Faux Bayh-Dole has been de facto federal research innovation policy now for thirty-five years. The real Bayh-Dole is sketchy enough, but the faux version is downright vile. Here are some “truths” of faux Bayh-Dole that are, in reality, simply not true. We could call them lies, except those repeating them don’t think they are untrue. They don’t have a regard for the truth–thus, it’s what Frankfurt would call “bullshit.”
As one senior university official blithely told me with regard to a compliance requirement (in that case, implementing the (f)(2) agreement), “We will do that when the feds tell us we have to.” And that was 1989. That’s the attitude university officials take to all of Bayh-Dole–the only parts they “enforce” are the ones that make it appear that they have a right to take any invention they want–either because Bayh-Dole requires or permits it, or because if they don’t then perhaps in some strange way they might not comply with Bayh-Dole, or because inventors might not realize that something is inventive and so would fail to comply with Bayh-Dole.
So the middlemen take all inventions, patentable or not, and pin it to compliance with Bayh-Dole. But it’s faux Bayh-Dole. And the feds have let it happen. Either this is what they, too, desire, or they just don’t care. Either way, it bodes ill for universities, research, and innovation.
Here are some common bits of faux Bayh-Dole bullshit. Together, the faux bits form something of a religious belief system, a sect of middlemen, as it were. The prophecy calls for economic prosperity, if only the middlemen have their way with inventions made by faculty receiving federal support. That prophecy has long ago failed. But they persist. Perhaps they won’t be rescued by UFOs from the planet Clarion, but they hold out that this middleman idea is still a really, really good message for all to benefit from. I can’t help but think this scam will come to an end soon and people will start thinking anew about the consequences of ramping up government funding for basic science and worry what Vannever Bush worried–that the opportunity for mediocrity and grandiose projects was huge–and worry also that wherever there is lots of money, the people who have made a commitment to an area of inquiry might be swamped out by those who have made an even stronger commitment to getting more research money than others. On the edges of this fray, patent middlemen have stepped forward at opportune times to declare that they represent the public interest, if only they can do their jobs with impunity. Sigh.
Anyway, here are those bits of faux Bayh-Dole, with commentary. I wish there were a “bald-face” font I could use instead of “bold-face.” But alas, typography has not yet developed this capability.
Bayh-Dole vests ownership of inventions made with federal support with the contractors that host the research.
Faux Bayh-Dole advocates could never get the story right. They said Bayh-Dole vested ownership, or gave a right of first refusal, or prohibited assignment except to the contractor, or changed the presumption of title, or made title vest upon election of title, or made title vest upon notice of election of title. Universities were authorized by federal law to “take” title or “retain” title, meaning “to have and to hold” title.
The Supreme Court by 7-2 rejected these claims in 2011: “Had Congress intended such a sea change in intellectual property rights it would have said so clearly—not obliquely through an ambiguous definition of “subject invention” and an idiosyncratic use of the word ‘retain.'”
Bayh-Dole changed the presumption of ownership.
This part is pure deception–deceiving while trying to be “truthful.” Executive orders discuss the “presumption of ownership.” But this is not a presumption of federal patent law or common law. This is a presumption of executive policy guiding the formation of federal contracts for research–either in the form of subventions (primarily with universities) or procurement (primarily with companies). The presumption that operates is whether the federal default in such contracting should be for government ownership (favoring the idea of procurement) or contractor ownership, with the government obtaining sufficient permissions for its own uses.
But even this presumption gets exploited in the case of subvention, as there’s nothing in a federal gift of support to faculty investigators that carries with it a requirement that a university that hosts (but does not commission and does not pay for the work) must own the patentable results of such investigations. It would be enough to require inventors to grant the government a “shop right” in exchange for support.
The idea behind the Institutional Patent Agreement, however, works against the idea of subvention and insists on creating a law for patent brokers. To do this, it must create a public covenant that limits the exploitation of patents covering an invention made with federal support. That’s the way that the “middlemen”–and they called themselves was just that–could make it appear that the public interest was served by making federal support for faculty research into a form of institutional procurement, invoking the “presumption” of executive orders that the government would procure title to subject inventions (subject to agency discretion whether to do so in any particular instance), and transfer that presumption to universities on behalf of the middlemen.
That is, under the IPA, the middlemen claimed that the federal government delegates to the university the government’s presumption of title, which the university then is permitted to “administrate” subject to limitations on licensing–in particular, no more than three years from the date of patent issue to achieve practical application, and with practical application no exclusive license running longer than three years from the date of first commercial sale or eight years from the date of the license, whichever comes soonest. The university then is required to obtain assignment of a subject invention only when the university decides to file a patent application on the subject invention. Otherwise, the presumption of title stays with the federal government, which deals with inventors directly under its funding agreement.
