The leaders of any status quo cannot self-propose changes to the status quo under which the leaders cease to be the leaders.
Or, put it another way, proposed changes in a status quo by those most benefiting from the status quo will be socially invariant.
Corollary 1: Change comes when the leaders screw up and propose a change intended to keep them the leaders only to find it does something else.
Corollary 2: Change comes from the outside and the edges and the unexpected.
Well, the National Research Council has deliberated on university technology transfer, and apparently can’t get past the linear model, the consensus view of patent licensing, commercialization as the high point of dishing public assets to the wealthy and powerful, and a mania for making tiny improvements to narrowly conceived processes. The bozonet rejoices.
Let’s look at some findings.
Finding 1: “The first goal of university technology transfer involving IP is the expeditious and wide dissemination of university-generated technology for the public good.”
Why should the first goal be expeditious and wide dissemination? Why not considered and focused? Why would wide dissemination be a first goal? Why not an ultimate goal of a few inventions a decade–like 30 years of university efforts have demonstrated is all that is possible? And if one wanted wide dissemination, why not start with broad, non-exclusive licensing? What a confused noise.
Widespread dissemination by using exclusive patent licensing where most patents don’t get licensed and most that do go moribund, but *this* is a good thing. If this were truly the first goal, then universities would license non-exclusively, royalty free. But they don’t. They say that the neat thing about Bayh-Dole is that it permits exclusive licenses, where the government wouldn’t do this often. This was the breakthrough. So from the university perspective, it should be considered and focused, not rushed and epidemic.
Or, is “expeditious” code for “brain-dead standardized licenses where a one-off generic biotech license becomes the grail for all licenses to follow”. Yes, you see, licensing is a commodity activity. They can all be the same. The report all but appeals: please make patent licensing a commodity activity done quickly by clerks.
Why should it matter whether there is IP? Should not technology transfer also take in NIPIA–non-IP intangible assets? Why should technology transfer be construed only as those parts involving IP rather than know how, expertise, insights, and data? The report is not going to help us here. These are Findings, dammit, not just some ho-hum punditry.
Why should the first goal of technology transfer conflate inventions and patents with technology and IP? It’s clear that universities are largely clueless about using copyrights and trademarks to disseminate “technology”, but apparently the report writers cannot get to that point.
And what other purpose is there beside the “public good”? Given that universities argue that making money = public good, what’s the point? If one wanted to argue, more royalty-free deals, that would be interesting. No one is going to go there in this report.
One might note that most university “technology transfer” is fixated on inventions and patents and is subject to Bayh-Dole. University practice is furthermore fixated on “commercialization”, meaning seeking to create products in lucrative markets, measuring success by royalty income. Yet apparently the report cannot recite the stated objectives of Bayh-Dole and say, perhaps, that the first goal of technology transfer is to use the patent system to promote practical application of federally supported inventions. The report does not get at this senseless conflation of commercialization with practical application.
In all, it’s a bland finding that doesn’t even state the obvious. How very disappointing.
Lots of questions remain from our variation on the theme.
What about F’s written agreement? Can F sign any documents for the company, such as the declaration and oath that F is a true inventor? F has an obligation to protect the government’s interest, but signing papers for the company doesn’t change this obligation.
What happens if F lets S do a bunch of work on the grant informally, because S’s company has an interest in the outcome, and S invents solely? Does the government have a claim on S’s title? Does the university? If so, is the university’s claim on behalf of the government or just in its own interest? Is making a claim by either a necessary bit of a broader effort to advance the use of inventions?
There are a lot of questions. Fortunately, no one seems to care about these. What’s remarkable is that folks are so fixated on just one of these–having to do with a joint invention where an investigator commits invention rights and then the university goes and accepts a grant directly within the scope of that commitment, and lets that investigator participate, and then folks want Bayh-Dole to cancel the prior commitment, not to protect the government’s interest, but to protect a private interest of the university.
We have looked at five ways Bayh-Dole can play out with a single faculty researcher, who we call F. These are government title, university title, third party title, inventor title, and public domain. We have emphasized that Bayh-Dole makes uniform the government’s management of its interest in the personal rights in inventions arising in federally funded research. We have emphasized that it is the written agreement that is to be required by universities of its employees that is the basis for conveying rights to the government at the government’s request. And we have emphasized that when a university elects to retain title, it intercepts, if you will, the government’s right to request title as a deliverable, and stands in for the government. The title a university (or anyone to which the university assigns the right to request title) may request is the personal right that the inventor holds by virtue of inventing.
Let’s look at a variation on the theme. Let’s add a company scientist, S. S is obligated by S’s employment agreement to assign all inventions S makes to the company, less any inventions that state law does not allow the company to claim. Let’s say S has a present assignment of all such future inventions.
Now F gets a federal grant at the university. While working on this grant, F has a conversation with S, and together they invent something. Now what? F has an obligation to report the invention to the university under the written agreement required by Bayh-Dole. S has an obligation to report the invention to S’s company. When S does so, under the present assignment S has agreed to, S’s company holds title to S’s rights in the invention. But what about F’s rights? Title follows the usual Bayh-Dole route. The university may elect to retain title, or any of three other ways. But the fifth way—public domain—can’t happen unless the company also chooses not to file. That’s because in a co-invention situation, each owner has an undivided interest in the patent, regardless of which co-owner files the patent application.
