The monopoly meme, 5

Now let’s turn to the Bayh-Dole Act and see how it works with the monopoly meme. Short form, if you don’t want to bother, is that Bayh-Dole doesn’t follow the monopoly meme in its gestures, but because these gestures never get used, in effect Bayh-Dole enables the monopoly meme. Here’s how.

The monopoly meme argues that the true purpose of the patent system is to enable the exploitation of patent monopolies–the right, for limited times, to exclude others from practicing an invention. Crucially, the monopoly meme argues that without such exclusion, new inventions will not be used, will not be developed, and the public will not benefit from them. If the public has to pay high prices for inventions for a period of time as a result, that is the price that must be paid. The alternative, according to the monopoly meme, is not to have any new products at all. This, then, is what university administrators mean by “commercialization.” If each invention made in university-hosted research and is not patented for the purpose of excluding all others in favor of a company willing to develop the invention into a commercial product–but only on the condition of exclusivity–then the public will not benefit from federally supported research. In this way, the monopoly meme insists that federal support of faculty research must necessarily be to create patent monopolies for companies. There’s no other point, then, to Bayh-Dole but to enable such monopolies, with their pricing and suppression of competition and suppression of research and professional use.

On the face of it, however, excluding others from practicing a university research-based invention runs counter to the public purpose of supporting research in which such inventions may be made. What’s the point, even, of publishing that research in the academic literature if the use of the results depends on the freedom to practice inventions, but that practice is nearly always excluded by a patent held by administrators looking for a single “commercialization” partner? An academic publication becomes a kind of propaganda, a nah-nah-nah about something new that’s been done that no one else can use. Perhaps the point of publishing research results that no one else can practice is so that others can quickly find ways to design around, obsolesce, undermine, block, fragment, and avoid using those research results. Otherwise, whatever the advocates assert is the “true” purpose of the patent system, in the area of research publication, the use of the patent system to exclude others from the immediate use of published research runs against pretty much everything we expect from university scholarship and the public role of the university.

All that aside, we might counter the “true purpose” of the patent system argument by pointing out that the express purpose of the patent system is to promote the progress of the useful arts–and “progress” here means “dissemination.” The grant of exclusive rights for limited times is the consideration for publishing what otherwise might remain unknown or held as a trade secret. Given that universities don’t generally hold trade secrets and university faculty gain much of their standing by publishing new results with priority, the right to exclude all others is a rather awkward right for university faculty and administrators to seek out or to be stimulated by.

The monopoly meme, however, argues that whatever the mores of the university might be, the creation of patent monopolies to stimulate profit-seeking is the primary public purpose that should displace all others–whenever an invention is owned by a federal contractor. Universities, the meme argues implicitly and many technology licensing officials argue openly, must change its mores. Porsches in the faculty parking lots should inspire other faculty to pursue inventions, desire the money that comes from patent licensing, and especially to temper these inspirations and desires to accept what a university’s technology licensing program might be able to deliver. 

The basic gesture in Bayh-Dole is that the law preempts most other law when a contractor obtains title to a patentable invention made in performance of work receiving federal funding. When a contractor discloses the invention and elects to retain title in that invention, then Bayh-Dole requires that the government get a royalty-free, non-exclusive license to practice and have practiced the subject invention. In the executive branch patent policies dating from 1963 on “practice” means “make, use, and sell” and “government” means not only the federal government but the states and municipal governments as well. That’s a broad license–broader than even the manufacturing and use of 28 USC 1498–which permits government use but allows the patent owner to file a claim for reasonable compensation. In effect, the government license strips a substantial portion of the monopoly value of any patent.

However, the government generally does not exercise its broad non-exclusive rights–at best it acquiesces in not having to pay royalties for using subject inventions to the extent that it does actually use any subject inventions. No one to my knowledge has any data on that. Thus, the monopoly meme survives the government license, although from time to time companies seeking exclusive licenses (i.e., assignments) of subject inventions will ask about the effect of the government license. They are told, generally, that the government license doesn’t matter, or that if they sell to the government, they are to discount the price by the amount of the royalty otherwise due the patent owner–1% or 2% or whatever–and even that advice is wrong, since the government’s license to a subject invention doesn’t have anything to do with what the government purchases! Oh well.

