We are working through the NIH’s Institutional Patent Agreement master template to establish the context for Bayh-Dole’s use of “reasonable terms” in its definition of practical application, which in turn establishes the march-in threshold for 35 USC 203(a)(1), one of the four Bayh-Dole march-in conditions. Since Bayh-Dole is based on the Nixon executive branch patent policy, the implementing regulations for Nixon in the Federal Procurement Regulations, and the IPA, it’s worth seeing how these various instruments dealt with “reasonable terms” and affiliated gestures. Section (d) of the IPA provides the framework for when a contractor, having obtained ownership of a subject invention (the definition of subject invention is not the same as Bayh-Dole), may offer the invention under an exclusive license. In usual fashion, the IPA makes the default to be non-exclusive licensing (section (c)) and then walks it back (section (d)), leaving one with a first impression of openness and burying actual practice in a lengthy treatment of exclusive licensing.
Section (d) continues with its walk-back on the limits on exclusivity:
provided that the licensee shall use all reasonable effort to effect introduction into the commercial market as soon as practicable, consistent with sound and reasonable business practices and judgment.
We will go on a roundabout here. It’s important. The basic gesture here in the section (d) walkback is that despite exclusivity, the exclusive license should require diligence–so that the invention gets used sooner rather than later. That’s the gesture of Kennedy’s prefatory statement of policy: “The public interest in a dynamic and efficient economy requires that efforts be made to encourage the expeditious development and civilian use of these inventions.” Kennedy’s gesture is directed at defense-related inventions and recognizes the “dual use” concern of the day–since the great majority of federally held inventions at the time were indeed defense related. Why would anyone need to “encourage” development “for civilian use” other than that contractors ought to be pushed to find civilian applications for technology developed for military use–so sonar for fish finders rather than to detect submarines.
The theme repeats in the Kennedy patent policy. If the government acquires inventions, patents are to be “made available and the technological advances covered thereby brought into being in the shortest time possible through dedication or licensing.” The definition of “to the point of practical application” then makes sense–work the invention so that “its benefits are reasonably accessible to the public.” That is–the benefits accessible to the federal government are a matter of the federal government license and the deliverables the government expects from its research funding agreements, so that’s not the problem. If a reported invention has not been developed to the point of practical application for government purposes, the government may jolly well contract to get the work done. There’s no issue. It’s the “dual use” that’s the concern of the Kennedy patent policy–who is going to develop an invention made for government work in a direction that will benefit the public rather than the government. That’s the point of the definition–“benefits reasonably accessible to the public.”
A contractor has three ways of going about things.
(1) Do the development work itself–that’s what capable companies do, and what the Harbridge House report found was the most successful, most expeditious development of federally supported inventions. If the federal government wants invented stuff it supports getting used, then it should contract with companies with the greatest capabilities and established non-governmental (that is, not working as contract research organizations; not nonprofits for goshsakes) commercial positions.
(2) Dedication–license the invention non-exclusively and royalty-free, or on reasonable terms. These days, that’s known as FRAND licensing–fair, reasonable, and non-discriminatory licensing. FRAND licensing is at the heart of standards participation. Thus, dedication involves standards creation–ad hoc standards (or “platforms,” or “cumulative technology” or “libraries” of tools) as well as formal standards. If the contractor dedicates the invention to public access, then it can hold the principal rights to the invention for the full term of any patent. The definition of “to the point of practical application” does not come into play because, even if the invention is not being “worked,” the benefits of doing so are “reasonably accessible” to the public because the public is offered FRAND terms.
(3) Licensing–this sort of licensing involves exclusivity–it may be an exclusive license or an assignment. Kennedy uses “contractor, his licensee, or his assignee.” In such a case, the definition of “to the point of practical application” is in play, and the contractor/licensee/assignee has three years from patent issue to show benefits to the public or the government can march-in and revert the effort to “dedication” on whatever FRAND terms the government chooses. It’s clear that when the government screws up and contracts with a contract research organization that habitually gives up ownership of research inventions to its clients and has no commercial market interest but for doing other people’s research, or worse, contracts with a nonprofit to support a proposal from a faculty member who has even less profile for commercial activity, then this sort of licensing–exclusive, or assignment (that is, conveying all substantial rights in an invention, whether by assignment of title or exclusive license)–will be the thing.
