The Legal Context of University IP, Part 2 Revisited

In 2010, the National Academies and the National Research Council published a commissioned a report–The Legal Context of University Intellectual Property and Technology Transfer by Sean O’Connor, Gregory D. Graff, and David E. Winickoff.

Here are comments on the findings of Part 2 of the report, “Inventions and Bayh-Dole.” Open up a new window for the report and follow along.

Finding 5. Bayh-Dole defines a subject invention: 1) is or may be patentable; 2) owned by a contractor; 3) conceived or first actually reduced to practice under a funding agreement. 37 CFR 401.1 makes clear that the scope is the”planned and committed activities” and “does not diminish or distract” from these activities–not just anything done within the general scope of a funding agreement. If the planned and committed activities get done, it does not matter what else happens with “closely related” work. Bayh-Dole does not “cover” data or materials. It does not have anything to say about what might constitute an ownership claim on an embodiment of an invention or a best mode of invention or data supporting claims made regarding an invention. Nothing in the law has anything to do with research data or materials. This isn’t a finding. The funding agreement for universities includes provisions for data and copyright along with subject inventions. 

Finding 6. This finding repeats a confusion about inventions that’s made the rounds in technology transfer offices like a case of mononucleosis. One kiss and you are laid out for months. Here’s the deal: a patentable invention is “made” when it is both conceived and reduced to practice. Have only conception? Not a patentable invention–no demonstration that the idea functions in practice. Have only reduction to practice? Not a patentable invention–no recognition that it is an invention. Reduction to practice takes two forms. Constructive reduction to practice involves documenting the invention so that anyone with ordinary skill may practice it. Actual reduction to practice means building or doing what’s conceived and showing that the invention works. But to have a patentable invention, you have to have both conception and reduction to practice. Keep this in mind.

If you have both, then you have an invention that is or may be patentable. Now look to see if either the conception or the reduction to practice was a “planned *and* committed” activity of the project. There should be something in writing that matches the work–to solve a problem, to build a device. If so, then you have an invention within scope. If not, then look to see if the work–the idea, the build–“distracted or diminished” from the planned and committed activities. If so, it’s within scope. Now: Do you own it? If so, it’s a subject invention. If not, then not a subject invention. A subject invention has to meet all three conditions–patentable (conception + reduction to practice), within scope (planned and committed, doesn’t diminish or distract), owned by contractor.

Now, to the mononucleosis. Some university tech transfer folks, notably at the University of California, where I worked for six years, insist that conception or reduction to practice means that Bayh-Dole can grab an invention made prior to or outside of a federal funding agreement if the invention is first built with federal funding. They imagine that an invention can be made by conception alone–a free-floating idea. Then any federal grant under which something gets built will appropriate that invention.

They reason that a faculty member could consult with a company, assign to the company an invention conceived in the consulting, and then work with federal funds to build a first version of the invention. Kahpow!, the invention becomes a subject invention and the university must grant a license to the government–but the university doesn’t own the invention, so it will be in breach of its obligations or the university automatically owns the invention (another even worse bit of nonsense) and the faculty consultant’s assignment to the company is retroactively made void. This was almost the scenario in Stanford v. Roche and the Supreme Court tossed the argument–made by scores of universities, all deeply afflicted and supported by hordes of lawyers who (at best) could not follow the elements of the situation and (at worst) did not care.

There are variation on this doomsday scenario. A company could sponsor research in which an invention is “made” because it’s conceived and reported, and thus is obligated to the company under some pre-arranged licensing protocol, but then, say, months later, another researcher could build what’s invented, first actually reducing it to practice and woohoo!, the invention becomes a subject invention, subject to federal ownership and licensing claims. If the company sponsor had been offered a first right to negotiate an exclusive license, then that promise is breached when the invention becomes a subject invention.