The next challenge in Bayh-Dole to the monopoly meme has to do with the conditions under which the government can obtain title to a subject invention. There are only a few stipulated conditions: failure to disclose, failure to elect to retain title, failure to timely file a patent application (except for inventor-contractors, who are not forced to use the patent system under their patent rights clause at 37 CFR 401.9); failure to prosecute a patent application, or to maintain, or defend the validity of an issued patent. Except for a failure to disclose a subject invention, when the government obtains title, the contractor-owner of the subject invention is assured a broad non-exclusive license. This license does not show up in Bayh-Dole. It is an addition inserted into the standard patent rights clause, 37 CFR 401.14(e):

The contractor will retain a nonexclusive royalty-free license throughout the world in each subject invention to which the Government obtains title, except if the contractor fails to disclose the invention within the times specified in (c), above.

This contractor non-exclusive license appears symmetrical with the government’s license when the contractor retains title–but the “markets” are entirely different. The government’s license is restricted to practice by or on behalf of the government. The contractor’s license is the entire market, worldwide. Furthermore, the contractor’s license extends to all entities that the contractor was legally obligated to grant licenses when the funding agreement was made:

The contractor’s license extends to its domestic subsidiary and affiliates, if any, within the corporate structure of which the contractor is a party and includes the right to grant sublicenses of the same scope to the extent the contractor was legally obligated to do so at the time the contract was awarded.

If anyone had their wits about them in the university licensing business, they would sign up companies to a conditional license under all government-funded work to every invention that the university acquires and subsequently assigns to the federal government. That way companies, for a tiny amount of money, have an option to anything that the university gives up that the federal government subsequently patents and tries to license exclusively, cutting off access by the companies. But no one does this.

The effect of the contractor license is dramatic, however. It breaks up the monopoly right. The federal government is left with a compromised patent monopoly. Where the contractor is a company, the best the government can do, short of march-in, is to offer a sole license–there then are two licensees, the contractor and the government’s favorite company competing with the contractor. In theory at least, the monopoly meme fails.

If we work through the practice logic at a company, however, it’s difficult to imagine any time that a company willing to give up ownership of a subject invention to the government would have any use whatsoever for a non-exclusive license, especially if the government might screw it over by licensing the government’s ownership position exclusively. If there really is value in the patent to create a duopoly, then the company may as well gain that value. If the value to the company is that the subject invention is available to be broadly practiced by all, then the company may as well obtain the patent at the least expense, with the least ambitious claims, and license the subject invention to everyone. That disables the federal government’s claims to title and precludes the federal government from trying to set up a competing user of the subject invention. From a company perspective, then, the non-exclusive license it retains in (e) is meaningless. The monopoly meme survives yet again.

A similar analysis may be done for nonrpofits with regard to (e). Unless they set up a program of signing up companies to take options to license all the inventions a university owns and then decides not to own in favor of the federal government, a university has little need for a government license to practice an invention. Although there’s no research exception in patent law, there’s also little by way of damages that anyone might obtain for suing a university for infringement–and if it’s the federal government that would be enforcing the patent, it’s difficult to imagine such a thing ever happening. Again, while the gesture in (e) of a non-exclusive license for the contractor is a happy one, in practice it does not disturb the monopoly meme.

Oddly, the parties who might want ownership of an invention–or at least a right to practice it–are the investigators and inventors. They might want to practice the invention outside the university–in a company, say. But neither has any clear privilege under Bayh-Dole, at least under the faux version of Bayh-Dole as it is broadly “interpreted” (meaning, disregarded). The Supreme Court noted as much in Stanford v Roche. The inventor’s privileges come *before* an institutional contractor has acquired title to a given invention. That’s the role of the (f)(2) requirement–but contractors don’t comply with (f)(2) and substitute instead their own requirements that preclude inventor rights. As for investigators that aren’t inventors–there’s no provision whatsoever. Investigators don’t even get to experience that bitterness that comes from knowing you have a right to your inventions even under Bayh-Dole, but administrators and their monster lawyers have cut you out. For investigators, the monster lawyers did their work in setting up Bayh-Dole and have gone to other pastures.