This situation sets up some things worth noting. First, there is nothing wrong with this outcome under Bayh-Dole. There is no requirement that F keep research confidential or not involve anyone except university employees. A federal agency may impose non-disclosure, but for universities this generally requires defense classification and special handling. Universities also generally do not self-impose such requirements, other than those that are required by regulation (such as concerning release of personal information), or by external sponsors. Often sponsor requirements have to do with securing sponsor-provided technical data and review for patentable subject matter or use of proprietary information prior to publication.
Second, what the university gets if it elects to retain title is what the government is entitled to request–F’s personal interest, not title in its entirety, since the university has no claim under Bayh-Dole on S’s personal interest in title, which goes to the company. The government standard is that its interest is protected—not that the funded work is entirely controlled by the university, without any external involvement.
Third, if the company files a patent application on the joint invention, the university is not going to get an exclusive position to offer to monopolists for money. It can play the spoiler (such as making its undivided interest available at no charge), or it can work something out with the company to standstill on its undivided interest in any patent that issues. Or it can build a practice community or standard using its patent position (though this lies outside the expertise and interests of most technology transfer offices at present).
This is not commercialization the way it’s set out on in tech transfer diagrams about “the process”, but there may still be good work to do to help with implementation, quite apart from anything having to do with a patent license.
Fourth, if the university elects to retain title, it still has its Bayh-Dole compliance issues. But the company does not. Its interest has nothing to do with Bayh-Dole. Its employee S has no written agreement to protect the government’s interest. Bayh-Dole follows the disposition of the personal invention rights of F that arise in the federally funded research. It does not cut off co-inventor S from S’s rights when these lie outside Bayh-Dole, even if the invention is within the “planned and committed” work of the federally funded research at a university. Even if S used university facilities in making the joint invention, the matter is still not an issue for Bayh-Dole, since S is not a university employee. While the university may have some issues with S’s use of facilities, these don’t have anything to do with Bayh-Dole compliance.
Finally, what if the university elects not to retain title? Now the right to request title passes to the government. Again, it’s only F’s personal interest in the joint invention that is at play. If the government requests title, then it becomes a joint owner with the company. If the government declines to request title, then it’s up to F to request to retain title, with those minimum Bayh-Dole requirements. But what if F doesn’t request to retain title? Here’s the thing—F will still end up with title if the company files a patent application. F is a co-inventor, co-inventors are co-owners, title to an invention includes title to patents that issue on the invention.
From this it can be understood that for Bayh-Dole, retaining title means retaining standing to file a patent application. The written agreement required by universities under Bayh-Dole to protect the government’s interest is an agreement regarding who has standing, within the chain of government procurement of personal patent rights as deliverables, to file patent applications on subject inventions. If an invention is jointly made, and one of the co-inventors stands outside the scope of Bayh-Dole, then whatever happens under Bayh-Dole, that other co-inventor has standing to file patent applications and use the patent system in whatever way the co-inventor, or any successor in interest, chooses. In the case of joint invention, the emphasis of Bayh-Dole may shift from standing to file to who has an obligation to protect the government’s interest. If the university chooses not to retain title (since what’s the point? no monopolies here, move along), then it is up to the government. This is the government’s chance to secure for itself a royalty free license for government purposes. If it does not, then interest in title settles back to F. At this point, if the company is filing a patent application, F has title and so it would appear that the government, to obtain the non-exclusive license it desires, would need to require F to retain title.
Way 5. The university declines to elect to retain title, the government declines to request title, and F does not request to retain title. The invention enters the public domain (one year from publication or other statutory bar).
Bayh-Dole does not require anyone to hold title to inventions, or to use the patent system at all. Not the university, not the government, not the inventor. What Bayh-Dole does is this: if a university wants to use the patent system to promote the use of a subject invention, it may do so on uniform terms with regard to the interests of federal agencies setting the conditions for providing research funding.
If F simply publishes the results of F’s federally funded research, no one apparently cares. The government does not pursue faculty researchers for publishing what may otherwise be patentable inventions, so that these enter the public domain. Bayh-Dole is not an effort to force inventions into the patent system, to get more patents, or to help universities make money from patent licensing. It may be that others have these agendas. It’s just not in Bayh-Dole.
Another aspect of this is worth mentioning: in US law, an invention is not an invention until it is recognized as such by its inventor. See MPEP 2138.04 for the discussion. That’s the point of conception. It is also a reflection of differing expectations with regard to what the patent law calls non-obvious to one with “ordinary skill in the art”. If F doesn’t recognize work as inventive, then there is nothing to report, since there is *no invention*.
An invention is not an objective commodity to be harvested by corporate officers or administrative clerks. It is a recognition in the mind of the inventor. It is up to F to recognize whether something is inventive. This is fundamental to Bayh-Dole.
Universities have made up their own mythology of innovation and compliance, called the Little Linear Model. The LLM claims that the key to innovation is that research inventions pass through the hands of university administrators. This they claim is the big insight of Bayh-Dole. It’s like saying the secret to a great dinner is that the waiters have to handle all the food before its served. Fascinating, isn’t it?
How might anyone know that F recognizes F has made an invention? First, F might report it as an invention, either in research records or in a publication or in an invention disclosure. That would be a good start. At least F thinks it’s inventive. Second, F might prepare a patent application. That might also suggest that a patent attorney, perhaps, also believes there’s an invention. Absent either of these indications, how can anyone know if F recognizes some research event as an invention?
Yet universities tell investigators they need to report findings “just in case” they might turn out to be inventive and if so then somehow it would be a violation of Bayh-Dole not to have reported these findings. University administrators even try to take title to unpatentable inventions to prevent investigators from acting on their findings without paying something to the university.
