The Faux Public Covenant in Bayh-Dole
The government license forms an essential part of the public covenant on inventions arising from research or development supported in part by government funding. For the other parts of the public covenant, the patent owner must take action–must work the invention, must promote free competition and enterprise, must manufacture in the United States for any exclusive use or sale in the United States. In each case, the patent owner must do something. For the government license part of the public covenant, the patent owner must grant the license to the government–but that part means nothing unless the government acts on the rights it has obtained. We explore this idea in two ways. First we look at the policy argument of Bayh-Dole for the importance of the private marketplace. Then we look at how this argument is political cover for providing patent monopolies in areas of public health (or suffering) to the pharmaceutical industry, with the claim that only through commercial products sold with monopoly protection and pricing to reward investors (or speculators) could the public benefit from publicly funded research.
Bayh-Dole’s statement of policy makes it clear that the law is intended to benefit the public. The benefits of the use of subject inventions must be made available to the public, and on reasonable terms. The government is provided with the right to march-in and defend the public from nonuse and unreasonable use (quite separate from any antitrust or patent misuse actions already available to the government). Advocates of Bayh-Dole argued that the law would help American regain its leadership in technological innovation, would provide the public with the benefits of inventions that otherwise would “sit on the shelf” in federal government warehouses, doing nothing.
The foundation of Bayh-Dole, then, is that the law serves the public interest, and that institutional owners of patents on subject inventions will provide better service to the public than will federal agencies (this foundation, of course, is empirically not true, but no matter). The argument for allowing institutional ownership of inventions made in research projects receiving federal support has been based, since Francis Biddle’s report in 1947, on the concept that patent rights in those inventions is necessarily restricted. One of the key restrictions is that the federal government has a license to inventions made in any research or development project that the federal government supports financially. So far, so good.
Now consider: the argument that the federal government has a license to all such inventions is made to assure the public that the government obtains something of value in return for permitting contractors to exploit patent monopolies for private gain. But if the government never acts on the rights it has obtained, it is clear that those rights have no value whatsoever in the transaction. They are not valuable consideration for permitting the exploitation of patent monopolies. They are something else altogether–they are for show. The government’s rights are presented as if the public should believe these rights somehow should matter to the public, and in some way justify allowing a private patent monopoly, because somehow that patent monopoly is limited and therefore acceptable.
You can see, however, that if the government does not act on its license, the gesture is empty. The government gains nothing but paperwork; the public gains nothing but bears the cost of paperwork. Now, of course, many inventions are worthless anyway, and historically about 5% of patented inventions see some commercial activity. We might, then, expect that the government should use its license in at least 5% of its cases–and that rate should be much higher wherever a commercial product has been developed, since in those cases it is established that the invention is likely worth something. You see, the government license is not merely to apply to those cases in which an invention is made but never reaches practical application–the whole point of the government license is to provide the government the right to make, use, and sell exactly those inventions that do achieve practical application. The government funds research or development in projects that intend to achieve practical application; the government receives a very broad license to the inventions made in those projects; when those projects achieve their goal, then the public market benefits (if the public receives benefits of use on reasonable terms) and the government market also benefits–but only if the government acts on its license.
What do I mean by the government market? Consider military research and development. The federal government supports the development of a new weapon, a kinetic weapon made from the warehouses full of old metal typewriters. It does not matter what part of the research and development the government supports, it has rights in any inventions made in the performance of the work to develop this kinetic weapon. Once the weapon has been developed–to be able to launch typewriters from low earth orbit at unsuspecting digital server farms–the government can build these weapons itself, can commission others to build them for the government, and can purchase these weapons from those that build them–even if there is also a private market for weaponized typewriters in space, some target fun for space tourists and all. That might seem to be a silly example.
Consider then nuclear energy, space technology. Same deal–even if there is a “dual use” for such technology, private patent rights affect only the public market for innovation, not the government market. The government and its contractors and these that the government aims to benefit all enjoy the benefits of the government’s license. The premise in these cases is, generally, that of procurement. The government, as a provider of support, has the right to specify the deliverables, and the price it demands is a broad non-exclusive license to make, use, and sell for government purposes.
