Bayh-Dole makes a default public bargain for nonprofit contractors conducting research or development work that receives federal support. Nonprofit contractors may retain ownership of inventions made in such work that they acquire, but if they assign any such invention, the assignee must accept the nonprofit’s patent rights clause.
Work that through slowly.
When a nonprofit assigns a subject invention, the assignee becomes a party to the funding agreement–a federal contractor (35 USC 201(c)). Everywhere in Bayh-Dole and the standard patent rights clause has “contractor” now also includes “assignee.”
Any exclusive license to all substantial rights in an invention assigns the invention. Doesn’t matter that the document is called “exclusive license.” Doesn’t matter if it claims that title to patents is not transferred. Courts have consistently held that exclusive conveyance of all substantial rights in an invention (make, use, sell) conveys ownership–assigns the invention.
Bayh-Dole’s definition of funding agreement (35 USC 201(b)) expressly provides that contractors may add additional contractors to a funding agreement, including through “any assignment.” When a university grants an exclusive license to make, use, and sell, it assigns the invention. The assignee not only becomes a federal contractor but also the assignee must act as a nonprofit federal contractor.
Under Bayh-Dole, a nonprofit federal contractor must use any income earned with respect to a subject invention for scientific research or education (35 USC 202(c)(7)(C)). The contractor may deduct only expenses incidental to the management of subject inventions–including sharing royalties with inventors.
Every exclusive license granted by a university or affiliated nonprofit must be audited for Bayh-Dole compliance. First, establish whether the license operates as an assignment.
Does it grant exclusive rights?
Do those rights include make, use, and sell?
Does the license allow the licensee to enforce patent rights?
If the answer is yes to these questions, then on the face of it, there’s an assignment of the invention.
In Bayh-Dole, there’s no restriction on how a contractor assigns a subject invention unless the contractor is a nonprofit. Nonprofits must obtain federal agency approval for any assignment of a subject invention unless that assignment is to an organization that has as a primary function the management of inventions. See 35 USC 202(c)(7)(A). Even then, regardless of whether federal approval is required or not, the assignee must accept the nonprofit’s patent rights clause. Usually that will be the default nonprofit patent rights clause specified by Bayh-Dole, but since Bayh-Dole permits federal agencies to use different patent rights clauses, such as for exceptional circumstances, one always must confirm what patent rights clause actually operates in any given situation.
Thus, as a first audit question, one pulls all of a university’s licenses to subject inventions that involve exclusivity. One checks the subject invention for the controlling patent rights clause. If it is the default clause, then one checks for whether the license operates as an assignment. If it does, then one asks whether the assignee has as one of its primary functions the management of inventions. If not, then one has to check to see whether federal approval has been granted for the assignment.
There is no guidance in Bayh-Dole for what happens if a university ignores the federal approval requirement. Is the assignment void? May the federal government take action against the contractor? If so, in what way? It’s difficult to know. It’s almost as if the drafters of Bayh-Dole set it up so that there would be no consequence to the contractor for breach of the standard patent rights clause (except where there is an express remedy, such as for failure to timely disclose a subject invention).
Regardless of federal approval, Bayh-Dole requires all assignments by nonprofits of subject inventions to carry the nonprofit’s patent rights clause. If a university has made assignment of invention by means of an exclusive patent license, then it is not to be expected that the university would require what it characterizes as a licensee to accept the university’s nonprofit patent rights clause. An audit will not find an express statement of this requirement. But it is a requirement of federal law: “provided that such assignee shall be subject to the same provisions as the contractor” (35 USC 202(c)(7)(A)). The requirement operates as a condition of a federal contract whether the university includes that requirement expressly in its assignment documents or not. The assignee “is subject” to the nonprofit’s patent rights clause. It’s not that the assignee is subject to the nonprofit’s patent rights clause only if the nonprofit bothers to put such requirement into the assignment instrument but isn’t if the nonprofit blows off doing so.
The nonprofit patent rights clause follows ownership of the invention. It is a lien on title to the invention, a covenant that runs with ownership. Accept ownership, accept the covenant that runs with ownership.
The audit then must turn to the assignee. The assignee is necessarily by operation of federal law–Bayh-Dole–a federal contractor. Here, the audit must establish multiple compliance items.