It’s also the case that F might think something is an invention, but the work turns out not to be patentable. That is, while it may be new to F, the work might be anticipated by other reports that F is not aware of. A patentable invention requires absolute novelty, not just a local wow. One can hold title to an invention, but if it works out that the invention is not patentable–and that’s pretty much the only thing that matters with title to an invention under Bayh-Dole–then !poof! A lot of hubbub about nothing.
Bayh-Dole is not interested in inventions without patents. The point of the university taking title is to use the patent system to promote the use of the invention. No patent, and we are done with Bayh-Dole. So whatever we are working with, title to inventions is about using patents that might issue some two or three years later. Title to a research invention is prospective to gaining an ownership interest in a patent on the invention.
Let’s turn to Circular A-110, where Bayh-Dole is a part. Circular A-110 applies to a much broader array of intangible assets, not just patents, but including other intellectual property such as copyrights and trademarks. Bayh-Dole that is not about owning the inventions themselves, but about the use of the patent system to promote the use of inventions. Circular A-110 __.37 says, for any intangible assets that the recipient (for us, the university) acquires or improves in the course of the funded research, it will hold those assets in trust on behalf of the beneficiaries of the research. However the university comes to own intangible assets, it is to act as the steward of those assets.
This is a broader mandate for university technology transfer than those folks focused exclusively on start ups and product development licenses care to discuss. And the aim is to support the beneficiaries, not the university administration.
When investigators conduct research, they discover and realize and develop data and develop proposals. Call these research events. The documentation of these events may give rise to intellectual property. But the dissemination of what’s newly discovered is a matter of instruction. What researchers know generally is standard knowledge. What they know as experts in their fields is pretty special. What they have just realized in their research is potentially distinctive in the world. This is one of the most important of research outcomes. If a university administration is going to do something useful in this area, it will be to contribute to an environment in which these distinctive research events are taught to others. Some few of these events might be taught by way of a patent license. Most won’t be.
While research may be its own “mission” of the university, the reporting of results is a branch of instruction—and it is the instruction to capable audiences, prepared to act on what they learn, that creates the opportunity for transfer of “technology”. Yes, there may be bits of intellectual property hanging onto discoveries, and there may be more intellectual property to be created as folks explore what’s been discovered. The starting point, however, is instruction. Not business, not money making, not patents, not licenses. That’s what F needs to be considering, not whether or not the feds are coming because F publishes without recognizing an invention.
Whatever Bayh-Dole may take as its primary emphasis, for universities supporting the conduct of research in the public interest, Bayh-Dole is subordinate to the instruction in what has been discovered, so that it may be evaluated, studied, and practiced more broadly than the originating laboratory. In that, Bayh-Dole as policy is subordinate to Circular A-110, and all this regulatory framework has to be understood to be subordinate to a broader commitment to discovery and innovation. It is this commitment that universities in their programs have to reaffirm and make visible and contribute to every day. Their business is not to be consumed with fussy compliance with bits and pieces of regulations simply because those regulations exist. Their business is not grubby helplessly acquisitive money-making with patent-happy folks trying to shake down industry and snooker investors of their dollars. These things can indeed be part of the effort. And for all my choice of words—for a particular edge—there are good reasons for both compliance and for “value capture”. But the public interest in research is not, “whew, we are sure glad there is compliance, and boy oh boy, it is great when that compliance single-mindedly supports the university trying to make money by shaking down industry!” It goes all absurd when folks then try to make this *more efficient*.
Bayh-Dole has little to say about innovation. It does not specify what inventions should be managed, nor what agents should be involved, nor how patents are to be used to promote use, nor the incentives. These are matters left to universities—to faculty and students and staff, not just senior administrators, and certainly not just erstwhile business people jumping in to tell everyone how to make big bucks from patent licensing for product development.
Yet we find that community’s discussion consumed by the idea of “commercialization” as if that is the only purpose of Bayh-Dole. After 30 years of this, universities have taken one of the most exciting parts of research outcomes and turned it into one of the dullest things imaginable. Now it’s just a matter of making their dullness the default. Then there will be maximum compliance and some really keen efficiencies. Yeah, that’s innovation for you, university technology transfer style. It just doesn’t get any better than that. And that’s the great sadness.
Even where there are no patents with “commercial potential”, university research sparks with challenges and insights. Our researcher F deserves to have a technology transfer program that puts its effort into creating forums in which research discoveries are taught to capable audiences, and where there aren’t audiences that are capable, then there is even more work to be done.
Not everything good in the world comes about because there are markets, not every collaboration has to be so mercenary that something has to be sold, or even that there has to be a reciprocity of “fair value”. Not every contribution or advance can be reflected in a bean-counting balance sheet. In fact, not many can. In fact, only a very few can. Maybe a handful a decade. That’s not what the public needs from technology transfer, nor what investigators need.
Bayh-Dole does not set the agenda for publicly funded research. It is merely a tool to make one bit of the deliverable interface uniform. Patent licensing may be a part of technology transfer, and important in its way, but it’s a bit role in the big picture. Without university claims on patents–without waiters fingering all the food–there would still be a huge role for technology transfer in many forms, not just from lab to product, but from lab to lab, from community to lab, from practitioners to companies, or to other practitioners.
We can work a few variations now with faculty member F. The university has a standard IP policy with a promise to assign to the university. Let’s look at four basic possible outcomes. Then we can start to add variations.