But Bayh-Dole is not specific to government procurement contracts, where the Kennedy/Nixon executive branch patent policies were–they started with the idea of procurement contracts. The idea of government grants to advance science in the public interest was at the time a side-show, a little something pushed by Vannevar Bush in contrast to the vast government agency and laboratory missions to develop technology the agencies had set themselves to needing, whether as weapons or fertilizers or vaccines for soldiers. For Bush, frontier science would be handled by a specialized quasi-government organization, aloof from agency missions, pressing out into unknown areas, not under the management of what sounded pragmatic or even reasonable to institutionally bound agencies. Frontier science would be discovered by people aloof from established professional orders, not better managed and motivated by such professional orders. Frontier science would be in the public interest because the realm of science would be enlarged, providing new ideas and tools for anyone to use to imagine new technologies–the digital computer, the internet and web, gene splicing–drawning on these discoveries.
The federal government had no specific need to own such discoveries, other than to guard them against non-publication and non-availability to others. Publication and availability were the essence of the government’s non-procurement program of funding frontier science. The patent could be used to publish–publication is foundational to patenting–but also could be used to deny availability. The problem, then, in funding frontier science was to push availability for any purpose. As Bush argued, no one could predict the eventual use of such new science. The innovative use may well be a lateral use. Inventions in chemistry would matter in medicine. Inventions in mathematics may give rise to “information technology” (as we call it now). How would anyone know? Thus, Bush advocated that all the government needed, for such, research discoveries, was a non-exclusive license for government purposes–that is, the government ought to be the first entity that was granted access to whatever had been discovered. Others ought to also get that access, but the choice should be left to the inventors, following the free play of their free intellects. How those inventors allied themselves with their institutions, or with industry, or with the public was a matter of confidence in the future, not a uniform government policy imposed out of fear across all forms of granting and procurement contracting.
Bayh-Dole, however, conflates procurement contracts and public interest grants to researchers hosted by universities. This, apparently, is its great accomplishment. That since uniform imposition of government policy to own all inventions so made was (arguably) bad, a uniform imposition of government policy to allow institutions to take ownership of all such inventions was (they argued) so much better. The argument ignored the fact that the government policy permitted flexibility by situation–contractors in commercial business obtaining procurement contracts could keep their inventions if they wanted, so long as they used them in a timely fashion and didn’t interfere with government agency missions. But if the contractors were engaged to do something for a government agency’s public mission–whether building atomic power or space vehicles or improving public health and safety–then the inventions would be the public’s. For everything else, it was a matter of agency judgment whether to allow a contractor to play with patent monopolies, based on the circumstances.
The Kennedy/Nixon executive branch patent policy was pretty reasonable, really, other than that in some areas, some federal agencies took a long time to decide whether to let a private interest assert a patent monopoly–in the Department of Energy, for instance–and some agencies argued that private patent monopolies weren’t acceptable for the inventions they funded, such was the case at the Public Health Service (despite the NIH patent counsel feeling otherwise). Bayh-Dole addressed the delay in determinations by removing federal agency ability to decide, and imposed this removal government-wide, including procurement contracts and including frontier science grants.
Again, in frontier science grants, government ownership of inventions meant that everyone had access, for whatever future purpose or practice. For research practice. For industrial use. For the development of commercial products. Or not. With government ownership of inventions, the inventors had access to their inventions wherever they went–from one institution to another, or from institution to industry, as did their students (when the students weren’t also the inventors). Had Bayh-Dole limited only what the federal government could demand of inventors (directly or through contracts with the inventors’ host institutions), and only in the area of frontier science grants, Bayh-Dole might have been a good thing. At least in doing so, Bayh-Dole would have implemented what Vannevar Bush had advocated for in Science the Endless Frontier.
But this was not the purpose behind Bayh-Dole. The purpose of Bayh-Dole was to prevent the federal government from preventing universities (and their nonprofit fronts) from taking ownership of frontier science inventions and procurement-based inventions alike and passing these inventions as patent monopolies to their favorite companies, taking a share of the upside for doing so in the process. Thus, exclusive license was used as the means of assignment of invention, because the idea of a license suggested that there would be nonprofit oversight of the corporate use of the invention and exploitation of the patent monopoly. In practice, there is no little such oversight, despite the appearance (even threat) of such oversight. The result, then, of Bayh-Dole was to prevent the federal government from either (i) granting public access to inventions or from (ii) insisting that inventors should make those choices without institutional interference.