- proper disposition of income
- compliance with disclosure of new subject inventions
- compliance with sharing royalties with inventors
- compliance with the assignment approval clause in exclusive sublicenses and in mergers or acquisitions of the company or some part thereof
Let’s deal with each, briefly. First, has there been any income earned with respect to the subject invention? Sales, sublicensing, equity investments, and the like. If so, then Bayh-Dole stipulates the accounting required. Total income earned. Total expenses incidental to the administration of subject inventions may be deducted. Payment of a share of royalties to inventors may be deducted. Second, has the allowable net income earned been used to support scientific research or education? If not, we have a potentially huge problem.
When a company accepts assignment of a subject invention and becomes a federal contractor, any inventions it makes and acquires in the development of that subject invention are also potentially subject inventions. One test is whether those inventions, were they made at the university, even without direct use of federal funds, would be subject inventions when acquired by the university. If so, then they are also subject inventions when acquired by any other federal contractor operating under the same funding agreement and therefore involved in the same “work.” If these subject inventions have not been disclosed timely to the federal agency, then the federal agency has under Bayh-Dole the right to take title to these inventions.
Furthermore, the assignee-contractor must provide the same accounting for income earned for each of these additional subject inventions.
The audit must consider whether the assignee-contractor has shared royalties with the inventors. Bayh-Dole’s default nonprofit patent rights clause requires such sharing (35 USC 202(c)(7)(b)). For the assignee, “royalties” does not include any income earned but rather only consideration for a license. Where the assignee has acquired ownership of a subject invention under the guise of an exclusive license, the matter of royalties will show up in sublicensing. Any income from sublicensing is a royalty, and the contractor-assignee owes inventors a share of this royalty. This requirement has nothing to do with whether the original nonprofit owner of the subject invention shares royalties it receives from the assignee-contractor with the inventors. Each contractor operating under the nonprofit’s patent rights clause has a legal obligation to share the royalties it receives with the inventors. Yes, the inventors get to double dip or triple dip. Dip as much as any chain of assignment activity requires.
If an assignee-contractor has received royalties and has not shared them with inventors, there’s another matter to resolve. Inventors, one would think, would be happy to consider an additional source of income. And what should it matter to a university assignor, if it gets the royalty it has bargained for in making the assignment of the subject invention?
Next, an audit must consider how the assignee-contractor has managed its own assignment of the subject invention. There are two primary means of such disposition. The first is by exclusive sublicense. An exclusive sublicense for all substantial rights is just another version of assignment made using an instrument labeled “exclusive sublicense.” The same Bayh-Dole requirements apply–the exclusive sublicensee becomes a federal contractor, and operates under the nonprofit patent rights clause with respect to income earned, royalties to be shared (if it subsublicenses), and inventions to be disclosed. The second is by merger or acquisition of the assignee-contractor. That merger or acquisition will involve assignment of the subject invention in addition to assignment of any assignment (or exclusive license) agreement. Same deal as before. Was there federal approval needed? Was it obtained? Even if not, the nonprofit patent rights clause continues to apply. Was the merger or acquisition based in some part on the subject invention? If so, then some portion of the consideration for the merger or acquisition is “income earned with respect to the subject invention” and that income must be managed to comply with Bayh-Dole’s nonprofit patent rights clause.
There’s your primary audit trail. Hundreds if not thousands of university exclusive licenses for subject inventions are implicated here. The vast majority of such exclusive licenses assign the subject invention by conveying all substantial rights and grant the assignee the right to enforce patents. These nonconforming assignments fail to make express that the nonprofit’s patent rights clause applies to the assignee, that the assignee has become a federal contractor, and that income earned with respect to the subject invention and any other inventions made responsive to the work that has been federally funded come within the scope of Bayh-Dole’s control.
While the remedy for a nonprofit’s failure to obtain federal agency approval for assignments to organizations lacking a primary function in managing inventions is not clear, the remedy for failure to use income earned with respect to a subject invention for scientific research or education is–establish how much money is involved, and see that this amount is indeed used to support scientific research or education. We are talking billions of dollars that should have gone to the specified public purposes that has been converted to private purposes.