F obtains a federal grant. So Bayh-Dole provides for the standard patent clause at 37 CFR 401.14(a), which is implemented through Circular A-110 ___.36(3), which sets the terms for the grant to the university. The university implements Bayh-Dole by requiring F to protect the government’s interest through a written agreement by timely reporting inventions and executing the papers necessary to permit patent applications to be filed and to establish the government’s interest in these inventions. F reports the invention to the university. The university reports the invention to the government.
Way 1. The university declines to elect to retain title, the government requests title from F, and F assigns title to the government, pursuant to the written agreement required by the university to protect the government’s interest, in compliance with Bayh-Dole.
With this way, a university does not need to have a technology transfer office. It is the closest to a pre-Bayh-Dole implementation, but uniform and expecting the university to provide notice of invention and confirming that the university declines to elect to retain title.
Way 2. The university elects to retain title, the university requests title from F, and F assigns title to the university, pursuant to the written agreement required by the university to protect the government’s interest, in compliance with Bayh-Dole.
This is the way, presumptively, a number of technology transfer offices want things to end up. They see it as more efficient to have Bayh-Dole simply changed to require this outcome, without any action by the inventors (they never hold title) and without any choice by the university. The university’s choice to retain title is shifted to whether to keep it or assign it, not whether to request it or decline to request it. It is a huge difference, unless you don’t care about choice. Given that commercialization folks often refuse to acknowledge choices—the Little Linear Model is the only one that’s not sub-optimal—it is clear why they would like to suppress those choices. Not only is choice inefficient, but it also could create competition with their claims of how “technology transfer” works. Further, they wish to take Bayh-Dole’s regularization of agency protocols for procurement of patent rights as deliverables in federal contracts to start a regularization of the process by which inventions are “commercialized”. That is, take a procedure and make it stand for making mechanical how innovation arises. That also is a huge difference. Bayh-Dole’s brilliance is in allow a diversity of approaches. The commercialization crowd cannot tolerate such diversity. It is inefficient and worse it makes them look bad when they make general claims to their superiors, the faculty, and the public about how things work, when it is only how things work in their own heads, and that doesn’t work very well, and they cannot bring themselves to imagine that other ways might work better, but it would cost them their standing, and worse they would have to roll up their sleeves and learn something new in the world.
Way 3. The university assigns its Bayh-Dole interest to an affiliated research foundation, and that foundation elects to retain title, requests title from F, who assigns title to the research foundation, pursuant to the written agreement required by the university to protect the government’s interest, in compliance with Bayh-Dole.
In this approach, the university never holds title to the invention, nor even retains the title, because title is retained by the research foundation, and then is conveyed as requested through the obligations created by that same written agreement. Note that the university’s IP policy simply doesn’t operate with regard to ownership. It is superseded by the written agreement. Where the university relies on the IP policy as its implementation of the written agreement, it is only those parts of the IP policy that in fact implement the written agreement that operate.
Way 4. The university declines to elect to retain title, the government declines to request title, F requests to retain title, and the government after consulting with the university, concurs, citing the minimum conditions for its approval as set forth at 37 CFR 401.9.
Bayh-Dole provides for inventors to retain title. That is, in the absence of any claims on title as a deliverable, title is not a deliverable under the federal award, and title remains where it always is under any normal Patent Act, and that is with the inventors. One might ask, then, if the inventors can have valid, conditional agreements that obligate the title they may retain. Of course, provided these agreements are compatible with the Bayh-Dole obligations required of them in the agreement under which they are permitted to retain title.
As a special question, one might ask, if the university declines to elect to retain title under Bayh-Dole, can it then come back around and say to the inventors, “but under university IP policy we can still claim the invention, but now with only the minimum of requirements to the government, which is a better deal, so you are out of luck and have to assign.”? One might argue, if the university declines to elect title, that *does* have an effect not only of waiving its interest in title under Bayh-Dole *and* exhausting its interest in title under its own IP policy. That is, Bayh-Dole supersedes that IP policy in matters pertaining to the government’s interest, but not in matters pertaining the university’s interest.
To see how this works, recognize that for the university to obtain title under Bayh-Dole, it does so under the written agreement, which is a university required agreement to comply with a federal regulation. The university’s own IP policy is no good at this point, but for stipulating that a written agreement will be required pursuant to Bayh-Dole (repeating the obvious) and that employees will comply with that written agreement (yes, since it is required that they do so by federal regulation). The university is not at liberty to throw anything else into the mix. The written agreement sets out the terms, and that’s what is required, no more, no less. But the university’s decision whether to elect to retain title lies outside the scope of this written agreement. It is a university decision. It is not constrained one way or another by Bayh-Dole. It does not even have to be made. The university can default, not notify the government of a choice, and Bayh-Dole operates on its merry way. But when the university makes a claim to title, it does so as its own action, and likewise if it waives an interest in title, it does so under its own IP policy. Once waived, the opportunity to obtain title is exhausted. It stands as a quit claim, an action by the university under its own terms, not those of Bayh-Dole.
One might see in this how a conflation of Bayh-Dole obligations and those that arise from the university’s own administrative choices might serve the interests of folks who want to have power over decisions and actions but do not want anyone to recognize the basis for that power. It’s good enough if Bayh-Dole requires it. Here’s the rub: if Bayh-Dole does not require it, and a university’s administrators put out that it does, and they know otherwise, or should know otherwise, that is professionally incumbent on them to know otherwise, then it is malpractice if not a kind of fraud. Employees are being induced to take actions on improper grounds. Where this inducement comes from public universities acting as governmental agencies, one might even go so far to say that it is a corrupt practice to claim personal assets citing an inapplicable federal regulation or an improperly applied state or university policy regulation.
