In the past few months, I have spent a good deal of time on the Bayh-Dole Act. One of the great challenges of dealing with Bayh-Dole is to get past what university patent brokers say the law “is” and get at not only what the law says, but also what the law does–when implemented, and when ignored (and most university officials ignore large swaths of the law–call it ignore-ance).
Bayh-Dole is a set of requirements having to do with inventions and contracting regarding inventions, patents, licensing, and money from inventions and patents. University patent brokers have selected a few bits to emphasize. But the other bits are there, too, even if university patent brokers don’t call them out. And it turns out, the other bits are rather more important than the bits that university patent brokers focus on in their talks and their articles and in their policies and in their “training materials” and in their talks at AUTM.
If we listen to university patent brokers, we get repeated assertions that Bayh-Dole gives universities the right to take title to inventions made with federal support (which isn’t true–universities have a conditional contractual right, not a statutory right, to keep title to inventions if they obtain that title). We see articles and slide decks and policy documents that explain that Bayh-Dole gives universities incentives to commercialize inventions (though there’s nothing in Bayh-Dole about incentives and commercialization is not a stated objective of the law–Bayh-Dole places restrictions on university management of inventions and makes commercial use as important as commercial sale). The patent brokers mention the government license (without including “practice and have practiced” or what “practice” means) and the dread march-in procedures (which were designed not to work and in three decades have worked as designed and have never been successfully used). And we see odd bits and pieces about compliance, such as the time period in which to elect to retain title, when patent applications must be filed, and the substantial U.S. manufacturing requirement for exclusive U.S. licenses (often omitting the limitation “to use or sell”).
In fact, almost everything university patent brokers claim about Bayh-Dole turns out to be wrong, skewed, spun, repeated out of context, or just made up. It’s rather remarkable. I have been through hundreds of university policy and guidance documents–and written up my findings here at Research Enterprise. University patent brokers make a big deal out of Bayh-Dole compliance, and yet they show no evidence they understand the how the law operates. They ignore the standard patent rights clause. They ignore the (f)(2) written agreement to protect the government’s interest provision. They ignore the restriction on assignment by labeling assignments as exclusive licenses. They ignore the policy requirement to use the patent system “to promote the utilization” of subject inventions and instead sue companies for using subject inventions. Let’s call them “ignore-amuses”). They don’t promote “free competition and enterprise” and instead aim to preserve patent monopolies. They assert in policy and licensing documents that universities and research sponsors and licensees must comply with Bayh-Dole when Bayh-Dole applies to federal agencies, not any of these entities, who are constrained when they become parties to a federal funding agreement or accept obligations required by the standard patent rights clause, such as in subcontracting or the (f)(2) written agreement.
We’re not talking here about differing “interpretations” of the law. We’re talking about wholesale bungling, inattention, and in some cases persistent, deliberate misrepresentation of the big parts of the law. For all university patent brokers’ posing as experts, they are collectively a sorry bunch. But their assertions find their way into policy statements, into contracts, and even into federal agency guidance. Their assertions find their way into academic articles and law review articles, where they become taken as established fact by the folks who read such stuff uncritically or don’t have the time to check their work. They are a clever, industrious sorry bunch. Think termites, not beavers.
The challenge is to get at Bayh-Dole for what it does rather than for what university patent brokers say we should be fixated on. Just because university patent brokers don’t mention some things does not mean they are no longer in the law. Just because university patent brokers are dedicated to suppressing or by-passing or ignoring parts of the law does not mean the law operates only as they say it does.
Bayh-Dole stripped of its conditionals amounts to requiring federal agencies to present a choice to inventors using federal funds:
Assign your inventions to your employer or another invention management agent, or deal with the federal agency over title.