Consider, for instance, just the Emory University deal under which it sold its future royalty stream for emtricitabine for $525M. Emtricitabine carries with it a federal funding statement in its patents. The nonprofit patent rights clause applies. When Emory licensed to Triangle Pharmaceuticals–eventually exclusively–Triangle became a federal contractor, operating under Bayh-Dole’s nonprofit patent rights clause. When Triangle was acquired by Gilead Sciences, Gilead became a federal contractor subject to that same nonprofit patent rights clause. Each of these assignees owes the inventors a share of any royalties (regardless of what they owed Emory) and each of these assignees is required to use any income earned for scientific research or education. An audit would determine how much.
But now what about Royalty Pharma, an investment company that buys out future royalty streams? In essence, Emory assigned its future financial interest royalties paid on an assignment of a subject invention to Royalty Pharma. That assignment comes within “any assignment” of the nonprofit patent rights clause. Royalty Pharma becomes a federal contractor under Emory’s patent rights clause. Any money Royalty Pharma makes–earned with respect to the subject invention–beyond what it has paid out to Emory must be used for scientific research or education. This makes clear policy sense. Otherwise, it would be easy for a university to set up an affiliated foundation that bought out each subject invention royalty stream for $1, and then the foundation would be out from under any obligation with regard to how income from the subject invention was to be used.
The workaround is similarly clear for a royalty buyout from a for-profit firm. A company does not buyout a future royalty expecting to make less than it pays. Perhaps in a given deal it pays more than the patented products eventually earn, but across a series of such investments, the company expects to make a profit–that is, to pay less for the income stream than the income stream pays out. But Bayh-Dole encumbers that entire income stream for nonprofits. They don’t have the right to sell it off for less than it is and thus avoid spending money on scientific research or education. Royalty Pharma does not violate Bayh-Dole by buying out Emory’s future royalty stream. It violates Bayh-Dole if it receives more than it paid. It owes the income earned to scientific research or education, deducting only its expenses in administrating subject inventions. No one in any chain of assignments–any assignment of any interest in a subject invention–pertaining to an invention subject to the nonprofit patent rights clause can get free of the Bayh-Dole nonprofit patent rights clause.
The way out is not to take assignment. Go for non-exclusive license of rights or a limited exclusive position (such as by means of a sole license), or purchase of product.
Such an audit of Bayh-Dole is essential to the operation of the law. An audit would provide the basis to restore billions of dollars earned on subject inventions that have been diverted to private benefit rather than funding scientific research or education.
If the “success” of Bayh-Dole depends on the conversion of public benefits to private benefits and the suppression of any means of correcting such conversion, then we have a very, very twisted idea of “success” and of “public benefit.” There’s no doubt that enforcing Bayh-Dole via audit would send a chill through the organizations that have violated the law. They would turn from championing the law to demanding that it be revised to conform to their practice–“if we cannot get away with violating the law, then it must be because of mere technical details that can be changed to make the appearances of violating the law go away.” But to argue this way is to create a new public purpose for the law–that regardless of the rationale for providing public funds to support a research or development effort for a public purpose (such as, say, public health), the public benefits only when patents are exploited for private gain. That public funding functions as a subsidy for private wealth seeking at the public’s expense–expense in terms of cost of products, expense in terms of delays in providing access to those or any other products, expense in suppression of diversity of uses and formulations, expense in lack of competition and multiple sources of manufacture, expense in suppression of research on and using what’s been invented.
The primary result of nonprofit holding patents made in publicly funded work is that all companies not obtaining access have an interest in not using the inventions that have been made in part with public funds. The patents, if dedicated to exclusive licenses for speculative private exploitation, work to prevent widespread utilization. If Bayh-Dole’s public policy is to give patent speculators an unsupervised go at exploiting inventions made in work done for a public purpose–if that unsupervised go at exploitation is the public purpose, and the work done is just an excuse to cover the subsidy–then let’s out and state Bayh-Dole that way and see how the public debate then goes. Otherwise, Bayh-Dole is clear–when a nonprofit acquires ownership of a subject invention, any income goes to scientific research or education, and that applies to any future owner of that subject invention and to any future owner of any royalty stream arising from that subject invention, whether the owner is a nonprofit or a for-profit.