We are at the heart of the logic here–if the CRO or nonprofit acting on behalf of faculty is going to “dedicate” the invention, then the argument has to be that the CRO or nonprofit will do a better job dedicating the patent (and any technology covered by the patent) than would the federal government. There is an argument here–that the nonprofit could be in a better position than a federal agency to support the inventors (and ancillary personnel) in being available to teach the invention, to assist in downstream use and development, to provide backing technical data and ideas for what variations might work best for a given purpose. But this argument is rooted in making an invention’s benefits “reasonably accessible to the public,” not with “commercialization” or “generating revenue to support research that lacks a sponsor.” If a CRO or nonprofit can’t do dedication things better than the federal government, then there’s no point in allowing the CRO or nonprofit to obtain rights or keep them. Bayh-Dole, however, makes this decision arbitrary unless a federal agency goes out of its way to anticipate it before requesting proposals or bids. In either case, dedication is the heart of “technology transfer.”
There is a second argument here at the heart of the logic. This is the “patent monopoly is essential” argument. Without the exclusivity of a patent–a patent monopoly–no one will develop or use the invention. “What is available to all will be used by none.” This monopoly meme is pretty much always a nonsense claim, but it is always used as a generally applicable admonition, and for some reason, people accept it rather than laugh themselves silly and then resume a sane discussion.
Patent monopolies clearly are not essential for government practice of an invention that it wishes to practice. Neither the government nor the contractor has any need of patents with regard to the government’s practice of an invention, or the development of that invention to the point of practical application. The only thing at stake with a contractor’s patent is whether the contractor can set up a claim for compensation in the Court of Federal Claims. And that’s not at all an argument that without a patent monopoly no one will develop an invention or use it. It’s obvious that the federal government does not care about patents when it wants to develop an invention for its use–for any use within its scope of authority, including for the general welfare, the public interest.
This second argument, then, stipulates that a patent monopoly is essential to developing an invention (i) for a civilian use that’s different from the use that the government may have in mind and (ii) and that exclusivity will do a better job of expeditiously developing the invention for civilian use and make the benefits of that development “to the point of practical application” “reasonably accessible” to the public. It is not simply that proprietary development must outperform open development–but rather that proprietary development has only the roughly six-year window from invention to three years after patent issuance to get to “the point” of practical application and further, at the end of this window, the benefits of use have to be made “reasonably” accessible to the public. There would be absolutely no need for “reasonably” in the definition if any form of accessibility would do–would be better than “open” development (as, to create a standard). There would be no need for “public” in the definition if the expectation was that any one or few people having access was sufficient to do better than “open” development. Accessible “to the public” necessarily means to everyone of the public, on terms that every member of the public–individual and community–can judge to be “reasonable.” We are not talking here about what a patent holder might rationalize as “reasonable” to its interests, or what an exclusive patent rights holder (licensee, assignee) might determine to be in its “reasonable” commercial interests. None of that matters. What matters is whether proprietary licensing will do a better job than dedication/FRAND/open in “expeditious development and civilian use.”
The argument in the contractor-singular seems sound–a CRO or nonprofit contractor might indeed do a better job with dedication than the federal government could do. And even for licensing, there’s a logic–that if (big if) a patent monopoly really is necessary to support development for civilian use, then it makes sense for the contractor to retain ownership of inventions. But consider the situation in which a federal agency (or multiple federal agencies) spread research money around among CROs and nonprofits in the same area of work. Now we may see scores of inventions reported across scores of contractors–all in the same area of practice. Now the question becomes whether each contractor claiming its own ownership of its bit of the overall research effort, and determined to license exclusively or assign, could possibly be “more likely to serve the public interest” as expressed by the Kennedy patent policy? The obvious answer is “no.” Every contractor claiming ownership of inventions fragments the practice of the overall research development. Broad patent claims on these inventions restrict variations and functional equivalents. All someone can do is patent improvements and work arounds and applications and create a standoff that forces a cross license–and at that point the monopoly meme breaks and it becomes clear that when there are multiple organizations all working in the same area of research and development, each claiming exclusive rights in their bit creates gridlock, not innovation. In the commercial world, companies have learned that they have to work things out through cross-licenses, patent pools, and standards supported by FRAND licenses–or one company necessarily has to end up buying up or neutralizing all the fragments of IP that might block commercial use. In short, in any area in which the federal government supports research across multiple organizations, either all the inventions must be pooled in a single company (and that’s going to involve favoritism and anti-competitive gestures, at the least) or everyone must be compelled to follow the dedication/FRAND/open route–and do it better than would the federal government.
