You can do anything you want at Bayh-Dole’s restaurant, 2

We are dealing with the practice reality that under the Bayh-Dole Act: anyone can do pretty much anything they want. Federal agencies can do anything they want, provided they follow Bayh-Dole’s procedures (or work around them). Contractors can do antyhing they want (other than thumbing their noses at the most obvious paperwork requirements). And inventors can do anything they want (if they are constrained, it is the result of demands placed on them by contractors and federal agencies, outside of Bayh-Dole).

There are two reasons for this great freedom. First, Bayh-Dole does not dictate specific forms of practice. Bayh-Dole does not dictate invention ownership. Does not require licensing or commercialization. Does not require exclusive licensing. Does not require contractors to try to make money. Bayh-Dole establishes a default for patent rights clauses included in federal funding agreements for research or development work, but federal agencies may change that default or decline to enforce it or decline to act on the rights provided to them by the default. Bayh-Dole establishes that federal agencies may grant exclusive licenses–even to the point of assigning the invention involved–but does not require federal agencies to seek patents or to license inventions or license exclusively.

Nothing substantive in Bayh-Dole or in the patent rights clauses is specifies is enforced, acted upon, even attended to. Paper requirements–yes. Invention disclosures. Reports of inventions at grant close out. Government rights notice in issued patents on subject inventions. Paper version of government license. Reports of utilization. None of this has a speck to do with promoting the use of subject inventions. It’s waste energy. The government does nothing with this information. Bayh-Dole makes reports a government secret. Federal agencies decline to march-in even when reports show no timely practical application. Federal agencies decline to act on their government license to make, use, and sell and authorize others to do so for any governmental purpose. The entire public interest apparatus that rides on top of Bayh-Dole does not operate.

There’s no right of public appeal, no public oversight, no enforcement mechanism to enable any substantive element of Bayh-Dole or the patent rights clauses. No audit, no authority independent of the federal agencies offering out public funds for research or development to enforce the law and the patent rights clauses, no penalties for failure to comply with substantive requirements–even while a defect in paperwork requirements can lead to a contractor’s loss of invention ownership. And in practice, federal agencies don’t take ownership of inventions despite widespread failure to properly document and disclose subject inventions, especially those made by assignees of subject inventions via exclusive license–often for-profit companies (who become federal contractors when they take assignment).

Anything substantive in Bayh-Dole–having to do with any of the stated policy and objectives of the law–can be waived or ignored or conveniently misrepresented, and whatever else that’s required is drafted to be narrow, weak, ambiguous, or unworkable. If one set out to create a honey pot for private exploitation of publicly supported work using patent monopoly positions without public oversight, accountability, and right of appeal, then Bayh-Dole would be your inspired law.

On the one hand, Bayh-Dole is a mechanism for contractor and federal agency freedom. Under the pretense of a public interest apparatus, Bayh-Dole allows contractors and federal agencies to do whatever they want. We might think, then, that nonprofit contractors would act in the public interest, since Bayh-Dole does not require otherwise, and in so acting, would adopt a default of publication, public licenses–even royalty-free–and pursuit of cumulative technology platforms, standards, and broad public access, reserving exclusive licenses for narrow objectives, such as sale of a particular item that requires substantial post-invention development work and for which no coalition of partners has formed to take on that work.

The situation is much like publishing, where a nonprofit might grant a publisher an exclusive right to a particular form of a manuscript or media work–with plenty of editing and layout, permissions obtained for included content, indexed and blurbified. But that exclusive right does not need to require the nonprofit to give up as well the right to permit non-published uses of the material (including copying and open distribution of copies), or even to permit other publishers to make their own investments in the core manuscript or media materials and make their own editions. If open distribution of copies ruins the opportunity for a professionally edited edition with professional layout and blurbification, then there’s no real reason to bother with all that professional overhead–which just adds costs, delays access, and restricts rapid updating and correcting of the material. If the only way to justify the professional edition is to ban all non-published uses, and to ban others from making their own professional editions, then there’s something deeply wrong with the deal to start with. Such a deal runs against access and use, all the while claiming that without professional editing and layout and blurbification, the core material is not usable by the public. Just cannot be. If the public cannot use the material without editing and the like, then a publisher has nothing to fear from open access to that material. And if they can use the material without editing and the like, then there you are–there’s no obvious need for a commercial version of the material, at least not blurbified.