What will happen, then, if Bayh-Dole becomes a real law and is enforced for the public side of its bargain? Certainly it will not be that all innovation having to do with federally funded research will cease. That’s a nonsense claim. Instead, nonprofits will patent less because they won’t be dealing primarily with patent speculators. When a nonprofit does acquire a patent, it will default license non-exclusively or with limited exclusive rights. It will never sell its future royalty interest and it will always retain exclusive control over the enforcement of patent rights. No licensee will have the right to grant an exclusive sublicense. At best, a licensee may accept an option to an exclusive license that covers just the instances of the invention for which it achieves practical application and even then that option is conditional on the licensee being the first to achieve that practical application and that in doing so it can show it made a significant financial investment at risk. And even then, the term of its exclusivity can be no longer than reasonably needed to recover its investment with a reasonable profit–and even then, the price under which its products based on the application of the subject invention must be reasonable. Such an option to a limited exclusive license does not convey ownership of the subject invention, since only limited rights to make, use, and/or sell have been committed, not all substantial rights.
In this approach–the actual approach established by Bayh-Dole–nonprofits license non-exclusively by default, allowing conditional exclusivity where a company does get to the point of practical application first. Such a deal rewards timely development, gives limited exclusivity to the company that moves the most quickly to public release on reasonable terms, and still protects the public from unreasonable pricing. Furthermore, such a deal does not encumber all rights in a given subject invention, does not shift control of patent enforcement to for-profits (including patent trolls and pyramid scheme speculative investors), and leaves broad rights available to all to practice even if commercialization efforts directed at obtaining a limited exclusive license fail.
Here. I’ll do the editing work. I’ll use a University of Texas template license, since that’s the first document that comes up in a DuckDuck search. Changes in red, and I’ve substituted LICENSOR and added TERM (not otherwise a defined term in the template, imagine that!–clearly an assignment like exclusive license, license exclusively and walk away, and see article 7: “LICENSEE, at its expense, must enforce PATENT RIGHTS against infringement. . . .” Totally an assignment of the invention):
BOARD hereby grants to LICENSEE a royalty-bearing, non-exclusive license under LICENSED SUBJECT MATTER to manufacture, have manufactured, use and/or SELL LICENSED PRODUCTS within the TERRITORY for use within FIELD.
If LICENSEE establishes in a written report to LICENSOR first use of LICENSED SUBJECT MATTER within the TERRITORY for use within FIELD such that the benefits of such use are or will be made available to the public on reasonable terms, then LICENSEE shall have for TERM an exclusive license in and to such LICENSED PRODUCT.
There. Straightforward. Drop any right to enforce patent rights (in the UT template, that’s article 7!) One can wrap all sorts of anal retentive prudent language around the form of reporting, what constitutes “first,” how use is “established,” what constitutes “reasonable terms,” and whether it’s enough to establish use along with proposed terms or whether one has to demonstrate an actual offer of sale to the public. With any exclusive license to use or sell one can add the US manufacturing preference. One can change the royalty rate if a company wants its limited exclusive position. And one can implement a faster, less fussy private march-in if public terms aren’t reasonable–such as providing notice the terms aren’t reasonable and then if there’s not a prompt, thirty day revision to the terms, terminating the limited exclusivity.
But that’s all easy stuff, too. Anyone can get it. Anyone can do it. The core of the deal is simply–
“You lucky dog, you get a broad non-exclusive license. Go develop something. If you are first to do so, then you can get a limited exclusive position for exactly what you have got done. Good for you. If you want a broader exclusive license, go develop something more that justifies it. If someone infringes your use or product, let us know and we will beat them into a pulp after our fashion. But it’s unlikely anyone will infringe your product because they will all be working on implementations of their own where they can get also get limited exclusive licenses and ask us to smack you around if you mess with them.”
That deal complies with Bayh-Dole. It promotes utilization. It promotes free competition. It promotes maximum participation by small companies. It easily deals with the US manufacturing preference. It avoids all the problems that come with assignment. The nonprofit maintains control over the patent as an asset held on behalf of the public. This would be the approach nonprofits would take to patent practice if Bayh-Dole ever became a real law.