It is this last bit that Bayh-Dole craters with its default arbitrary provision for contractor ownership. Nonprofit and CRO administrators could look around and determine that since there were multiple organizations receiving federal grants in the same area of work as their own inventions (and why restrict to federal grants–any evidence of the same work in industry), they necessarily must follow the dedication/FRAND/open route. But they are helpless to make such decisions–they have to bite, like dogs eating chocolate–too good to be true, sickening results. Federal agencies, too, could–and should–determine exceptional circumstances any time any one or more of them recognizes that they are offering multiple contracts in the same area of study to different contractors. The exceptional circumstance is that Bayh-Dole assumes a single funding agreement with a single contractor in any area of federal research interest. That’s the default. It is a rare situation. The exceptional circumstance is therefore often a common circumstance, even if it lies outside the default expectation established by Bayh-Dole.
One of the most important things that can be done to salvage Bayh-Dole is for federal agencies to create a standard patent rights clause for the exceptional circumstance of multiple federal research contracts to CROs and nonprofits in the same area of study that permits contractor ownership but only on the condition that the contractor follow the dedication/FRAND/open alternative and do a better job than the federal government. This same condition should be imposed wherever an invention involves the development of a research tool or resource that would be used by other organizations regardless of whether those organizations received federal research funding. The precedent for doing this is already in Bayh-Dole–the Department of Energy negotiated into the law its own default exceptional circumstance for certain nuclear energy inventions, such as naval propulsion systems. NIST has worked overtime to confuse the standard patent rights clause architecture, but the architecture is still there. It may be that NIST has to be removed from its role in charge of Bayh-Dole regulations for things to get back on track.
One can see then that Bayh-Dole does not merely reverse a calculus of ownership for nonprofits, which under the Kennedy and Nixon patent policies were not assumed to have a right to retain principal rights in inventions even if they acquired those rights–simply because there was no equitable reason to assume that CROs and non-profits could do a better job with dedication than the federal government. Contractors had to make a case for doing a better job than the default federal government dedication/FRAND/open development. Bayh-Dole does away with the need for a contractor to make a case–unless a federal agency were to add that back in as a matter of “exceptional circumstance.”
Oddly–and revealingly–on the other side of Bayh-Dole, having to do with the disposition of inventions owned by the federal government–federal agencies are required to follow an extensive checklist to justify deviation from dedication/FRAND/open and potential federal licensees are required to submit extensive plans for development as a precondition to receive a license. Why is it that contractors get a free ride in adopting the patent monopoly position if they secure ownership of a research invention made with federal support, but that is not the case for companies seeking an exclusive license from the federal government? You see the contradiction. If it is by default in the public interest that a company holds a proprietary position in an invention (the “monopoly meme”), then there should be no apparatus for federal exclusive licenses–whoever shows up first should get an exclusive license, no plan necessary–just as Bayh-Dole provides for contractors. Whoever gets the research contract also then gets to take a proprietary patent position, no questions asked. It cannot be that in one case, there must be piles of administrative paperwork (or at least the appearance of it) and in the other contractor proprietary positions are by default in the public interest. It just doesn’t work. It is a nonsense distinction.
If we wanted to push the logic–reason, such a drat thing–then we would make Bayh-Dole’s default in practice, using the exceptional circumstance determination, that nonprofits, if they elected to retain title, must follow dedication/FRAND/open practices for all inventions that are made in a research area where multiple organizations receive federal research awards in a common area of study. For any other funding agreements, each agency would establish a default time frame for practical application. For non-medical inventions, that might be the Kennedy default–three years from patent issuance. That’s more than enough time in most areas of information technology, for instance. Any longer suggests patent trolling rather than expeditious development.
