Previous parts of this essay (I, II, and III) have taken up the structural shift of Bayh-Dole from the Institutional Patent Agreement, removing assignment to the university but also relaxing licensing oversight to the point of non-consequence. What is left is to look at the march-in provisions in Bayh-Dole and with them the reporting requirements. The march-in requirements have never operated, and the reporting requirements, well, because of FOIA, are just another form of administrative burden.
The march-in requirements in the IPA are found in Article XII, labeled “Additional Licenses.” Let’s look at part (a) first.
The Article uses “Grantee or its licensee” to suggest that the issue of additional licenses likely only applies in the case of an exclusive license, so long as non-exclusive licenses are still on offer. They would not be, for instance, if the Grantee has offered a limited number of what have been called “co-exclusive” licenses, or has agreed to end its non-exclusive licensing program as a condition of a final non-exclusive license, or a sole license that grants all remaining non-exclusive licenses a Grantee has left.
More important, this Article gives a term–three years from date of patent issue. That’s a nice objective starting point for the clock. The Grantee has three years to get to practical application–here a term without a definition, remedied in Bayh-Dole–or in the alternative offered the patent for license on FRAND terms–or in the alternative able to “show cause” why the Grantee or licensee should continue to have a monopoly, the federal agency can step in and require:
It is a clear, direct statement of what the federal government can do. It should be clear from the IPA that while a university may own a patent on a subject invention, it does not have the freedom of any patent owner. The university owns such patents as a trustee rather than for its own benefit. It does not have, as one might say, quiet possession. Anything but, actually.
In addition to the march-in focused on exclusive licenses, under the IPA the government also reserves a general right to march-in for public purposes:
These provisions should look familiar to anyone who has reviewed the Bayh-Dole march-in provisions at 35 USC 203. What is different, however, is that the objective time frame for practical application has been dispensed with, replaced by qualitative parameters that sound nice but are difficult to establish. As well, under the IPA, the government may grant licenses without first requesting that the Grantee or assignee do so and getting refused. What in the IPA is a standalone freedom to act, with a 30 day notice period, is turned into a 60 day period to respond which includes a clear pathway to file a claim in court. So much for the agency making a determination. It will get made by a court, and it will take years not weeks, and by the time the case is heard, likely the need will be moot.
The strongest provision in the IPA is that a license can be granted if “the public interest would otherwise suffer.” That provision is entirely written out of the Bayh-Dole march-in text. The three year period to show practical application or show cause is replaced by “a reasonable time” in Bayh-Dole, with no guidance as to who determines what is reasonable. The effect is to create a march-in provision with all of the trappings of concern for public welfare and nothing by way of firm requirements or remedies. As a result, march-in has never been used in the thirty plus years of Bayh-Dole. It is not because universities have operated spectacularly public-spirited licensing programs, but rather because there has been no oversight, limited reporting, and no clear pathway by which government march-in can be made routine. In short, with regard to faculty inventions made with federal support, the university stewards would be kings, and have worked the federal legislation around to nearly accomplish this.
The expectation that universities should act as stewards should the acquire ownership of patents was not entirely eliminated when IPAs were rewritten with Bayh-Dole. Stewardship is also required by the handling of intangible assets in federal funding agreements with universities. Those funding agreements are set forth by standard terms in 2 CFR 215, formerly known as OMB Circular A-110. The Bayh-Dole element is addressed at 2 CFR 215.36(b). The next provision, .37, however pertains to intangible assets which include patents, per the definition at 2 CFR 215.2(s):
(s) Intangible property and debt instruments means, but is not limited to, trademarks, copyrights, patents and patent applications and such property as loans, notes and other debt instruments, lease agreements, stock and other instruments of property ownership, whether considered tangible or intangible.
Here is .37:
§ 215.37 Property trust relationship.