Thus, while in theory Bayh-Dole appears to require patent monopolies to be broken up between a contractor and the federal government, regardless of who ends up owning, in practice, the monopoly holds.

Now, what about subject inventions that the government acquires? Bayh-Dole section 207 authorizes the federal government to deal these inventions as it will, provided, under section 209 that for exclusive licenses it demand a “plan” from potential licensees and go through administrative procedures to document the basis for granting an exclusive license.


Each Federal agency is authorized to–

(2) grant nonexclusive, exclusive, or partially exclusive licenses under federally owned inventions, royalty-free or for royalties or other consideration, and on such terms and conditions, including the grant to the licensee of the right of enforcement…

That last bit about enforcement implies that federal agencies can assign inventions–because the owner of an invention has the right of enforcement of patents under patent law, and Bayh-Dole is part of patent law. Here the federal government is authorized to do anything it wants, but the monopoly meme says that there is only one thing that will “work” and that is to assign inventions to companies. Yes, in theory a federal agency could do anything. But in practice, assignment (exclusively license all substantial rights) is the only reason that the monopoly meme supports. If federal agencies did anything else as their default, then the entire apparatus of Bayh-Dole fails. The agencies would be back to licensing non-exclusively, as they were before, and Bayh-Dole (and the IPA program before it) exist because patent advocates claimed that non-exclusive licensing meant that federal research results never benefited the public and only through patent monopolies might the public benefit and America restored to global technology leadership.

Let’s look at a third challenge in Bayh-Dole to the monopoly meme. This one is more serious in theory, but just as empty in practice, so not to fear. Under section 203, the federal government may “march-in” on the rights of the owner of a subject invention on any of four conditions–nonuse, health or safety needs not reasonably satisfied, regulatory needs not reasonably satisfied, or breach of section 204’s substantially manufactured in the United States requirement. With march-in, the federal government can require the subject invention owner or exclusive licensee to grant licenses–

require the contractor, an assignee or exclusive licensee of a subject invention to grant a nonexclusive, partially exclusive, or exclusive license in any field of use to a responsible applicant or applicants, upon terms that are reasonable under the circumstances,

For nonuse (the failure to achieve the appearance of taking effective steps toward “practical application”), the monopoly meme may continue, since the federal government can require the contractor or exclusive licensee to grant an exclusive license and that would, in theory, preserve the patent monopoly and of course then preserve also the monopoly meme–that without a monopoly no one will develop any invention. Under the monopoly meme premise, the government will never grant non-exclusive licenses because no one will ever take a non-exclusive license to a subject invention unless the invention has already been developed into a commercial product and the only right the licensee needs to make or use or sell that product is the license to the subject invention. That’s a near impossibility. Perhaps if the invention rides as a feature on a product that is otherwise in the public domain or for which rights are available through a standard or a massive cross-licensing scheme (such as open source or open innovation). Otherwise, just no. It’s an elephant idea with no legs. Can’t live.

Of course, march-in has never been used–not with Bayh-Dole, not with the IPA program, not under the executive branch patent policy since 1963. Thus, when advocates of Bayh-Dole make the claim that no invention ever “taken” from a contractor by the federal government has ever become a commercial prescription drug, the effectiveness of the claim depends on everyone not knowing that there have been no such “takings”–and thus the claim is vacuously true, all the while accomplishing its intended purpose to allow people to deceive themselves with regard to the proper course for federal policy.

A fourth area of Bayh-Dole that might challenge the monopoly meme is the section 204’s restrictions on exclusive licenses. Section 204 requires licensees holding exclusive licenses to use or to sell in the United States to source product manufactured substantially in the United States. That in turn requires the owner of a subject invention either to manufacture in the United States itself, or to grant at least one license for another party to do so. Of course, if the owner of a subject invention grants an exclusive license to make, use, and to sell all to the same party, then section 204 doesn’t apply because the exclusive licensee is then in fact an assignee and owner of the subject invention, not merely an exclusive licensee of a right to use or to sell.