Or, even if a patent application has been filed, or even if a patent has issued on the invention–so, years after the disclosure and license to the company sponsor, someone could first build the invention with federal funds and utterly change the nature of the invention, forcing the university to grant a non-exclusive license to the federal government. It gets worse, with the same reasoning, the someone building the invention for the first time does not have to be at the same university where the original work was done. Someone at Stanford could build for the first time an invention made five years earlier at Berkeley and poohaw!, Stanford has control over the invention and the government must get a non-exclusive license.

The upshot claim for university patent policy then is that the university must own all inventions, whether federally supported or not, to make sure that the university can comply with its obligation to grant a license to the government. Taking it further, who knows whether something new is an invention, or if the invention is patentable, or if a patentable invention is a subject invention. Subject inventionship could happen later or elsewhere, and a university has to be ready, has to make its risk minimal. Thus, a university has to own everything–anything new that remotely might be an invention, just in case. Otherwise the university will likely violate federal law or end up double licensing.

This is plain goofiness. In 2010, even, the authors of the report, with a some careful reading and putting on their reasoning hats, could have seen through the problem and offered some decent legal commentary. If the invention was not made with federal funds (that is, patentable invention in hand, either conception or first actual reduction to practice is within the planned and committed activities of a funding agreement), then a subsequent build of the invention is not a reduction to practice that matters for patent law or for determining whether an invention is “subject.”

You invent and get a patent. I read your patent and build it before you do. Hey, I’ve first actually reduced it to practice! But I can’t possibly be a co-inventor. I don’t have any patent rights to assign. But the goofy university administrator argument is that if federal funds are involved, I must somehow be a co-inventor, I do have patent rights, and the university must have them to comply with the law–or worse, that the university owns the invention outright, whoever thought they owned it previously by operation of law. You could ride your unicorn on a magic carpet if you believe this argument.

The report adopts this same argument for itself, having stated it as a problem:

Contractors have received federal funding after a particular patent has been filed for, or issued, and assumed that they owned the invention and patent free and clear of the federal government. However, if the invention had not yet been actually reduced to practice, and was then first actually reduced to practice under the new funding agreement, then government rights under Bayh-Dole subsequently attach to the initially “free and clear” invention/patent.

Now you can see the illogic. Filing a patent application constitutes reduction to practice. A later first build cannot possibly be a first actual reduction to practice. That first build is not the reduction to practice that matters. The argument is so naive and unfounded that one wonders how the authors of the report could accept it at face value. It would seem the authors *want* to find arguments to support university ownership of faculty work and thus don’t challenge the goofy arguments put up by university administrators.

Finding 7. The current standard patent rights clause at 37 CFR 401.14(a) is *not* “quite similar” to the IPA. Bayh-Dole guts public accountability. (No review for adequate patent policy; no requirement to prefer non-exclusive licensing; does not require reporting on use; exempts any such reporting from public disclosure; impossible march-in procedure; nothing to prevent restrictions on publication by patent owners). Bayh-Dole does not require institutional ownership, narrows the scope of interest, and allows exclusive licenses of longer than 8 years. Bayh-Dole eliminates flexibility. Bayh-Dole requires royalty sharing. See this comparision. The only similarity worth noting is that both the IPA and the standard patent rights clause authorized by Bayh-Dole are in federal contracts with universities–not a matter of law, and not contracts with the people actually involved in the research. May as well say that an elephant and a mouse “quite similar.” Well, yes, they are. But it’s wasted breath.

Finding 8. Yes! Something the report gets right. Bayh-Dole does not mandate contractor ownership. But then the report screws things up immediately. This is bad, folks:

Initial title allocation can go either to the federal government or the contractor.

Bayh-Dole does not mess with federal patent law. Initial title is with the inventors. Bayh-Dole does not say anything about initial allocation of title. The standard patent rights clause does not say anything about initial allocation of title. Title can stay with the inventors, or the inventors can assign title to their employer, or the inventors can assign title to an invention management organization, or the inventors can assign title to the government, or the inventors can assign title to whomever they please. If the university gains title, then it is restricted in how it can assign title. Under the standard patent rights clause, initial title is with the inventors and is allocated nowhere. Once an invention is reported, it is up to the inventors whether to assign to an approved invention management agent. If they don’t, then it is up to the agency whether to request assignment from the inventors. If it doesn’t, then the standard patent rights clause at 37 CFR 401.9 applies, and that does not involve allocation of title.