Real property, equipment, intangible property and debt instruments that are acquired or improved with Federal funds shall be held in trust by the recipient as trustee for the beneficiaries of the project or program under which the property was acquired or improved. Agencies may require recipients to record liens or other appropriate notices of record to indicate that personal or real property has been acquired or improved with Federal funds and that use and disposition conditions apply to the property.
University patent administrators dismiss this provision, but they should not, especially if they claim the right to assignment of inventions based on the use of university facilities that are paid for by federal funds–either directly or through indirect costs. Such inventions are to be held in trust for the beneficiaries of the grant. The university is to act as trustee. This language parallels that of the IPA, which requires the university to administer subject inventions “in the public interest.” Trustee is much the same thing. Unless the university is named as a beneficiary of a federal award, it would appear that the university as patent licensor is limited to its role as trustee. That would mean, among other things, that its own financial self-interest is not relevant, beyond prudent recovery of costs, to the management of patents. But all this slips by most university patent administrators. I have never seen a reference to 2 CFR 215 in a university patent policy, though there are often gestures to public benefit in such policies, usually in a preamble that is lofty in style and quickly forgotten, if not undermined by the remainder of the policy.
One reason for this denial of trusteeship is perhaps that the advocates for Bayh-Dole worked hard to get free of such oversight, both in terms of restrictions on licensing, and in reporting of licensing activity. Another reason is that, perhaps, few take seriously the idea of a university acting as a trustee, though that is just what efforts such as the humanitarian licensing programs championed by Dr. Carol Mimura at UC Berkeley seek to establish. What is strange is how hard they have to work to get a hearing (I remember the effort Carol made to get the trustee element into the Nine Points to Consider document). This stuff should be anchored in federal contracting. It should be the default. It should be a matter of compliance. Compliance officers and audit personnel should be all over this, fretting that bad things will come if universities become cavalier about it. And while there is fretting over such things as making sure there’s a federal funding statement in every patent on a subject invention, something that should stand as a badge of the university’s responsibility as a trustee–or post-IPA, the government’s role in holding universities to compliance in the public interest–the compliance folks are largely asleep at the wheel regarding the fundamentals of university administration of patents.
Turning to Bayh-Dole, we find that as the march-in provisions at 35 USC 203 are translated into implementing regulations, even greater administrative burdens are introduced. There is a pre-march-in notice, requirements for the protocols of march-in, an argument period in opposition to march-in, fact-finding with use of counsel, with transcribed records, hearings, and testimony, all of which shall be “closed to the public, including potential licensees.” Then there is a determination based on the facts and referencing the statement of objectives at 35 USC 200–the only place that the objectives appear to be referenced in the whole body of text comprising the Bayh-Dole Act and the implementing regulations. If the agency wants to stay with it (there is an agency opt out clause, as if it did not have that, then the march-in decision “shall be held in abeyance pending the exhaustion of appeals” such as those under 35 USC 203–the Bayh-Dole march-in appeal.
The implementing regulations here introduce such a thicket of requirements between march-in and the appeal set forth in the law that a federal agency would have to have a powerful reason to go after a university. It would not be merely because the “public interest would otherwise suffer.” That’s a low threshold, and a sound threshold, from the IPA days. It’s clear that someone worked hard to get it removed. It is hard to believe that someone worked for the federal government.
Tied to march-in procedures are the reporting requirements. In the IPA, Grantees are required to report their patenting activity.
These reports are required for each subject invention, each year. By the time this provision gets to Bayh-Dole and the SPRC, it has been watered down. Under 35 USC 202(c)(5), federal agencies have the “right” to “require periodic reporting”:
The right of the Federal agency to require periodic reporting on the utilization or efforts at obtaining utilization that are being made by the contractor or his licensees or assignees: Provided, That any such information as well as any information on utilization or efforts at obtaining utilization obtained as part of a proceeding under section 203 of this chapter shall be treated by the Federal agency as commercial and financial information obtained from a person and privileged and confidential and not subject to disclosure under section 552 of title 5.