Malpractice, fraud, corruption. These are big words. But folks shouldn’t be in such denial that it’s not possible. And one should at least recognize that triggering these concerns might at least reflect that there’s a potential breach of good faith, of integrity, of reciprocity that otherwise ought to be present between university inventors and those asked to work with them to realize a public benefit.
Maybe it’s all too clear. In the case of an employee bringing the scope of authorized private consulting into federally funded research, the issue is not the mechanism of the obligation to convey title, but the scope. The problem is not in Bayh-Dole, but in the university’s compliance with the law, and with the implications for a company claiming the benefit of the mechanism by which it expects to obtain title.
If an employee comes back to the university with a valid obligation to convey title to a company, it is up to the university to require the written agreement as Bayh-Dole has it. If the university cannot do that, then either the award should be rejected or the employee should be prevented from participating in it. If the university accepts the award and permits the employee to participate, and there is yet is a way that Bayh-Dole can be complied with, then that is the route the university is constrained to take. Not try to change 30 years of practice and implement the AbyNormal Patent Act, as AUTM and others are attempting, forcing previously valid agreements with reliance and consideration to become void at the university’s whim. And even if one succeed in doing that, there are all sorts of other consequences that are worse and worse. More later, I’m sure.
Another thought experience. Imagine if you will a university faculty member, who we will call F. F’s university requires F to agree to a patent policy that used to say, F promises to assign title to all inventions that the university has an interest in to the university. Now the university says, we will fix this hole and require F to make a present assignment of all future inventions F makes in which the university has an interest. This last bit is necessary, because states have laws that make an obligation to assign in an employment agreement void if it over-reaches the reasonable bounds of the employer’s interest.
This should fix the hole, says the university, thinking that the hole is in *their IP policy* (which better than thinking it is in an attack on Bayh-Dole).
Let’s see. The university has a policy that allows, even encourages F to consult with industry. In this consulting with a company, F agrees to a present assignment of future invention to this company. Is this a problem?
Generally, no. There is no collision between the present assignment to the university and the later present assignment to the company because the scopes do not overlap. The university has agreed that F can consult, so long as the work does not involve university facilities, is not commissioned by the university, and is not the subject of extramural research contracts that require other disposition of IP. The consulting doesn’t do that, so the scope is good, the present assignment is good, it’s all good.
The key is: the scope of the present assignment to the university has to change with the conditions under which the university permits personal consulting. If the IP policy said, no consulting allowed, or no IP disposition in consulting, that would be another thing. But most university IP policies, for very good reasons do not, even cannot go there. Again, the key is the scope of the claim, not the mechanism of the claim. It really doesn’t matter, looked at this way, whether it is a promise to assign or a present assignment of a future interest. What matters is the scope of interest in that commitment.
So far, so good. Now consider what happens if F likes the consulting work a lot and decides to apply for a federal grant to continue the work at the university. F writes a proposal, and the proposal is in the scope of the present assignment to the company. The grant is awarded to the university, and now it is time for the university to follow Bayh-Dole. What is required?
Right! The university must require a written agreement from its employees on the grant, namely F, to protect the government’s interest. That interest is (1) in timely disclosure *to the university* in reasonable detail and to execute papers necessary (2) for patent applications to be filed and (3) to establish the government’s interest in the inventions.
What does F have to do to comply? To meet (1), F will have to make sure that the company waives any non-disclosure arrangements and that the university accepts a non-disclosure agreement. (2) is easy. The written agreement in Bayh-Dole doesn’t specify who provides the papers to sign. (3) Would require that whoever is taking title, if it is not the government, has to comply with the government’s interests as these may be established by Bayh-Dole. There is nothing in Bayh-Dole that precludes title in inventions made in the research passing to the company.
Of course, there may be other things that would stand in the way. Agency approval would be required if the company did not have invention management as a primary function. But since it was smart enough to get a present assignment, and astute enough to hire F as a consultant, that sounds like it’s okay.
Or the university could refuse to assign its Bayh-Dole interest to the company. That’s not a function of Bayh-Dole; that’s a function of the university’s position. But if the university found itself in this situation, and the only way to comply with Bayh-Dole was to assign Bayh-Dole obligations to the company, and the university did so, then Bayh-Dole *would be fine*. It is an outcome that is not only possible under Bayh-Dole, one might even think it would be desirable! The university already have the commercialization partner, has it without having to do a license, and the partner has taken on Bayh-Dole compliance.
Of course, this would produce only a cool industry partnership, which is fluffy and uninteresting, even if it is an express objective of Bayh-Dole, rather than a tough sounding, royalty-bearing license for development of commercial product.
What about university IP policy? Even if the contractual obligations can be satisfied, isn’t the research using university facilities? Yes, of course, but again, the issue is whether the university in accepting the grant and permitting the services of F has waived that claim on its facilities use. The government, after all, is obligated to pay the direct and indirect costs of the research and so presumably is covering all the costs of those facilities. In some cases, yes, there may be cost sharing. For another time. In essence, the university isn’t out anything doing the research, or if it is, it is its own durn fault for not negotiating a proper F&A rate or finding a way to damp its facilities costs.
By the way, I have yet to see a university IP policy that claims patent rights on the basis of significant use of university administration. It’s always facilities. Too funny.