To make this switcheroo happen, the middlemen had to construct a narrative in which universities would undertake the same public covenant expected of federal agencies, but with the permission to use just a bit of selective, limited exclusive licensing, all with federal agency oversight. That allowed the change through the APA contract of the “presumption of title.” But to make it stick, the middlemen had to argue that the overhead for all this federal oversight was too much for the poor inventor, so the IPA had to require universities to require assignment from inventors of their interest in subject inventions. The middlemen would do the government’s work better than the government could, and so needed the power of the government and for that took on a public covenant, required it all in the IPA, and thus diverted to brokers inventions that would have gone to the federal commons, to the public domain, or remained with the inventors (with the government’s approval) to be developed by the inventors or by one or more companies or through an invention broker of the inventors’ choosing.
The IPA was a nasty bit of work. Bayh-Dole is even worse, as it tried to make the switcheroo in government contract rights a matter of patent law while at the same time gutting the public covenant that ran with the IPA contract and obscuring the link between the new guidance, now in federal patent law as a statement of policy and objective, and the limits of the personal property rights permitted for patents on subject inventions.
There never was federal patent law that simply vested ownership of inventions made in federal contracts with the government. Executive orders could not do that. The Supreme Court made clear that even Bayh-Dole does not do that, despite the crowing–turning to howling–that faux Bayh-Dolists put up about it. All that changed was executive branch policy regarding when federal agencies should require federal ownership of patents on inventions made with federal support. For that, Bayh-Dole changed the default from federal ownership to only a royalty-free government license. Agencies can still require assignment to the government, but must show “exceptional circumstances” to do so. Under the Kennedy Statement of patent policy, it was the other way around–the government should take title to inventions (whether it chose to seek patents was another matter) except when it was in the public interest not to. Mostly, that public interest was reflected by the commercial position–the “equities”–of a commercial contractor that already had expertise and products in the area of research in which the government was contracting for procurement.
What has been entirely wiped out by Bayh-Dole (and later by Stevenson-Wydler) is the public covenant governing the federal patent commons–that federally held patents should be made broadly available. The middlemen were aghast at this waste of the opportunity to profit from the disruption possible in creating monopolies in the middle of basic science projects. They could not imagine a use of the patent system exactly to prevent them from operating–they, like undertakers with a deep dissatisfaction with the waste of living when there is so much profit to be had in death.
So Bayh-Dole did change the presumption of ownership, but not in the way that faux Bayh-Dolists want folks to understand. Bayh-Dole changed agency defaults for federal ownership to a shop right.
One more thing about presumption of ownership. There are some vesting statutes out there–AEC and NASA among them. These statutes do (or did) vest ownership of inventions within a statutory scope with the federal government. But there are two huge differences between doing so and what the faux Bayh-Dolians proposed. First, those vesting statutes apply not only to contracted research but to anything else relating to the contracted research–that is, these statutes change the scope permitted to personal property rights in inventions, just as the government does so for laws of nature or surgery techniques. Laws of nature, even when discovered, are not patentable. Surgery techniques may be patentable, but the property right in the patent does not permit the exclusion of surgeons practicing the technique. Similarly for atomic energy and bombs, or for rockets and pigs in space. These are utterly unlike the idea that there is patentable subject matter with a profile for desirable public use, and that the rights to such inventions should be relocated from the inventors–where the Constitution puts the federal authority for patents–to mere contractors who receive money from the government earmarked as a subvention to assist faculty research.
To cover this bit of nastiness, the middlemen added in small businesses, to make it appear that all federal subventions to universities should have the appearance of government procurement deals with companies. Small businesses operate like contract research organizations. They get assignment of all inventions from their employees and pass the inventions along to the commissioning party–in this case the government. Bayh-Dole then allows them to “retain” the rights and not pass them along, if they decide to patent the inventions. Then universities get the same treatment–first, contract research organizations under a procurement deal which requires them to procure from their faculty and students on behalf of the government. Then the deal switches to procuring just for the university’s sake. So nice.
The university, then, was made to appear to adopt the profile of a company, having employee servants who owed their inventions to their employer so the employer could pass them along to the government, but then doesn’t have to–rather than as a servant of the public to assemble and support faculty who then could conduct better research and provide better training in the public interest. Bayh-Dole, by contrast, does not dictate that inventors must assign to the universities that host their research. If an inventor does not assign an invention, then federal agency regulations control–the agency could request assignment and seek a patent, request assignment and place the invention in the public domain, or allow the inventors to retain their ownership, subject to the standard patent rights clause in 37 CFR 401.9.
