be used for the things a patent gets used for–excluding others, providing incentives to invest, exploiting potential markets, licensing uses, selling out.

If we pull these ideas together, we see a strategy–if an invention is made with federal assistance, then it isn’t patentable unless it comes within the requirements of this new Chapter 18. There are still the two prongs–subject inventor and government–but if someone plays outside the new law, they don’t have patent rights. It’s not that the federal government has the right to take away patent rights–it’s that there are no patent rights for the government to grant in the first place. You have to have property before the government must follow eminent domain procedures to take it. In the case of real property, the government does not charter the property in the first place. In the case of patents, however, the government creates the property–and decides what property may be so created, and decides who will own that property. Thus, it’s interesting, at least–the Dole bill suggests that inventions made with federal assistance might not be patentable, but for the proposed provisions, should those provisions be made law.

There’s not much else in the Dole draft that develops this idea, but it’s there at the start, and there in the proposed title to Chapter 18. Instead, Bayh-Dole carries the title “Patent Rights in Inventions Made With Federal Assistance.” The Thornton bill, by contrast, uses a rather clearer title (Title III, as it was proposed): “Allocation of Property Rights in Inventions Resulting from Federally Sponsored Research and Development.”

It would appear that roots of Bayh-Dole lie in the idea that Congress might withhold the grant of exclusive rights to inventors supported by the federal government, but for their agreement to practice the invention according to special rules–a working requirement, for instance, and reasonable pricing. In such an approach, march-in is not so much the taking for public use rights held by citizens (er, persons) but rather is the refusal to extend rights beyond the attributes of the property right that has been created by the government. Unused or misused inventions aren’t patentable. Bayh-Dole, however, backs away from this idea and leaves it at “patent rights”–rather than “patentability.”

If we dropped the idea that federal agencies should deal in exclusive rights, should be in a position to threaten citizens and companies with the full force of the federal government to sue for infringement, then the patentability idea becomes viable again. If the federal government took ownership of inventions–and even applied to issue patents to itself–but only to insert inventions into the federal patent system as publications and to make clear their position as providing prior art to expand the public domain, then the only issue would be whether inventions made with federal assistance were patentable when held by an inventor or assignee of an inventor. The government “patent” would be a face only, and otherwise would be empty of exclusive rights. The “subject inventor” patent would be patentable, too, but it would be a different sort of patent, with a working requirement and a requirement for reasonable pricing and availability–stuff not otherwise in federal patent law. With these requirements, too, would come questions of just how such a patent might be enforced–remedies for infringement would be conditioned on compliance with these requirements. March-in would in such a case be more along the lines of precluding infringement litigation that attempted to extend a patentee’s Bayh-Dole-limited patent property right as if the invention were an ordinary invention.

If we took this approach, then inventions made with federal assistance would not be patentable except if assigned to a party to the funding agreement. Inventors could own inventions all they wanted, but those inventions would not be patentable. If an inventor wanted the benefit of exclusive rights, the inventor would have to find a party to the funding agreement and work a deal. That party could not be the federal agency–if if the federal agency could patent the invention–because the federal agency would not have standing to enforce that exclusivity. In this approach, then, inventors would have to seek out a non-governmental party to the funding agreement. Or, of course, a party to the funding agreement could make the inventor a party to the funding agreement and then the inventor would be set–depending on the deal that the party requires to bring the inventor to the party, too, as it were.

This approach is pretty much what inventors faced before universities got the idea to try to make invention management into a bureaucratic system. If an inventor wanted to patent an invention, he or she needed to persuade someone at Research Corporation or another of the patent development firms. It wasn’t quite a matter of patentability–rather it was a matter of the viability of patenting. But it amounted to the same thing. Sure, the inventor might put up his or her own money to file a patent application, but university faculty did not in general look fondly on patenting and keeping an academic position.

And oddly, some part of this approach is also buried in Bayh-Dole’s standard patent rights clause, despite NIST having no clue about it. The standard patent rights clause requires organizations to make their inventors parties to the funding agreement. In Bayh-Dole, parties get their own patent rights clause based on their status. Small businesses get one version. Nonprofits get another. And inventors get a third (at 37 CFR 401.9). When inventors become parties to the funding agreement, they operate under their own patent rights clause, not under the nonprofit (say) patent rights clause of organization hosting the federally supported work. And that inventor patent rights clause says that inventors should be treated as small businesses, but without the obligation to file a patent application. Thus, an inventor may choose to dedicate an invention to the public domain rather than start the patenting process, under which, if the inventor does not complete it, the federal government may take ownership of the invention–and to do what? exclude the inventor from practicing the invention in favor of a favorite company or venture speculator? That hardly appears to be sound public policy directed at promoting use of the invention.

Given that investigators beg the government to support their work–submitting competing proposals through their organizations–it’s not at all unreasonable that investigators agree that their inventions aren’t going to be patentable, unless they agree to working the invention and providing for reasonable availability and pricing–and requiring any other party to the funding agreement that acquires ownership of such inventions to hold to those same requirements.

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