Section 204 does not actually require an owner of a subject invention to license anything, nor if licensing to prefer breaking up exclusivity between makers, users, and sellers. Instead, it operates to introduce a virtuous-sounding poison pill into transactions involving subject inventions. If a subject invention is assigned, section 204 doesn’t operate. If however a weak exclusive license is granted, then everyone has to find a way to manufacture product for that license in the United States. It’s a rather strange twist, given that Bayh-Dole was premised by Bremer and Bayh on the claim that it was needed to rescue American global technology leadership, which was in “peril.” We might expect Bayh-Dole to go full protectionist–anything used or sold worldwide based on a subject invention had to be manufactured substantially in the United States. Bayh-Dole would mandate exports. But there’s nothing like that in Bayh-Dole. Section 204 is an empty gesture–as well as an added burden to make unattractive any limited exclusive licensing that breaks up a patent monopoly. Thus, section 204 works to prop up the monopoly meme.

Despite the monopoly meme’s control over Bayh-Dole, agencies still work to undermine it. One way to do so involves a determination of “exceptional circumstances” under 35 USC 202(a) and (b). In those circumstances, a federal agency can constrain the right of a contractor to retain title to an invention the contractor has acquired, or place limitations on the contractor’s ownership of the invention. Preventing a contractor from retaining title preserves the monopoly meme–dealing in monopolies just moves from private hands to federal agency hands. But limiting a contractor’s ownership rights challenges the monopoly meme. If a contractor must grant non-exclusive licenses, even to make and use, then the meme is up.

Consider a research consortium with industry members. Companies join the consortium, contribute funding, technology, and talent, help to specify research to be done and technology to be developed, and one would expect that each would want to have access to anything that the consortium develops. That is, consortium members would expect a broad non-exclusive license to make, use, and sell any invention made in consortium activities. Now according to the monopoly meme, industry research consortia cannot happen. No company would ever develop anything based on a non-exclusive license to a patentable invention. We are left in stunned silence, then, at the NSF cooperative research center programs established under the authority of Stevenson-Wydler and using Bayh-Dole requirements for the management of inventions made in consortium activities. Perhaps the NSF CRCs exist only to train students in how to conduct inconsequential research that no one will ever use but still extracts federal funding.

The NSF avoids the Bayh-Dole requirements for exceptional circumstances in its CRC programs by insisting that universities applying for grants to run a CRC include the NSF supplied “model” IP agreement as part of their proposal. If they don’t use the model IP agreement, then the IP agreement will be reviewed as part of the proposal review–and you can assume that anything other than the model IP agreement will hurt the proposal’s chances of getting funded. University folks do the mind experiment and then include the model IP agreement, even though it is a rather sucky thing. Rather than imposing the IP agreement, the NSF lets university administrators work through the problem. Intelligent administrators figure it out and *they* make the determination that exceptional circumstances should apply, avoiding all the Bayh-Dole procedural nonsense.

The NIH, which was where Bayh-Dole was originally drafted, is not half so clever as the NSF. The NIH problem involves research tools that might be inventive. The NIH would like to see research tools developed with federal support made available to others for use in research. Makes sense. But Bayh-Dole’s default to the monopoly meme undermines the effort. The NIH would have to determine exceptional circumstances for each grant, define upfront in some comprehensive way just what constitutes a research tool (or “unique research resource” or whatever), and what constitutes “research,” and then shape a revised invention ownership clause that goes no further than necessary toward requiring non-exclusive licensing for research purposes.

The procedures regarding exceptional circumstances are as cumbersome as those regarding march-in, and the march-in procedures were designed not to operate, according to Bremer. The NIH, then, merely begs and pleads with grant recipients to make research tools generally available. It is rather sad for research. But the monopoly meme lives to fight another day. And that’s Bayh-Dole working as it was designed to work–preempting federal agency objectives in favor of the “free enterprise” of private owners of inventions made with public support. In this design, the stimulus of making money from monopoly patent positions preempts the stimulus of making something available for others to use. These two objectives converge only where a university finds a company willing to take an exclusive license to create a commercial version of the research tool, which other research organizations may then purchase at whatever price the market will bear, at some point in the future–typically years.

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