But what about a more general claim to IP under university policy, and that present assignment of future inventions. First, there is the problem of scope. The scope that controls was created by the second present assignment. In allowing the consulting, the university cedes scope to the company’s present assignment. In allowing the research to continue in the university, the university confirms that scope is still in play. Otherwise, the university would not allow the research. If F cannot agree to the Bayh-Dole written requirement, then F cannot be involved in the research as an employee of the university. That is the university’s obligation. If F can agree to the written requirement but the outcome doesn’t give the university title to inventions, but there’s a way that Bayh-Dole is satisfied, then that’s a university problem, not a Bayh-Dole problem.
Of course, in this, the company may not want to comply with Bayh-Dole, what with all royalties going to scientific research and education. Maybe the company would cut a deal to waive its present assignment and have instead a license that limits its exposure.
Anticipating such a thing, the company may insist on a research relationship with the university going into the research using federal funding. That relationship could provide for the license, the consideration, and all things juicy.
Or, not wanting anything like this, the university could prevent F from participating. Even if there were a pathway under Bayh-Dole for compliance, it is up to the university to decide to do it. It can even refuse to accept the award.
What does not at all seem reasonable is the argument that the university can accept the award, knowing about the present assignment F has with the company, and simply argue that this assignment is canceled automatically by the operation of Bayh-Dole, and that neither F nor the university has any obligation to the company, regardless of the value the company has provided to F, and the reliance the company has made of the university’s acceptance of the scope of obligations as between the present assignment to the university and the later, and university-authorized present assignment to the company.
And what about university IP policy generally? Bayh-Dole does not acknowledge university IP policy. The only things that matter are stated in the Act. The Act doesn’t say, do these things this way unless you have your own IP policy and then do things however you wish under that policy. It doesn’t say you have to have an IP policy that tells people to do things to comply with the Act but throw in anything else you want at the same time and make your employees agree to that stuff, too, as a condition of being able to agree to the written agreement required by Bayh-Dole. Nope, none of this. One might say, under Bayh-Dole one does not need to have much of an IP policy, because Bayh-Dole is self-directing. It’s all there. The IP policy for Bayh-Dole is to do those things required, and don’t make anything that isn’t required as a condition of doing those things that are required.
I know, been through this already. It’s like training terriers–it takes a lot of reps to get there. Given how many things have been put forward repeatedly in the name of Bayh-Dole, perhaps it is a necessary counterpoint to repeat things in various ways.
We are working through what Bayh-Dole did and didn’t do in managing federally funded inventions made at universities. If you are with me, we did a “thought experience” (not “experiment”). Then I got all grumpy about rhetoric. But that’s out of my system now, and we can review. In the world before Bayh-Dole, with a normal Patent Act, inventions made in research are personal to the inventors, they owe the government title on request as a condition of the award, and they convey title by a written assignment as provided by the Patent Act.
Universities gain the benefit of Bayh-Dole. They believe that exclusive patent positions are necessary to work the Little Linear Model. While the government wants the personal title one has, the university wants all the title there is. What’s at dispute is whether Bayh-Dole reflects the government’s interest, or the university’s interest. That is, in electing to retain title, are universities accepting the government’s interest, or are they claiming that the government’s interest is that they do whatever they will, with the rights they want, not the rights that the government otherwise would obtain as a deliverable?
The universities are making the argument that Bayh-Dole sneakily changed the Patent Act (the abynormal Patent Act), and changed the government interest to be whatever the universities want, and what they want is to make money licensing rights to monopolists, so that is the secret purpose of Bayh-Dole. (Oh, if this is twisting it–go back and read Dag Wags Dog, or Doping the System). The changes in the Patent Act were so sneaky that the Patent Office never realized it, universities never realized it, and only now is it coming to light how abynormal it all is. All this to try to *defend* Bayh-Dole.
Now, some more thought experience. Let’s say that Bayh-Dole aims to normalize the delivery of title and management of university requests to step in and take over what otherwise would be the government’s to dispose of. The aim then would not be to change things, but to make the treatment of these things in federal funding agreements–grants and contracts–uniform.
For this to happen, Bayh-Dole has to preserve a default that if the university doesn’t step in, then agencies get the opportunity to request the title as usual. This is the pathway if the university does not step in.
And it has to provide a protocol by which when a university requests to stand in for the government, it gets to do so *automatically*. That is, it does not obtain title to inventions *automatically* (since the government didn’t), but rather that the agency does not have to review the request with diligence before taking an action to allow it. The notice of election to retain title serves this function. It means: rather than title passing from inventors to the government at the government’s request, the university may make that request, and dispose of title according to the protocols of Bayh-Dole.
How does Bayh-Dole manage these two pathways? It really is genius. Bayh-Dole requires universities to require written agreements from employees–the research employees–to protect the *government’s* interest. That interest is in obtaining title (if need be) or in obtaining a royalty free license (otherwise), or in working directly with the inventors (if it comes to that), and in any event seeing that the objectives (all of them, not just one) of Bayh-Dole are met. That written agreement says, report inventions timely in reasonable detail, and execute all papers necessary as requested to allow patent applications to be filed and establish the government’s rights in the inventions. That agreement is personal to the investigators, to the inventors-who-may-be. That agreement reflects the special performance due the government. That agreement is specifically to protect the government’s interest, and to allow the government to establish its rights in inventions. The university has this obligation, but it is not about the university’s benefit or the university’s interest in inventions, patents, money, policy, compliance.
It is all there in 37 CFR 401.14(a)(f)(2).