(There is actually more to it–inventors are not contractors, so what they invent is not a subject invention, not subject to reporting, not subject to uniform treatment required by Bayh-Dole. The standard patent rights clause patches this problem in the law by requiring universities to bring potential inventors within the scope of the definition of contractor. This is the (f)(2) agreement that is directed not to assignment of inventions to the employer-contractor, but embodies a conditional delegation of responsibilities to the potential inventors as if they were small business concerns operating under a subcontract. If a potential inventor invents, she becomes a contractor, the invention is subject, and title is with the inventor-contractor. If the inventor-contractor does not assign, then 37 CFR 401.9 becomes the patent rights clause in the funding agreement with the inventor-contractor.)

At its simplest, the standard patent rights clause says, “if you manage to get title from your inventors, you can keep it, but some of us don’t trust you with it, so you will have to jump through hoops, but others of us have made the hoops work only for show, so aren’t we sweet?”

Bayh-Dole thus became a dare for universities to take as much as they can from their faculty under the ruse of complying with federal law. At the same time, Bayh-Dole’s stated objective to lower administrative costs results in huge new costs for universities that take up the dare. In terms of administrative cost control, Bayh-Dole is again a dismal failure. All that paperwork for tens of thousands of inventions that go nowhere–but have to be reported, elected to retain, updated for patent bars, licensed to the government, patented–with a federal funding statement, and reported as to status.

Consider how much simpler it would be if universities did not claim ownership of anything made with federal funds–and therefore greatly reduced the number of subject inventions for which they had to file paperwork. Then it would be up to agencies to decide what to request and what to leave with the inventors. If the inventors then pursued patenting, universities could look at the equities (under the policies they had for decades before Bayh-Dole) and allow a faculty committee to recommend what if any financial interest for the university. A university would then only administrate the inventions that faculty chose to patent, and would have only the financial interest (liability, benefit) recommended based on a faculty finding of university contribution. This, of course, is not a finding of the report, as it would reduce the need for attorneys working for the university.

Finding 9. After Stanford v Roche it is clear that Bayh-Dole does not have anything to do with invention ownership by act of law. The report seriously misstates the effect of the appeals court ruling in Stanford v Roche as the case made its way to the Supreme Court. The appeals court found that a prior assignment meant that a later attempt at assignment was not effective. In the case, Stanford had approved the prior assignment. The prior assignment was entirely acceptable under Stanford’s patent policy. Stanford then tried to invoke a version of the goofy patent argument discussed in Finding 6, that Bayh-Dole gave the university ownership of the invention, voided the prior assignment, and gave effect to the later assignment. The Supreme Court tossed the university argument. The upshot is not that universities must have a present assignment that grabs all future inventions the moment they are made. The upshot is that universities do not have to take ownership of any invention to comply with Bayh-Dole or the standard patent rights clause. The upshot of both the appeals court and the supreme court rulings is that ownership is not a matter for Bayh-Dole (and its standard patent rights clause) until the contractor has ownership.

The deeper foolishness–and this shows up a deep lack of legal awareness–is that the present assignment as used by companies, and the analogous present assignment used in copyright transactions such as publication agreements, is bounded by a scope definition tied to the contractual obligations of the potential inventor or author. That is, a present assignment cannot be unbounded in an employment relationship. A number of state laws directed at exactly this issue, including California and Washington state. The requirement to assign must be within the scope of employment, or use employer’s facilities, or be in the business direction of the company. For a university, “employment” of faculty for the purposes of either patent law or copyright law is not a simple matter of asking if someone is “employed” for, say, tax purposes. “Employment” means something in the cases of patents and copyrights.