The public policy wrapped up in this core bit of Bayh-Dole is that assignees of subject inventions should have “certainty of title” but inventors themselves should not. As Senator Bayh had it, inventors should be “last in line ” after the middlemen and the federal agencies, even though federal patent law puts them first in line and in fact does not provide even for “invent for hire” as a matter of employment. We might ask, what would Bayh-Dole be like if it required federal agencies to assure inventors of “certainty of title” rather than middlemen? if nonprofit inventors at least were the ones to notify a federal agency that they “elected to retain title” and then the inventors were responsible for seeing that they used the patent system to promote the use of the inventions? In such an approach, university inventors might recruit agents to assist them with patenting and licensing, but no university would have a “right” to “take” inventions or to claim that the law already “gave” them inventions because it used the word “retain” and to “retain” something one must already magically have it.
Of course, no change in Bayh-Dole is required to implement an inventor-first approach. It is right there in the law and standard patent rights clause. But if NIST has its way, then the (f)(2) written agreement in the standard patent rights clause will be turned into an assignment requirement–without any authority in Bayh-Dole or federal patent law to do so–and then, of course, no inventor-first approach is possible, as Bayh-Dole will amount to a vesting statute, requiring employers to take ownership of inventions made with federal support. But as it is, university patent brokers cannot find a way (even those that want to) to break the pile of rhetoric, policies, contracts, and articles that all manage to misrepresent and bungle Bayh-Dole.
There is another way to look at Bayh-Dole afresh. This direction looks at how the patent system is used rather than on who owns rights in inventions made with federal support. Here is the bare framework:
35 USC 200 sets out a “policy and objective” of Congress, including:
Use the patent system to promote the utilization of inventions arising from federally supported research and development.
Ensure that inventions made by nonprofit organizations and small business firms are used in a manner to promote free competition and enterprise without unduly encumbering future research and discovery.
35 USC 202(c)(7) requires federal agencies to place limitations on the assignment of inventions by nonprofits:
a prohibition upon the assignment of rights to a subject invention in the United States without the approval of the Federal agency, except where such assignment is made to an organization which has as one of its primary functions the management of inventions (provided that such assignee shall be subject to the same provisions as the contractor).
If we take these three statements together, we get a two-part requirement. Here’s the first part:
Use the patent system to promote the competitive use of each subject invention
The “competitive” use derives from “free competition”–the users are free to compete with each other and with the owner of the patent on a subject invention. That’s because the owner of a patent on a subject invention must use the patent system to promote utilization–that is, practical application. The owner cannot use the patent system to prevent utilization. Hence, the competitive element involves both multiple licensees or a licensee in competition with the patent owner.
Now here’s the second part of the requirement, that makes this clear:
by breaking up the monopoly right of the patent on the subject invention
This second part of the requirement derives from the restriction on assignment to be placed in the standard patent rights clause. When an owner of a patent on a subject invention grants permission to another to make, use, and/or sell an invention, that transaction is a license. The license breaks the monopoly–the patent owner has substantial rights and the licensee has freedom to practice (at least with regard to the patent owner). There’s at least one other entity that can practice the invention, and so there is the prospect of competition between the patent owner and at least one licensee. If the patent owner grants multiple licenses, then of course each licensee competes with the others, and all of the licensees compete with the patent owner. That’s “free competition and enterprise.”
But consider what happens if the owner of a patent on a subject invention assigns the invention (and patent rights). Then Bayh-Dole stipulates that federal agencies must require the patent owner (if a nonprofit was the contractor for the research) to also assign the nonprofit requirements that run with the title to the invention–that is, the standard patent rights clause plus the requirements that comply with 35 USC 202(c)(7). In the case of an assignment, the monopoly rights are not broken up, and the assignee has substantial rights and the patent owner is not in a position to compete with the assignee.
An exclusive license is generally an assignment when it grants rights in the patent rather than granting permission under a patent right. An assignment covers make, use, and sell, permits the assignee to enforce the patent, and allows the assignee to grant its own licenses (called “sublicensing” when the assignment adopts the wording of an exclusive license). An assignment does not break up the monopoly right in the patent, and so does not use the patent system to promote use or free competition and enterprise. An assignment just shifts who has responsibility for meeting the policy and objective of Bayh-Dole. Hence, Bayh-Dole requires nonprofits to pass through their obligations in any assignment of rights to a subject invention. The assignment restriction is not just with regard to “title” to a subject invention but to “rights.”