It is by this written agreement that everything having to do with title happens in Bayh-Dole. If the university does not elect to retain title, then this agreement is the means by which the government may request title and get it as a deliverable. And if the government chooses not to request title, then no title has changed hands. It is with the inventors, and the government may allow the inventors, at their request, and with some delay to talk the tech transfer office down from the ceiling, to retain title to their inventions, per 37 CFR 401.9.
If the university does give notice that it elects to retain title, then what it gets is the benefit of that same written agreement. The university may request from its employee-inventors the title they otherwise would owe to the the government. That is what retaining title means. It ought to be very clear to anyone not paid to murky it. Retain as keep it back from passing to the government, rather than retain as in hold it automagically under the abyNormal Patent Act even before electing to retain title, only to have it spring up automagically with the inventors if one doesn’t elect title and the government waives its interest in title.
What the university gets is what the government would get. What is not clear at all is whether anything else that the university claims under its own IP policy it should get. Bayh-Dole says what the deliverable is, as a matter of federal law, and what is to be obtained and how disposed. Does it not pre-empt university claims on title? If Bayh-Dole says, if the university does not elect to retain title, and the agency chooses not to require the conveyance of title from the inventors, then the inventors may retain that title. That’s the federal position. It doesn’t say, the inventors may hold that position, but if the university’s IP policy claims the rights anyway, then so much for 401.9, and we have a second right of refusal, better than the first, since under a post 401.9 claim, as soon as the university gets title (under its own IP policy claim), it owes the government only what is assignable to it by the inventors–a much more limited set of obligations than under its own election to retain title via Bayh-Dole.
We may underscore: Bayh-Dole allows a university to elect to retain title. It does not say title vests upon invention with the university. It does not say that by electing to retain title the university has title. It does not even say that by electing to retain title the university must actually obtain title (since the university can assign its interest in obtaining title to, say, a research foundation, as a number do) and never hold title at all. Electing to retain title cannot possibly mean that the university already had it, automagicaly, under the abyNormal Patent Act.
Is it clear now? Is it clear the angle AUTM and others are taking? Why I call it the dag wagging the dog? Rather than look straight at the law and how it operates–which folks should know even by looking at their own practices–they drop out whole sections of the Act, such as the written agreements and the nature of electing to retain title. Instead they reason from bits of wording, and would turn the whole Patent Act on its head (and put it in a bottle labeled AbyNormal no less), and cancel otherwise valid private agreements after the fact in order to deal with their problem having to do with personal disposition of rights. Confiscate those rights! Make it as if they never were!
I call this position inventor-loathing. It’s short hand for failing to respect the idea that invention is a personal right established and guaranteed by the Constitution, and we ought to frame our practices around that, so long as we honor the Constitution. So I don’t give a rat’s ass about administrative efficiencies or presumptions or what would solve some bureauklept’s desire for money or satisfaction or power or convenience. Get it right, honor the inventor, think hard about those beloved processes and charts with boxes and arrows before you think invention is a commodity, that applying the process is better than any other prospects the public might have for promoting the practical application of each invention.
Okay. Next.
In semiconductor theory, adding a few impurities into the system may actually improve performance. It’s called doping. Yeah, there are overtones. With regard to research inventions, what happens when the impurity is the university administration seeking title, rather than the government? Does this impurity improve the system? That’s the premise of Bayh-Dole, no?
Based on the thought experience, it is clear that patent rights in research would be personal, but for, potentially, the impurity of university administrative claims on title in place of the government’s.
The university has, apparently, a different purpose in obtaining title to patent rights than the government may have in seeking personal title as a deliverable. The Little Linear Model says, exclusive patent rights in research inventions are the best, highest value, if not the necessary rationale to induce investment to create new products to benefit the public, and prove Bayh-Dole successful. Anything else, in this formulation, is “just a tax”. Exclusive licenses are the reason for being. Monopoly is the ground truth of the patent system.
Thus, non-exclusive licensing cannot be considered, and certainly is not best practice (so: forget, or at least suppress or badmouth standards, platforms, libraries, testbeds, consortia, commons, internal practices, open source, crowd sourcing, user-based innovation, and research exchange–makes one wonder why Bayh-Dole proposed collaboration with industry when that apparently only means “collaborate with industry by licensing patent rights exclusively to a single company that will monopolize the market, exclude everyone else who would otherwise want to work with our lab, and sell to consumers at premium price points so as to maximize the return on investment made possible by the patent position.” This is what is meant by “commercialization” in the Little Linear Model.)
Pretty compelling focus for an impurity in the research system. It’s at best the echoes of a sell job. More likely, it’s something repeated so often it is necessary to continue to repeat it, regardless of what one might believe.
So one can expect that there is a lot of pressure to come up with “data” that supports the proposition that Bayh-Dole is successful at producing the outputs of the Little Linear Model. Everyone knows the answer, politically, always has to end up that Bayh-Dole is a success, even there’s only one pull quote from one weekly news magazine from years ago to hang one’s hat on. Any other answer is not with the program. Truth is not on the table. Or, if you wish, the truth on the table is the sincere aspiration to make Bayh-Dole = success = Little Linear Model true. The effort then is to define what success is, because that will make “true” a lot easier to get to, politically. If it’s not licensing to industry, then it is making companies to sell to industry. Worthy stuff. The thing, though, is not what is happening, but what one says about why and how.
This situation leads to the absurdity that university technology transfer is never sub-optimal, because it cannot be, politically. At each point, it is always optimal. The absurdity then is that if university technology transfer is not sub-optimal, then it must be optimal, and thus really doesn’t have any place to go with improvement. All it can do is expand its “best practice” everywhere so more and more universities can participate in the never sub-optimal practices already developed.