Bayh-Dole applies directly to universities and inventors.
Bayh-Dole applies to federal agencies and to patent law. It does not apply to universities or inventors. Bayh-Dole supersedes executive orders (which permitted flexibility and recognized special situations) and pre-empts laws and regulations by requiring federal agencies to use a standard patent rights clause to be created (now, by the Department of Commerce). Bayh-Dole states some but not all of what must be in that clause, and places limits on how agencies may modify or enforce the clause. Bayh-Dole also changes federal patent law to create a public covenant that runs with any patent on a subject invention, regardless of who owns the invention.
Nothing in Bayh-Dole reaches directly to universities or to inventors. Ownership of inventions is not altered–inventors own subject inventions just as they own ordinary inventions. Universities have no special right to own subject inventions, no special advantage in the law to take ownership, no mandate to profit from subject inventions or to attempt to commercialize those inventions. Any effects of Bayh-Dole on federal agencies are transmitted to universities through the inclusion of patent rights clauses in federal funding agreements. That is, whatever a university is required to do comes via the federal contract–typically 2 CFR 215 in the case of grants to universities. These requirements reach inventors when a university complies with the standard patent rights clause.
Bayh-Dole does change federal patent law. Thus, when an inventor or university owns a subject invention, exploitation of the patent right is constrained by the property rights available for a subject invention. But Bayh-Dole doesn’t apply to universities.
Bayh-Dole requires commercialization.
The standard in Bayh-Dole, as in the IPAs, is “practical application” not commercialization. Practical application is defined as use “under such conditions that it can be established that the invention is being utilized and that its benefits are, to the extent permitted by law or government regulations, available to the public on reasonable terms.” There is no mandate to “commercialize” though commercialization is one way to make the benefits of an invention available to the public. Commercialization, however, is a choice. It is not always necessary, nor is it in some cases even possible or desirable. For instance, an invention may become a standard and never become a commercial product. Or an invention–such as a method embodied in software–may be widely practiced without the need for a “commercial” version, even if commercial versions are also introduced.
There is no obligation in Bayh-Dole for universities to take ownership of subject inventions, to license subject inventions exclusively, to license subject inventions for profit, to commercialize inventions, to have a patent licensing capability, to have a patent policy, or to have patent agreements with employees. There is no hazard to universities, even, to fail to comply with the standard patent rights clause–as most universities have been willfully, openly breaching that clause for years with no ill effects. Failing to place federal funding notices in patents, failing to timely elect to retain title (even Stanford screwed this up), failing to implement the (f)(2) agreement, failing to reimburse only the costs of managing subject inventions from the royalties on subject inventions, failing to use the balance of royalties after costs only for scientific research and education.
Any audit will show wanton disregard. But no one cares. No one bothers. Faux Bayh-Dole is a law to give middlemen free access to the personal property of faculty inventors working under a professional expectation of public service, allowing the middlemen to exploit that property any way they wish, including sitting on it until it is worthless, and they gain the buy-in (or sell-out, as it were) of faculty by offering all of them a share of the income and calling it all “in the public interest.” What killers.
In faux Bayh-Dole, “commercialization” means any exploit of a patent position for profit–speculation, trolling, withholding from research competitors, and even the preparation of commercially available product. In every ounce of poison, a grain of truth.
The property rights in patents on subject inventions under Bayh-Dole are pretty much the same as those for ordinary patents
Faux Bayh-Dolists act as if a university can do anything with a patent on a subject invention that one could do with an ordinary patent, one that isn’t based on federally supported research, wasn’t owned by a university, and was not positioned on a frontier of science or to address public health, safety, or welfare. They believe the only conditions on their exploitation of patents on subject inventions comes in the form of the requirements to give preference to small businesses (which they interpret as licensing to their own startups rather than to small businesses) and to require substantial manufacturing in the U.S. for exclusive licenses. They do not believe that 35 USC 200 does anything other than give happy rationalizations for why Bayh-Dole exists. That is, they read 35 USC 200 as surplusage, having no effect on federal patent law, and therefore having no purpose in being part of federal patent law.