The framework of Bayh-Dole then becomes clear. It is an engine to deny public access and inventor control in favor of an institutional role in creating patent monopolies on behalf of corporations, especially in matters of public health. The intended effect is to prevent general access to such inventions, thereby increasing the speculative value of patents on these inventions, and thereby increasing as well the cost of obtaining access to these patents, and thereby positioning publicly funded inventions to be acquired only by people with substantial money to spend–and especially the pharmaceutical companies, and later, venture-backed companies aiming to sell out to the pharmaceutical companies or more rarely to become big companies in their own right.
Bayh-Dole does everything it can to deny inventor ownership of inventions and access to those inventions in favor of institutional ownership for the purpose of dealing private monopolies to favored corporations speculating on public health. The implied federal policy is that there should be no non-commercial development of research inventions. Not for research use. Not for industrial use. All technological innovation from federal research must move through an initial commercial product, and the maximum value for doing so is obtained by creating patent monopolies and conveying those monopolies to single organizations able to “invest” their “private risk capital.” In other words, plainly, reserve patent rights for speculation on the value of alleviating public suffering by means of commercial products. Put this way, the policy vision of Bayh-Dole is ugly. To make Bayh-Dole politically palatable, its designers insisted that the law should cover all forms of contracting for research, as if technological innovation arose only (or best) through monopoly development of commercial products. We know that’s not true.
To further sweeten the politics, Bayh-Dole was restricted to nonprofits and small companies, to remove the apparent distaste for giving large companies ownership of federally supported inventions–even though federal policy followed the practice of providing large companies with ownership of principal or exclusive rights in inventions in areas other than in matters of public health. What was distasteful–objectionable–was the idea that public money was to be used to subsidize the speculative profit-seeking interests of big pharmaceutical companies, who had plenty of behavior management problems (as the extensive federal investment in regulating them demonstrates). When questioned directly on this issue, advocates for Bayh-Dole admitted that universities did a poor job commercializing inventions and that the federal government would do better passing ownership of inventions directly to big companies, but that the public was not ready to accept such a thing, and so things had to be done this roundabout way, through nonprofits that the public would trust to do the right thing–even if their recent past history in the IPA program showed that they rarely did the right thing.
Thus, universities and nonprofits were used as a cover for this essential transaction–publicly supported inventions made in work directed to the public interest passed as patent monopolies to profit-seeking corporations with the claim that doing so was the only (or best or necessary) way in which the public would see any benefit from such inventions. Public money would be wasted on research unless speculative investors in commercial products had the only (or first) access to inventions and discoveries. Universities and nonprofits would look out after the public interest, would demand diligence of use and development, would require reasonable terms of public access, would make things available non-exclusively and only use exclusive licenses when absolutely necessary, and only with restrictions (at least when exclusively licensing big companies), and never assign any such inventions to big companies except with federal approval, and only then if the companies accepted the same conditions as those accepted by the nonprofits (including sharing royalties with inventors and using any profits after the cost of managing the inventions for scientific research or education–that is for a public purpose, not to reward speculative investors).
From this account, we can see why there is at once a public covenant in Bayh-Dole–it’s real, it’s substantial, and were it implemented would make some sense of an otherwise ugly, objectionable federal policy. We can also see why the public covenant has been suppressed, unenforced, uncomplied with, actual practice kept a government secret, reporting not required, most substantive positions waivable, procedures to march-in lacking any ability for the public to initiate, and procedures for public protection so burdensome that they would never be used and even if used would be delayed and largely ineffectual.
Only one part of the Bayh-Dole public covenant by-passes the faux Bayh-Dole Act, with its effort to mask the delivery to pharmaceutical companies (and startups aiming to sell to those companies) of patent monopolies based on publicly funded research directed at matters of public health: the government license.