In copyright, work for hire does not arise because an employer is an employer, but because the employer meets the conditions of author–if the employer does those things that characterize authorship, then the employer, not the employee, is the author. An employer can decide such things as whether a work of authorship meets its requirements for being complete, and so can in this sense author the work. The problem is deeper for inventions, however. While a company can direct the fixing of a work in a tangible medium of expression, it cannot conceive of an invention. It has no mind. It cannot invent. The company can only require assignment of inventions. The issue for public policy, and for employment law, is what is the proper reach an employer may have in demanding, as a condition of employment, assignment of an employee’s inventions. The Supreme Court in Stanford v Roche reaffirmed that employment alone was not sufficient to establish an employer’s interest in an employee’s inventions. There has to be a patent agreement.

Now we come to how universities handle the patent agreement. Most don’t. Instead they have a policy statement. And then in a fit of cluelessness, they add to their policy statement that it applies to employees, students, visitors, and even non-employees. That is, somehow a policy statement that delegates authority and authorizes use of resources and sets out procedures that employees and the public might rely upon becomes instead a binding contract under which the university obtains ownership of most any invention it wants.

There are various ways of breaking down the arguments supporting the idea that administrators can take ownership of inventions this way. One is to examine what employment means for university faculty. Faculty are appointed. A university does not control what a faculty member studies, where the faculty member does her research, who she collaborates with, how she conducts her research, when and how she publishes her findings. The faculty member does not conduct research for the benefit of the university, nor have any obligation to the university to account for time spent in research or to report findings to the university for review, approval, or university use.

For the purposes of most all faculty research, a university is not the employer. Faculty may well be employed. But not for the research that a university patent policy is created for. The scope of employment for a typical company employee may be drawn as broadly as the business of the company. But the scope of employment for a faculty member is not even as broad as the faculty member’s appointment, or field of expertise, or even what the faculty member chooses to study, or even what of that studying is done at the university. The scope of employment is those activities over which the university is the master and the faculty member is the servant–where the university requires, directs, and approves the work of the faculty member.

Thus, a present assignment added to a university policy, or even put into a separate document that references university policy, and which claims all future inventions within the scope of employment (or the scope of the university policy) still leaves ambiguous just what is meant by a faculty member’s *employment*. If the statements of academic freedom found in most every university policy are not void, then faculty employment is narrow, and the present assignment applies only to those areas in which the faculty member is *employed* by the university.

Now turn to federal funding agreements. Bayh-Dole and the standard patent rights clause refer to employees–and indeed limit federal agency interest to the work only of employees, not visitors, contractors, volunteers, students. But Bayh-Dole does not define employee–another shortcoming of the drafting. The confusion gets worse because small businesses were added to the law to make it politically more acceptable. “Employee” is nearly okay for small businesses, except that owners and directors of a small business may be involved in day to day operations and not be employees of the business. But “employee” is only jargon for faculty working on a project with federal support. Even the implementing regulations for Bayh-Dole recognize that a faculty member is to be treated as a small business concern if he doesn’t assign invention rights–not as an employee of a small business concern!

For federally funded research at universities, where faculty initiate the project idea and apply for support, faculty are in fact subcontractors, not employees, when it comes to their inventions. That’s the recognition of the (f)(2) agreement. That’s what makes Bayh-Dole, that awful patchwork of drafting, actually hold together. That the employees who might invent are for the purposes of their inventing subcontractors. Whatever a university might want out of them, it cannot coerce it as a condition of employment or as a condition of federal funding. The (f)(2) requirement trumps both. Very clever. No wonder the folks who want to dominate faculty work–call it fascism if you must–have refused to implement the (f)(2) agreement. How odd for all the legal planning, that no university lawyer has the courage to require compliance with this part of Bayh-Dole (patent rights clause) compliance.

The result, then, is that present assignments imposed by universities on their faculty generally fail for lack of definite scope. They amount to an agreement to agree–or, put another way, faculty are told they have assigned whatever an administrator decides they have assigned. If a university policy requires disclosure and review before a determination on ownership is made, then adding a present assignment makes a confused nonsense of the policy. Either there is a review and determination prior to assignment or there isn’t. It cannot be that the policy is turned on its head, so that the review and determination is what to assign *back* to the inventors (generally, nothing, never). The only trouble is that universities have unlimited budgets for legal fights, and faculty typically do not. It costs somewhere north of $150,000 for a university faculty member to prove administrators wrong. University administrators therefore can operate pretty much with impunity. Another reason why a primary role for university IP policy should be as an inventor’s bill of rights–restraining university administrative powers, not enlarging them.