Bayh-Dole supersedes a string of executive orders that established federal policy with regard to inventions made in federally funded research. That policy set forth what amounts to a public covenant that ran with the federal contract. Bayh-Dole broke up that public covenant into two parts. One part followed the standard patent rights clause; the other part ended up in 35 USC 200. The faux Bayh-Dolists then went to work on the standard patent rights clause and march-in rights to weaken the public covenant as much as possible, removing any time requirements on achieving practical application and on the duration of exclusive licenses. They also gutted public accountability by making reports on usage secret and making march-in procedures so difficult and time-consuming that no one has been able to make them work.
The part of the public covenant that ended up in federal patent law at 35 USC 200 places limits on the patent property rights in subject inventions. This is very different in action from the Kennedy Statement of government patent policy that controlled what a federal agency could do with an IPA. The IPA system ran within a framework of executive orders and federal contracting. There, a “subject invention” is a special class of potential contract deliverable. By contrast, the standard patent rights clause exists within a new framework of patent law created by Bayh-Dole–a new class of patentable inventions called “subject inventions” with special property rights defined by 35 USC 200. While the standard patent rights clause and the various administrative regulations surrounding its enforcement–reporting, electing, march-in, and the like–concern federal contracting, 35 USC 200 concerns the boundaries of the patent right granted.
An owner of a patent on a subject invention has property rights no greater than those established by 35 USC 200. The patent system may be used to “promote the utilization” of subject inventions. There is no authorization to use the patent system to “prevent the utilization” of subject inventions. This *is* part of federal patent law, not merely federal contracting. It changes what can be done with patents on subject inventions. Federal patent law is clear that the property right in a patent is “subject to the provisions” of federal patent law. For subject inventions, 35 USC 200 sets out those property right boundaries. This part of the public covenant runs with the patent, not with the contract. This is one of the key changes brought about by Bayh-Dole.
A patent portfolio approach to licensing is perfectly acceptable.
The faculty-led approach to patents made in university research was that of the invention management agent. Both Research Corporation and WARF (and its imitators) were set up on this model. A faculty member would present an invention for management, and the agent would decide whether to take it on. If so, there was a deal between the inventor and the agent involving strategy, diligence, rights, and royalties. Inventions were managed case-by-case, agent-by-agent. Consider a sports agent. If the agent takes on a client, the client expects the agent to find a team for the client to play for and negotiate a contract. Same thing with a band and a music publisher. But what happens if an agent accumulates a bunch of clients under contract and makes money selling the agency contracts or a future interest in the agent’s share of future income rather than making money when the clients get paid for their services? The agent makes money and the clients get nothing. This is portfolio thinking. One can do this with a bundle of investments, for instance. Own a bunch of things and see which ones pay off. One can even do this as a broker for others. But it doesn’t work for getting inventions into use–though it does present a way to be financially successful while putting on the appearance of trying.
Bayh-Dole does not permit a portfolio approach to subject inventions. Patents on each subject invention under management by a university must be used to promote the practical application of each acquired subject invention. The standard patent rights clause operates funding agreement to funding agreement. The same deal repeats for each subject invention when acquired by the university. Bayh-Dole requires an agent approach, not a portfolio approach. If one adopts a portfolio model, then one must create a commons–just as the federal government was doing until Bayh-Dole (and then Stevenson-Wydler) destroyed the effort.
There is no property right under 35 USC 200 that permits a university to hold a bundle of patents on subject inventions with the expectation that only one or two in a hundred–or a thousand–may pay out significant royalties. If a university takes title to a subject invention, then it has an obligation of good faith to be diligent in using the patent system to promoting the use of the invention. The financial success of a portfolio has nothing to do with it. Nor does the activity that contributes to the portfolio–number of inventions reported, patent applications filed, patents issued. Nor does the activity of licensing–the number of licenses granted, the number of startups. These are portfolio numbers and fail to show the status of each subject invention–has practical application been achieved? what is the date of first public use or commercial sale? How are public benefits on reasonable terms established?
But faux Bayh-Dolists don’t care. They report “success” with a handful of instances of inventions with “potential” because the university has signed a license and received payment “up front” with promises of more. When they license to a startup, they take equity as some or all of the royalty. That equity then takes on value with the development of the company, not necessarily with the development of the licensed subject invention. In fact, development of the licensed subject invention may diminish the value of the equity–other products might be faster to market or more lucrative targets. The startup may be sold long before it creates any product, and the university’s income from selling its equity has to do with trading in a secondary market–in a futures market for the possible value of the company’s holding an exclusive license in a subject invention. Investors may see the license as “insurance” or “collateral” that can be exploited if the company runs out of money. “We can always sue industry for infringement to recover our investment if the company fails.”