One might see in this assumption a desire to make everything related to patent title to feed into an exclusive university position on which to trade to set up monopoly positions with commercial partners. That’s the essence of it. Making the feeding more efficient will improve the metrics. Nothing else in the Little Linear Model is even close to being as sweet as royalties from a monopolist. So, more disclosures, more patents, more faculty training, and better agreements to knock up the monopolists. If one is going to have monopolists, and that’s the assumption of this kind of patent commercialization, then best if they are sufficiently cornered that they pay for the party.
One might also see in this how difficult it might be to find tamable monopolists, and how “negotiation” fails to get at the discourse necessary to accomplish this. It’s like a bunch of cowboys sitting on the wood rails trying to talk a bronco into being ridden. “C’mon horse, stop buckin’ and accept it!” And then some more folks come along–management advisers, say–listen to the cowboys for a while, and come up with the idea, hey, we can make that chatter *more efficient*. Say something once, why say it again? And this is an advance in negotiation practice.
This is what this particular version of commercialization requires. It is the sweet spot, it is held up as the most important thing, all the success stories that matter make the monopoly position so important, even if it wasn’t really for, say, Cohen-Boyer or the MPEG patents. It is not that Bayh-Dole doesn’t permit this–it does–but that this is what Bayh-Dole intends, the best thing, the only thing, the necessary thing. Which isn’t true.
When things are presented this way, some folks will object, saying, but that’s not what we mean, you are twisting things. My response, from having had to deal with this kind of thing for years and years, is: you are right, it is not what you mean, because in reality you don’t mean anything. It is spout. That’s all it needs to be. There are bits of true in it. That’s enough. Those who are in–they know why you do it, and those who don’t know, generally don’t have a clue, so it’s all good.
What you mean is that you want to be seen as knowledgeable, successful, in control. What you say is dedicated to that purpose. You don’t want someone to take what you say literally, or to ask what it pre-supposes or implies. There are no underpinnings. Just a set of success stories that are to end discussion of the point. All understandings of what you say have to be corrected if they slip toward “this is goofball” or “that is unsettling” or “I don’t see the point”. Only the non-clueless can see how the system is doped.
That’s a problem, folks in the university technology transfer business who don’t mean anything and just spout buzz words and attack whatever doesn’t conform. Folks doing this don’t have any understanding what underlies their discourse. They don’t have to. It’s the discourse everyone uses, about technology transfer, commercialization, innovation, economic development, success. These folks don’t like anyone pointing it out. I understand. It’s not the funnest thing in the world to do, believe me. There are few conversion experiences.
No matter what it is that gets pointed out, it’s twisted, it’s lacking in experience, it’s just meanness. One sees this in the recent AUTM publicity campaigns in Stanford v. Roche and use of innovation agents to deploy research findings. Why? Because the point of the discourse about this sort of commercialization is to confirm that things are successful as they are, with the folks who want to control the public discourse on innovation. Nothing more.
As far as they are concerned, they *already understand* what technology transfer is, how it is done, and who should do it. No mysteries left there. All that is needed is to do it. And it would be done, but for lack of resources, ever present funding gaps, lack of cooperation by industry, the lack of innovation capacity, need to change university culture and train faculty to accept the system, and of course the wasted effort where things could be standardized. It couldn’t possibly that the whole idea of the Little Linear Model is so poorly adapted to the innovation potential of community, including its research activities, that it operates maybe a handful of times in a decade, over billions of dollars of research funding. That may be enough to spill lucre of royalties over a given university for 20 years, but it sure as heck isn’t what I say the purpose of Bayh-Dole is all about. And it sure as heck isn’t to make the effort to improve outcomes from one handful a decade to two, if the world would just change in its entirety and prove the model out.
This is the sort of rhetorical environment in which folks invoke Bayh-Dole. If one believes that the purpose of Bayh-Dole is to provide universities with exclusive title to patent rights so they can fuel the next generation of monopolists, and that’s the vision of public policy that binds research support to innovation, then one can see what’s at stake in suggesting Bayh-Dole doesn’t do that. One is not only challenging a public claim made on Bayh-Dole but also the substance of the claim made to the public with regard to the worthiness, if not intellectual honesty, of university technology transfer.
If this situation can be untangled at all, it goes something like this: the transfer of technology, with or without IP, is a worthy thing. Teaching and enabling new things in the world arising from research or benefiting from research are worth activities. Those that commit time and effort and resources to this are worthy folks. I have been part of that effort, continue to be, and am proud to be part of it.
All that has nothing to do with being complicit in making one model (the Little Linear Model) stand for all the ways that technology transfer comes about, nor with whether any one organization or university or coterie of universities speaks for all the varied circumstances and efforts to engage research and community and enterprise. Nor with the idea that Bayh-Dole exists for the primary if not only purpose of delivering patent rights to monopolisits to create products sold at a premium to be shared with universities, as if monopolies based on research findings are the primary agent of innovation in the public interest.
Yes, Bayh-Dole asks that the patent system be used to promote the practical application of federally supported inventions. Yes, Bayh-Dole includes one objective among others that anticipates commercialization in the form of products sold in a market place. Yes, Bayh-Dole includes a provision that royalties be shared with inventors (interestingly, as a *cost*). But it does not mean that Bayh-Dole’s secret and best agenda was to have universities grab up title to all inventions, require any licensee to sell product, or try to make money from every such situation.
Okay, had to get that out. Next, how university doping of research deliverables comes into play in the discussions around Bayh-Dole and present assignments.