Finding 10. Now we come to the (f)(2) agreement. The finding here simply ignores the substance of 37 CFR 401.14(a)(f)(2):

While the language of the regulations does not expressly state that universities must require employees to assign their inventions, it would be very difficult to craft a patent agreement that leaves title with the employee-inventor while protecting the title allocation, licenses, and other rights of the federal government and university as required by the Act and the implementing regulations.

So very much is wrong with this statement. Yes, it’s prior to Stanford v Roche. But it was stupid then, and in any event, no one has bothered to go back to revise or retract it. Bayh-Dole does not require a patent agreement. Bayh-Dole does not impose anything on inventors or universities as a matter of statute. Bayh-Dole applies to federal agencies. It requires agencies to use a standard patent rights clause. The patent rights clause as it may be modified is used in funding agreements. The agreements convey obligations to the universities who have agreed to support faculty who have been released from other duties to work on a project they have proposed to the government. The faculty involved are free of obligations to the government.

The (f)(2) agreement changes this. It does so in the same manner as the subcontracting provision operates. The university delegates specific work to the potential inventor, and as a condition of that delegation does not have the right to claim ownership of what the potential inventor might invent. See the subcontracting clause at 37 CFR 401.14(a)(g) for the language. The (f)(2) agreement is not the “patent agreement” that the report so dearly wishes for and fusses over. The (f)(2) agreement makes potential inventors conditional contractors–if they invent, then they are delegated by the university to report the invention, to sign paperwork (as may be requested by anyone with standing to request) to allow patent applications to be filed, and to sign paperwork to establish the rights of the government in the invention–that may be a non-exclusive license or assignment of title. These duties are material to the federal contract. The university in delegating these duties cannot at the same time un-delegate them by policy. The (f)(2) agreement trumps university policy–both because it is in its nature a federal contract and because the funding agreement requires the university to require it–it is not imposed directly by regulation on potential inventors, it arises because a university requires it, and therefore the university cannot require anything different.

The (f)(2) agreement is not “very difficult.” It is simple. One sentence. Put it on the form that anyone signs to accept federal funds from a grant account. Why can’t legal planning for technology transfer just do this? Oh, yeah, because it’s all so friggin’ complex, once university administrators and their legal advisers have had their way.

The provisions of Bayh-Dole do not envision any initial title allocation. It is not a matter of dividing stuff up between the contractor and the federal government. There is not even “the” in contractor. By the time a university has implemented the (f)(2) agreement, there are potentially many contractors. Bayh-Dole imposes on federal agencies a uniform practice of invention management. There is no “de facto requirement for universities to secure subject inventions from research personnel.” That was nonsense then. And it’s clearly not the law after Stanford v Roche. The only thing that matters in the standard patent rights clause is the (f)(2) agreement. It’s one of the very few things in the patent rights clause that’s not in Bayh-Dole. It is the flow down that enables the law without eminent domain.

Think of it this way. A federal agency could contract with a university for invention rights. It could (and did) require the university to obtain ownership of inventions the agency wanted. Or it could (and did) contract with a university for invention rights by contracting with each person working with federal funds to assign rights to the government. But these are contracts–they are voluntary.

Now someone gets the idea to make a law of it, because some agencies demand ownership and some don’t and this pisses off some external research foundations in Indiana and Wisconsin and Kansas. So the draft up a law, but now they are dealing with the power of the federal government. That means they are in the teeth of the fifth amendment to the US Constitution. If the government by law takes personal property for public purposes, then it must provide due process and just compensation. Same then if the federal government takes the property and hands it to another–say, a university–for disposition. So they can’t draft the law to do what the IPA did. They can’t require individuals by law to assign their patent rights to their employer. There is no work for hire in patent law. And patent law derives as well from the US Constitution, and clearly there, the rights are with inventors, not with the employers of inventors. An invention is personal property. US patent law affirms this. See 35 USC 261.