However, any information reported in this way is exempt from FOIA. By the time this provision makes its way to the SPRC, reporting is nearly optional, and gross sales by the licensees has been eliminated from the list:
The Contractor agrees to submit on request periodic reports no more frequently than annually on the utilization of a subject invention or on efforts at obtaining such utilization that are being made by the contractor or its licensees or assignees. Such reports shall include information regarding the status of development, date of first commerical [sic] sale or use, gross royalties received by the contractor, and such other data and information as the agency may reasonably specify. . . . As required by 35 U.S.C. 202(c)(5), the agency agrees it will not disclose such information to persons outside the government without permission of the contractor.
Under the IPA, there is a presumption that the university acts as trustee, acting for the public benefit, that the default licensing position will be non-exclusive, if not also royalty-free, that there is a short period allowed for monopoly efforts after which everyone should have access, and the reporting of efforts and outcomes is mandatory, without secrecy. Under Bayh-Dole, a great effort has been made to declaw the requirements and detooth the march-in. Such an effort only matters where there is an intent to hold a monopoly for an extended period, and to exploit the patent position for income even if there is no actual sales of product–the premise for which an exclusive license actually means anything.
The IPA required assignment of inventions to the university. Bayh-Dole did away with this and left inventors with a limited set of choices. This change was necessary to accommodate universities without patent policies and universities that had no interest whatsoever in taking ownership of inventions from their faculty or getting mixed up in patent licensing. Bayh-Dole was not going to pass with such a clause. So the advocates for Bayh-Dole had to backfill to undo the licensing restrictions, relax reporting, and make march-in provisions–also necessary for Bayh-Dole to get through Congress–put in such a condition that they would never operate effectively. These things largely got done. The result is an ungainly law and implementation with many gestures toward the public interest but in the end, the only bargain left is that federal funding is to be used to fuel a profit-seeking enterprise under the cover of public benefit, but without meaningful public oversight.
In Ivory Tower and Industrial Innovation, David Mowery and his co-authors comment that the universities have been “cavalier” about the evidence they put forward for their licensing efforts:
Remarkably, none of these characterizations of the positive effects of the Bayh-Dole Act cite any rigorous evidence in support of their claims. This cavalier treatment of evidence, of course, is not unprecedented in the political history of the Act. We noted above that the legislative debates over the bill’s drafting and passage relied on little if any evidence that patents and licenses are necessary for the commercialization and development of university technologies. (94)
It may be that we have no rigorous evidence because the university administrators working for the passage of Bayh-Dole worked diligently to ensure that reporting requirements were relaxed, and any information that was provided was to be kept secret. There is no public record that shows even the barest aggregate information specific to the Bayh-Dole Act: how many inventions have been claimed under the Act, how many of these have had patents issue on them (not the total number of patents, which may multiply like hamsters), how many of those with patents have been licensed (again, the number of licenses in aggregate is not relevant, as a few inventions with many licenses easily swamp out the shape of the activity overall), and of those that are licensed, how many have resulted in products on the market (again, the total licensing revenue is irrelevant, as it is likely based on an exponential distribution in which a few deals render the contribution of the rest statistically meaningless). We are not even told what income comes from a royalty on the sale of products and what comes from other fees, which would distinguish between income from product on the market and income from speculation on the potential for future products, or from trolling industry based on patent positions, with no intent whatsoever to attempt to build product of one’s own.
It may be that may well be that the Bayh-Dole data is not good. It may well be that “public interest would otherwise suffer” many times since Bayh-Dole has been in effect, but again, formally, we would never know. From the inside, it appears that universities have accumulated substantial patent holdings, have been able to license only a smattering that have resulted in products, and have largely profited from a significant revenue generating license every decade or so, supplemented by a trade among speculators, especially in startup ventures, which may never develop a product based on a licensed subject invention.