When a university grants rights in a patent, rather than grants permission under a patent right, the university preserves the monopoly in the patent and assigns rather than licenses. The recipient of the grant has the right to practice the invention because the recipient is an owner of the invention (co-owner, or conditionally the exclusive owner, or outright the exclusive owner). Assignment can take place any number of ways–by sale, gift, bequest, or under the cover of license. Bayh-Dole requires that for any assignment of rights in a subject invention, the nonprofit requirements must also be assigned–including restrictions on further assignment, use of royalties and other income, and sharing royalties with inventors (that is, any money from licensing, but not other income such as money from use or sale).
Thus, when a nonprofit assigns rights in a subject invention, the nonprofit must also require the assignee to be subject to the same requirements as the nonprofit: to use the patent system to promote the competitive use of the invention by breaking up the patent monopoly.
It is entirely possible to use an invention with public benefit without granting licenses to anyone else. A company might do such a thing, and use the patent system to exclude other users of the invention. In such a case, the patent is useful in promoting the patent owner’s own use of the invention–with monopoly might come money. A nonprofit also might exclude all others and use an invention such that there are public benefits on reasonable terms. It could make a product or practice a method that did everyone good, even without selling product or licensing others to practice the method. But this is a rare, odd thing. It’s more likely that a nonprofit will have to rely on others to make and sell product and for that will have to grant licenses, or will allow others to practice methods (as in standards, or to produce products or provide services). Either way, the nonprofit must break up the monopoly in the patent, not merely pass that monopoly along in return for a payment.
Thus, for nonprofits, this is the essence of Bayh-Dole when it comes to the law’s policy and objective:
Use the patent system to promote the competitive use of each subject invention by breaking up the monopoly right of the patent on the subject invention.
Until the monopoly right is broken up–distributed via licenses so that there is “free competition” between the patent owner and licensees, or among licensees–Bayh-Dole’s policy and objective has not been met by nonprofit owners of patents on subject inventions.
We are now in a position to state simply what Bayh-Dole requires federal agencies to do:
Allow each inventor to assign their invention to their employer or another invention management agent or the federal agency. (Supreme Court in Stanford v Roche, 37 CFR 401.14(a)(f)(2))
If an inventor does not assign, then the federal agency may negotiate for rights, treating the inventor as small businesses if the agency decides to permit the inventor to retain title to their invention. (35 USC 202(d))
If the inventor does assign their invention, and the contractor for the research is a nonprofit, then the nonprofit, if it does not achieve practical application itself, must use the patent system to promote the competitive use of each subject invention by breaking up the monopoly right of any patent on the subject invention. (35 USC 200, 201(e), 202(c)(7))
The nonprofit must share royalties with the inventor, may recover only its expenses incidental to the management of subject inventions (and not other inventions, or other activities), and must use royalties and other income from subject inventions for scientific research or education. (35 USC 202(c)(7))
If the nonprofit assigns any right in the subject invention (whether by sale, gift, exclusive license, and the like), the nonprofit must in turn require the assignee to use the patent system to promote the competitive use of the subject invention by breaking up the monopoly right of any patent on the subject invention. (35 USC 202(c)(7))
An assignee is subject to the same requirements as the nonprofit: the assignee must share royalties with the inventor, may recover only its expenses incidental to the management of subject inventions (and not other inventions, or other activities), and must use royalties and other income from subject inventions for scientific research or education. (35 USC 202(c)(7))
If a prospective assignee were a commercial concern, then it might well reject an assignment and demand, instead, simply an exclusive license–with no right to enforce the patent, no right to sublicense, and perhaps even only a non-exclusive right to make and use the invention, focusing exclusivity on the right to sell (and perhaps to import). In such a case, the exclusive licensee’s sales would compete with everyone else’s right to make and use the invention as they pleased. They could buy product or not. That, too, is “free competition and enterprise” and meets the policy and objective of Bayh-Dole. If one wanted to put this simply as a positive obligation for nonprofits, it would be along the lines of:
Break up the monopoly right in a patent on a subject invention to promote competitive use by granting no more than an exclusive license to sell and import while granting fair, reasonable, non-discriminatory non-exclusive licenses to make and to use, if not also to sell and import.