So they are screwed. All they can do is force the agencies to use a standard clause–that is, eliminate the flexibility that had existed under a series of presidential executive orders. Then delegate the responsibility for drafting the actual standard patent rights clauses to an agency (now the Department of Commerce). Then hope they could get that agency to re-introduce the requirement that every university had to get assignment up front from all their inventors as a condition of federal funding. I am told there was a huge cat fight over the drafting among the agencies. The result was not the IPA’s demand for university ownership, but rather, apparently, the (f)(2) agreement.

Why? The IPA demand for ownership would not work with the language of Bayh-Dole. Everything depended on an invention being within the definition of “subject invention.” But that meant the invention had to be owned by a contractor, and the inventors were not contractors. So how would inventions be reported if they weren’t subject inventions? How would a requirement to assign all subject inventions be implemented if all inventions made with federal support, before they were assigned, were *not* subject inventions?

There is not a de facto requirement in the law that universities must get assignment. The law is so badly drafted (yes, Senator Long was right) that de facto universities had no authority to demand assignment of anything but subject inventions, and no invention made under a federal funding agreement would be a subject invention. Oh, unless universities owned everything done by faculty outright anyway–the present assignment fetish. That would be the de facto requirement of the law–to require universities to take ownership of personal property as a condition of complying with the law, without the law actually stooping to state this. It’s still eminent domain, if it is the law. And if it’s not the law, then it’s also not de facto the law.

The advocates for university ownership (or research foundation ownership–as directed by universities with ownership) failed to get the law drafted the way they wanted. And they failed to get the standard patent rights clause drafted the way they wanted. They could not do with law what federal agencies had done with contract, and once they had the law, the federal government lacked the authority to do in contract what it had previously done. Screwed both ways. So instead they just declared that they had got what they wanted and explained it to the university licensing officers that the (f)(2) agreement was really a federal requirement that inventors assign, and that all a university had to do was to “elect” title in order to have it, and there you have it, for the authors of this NAS report to repeat uncritically but with plenty of sophistication years later. May as well title the report “Documentation of the Myths, Rationalizations, and Psychosis of University Technology Transfer Administrators and Their Hapless, Complicit Legal Counsel.”

Finding 11. Bayh-Dole does not establish an inventor’s right to share in licensing revenues. Bayh-Dole makes a conditional requirement on how agencies may permit a university to retain ownership of subject inventions. The university has an obligation to share royalties. It is not an inventor’s right to receive royalties. Splitting hairs? Not really. The royalty sharing provision only applies to university and other nonprofits retaining title. If a small company retains ownership, it does not have to share royalties. If the federal government obtains ownership, it does not have to share royalties. If the inventors retain ownership, obviously they do not have to pay themselves anything if they don’t want to. So the assertion of the finding is simply wrong.

The universities are subject to the patent rights clauses of their funding agreements. They are not subject to Bayh-Dole. The sole procedure by which university employee-inventors retain title is that they do not assign title to anyone else. A university has no mandate to require assignment of title, and there is no de facto obligation in the law or the standard patent rights clause to obtain title. The only thing left is how it comes to be that a federal agency has standing to require the inventors to assign title. The only thing that reaches to the inventors is the (f)(2) agreement. If that has been implemented, then there is a promise by the inventors to establish the federal government’s rights–but there’s nothing to indicate what those rights are. Even 37 CFR 401.9 simply proposes a patent rights clause for dealing with inventors, and that clause is only partially constructed. One might argue that federal agencies have only the right to request, but not require, assignment from inventors. If, however, inventors have also become contractors by means of (f)(2), then they would have the same obligation to elect to retain title, but now without the pre-approval granted to universities and to invention management organizations.