Copyright infringement requires access to the work and substantial similarity to that work. Patent infringement, by contrast, has no access standard–any practice of the claimed invention may be excluded, even if that practice has arisen without any knowledge of or reference to the patented invention. Further, unlike copyright, there’s no “fair use” or “research exception” in patent law. A patent holder can forbid even research uses. It’s one thing to fuss about “free riders” who copy an invention once they see it. It’s another to go after the folks who are clever enough to figure the same thing out independently (meaning, perhaps, that the thing itself wasn’t all that clever anyway)–they are not free-riding, as it were–they are just working the same area, and making the same observations, coming up with similar practices. In a distributed research environment such as the one created by multiple federal agencies operating in overlapping areas, each spreading work around to multiple contractors, any contractor obtaining a patent can disrupt the research activities of other contractors.

Sure, there’s the argument that those other contractors can rely on a government license or at least on the government to cover the costs of the contractor’s infringement, but those other contractors have no technology transfer pathway with regard to their work. That pathway is foreclosed by the patent holder–and even if that patent holder provides a non-exclusive license, if that license demands royalties or a license back of “improvements,” then the other contractors’ technology transfer pathways are burdened with expenses that the first contractor does not have. Now do the integration of this situation across all contractors to all federal agencies in some area of research, and everyone’s technology transfer pathways are foreclosed or burdened with demands made by the patent positions on claimed inventions of all the others. Everything might be delivered to the government, but the whole point of Bayh-Dole, one would think, would be to enable each contractor to have a meaningful prospect to use the patent system to promote the utilization of the inventions acquired by the contractor and made under federal contract.

But if every contractor adopts even a royalty-bearing non-exclusive license, and new products require multiple such licenses, then the royalty burden on any given product runs up the costs to the public with a royalty stack or delays the introduction of such product as those royalty stacks must be negotiated down to a single tolerable royalty. No, ten or thirty nonprofits don’t each get their 2% royalty on sales–which might add 20% to 60% to the cost of the product. Instead, they will share 2%. And you’ve never seen a food fight until you see even three nonprofits fussing over their share of some tiny amount of future money. In the unrealistic, counter-factual, and therefore seductive scenario in which a nonprofit acquires an invention made without any practice context–entirely new with no others working in the same area, no industry involved, and the like–contractor freedom to do anything sounds really good. Add in context and multiply by all the contractors involved, and one gets gridlock, gets contractors disrupting other contractors, gets patent trolls, gets delays, gets nonprofit money interests interfering with not only research progress but with collaboration, pre-commercial development of technology platforms, patient care, and technology transfer pathways of other research organizations.

Consider, then, the effect of a huge amount of federal funding directed at some new, emerging area of technology. Those involved in that area–maybe a small company here and a university lab over there–get swamped out as the federal money flows in and nonprofits take out patent positions on every possible improvement, variation, application, method of production, system of use, article of manufacture that they can justify. Not only does the Bayh-Dole approach permit contractors to disrupt the technology transfer pathways of other federal contractors but also it serves to swamp out if not destroy the development pathways of folks not dealing in federal money–of entrepreneurs, private investors, nonprofit foundations, independent research labs. Without such influx of federally funded work to any area that suddenly becomes trendy, those involved in the area might have the room to develop their work, establish core patent positions (if that’s what they want) and make a go of commercial development (or lead standards development, or publish openly–whatever makes them tick). But with the non-profit gold rush mentality in which every nonprofit stakes out hundreds of additional claims surrounding and building on everything done independently, federal funding effective kills not only the public domain but private technology development in any area that becomes trendy for federal funding.

We saw something akin to this when the Gates Foundation, loaded with cash, moved into areas that were being funded by nonprofits with much less means. The nonprofits’ work was swamped out. It was as if it didn’t matter. Whatever would happen next would be decided by Gates Foundation funding. Nonprofits begged the Gates Foundation to lay out a road map to its funding so the smaller foundations could *avoid* funding in areas that the Gates Foundation staked out. That’s right–the big money chases out all sorts of other support. If no matter what you fund, some version of the Gates Foundation or a federal agency shows up with ten or a hundred times more money, spreads it around to ten or a hundred researchers, then everything you have worked for gets co-opted, overridden, forked, disrupted, burdened, swamped out in the published literature, screwed over. It’s like a beautiful park getting overrun by a ten-year-long free beer festival. It’s great for the beer drinkers but it’s no longer a commons serving multiple purposes.