This is then the core of Bayh-Dole, including its conditionals. The rest of the apparatus–the requirements on disclosure, filing patent applications, election to retain title, the notice of government interest in patent applications, the government’s license to practice and have practiced, the conditions under which the government can take title or intervene to require licensing, the reporting, the restrictions on exclusive licensing in the U.S., the hand-waving preference in licensing for small businesses by nonprofits–all of these things are window-dressing on the primary purpose of the law, which is to break up a patent’s monopoly rights and by doing so achieve the practical application of each subject invention–so there are public benefits from its use, on reasonable terms.
If university patent brokers put Bayh-Dole this way, then they might feel a different set of compliance obligations. Instead of demanding ownership of all inventions outright (and for no valuable consideration, apparently), patent brokers would accept ownership only when they felt assured that they could break up the patent monopoly and by doing so promote practical application. If they merely passed the patent monopoly along–often a profitable thing to do if one buys low and sells high–they would make sure the new assignee was held to the same obligations that the nonprofit had. After all, if compliance with Bayh-Dole matters (and it should, though at present it doesn’t), then the obligation to comply should start with the primary requirements of the law, not with piddling administrative details. And the primary requirements have to do with how the patent system is used, especially by nonprofits, and what the patent system used this way is to accomplish.
Universities report the number of inventions they produce, the number of patents they obtain, the number of licenses they do each year, and how much money they receive. They don’t report how many subject inventions are involved in this activity. And they don’t report the primary outcome that matters: practical application. They take advantage of Bayh-Dole to create livelihoods for patent brokers, and (sometimes, at least) positive revenue from patent transactions. But they report practical application only as selective “success” stories, without looking at what happens across their entire “portfolio” of subject inventions. It is as if Bayh-Dole exists for patent brokers to aggrandize their practice, and not to achieve practical application for each subject invention taken under management.
University patent brokers misrepresent the law and recite distracting details to make it appear that the law is complicated and the only ones qualified to address it are, er, the priestly patent brokers–who time and again get the law so very wrong, if not in its details then certainly in its major requirements. Scores of university attorneys participated in amicus briefs in Stanford v Roche at multiple levels as the case made its way through the courts. They consistently got the law wrong, argued against the interests of inventors, and claimed that Bayh-Dole was about them and their freedom to do what they wanted with subject inventions, including (and this was the great irony) suing companies who had achieved practical application even while the university had done nothing in the same time to achieve practical application. Those attorneys are only the cold, hard, dirty tip of the iceberg of university misunderstanding and blundering when it comes to Bayh-Dole.
Bayh-Dole requires federal agencies to let inventors decide what to do with their inventions. The law dictates a standard patent rights clause with conditionals, depending on what inventors decide. If inventors assign to their nonprofit employer, then the conditionals kick in that involve administrative details but most importantly require the nonprofit to be a trustee for the patent–break up its monopoly, share the royalties, deduct only the incidental expenses for managing such patents, and use the balance for scientific research or education. If folks wanted to report Bayh-Dole success, these would be the metrics:
- how many subject inventions with one or more patents
- how many of these patent rights have been broken up
- how many assigned
- how many subject inventions with one or more patents have achieved practical application
- how much royalties and other income received
- how much royalties shared with inventors
- how much expense deducted
- how much money made available for scientific research
- how much money made available for education
With these metrics, we could start to see whether university patent brokers understand Bayh-Dole well enough to do those things necessary to achieve the impact that Congress intended when it passed the law.