The rest of Finding 11 gets things ass-backwards. The standard patent rights clause envisions that inventors own their inventions unless and until they assign title. There is no obligation that an inventor assign title to the university-contractor or to an invention management organization. There is no obligation for the federal government to take ownership of inventions that inventors do not assign, and certainly nothing that vests ownership outright without an assignment. In this regard, Finding 11 is good–it takes an act of assignment to assign. Excellent. Patent law survives intact. The employee-inventor only needs title assigned back from the university when the university receives assignment, and chooses not to pursue a patent, or maintain a patent.

At that point, the standard patent rights clause does not indicate what may happen. University administrators, untrustworthy as they are in these matters, argue that they have to offer the rights to the federal agency, and cannot assign to the inventors. And there are provisions that require a university to convey title upon request to the federal agency–37 CFR 401.14(a)(d). The university has an obligation to notify the government, but does not have an obligation to offer title to the government. It is the government’s obligation to request title. If the inventor has been made a contractor via (f)(2), then the inventor would have these same obligations until 37 CFR 401.9 becomes effective, once the federal agency determines it will not request title.

A university may assign subject inventions to an approved patent management organization. That assignment includes the provisions of the patent rights clause as well. An approved patent management organization is any organization that has as a primary function the management of inventions. Or any other entity that the federal agency approves. So a university could assign a subject invention to a company set up by the inventors expressly to manage the invention without agency involvement, or directly to the inventors with agency approval. The university does not have to offer title back to the federal agency. It’s not in the standard patent rights clause. It should not have been found in Finding 11. It was, in fact, “repeated,” or “concocted,” rather than “found.”

So title starts with inventors, and stays with inventors until they assign. Inventors agreeing to (f)(2) requirements have obligations to report and patent, and if they don’t assign to an approved invention management organization, then the university will notify the federal agency that it does not elect to retain title, and then it’s up to the federal agency to request title from a contractor–in this case, the inventors, who through (f)(2) are also contractors. The “the” in “the” contractor is a most misleading word in the standard patent rights clause.

Finding 12. A tangle of technicalities for something that has rarely arisen. Depending on the status of the subject invention that has been patented by some entity, a federal agency could also march-in and require granting licenses and avoid the Government Use Statute. Given that most university research results in no meaningful actual damages to a patent holder, and the cost of a patent infringement litigation can easily run $500K to $1M, there’s not a whole lot of incentive for the worry of Finding 12. But it’s arcanely interesting. The real problem with all this is that Bayh-Dole makes subject invention status reports optional and excludes subject invention reports from public disclosure. Other than the federal funding statement in issued patents and patent markings on commercial products, how would anyone know what’s a subject invention, or what’s been developed or licensed or being sold? What a pile of crap! Every subject invention assigned to a patent management agent should have to be registered publicly, and that registry should have to be updated to show that title has been retained, that a patent application has been filed (one or more), that patent(s) have issued, that a patent has been exclusively licensed, and date of first commercial sale or use. Then we could see what patents were in play, and the effectiveness of Bayh-Dole’s genius fragmenting of federal research to competing speculative monopolists.

Finding 13. Yup. But this is just reciting 35 USC 202(c)(5) and 35 UCS 205. Not much of a finding. The big problem is that the federal government failing to report to the public creates all sorts of opportunity for exploitation, deception, and fantasy.

Finding 14. Yes, the Department of Commerce does have “limited rulemaking authority”–but the standard patent rights clauses, in particular, are at the heart of federal contracting with universities. What’s missing is that while Bayh-Dole is added to patent law, the thing that matters for university technology transfer–and the report is about the legal context for university IP and technology transfer–is the federal funding agreements. The primary one used for federal grants to universities is at 2 CFR 215. There, two sections deal with IP–.36 and .37. Of these, .37 is the most important, as it places a substantial condition on universities that acquire or improve intangible assets (expressly including patents and patent applications) with the use of federal funds. The university must serve as a trustee and use the intangible property for the beneficiaries of the funded project. That obligation is routinely ignored both by university administrators and by the academics that write about the legal context for university IP management and technology transfer.

This entry was posted in Bayh-Dole, Policy, Present Assignment. Bookmark the permalink.