It’s just weird then. Bayh-Dole sets out a complicated but useless and ineffectual apparatus of public protections–a sort of working requirement that doesn’t either work or require. But this very freedom that lets contractors and federal agencies do whatever they want does not in practice work. The contractors acquire inventions, take out patent positions, and hold out for exclusive licenses (and assignments for commercial exploitation) as the defaults. Federal agencies ignore the public protection apparatus that would limit the use and scope and term of exclusive licenses, and federal agencies also adopt the same default of exclusive licensing and don’t police themselves for ignoring the public protection apparatus that all but prohibits default exclusive licensing positions. It’s that “all but” that ends up meaning that federal agencies hold out for exclusive licenses rather than default to royalty-free non-exclusive public licenses and only move on to exclusive when no one bothers to take the public licenses, or if they do, they don’t do anything with them.

Under Bayh-Dole, federal contractors and federal agencies can do anything they want. But they don’t then do a whole lot of different things. They do pretty much all of them the same thing–hold out for exclusive licenses or don’t license at all and end up trolling industry if they bother with old patent positions at all.

When I was working on my Kauffman Foundation grant, I once tried to get the University of Washington to adopt a remedial practice for ageing unlicensed patent positions. There were a set of patents on implicit solid modeling techniques that had sat unlicensed for over a decade. I did a patent dependency search and saw that these patents were cited by a who’s who list of industry tech companies. Rather than sending out nastygram notices to these companies that they should consider taking a royalty-bearing license (i.e., the first step in a patent troll play), or worse, attempting to engage one company to take an exclusive license to the ISM patents–and then let that company go after the others (a different first step in a patent troll play), I suggested that the University send to each company that had cited the UW’s ISM patents a simple (one page) paid-up, royalty-free license with an offer for the UW lab to provide assistance (and codes) if a company wanted to learn more or collaborate.

Such a move would reduce any implied threat that the UW would “enforce” its patents against the company (and since the company’s patents cited the UW patents, the company was exposed to a claim of willful infringement and treble damages and attorney’s fees). That is, even where a company was most vulnerable, the university move would be to remove that vulnerability up front rather than exploit it, even in negotiations. The move also would reduce the incentive–indeed, a default commitment–to design around the ISM patents, to make the work obsolete, to lock out pathways of improvement. With a license to use these core ISM patents spread to all the major players, the university had the greatest chance of promoting use of the inventions. The university also improved the changes of bringing industry collaboration to its ISM lab work. The university also improved the chances to spring a modest industry consortium built around services flowing alongside a license-tech transfer services–rather than a consortium built around research services (the conventional form of consortium, but way upstream from the condition of the ISM work at the university’s lab).

The university would have none of this, of course. There wasn’t even much discussion–the idea was dismissed out of hand–even when the ISM lab was ready and willing to go with it. Instead, the patents (as far as I know) went unlicensed until they expired, thoroughly wasted assets.

The practice reality is that although federal agencies and universities can do anything they want under Bayh-Dole, they don’t. They have all–with a few wonderful rogue exceptions–adopted a default exclusive license or nothing approach. In this sense, then, Bayh-Dole legitimizes under a pretense of freedom and public purpose a uniform practice of patent exploitation based in exclusivity. That uniform practice in turn defies federal agency and nonprofit statements of public mission, of academic freedom, of the importance of freedom of publication and research and insists that a “new, entrepreneurial” culture must displace the old fuddy-duddy idea of public mission and the free play of free intellects. What’s meant, however, is a new culture–both federal and nonprofit–dictated by dealing in patent monopolies for the sake of dealing in patent monopolies, as if bureaucratic dealing in patent monopolies is what the public craves. The result has been a continuous, widespread destruction of not only a research public domain but also of private initiative to create anything in an area that might become trendy for federal funding. These are absences. One does not easily obtain metrics for what hasn’t happened.

Thus, in its way, Bayh-Dole allows a diversity of practices but in practice everyone adopts a single practice of default exclusive licensing. By any standard, Bayh-Dole patent practice has been woefully ineffective when not outright damaging.

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