The model proposed in Bayh-Dole is that of an agent, under which the patent system is to be used to promote the use of each subject invention for which the agent accepts ownership. The universities instead have adopted a portfolio model, in which all inventions are appropriated in the hope of catching a “lucky” one that will justify the overall operation, or in like manner, ensuring that no university inventor or invention management agent could secure development of an invention that would result in a lucrative deal not shared by the university, or much worse, might result in widespread use without pressing money to travel through a patent position. Such a dire outcome, regardless of the wealth creation involved or public benefit, would represent a “race to the bottom” that would undermine the patent positions of the entire university portfolio and might well prove David Mowery’s point, that patents and licenses are not particularly necessary for the development and commercialization of university-hosted inventions.
Overall, other than the demand within the IPA that a university take ownership of all inventions, the IPA presents a much more powerful approach than that of Bayh-Dole: it has clearer terms, stronger accountability, greater restrictions on licensing programs, and a simple mechanism by which the federal government can ask a university to get on with it or get out of the way. Were an IPA positioned to permit faculty inventors to decide between their university, an invention management agent, the federal government, or the public domain, there is little else that would need to be altered to provide a robust invention management environment led by faculty investigators. There is no doubt that Bayh-Dole can be reconditioned–make clear that the (f)(2) agreement places initial choice of agent with the inventors, make subject invention reporting mandatory and public, report by invention not in aggregate, and place restrictions on the use of litigation to resolve infringement disputes, something that would likely be addressed by restoring for most subject inventions the 3 yr/8 yr deadlines for exclusivity and the much lower–and perfectly appropriate–threshold of the “public interest would otherwise suffer.”
In the meantime, there are ways to pressure universities into compliance and accountability:
- First, use state public disclosure laws to get Bayh-Dole information out of public universities that is otherwise protected at the federal agency level by the Bayh-Dole FOIA exemption.
- Second, use audits to determine compliance with (f)(2), reporting, and licensing practices. Audit as well claims made for “successful” operations, looking not only at financials but also at public claims regarding numbers of licenses and startups.
- Third, push for “freedom to invent” legislation at the state level, barring public universities from requiring assignment of inventions as a condition of employment, use of resources, or participation in extramural research.
- Fourth, encourage faculty governing bodies to get back into the discussion. Faculty governance of invention equities was long a foundation of invention management at universities. There is no reason it should be so again. As a start, faculty should form an intellectual property and innovation policy committee charged with maintaining information with regard to university practice and advocating for changes that restore faculty to a leading role in the deployment of their scholarship, inventive or otherwise. Faculty should demand that invention policies restrict their scope to patentable inventions, and that university financial interest in faculty work, or student, or staff work, be decided by a committee of peers if the inventors choose not to use an invention management agent with an arrangement with the university.
- Finally, if none of this gets desired results, we must raise the prospect of class action litigation for conversion of inventors’ property based on deceptive or willfully negligent recitation of federal law, if not also a violation of university inventors’ constitutional rights under the Takings clause of the Fifth Amendment, for which universities should owe compensation to inventors that represents the value of inventions at the time of the appropriation of inventions, regardless of what has happened, or more likely, has failed to happen after.
The best way of all, of course, is for university administrators to open a dialog with faculty, including governance bodies, investigators, inventors, and faculty generally. Inventions are not merely the domain of the sciences, medicine, and engineering, and inventions are scholarship and that does not change with the introduction of an ownership claim or the apprehension of substantial value. A movie or game version of a novel may carry equal or greater value. And for all that, value may be better measured in social impact than in money returned to a university, especially money paid under threat as tribute not in recognition of fealty. The primary way for universities to use the patent system to promote use is to foster goodwill and create an environment in which creative people can pursue their findings with resources and responsibility, subject to disclosure and the guidance of their peers. It may sound fluffy, but it is possible, and, for all that, it was the system that provided the foundation for Bayh-Dole. Given the limitations of Bayh-Dole, the least we can do is work back to a known optimum and adapt for the new circumstances we face going forward. Until then, I want my IPA.