In general, a royalty is any consideration for a patent license. But patents on subject inventions are not ordinary patents. Practical application is the limiting scope of the property right of a patent on a subject invention. Thus, arguably, the only royalty that is acceptable in an exclusive license is one that reflects use, not one that reflects future value or speculation or a secondary market that trades in the future value of the licenses that have been granted. That is, 35 USC 200 and the standard patent rights clause require a royalty to be paid based on actual use or sale of product or services embodying the subject invention–not paid for the acquisition of the license, not proxy for the success of the company, not for the benefit of excluding competitors and delaying development of the invention, not for separating gullible investors from their money with the expectation that they will find even more gullible or desperate or disrupted investors to whom to sell their shares.
Compliance with Bayh-Dole requires an agent model, not a portfolio model. It requires practical application of each subject invention acquired for management by a university. The compensation that a university receives must be a royalty based on successful practical application, not speculation on the future value of the patent, used any which way one might use an ordinary patent. Faux Bayh-Dolists are in deep denial about all this. They believe they can acquire everything and their work is done if they are financially successful and report one happy deal in a hundred inventions, or even one in a thousand, once a decade, and otherwise how busy they are. It’s all bullshit. Worse, it’s simply not compliant with Bayh-Dole or the standard patent rights clause. And worst of all, it’s a disaster for the research commons and the flow of new discoveries, tools, and data for public use on non-discriminatory terms.
Royalty income from the exploitation of patents on subject inventions may be used for anything in “higher education.”
Bayh-Dole places three requirements on royalties. Royalties must be shared with inventors. Royalties may be used only for reimbursing the expense of managing subject inventions. Any remaining money must be used for scientific research or education. The faux Bayh-Dolists don’t see it this way. They do share royalties with inventors, of course, but they don’t understand that this sharing was expected to take the form of a negotiation with inventors (according to Senator Bayh). In the system of invention management agents and university patent policies and university committee review for equitable interest, inventors negotiated with universities and invention management organizations based on circumstances and prior agreements–the public covenant was created in those negotiations, based on the commitments of those involved. The IPA system distorted these negotiations by forcing assignment to the university. Faux Bayh-Dole tried to do away with the negotiations. Outside the IPA, a share of royalties was consideration for the inventor’s license or assignment. Within the IPA, a share of royalties was a function of the university’s policies but capped at no more than 15% after $13,000. In Bayh-Dole, franken sausage law that it is, royalties are shared as a requirement of federal patent law transmitted by federal funding agreement. The sharing, whatever it is, is not consideration for anything. It can’t be if it is required by law. What’s required by law cannot also be consideration for a contract.
But as for the use of royalties beyond the share with inventors, universities don’t care. They do not separately account for the expenses incurred in managing subject inventions. Typically about 60 to 75% of university extramural research support comes from the federal government. More research is conducted using donated funds–which are not booked as extramural research–and conducted without using any funds at all. And university faculty may consult one day a week (typically) and that work, too, may result in inventions. Often, a university will by policy take a share of royalties–say 20%–and assign it to the budget of the technology licensing office. The standard patent rights clause forbids this practice, as it permits subject invention royalties to fund invention management for inventions not made with federal support. But faux Bayh-Dolists don’t care, as they have never (to my knowledge) been audited on this point. Bayh-Dole, you see is a non-law for middlemen–it only operates to strip inventors, the government, the research community, and the public of access to inventions made with federal support so that the middlemen have a shot at profiting from patent positions. What a killer.
Similarly, universities do not separately account for the balance of royalties after sharing with inventors and recovering costs of managing inventions. Bayh-Dole makes a specific requirement: use for scientific research or education. Not for “higher education” in general. But universities routinely do things such as investing income rather than spending it–so the use becomes, say, speculating with the royalties to make more money. The aim might be worthy (“more money to spend in the public interest”) but the use itself violates the standard patent rights clause because the use is investment, not expenditure. Even if all the proceeds of the investment were directed to scientific research or education, the effect of the investment step (and the money shaved off to recover the costs and pay the fees of such investments, including any losses) is to delay the expenditure pending playing with the money for other purposes. Other uses of royalties–to pay administrators, to build hospitals and stadiums, say–is not scientific research or education, though anything a university might do, to some hardened administrator, “ultimately” supports scientific research or education. If that had been Congress’s intent, there would be absolutely no need for the restriction on the use of royalties stated in Bayh-Dole. In faux Bayh-Dole, however, no one cares, so no one accounts for the use of profits on subject inventions.