Where a contractor does not have capability or an established non-governmental commercial position–a CRO or nonprofit–the federal agency in determining principal rights in an invention should consider what would “most likely serve the public interest as expressed in this policy statement.” It’s clear that the public interest is in the expeditious development for civilian purposes and use of the invention so that there are public benefits–“reasonably accessible.”
Now back to the walkback in section (d) of the IPA master agreement–it makes a gesture at requiring diligence, which would be expected to comply with the Kennedy patent policy. That’s a positive requirement (such as it is) rather than just setting a minimal standard of effort beyond which the government might march-in.
Under section (d), exclusive licenses should require “all reasonable effort”–not best effort, say–to introduce commercial products–but only “as soon as practicable” (meaning whenever), and all made then essentially at the discretion of the licensee–“consistent with sound and reasonable [because sound stuff may not be reasonable, apparently] business practices and judgment.” So, basically, it’s up to the exclusive licensee without any particular extra diligence reflecting a public interest need for rapid introduction (as might be thought to be the case with new medicines), and the limits on the term of exclusivity are subject to the discretion of the NIH:
Any extension of the maximum period of exclusivity shall be subject to approval of the Grantor.
And finally, what happens if exclusivity ever does end:
Upon expiration of the period of exclusivity or any extension thereof, licenses shall be offered to all qualified applicants at a reasonable royalty rate not in excess of the exclusive license royalty rate.
The idea in big terms is that once the exclusive licensee has done the development work and for a reasonable time enjoyed the benefit of commercial sales in whatever way it chooses (there’s no reasonable standard in the IPA master agreement for exclusive licenses), then everyone should have an opportunity to make, use, and sell product based on the invention. That’s a nice gesture. Again–a private licensee undertakes development of an invention on behalf of the entire industry as well as for the public benefit. In return, the private licensee gets a reasonable period of exclusivity to recoup its development costs, and then everyone happily competes, giving the public choice, price competition, and general “reasonably accessible” benefits, even if the exclusive licensee is not constrained to “reasonably accessible” benefits during the term of its exclusivity (unless somehow the nonprofit licensor determines that such constraints are necessary to comply with its mandate to administrate inventions in the public interest).
Beyond this big idea is silly nonsense that undermines the idea into pretty much nothing. In real world practice, the royalty rate for non-exclusive access should be substantially less than that of any exclusive license. The IPA master agreement here stipulates not that a reasonable royalty rate must be charged (which would be odd–one might expect royalty-free) but that the royalty rate of the exclusive license is, as a matter of federal contract, a reasonable royalty rate for non-exclusive licenses. That’s pretty ballsy. The Kennedy patent policy, again, has it “royalty free or on terms that are reasonable in the circumstances.”
All of this licensing stuff in the IPA is subordinate, however, to the Kennedy patent policy requirement that the contractor/licensee/assignee has three years from patent issue to show practical application or license non-exclusively or show cause why the contractor/licensee/assignee should retain principal rights in the invention. Thus, the limitation on exclusivity–three years from commercial sale or eight years overall–may lie entirely outside the Kennedy limitation on a contractor/licensee/assignee holding principal rights. Inventions are rarely licensed immediately–companies like to wait to see that relevant claims will be allowed–and so might string a license negotiation out for three years or more. Commercial sale may take years more to achieve. One might be talking most of the way through the life of a patent before one is even up against the IPA master agreement’s 3/8 limits–even without the NIH available to approve extensions.
The unwritten necessary policy in the IPA master agreement is that the NIH will decline to take ownership of inventions or march-in within the 3/8 limits rather than the three years from patent issuance stipulated by the Kennedy patent policy. If a nonprofit merely licenses an invention to a company, that is sufficient “effective steps” to waive the Kennedy three year clock–even though a patent license merely conveys a promise not to sue and does not show that anything has actually been done to “develop” the invention or enter a “commercial market.” A clever policy workaround. In essence, the 3/8 limits are a restatement of the contractor/licensee/assignee showing cause why it should retain principal rights for more than three years from patent issuance–essentially, for the life of the patent, without so much as saying that.